1 INFO-VAX	Fri, 27 Apr 2001	Volume 2001 : Issue 235       Contents:, Re: 80% is Good Enough for M$. Now it's 100%, RE: 80% is Good Enough for M$. Now it's 100%, Re: 80% is Good Enough for M$. Now it's 100%, RE: 80% is Good Enough for M$. Now it's 100%3 Announcing New Oracle Rdb Technical Forums for 2001 0 Re: Anyone have a Modular Data Router installed?0 Re: Anyone have a Modular Data Router installed?3 Broadcast messages while waiting on INQUIRE command 7 Re: Broadcast messages while waiting on INQUIRE command  Changing UIC of running process  Compaq and SSH Re: Compaq and SSH Re: Compaq and SSH Re: Compaq and SSH/ Re: DCL To Compute IP Address From Hex Number ? / Re: DCL To Compute IP Address From Hex Number ? / Re: DCL To Compute IP Address From Hex Number ? ) Duplicate procedures in EVE source code ?  Re: Fall of the Roman Empire Re: Fall of the Roman EmpireG Re: Hackers crack challenge within 24 hours, and Solaris gets the blame G Re: Hackers crack challenge within 24 hours, and Solaris gets the blame G Re: Hackers crack challenge within 24 hours, and Solaris gets the blame G Re: Hackers crack challenge within 24 hours, and Solaris gets the blame G Re: Hackers crack challenge within 24 hours, and Solaris gets the blame G Re: Hackers crack challenge within 24 hours, and Solaris gets the blame G Re: Hackers crack challenge within 24 hours, and Solaris gets the blame G Re: Hackers crack challenge within 24 hours, and Solaris gets the blame G Re: Hackers crack challenge within 24 hours, and Solaris gets the blame G Re: Hackers crack challenge within 24 hours, and Solaris gets the blame G Re: Hackers crack challenge within 24 hours, and Solaris gets the blame G Re: Hackers crack challenge within 24 hours, and Solaris gets the blame G Re: Hackers crack challenge within 24 hours, and Solaris gets the blame  Re: IBM bought Informix  Re: IBM bought Informix  Installing OpenVMS - HELP!!  Re: Installing OpenVMS - HELP!!  RE: Installing OpenVMS - HELP!!  Re: Installing OpenVMS - HELP!!  Re: Installing OpenVMS - HELP!!  Re: Installing OpenVMS - HELP!! 7 Re: Mounting and Management station: comments please...  network card on win2k  Re: network card on win2k A newsgroup manners (was: 80% is Good Enough for M$. Now it's 100%) E Re: newsgroup manners (was: 80% is Good Enough for M$. Now it's 100%) / Re: OpenVMS article - please explain last line! / Re: OpenVMS article - please explain last line! / Re: OpenVMS article - please explain last line! / Re: OpenVMS article - please explain last line! / Re: OpenVMS article - please explain last line! / Re: OpenVMS article - please explain last line!  Re: OpenVMS Hobbyist version  Re: PCs Need to Query RMS files.  Re: PCs Need to Query RMS files.2 Re: R.I.P. FTP.WKU.EDU.  Long live FTP.PROCESS.COM, Re: Random access to 2 bytes in a large file, Re: Random access to 2 bytes in a large file, Re: Random access to 2 bytes in a large file, Re: Random access to 2 bytes in a large file= Re: Reply before or after? (was: seeing escapes is believing) 	 Re: SAMBA 	 RE: SAMBA 	 re: SAMBA 	 Re: SAMBA  Samba vs NFS Re: Samba vs NFS Re: Samba vs NFS Re: Samba vs NFS Set RADIX and OCTAL arithmetic. # Re: Set RADIX and OCTAL arithmetic. # Re: Set RADIX and OCTAL arithmetic.  Sync Comm - PCI Card Re: Sync Comm - PCI Card Re: Sync Comm - PCI Card Re: Sync Comm - PCI Card Re: Sync Comm - PCI Card Re: Sync Comm - PCI Card Re: Sync Comm - PCI Card Re: Sync Comm - PCI Card' Re: Tabular Timeline Tracking Tool (T4) ' Re: Tabular Timeline Tracking Tool (T4) ( Re: TCP/IP 5.1 Allow Host Table To WipedJ Trying to connect HSG80 console port to Bay Networks Annex terminal serverG Re: using a pipe command into a CP creates DCL$OUTPUT_xxxxxxx.LOG files , Re: Re: V7.3 in clusters w/earlier versions?, Re: Re: V7.3 in clusters w/earlier versions?, Re: Re: V7.3 in clusters w/earlier versions?, Re: Re: V7.3 in clusters w/earlier versions? Re: VMS and Fast Etherchannel? Re: VMS and Fast Etherchannel? Re: VMS and Fast Etherchannel? Re: VMS and Fast Etherchannel? Re: VMS and Fast Etherchannel?G Re: VMS gets positive mention in Compaq Q1 financial analystsconference G RE: VMS gets positive mention in Compaq Q1 financial analystsconference G Re: VMS gets positive mention in Compaq Q1 financial analystsconference G RE: VMS gets positive mention in Compaq Q1 financial analystsconference G Re: VMS gets positive mention in Compaq Q1 financial analystsconference G Re: VMS gets positive mention in Compaq Q1 financial analystsconference G RE: VMS gets positive mention in Compaq Q1 financial analystsconference % Re: VMS Loyalist dies from starvation % Re: VMS Loyalist dies from starvation  Re: VT POP3 client? ) Re: Warning - Charon-VAX Hobbyist Edition ) Re: Warning - Charon-VAX Hobbyist Edition - Warning: EV68/833 and Oracle Rdb incompatible ! Re: What is better?? VMS or Unix? 2 Re: www.compaq.com/openvms news from Sue Skonetski( [INFO] PCSI for Newbie version x0.0-0.00: Re: [INFO] SYSTEM-F-INSFSPTS, insufficient SPTEs available: Re: [INFO] SYSTEM-F-INSFSPTS, insufficient SPTEs available  F ----------------------------------------------------------------------   Date: 27 Apr 2001 09:27:44 CDT= From: wayne@tachysoft.xxx.065234.killspam.015d (Wayne Sewell) 5 Subject: Re: 80% is Good Enough for M$. Now it's 100% . Message-ID: <3GJZVTREJtgi@tachxxsoftxxconsult>  [ In article <3AE87870.EE6B82F7@fsi.net>, "David J. Dachtera" <djesys.nospam@fsi.net> writes:  > John Eisenschmidt wrote: >>  G >> Christ, I don't know which is sadder: The software or the people who  >> make it.  > [snip] >> For god sakes   > E > Just my opinion, mind you, but I do take umbrage to such use of the J > Lord's name(s) in vain. Other folks react badly to obscenities and other > profanities. Watcha gonna do?  >   N Personally, I don't think religion belongs in a technical forum in *any* form,M pro or con.  I find religious quotes in sigs, such as those found in posts by I that speakeasy guy, to be every bit as offensive as you find the above.          --  O =============================================================================== M Wayne Sewell, Tachyon Software Consulting  (281)812-0738  wayne@tachysoft.xxx : http://www.tachysoft.xxx/www/tachyon.html and wayne.html  K change .xxx to .com in addresses above, assuming you are not a spambot  :-) O =============================================================================== B Jed Clampett, checking into hotel: "This place got a cement pond?"+ Ellie May: "And do yuh let critters in it?"    ------------------------------  % Date: Fri, 27 Apr 2001 10:11:32 -0500 + From: Christopher Smith <csmith@amdocs.com> 5 Subject: RE: 80% is Good Enough for M$. Now it's 100% L Message-ID: <3B55D7F383B0D31197D9009027541CBF0D9D1DC4@cmiexch1.cmi.itds.com>   > -----Original Message-----0 > From: wayne@tachysoft.xxx.065234.killspam.015d  > > In article <3AE87870.EE6B82F7@fsi.net>, "David J. Dachtera" ! > <djesys.nospam@fsi.net> writes:   G > > Just my opinion, mind you, but I do take umbrage to such use of the 7 > > Lord's name(s) in vain. Other folks react badly to   > obscenities and other ! > > profanities. Watcha gonna do?   < > Personally, I don't think religion belongs in a technical  > forum in *any* form,> > pro or con.  I find religious quotes in sigs, such as those  > found in posts by ? > that speakeasy guy, to be every bit as offensive as you find   > the above.    I Personally, I think the quotes in somebody's signature are the last thing J you ought to complain about. :)  Signatures are generally pretty personal,G so there is no problem, as far as I'm concerned, with putting religious  references or whatever in them.    > --  @ > ============================================================== > ================= < > Wayne Sewell, Tachyon Software Consulting  (281)812-0738   > wayne@tachysoft.xxx < > http://www.tachysoft.xxx/www/tachyon.html and wayne.html  ? > change .xxx to .com in addresses above, assuming you are not   > a spambot  :-)@ > ============================================================== > ================= D > Jed Clampett, checking into hotel: "This place got a cement pond?"- > Ellie May: "And do yuh let critters in it?"   6 .. On the other hand, rednecks kind of offend me... ;)   Regards,   Chris   ! Christopher Smith, Perl Developer  Amdocs - Champaign, IL   /usr/bin/perl -e '? print((~"\x95\xc4\xe3"^"Just Another Perl Hacker.")."\x08!\n");  '    ------------------------------  % Date: Fri, 27 Apr 2001 19:27:48 +0100   From: Paul Sture <paul@sture.ch>5 Subject: Re: 80% is Good Enough for M$. Now it's 100% + Message-ID: <VA.00000377.214ff5f7@sture.ch>   F In article <87k847wmga.fsf_-_@prep.synonet.com>, Paul Repacholi wrote: > Newsgroups: comp.os.vms 7 > Subject: Re: 80% is Good Enough for M$. Now it's 100% . > From: Paul Repacholi <prep@prep.synonet.com>" > Date: 27 Apr 2001 02:22:13 +0800 > ) > Alan Greig <a.greig@virgin.net> writes:  > H > > And a big problem with NT Terminal server edition (multi user NT) isF > > that a single user can bring the entire machine to a halt with oneC > > command in Excel. We have users occasionally do something which G > > causes Excel to try and allocate a few gigabytes of memory. Machine G > > just hangs to the point where even the console doesn't respond. And H > > there's no quota system to prevent this. Problem not addressed under > > Windows 2000 either. > D > Atrittion.org has opined that Win2K is less secure for Web serving
 > than NT. >  > But is gets even better... > H > M$ have pulled all there security fixes, and they are not available at8 > the moment. Seems they have this slight virus problem. > 3 > See the Reg for details, if you have the stomach.  >  I did, and was nearly sick.   < Tie it in with a previous Reg article and it's even nastier!   19-Apr-2001   G http://www.theregister.co.uk/content/8/18370.html points out a massive  A security hole in LanMan. So off the MS site to read about it and   potentially download a fix.    25-Apr-2001   ? http://www.theregister.co.uk/content/8/18516.html reports that  F "Microsoft expects the FunLove infection period spanned approximately < two weeks, from Friday, April 6, to Friday, April 20, 2001."  E So that's at least 24 hours when folks alarmed at a serious security  ( hole could have been dowloading a virus.  F Does anyone know if the LanMan fix concerned contained the virus? The % timings strongly suggest that it did.    ___ 
 Paul Sture Switzerland    ------------------------------  % Date: Fri, 27 Apr 2001 10:47:54 -0700 ! From: Shane.F.Smith@Healthnet.com 5 Subject: RE: 80% is Good Enough for M$. Now it's 100% D Message-ID: <OF00D573B0.4D663818-ON88256A3B.00613705@foundation.com>  D Ah. Maybe I'd better qualify my "ditto". Most religious sigs are OK,J because most are considerate of people's feelings. But there are some thatJ really raise my hackles. Worst exaple I saw was "The fool hath said in hisE heart, there is no God." Or, to put it another way, "All atheists are ; fools." As an atheist, that makes me want to thump someone.   G Hey, choose to believe what you want, just let me do the same and don't 8 call me an idiot if I don't make the same choice as you.   Shane           ? Christopher Smith <csmith@amdocs.com> on 04/27/2001 08:11:32 AM   7 Please respond to Christopher Smith <csmith@amdocs.com>    To:   Info-VAX@Mvb.Saic.Com  cc:   6 Subject:  RE: 80% is Good Enough for M$. Now it's 100%     > -----Original Message-----0 > From: wayne@tachysoft.xxx.065234.killspam.015d  = > In article <3AE87870.EE6B82F7@fsi.net>, "David J. Dachtera" ! > <djesys.nospam@fsi.net> writes:   G > > Just my opinion, mind you, but I do take umbrage to such use of the 6 > > Lord's name(s) in vain. Other folks react badly to > obscenities and other ! > > profanities. Watcha gonna do?   ; > Personally, I don't think religion belongs in a technical  > forum in *any* form,= > pro or con.  I find religious quotes in sigs, such as those  > found in posts by > > that speakeasy guy, to be every bit as offensive as you find > the above.  I Personally, I think the quotes in somebody's signature are the last thing J you ought to complain about. :)  Signatures are generally pretty personal,G so there is no problem, as far as I'm concerned, with putting religious  references or whatever in them.    > --@ > ============================================================== > ================= : > Wayne Sewell, Tachyon Software Consulting  (281)812-0738 > wayne@tachysoft.xxx : > http://www.tachysoft.xxx/www/tachyon.html and wayne.html> > change .xxx to .com in addresses above, assuming you are not > a spambot  :-)@ > ============================================================== > ================= D > Jed Clampett, checking into hotel: "This place got a cement pond?"- > Ellie May: "And do yuh let critters in it?"   5 . On the other hand, rednecks kind of offend me... ;)    Regards,   Chris   ! Christopher Smith, Perl Developer  Amdocs - Champaign, IL   /usr/bin/perl -e '? print((~"\x95\xc4\xe3"^"Just Another Perl Hacker.")."\x08!\n");  '    ------------------------------  % Date: Fri, 27 Apr 2001 11:03:42 -0400 2 From: norm lastovica <norman.lastovica@oracle.com>< Subject: Announcing New Oracle Rdb Technical Forums for 2001* Message-ID: <3AE98A4E.E4F7F23A@oracle.com>  2 Rdb Product Management and Engineering are pleased, to offer a series of Rdb Technical Forums in- various locations around the world for 2001.  1 These Forums are offered at no charge and present . a series of sessions conducted by acknowledged2 technical specialists from Rdb Engineering and the0 Rdb community at large. The Rdb Technical Forums- should be considered a requirement for anyone 0 wishing to remain completely up-to-date with all, the latest enhancements and optimizations in Oracle Rdb.   / Forums will be held in Stockholm on June 14 and 1 15, Paris on June 18 and 19 and Berlin on June 21 - and 22.  Both the Paris and Berlin forums are 1 being held in conjunction with Oracle Open World. / Forums will also be held in Australia (date and / location yet to be finalized) and in the United  States (December).    /         Stockholm, Sweden       14-15 June 2001 /         Paris, France           18-19 June 2001 /         Berlin, Germany         21-22 June 2001 5         Australia               To Be Determined 2001 -         USA                     December 2001   , For additional information and registration 1 instructions, please visit the Oracle Rdb site on - the 'web at http://www.oracle.com/rdb/ and we & look forward to seeing you at a forum.   ------------------------------  # Date: Fri, 27 Apr 2001 10:38:22 GMT $ From: Scott Vieth <svieth@wi.rr.com>9 Subject: Re: Anyone have a Modular Data Router installed? ) Message-ID: <3AE94C78.1DCC9546@wi.rr.com>    John:   M Thanks for the idea.  I hadn't thought of doing it that way.  We normally use S our DHCP servers just to hand out IP addresses (and gateway info, DNS info, etc) to  PCs  on the network.   P I'll check to see if our DHCP servers can accept a "permanent" entry for the MAC address of the MDR.   
 -Scott :^)   John Santos wrote:  & > On Thu, 26 Apr 2001, JF Mezei wrote: >  > > Scott Vieth wrote:Q > > > I also can't believe that the device has an option to use DHCP.  How do you  > > > manageO > > > a network device who's TCP/IP address could change every time it reboots?  > > O > > You can program a fixed IP address for a device in the DHCP server. The big O > > advantage of such a setup is that your device will get the remainder of its H > > TCPIP config from the DHCP server (DNS, various TCPIP settings etc). > F > I think the general idea is that you can do most/all of your networkE > configuration management by modifying the DHCP config files on your G > DHCP server, rather than by individually configuring lots of devices.  > E > DHCP allows you to specify an explicit IP address for a given host, H > based on its MAC (ethernet) address.  I don't know if all DHCP serversF > support this (in fact, I'm pretty sure the DHCP server in my LinksysG > router/firewall doesn't), but non-toy DHCP's should.  I messed around G > some with the DHCP server in TCPWare several years ago and I remember  > it doing this. > E > This way, all your network configuration ends up on your VMS backup2C > tapes instead of being scattered around the building on a varietyi* > of unbacked-up and unbackupable devices. >'D > I think the original question in this thread was was there any wayE > to specify a default route for this device.  One of things DHCP cansE > supply is a default gateway, so configuring it to use DHCP might doe > the trick. >a > --
 > John Santos  > Evans Griffiths & Hart, Inc. > 781-861-0670 ext 539   ------------------------------  # Date: Fri, 27 Apr 2001 10:41:48 GMTu$ From: Scott Vieth <svieth@wi.rr.com>9 Subject: Re: Anyone have a Modular Data Router installed?C) Message-ID: <3AE94D46.B138DF5E@wi.rr.com>    Oliver:C  J > If you use a fixed IP on the MDR and configured on the DNS server then I- > don't believe you need a gateway (I think).d  G That's incorrect.  If the MDR is one a different subnet from the systemr> (whether that's a PC running SWCC or a StorageWorks Management? Appliance) that is monitoring the MDR, then you need to program 2 a default gateway on the MDR.  DNS won't help you.  M > I had a quick chat with some people and rumour has it that the new firmwarel. > might contain gateway support, I found this: >iB > http://www.compaq.com/support/files/server/us/download/9550.html  K It's interesting that the version is 1170A.  The version in my MDR is 1170.iF The gateway support might be in there but wasn't "important" enough toL warrant a mention in the release notes.  I'll load the new firmware into the2 MDR and see if something like setGateway shows up.   Thanks for the URL.n  
 -Scott :^)   ------------------------------  , Date: Fri, 27 Apr 2001 17:12:29 +0200 (CEST)> From: =?iso-8859-1?q?Sandro=20Polato?= <robertpiatto@yahoo.it>< Subject: Broadcast messages while waiting on INQUIRE command@ Message-ID: <20010427151229.87701.qmail@web14307.mail.yahoo.com>  	 Hi list !t/    Could you help me with this simple problem ?t5 I build a simple menu system based on a DCL procedure!5 : it displays a list of option and then wait the usere  choice with the INQUIRE command.2 It works well but when a broadcast message arrives4 (i.e. "you have a new message") the top lines scrool* up and the choice list appear disordered. ' In this case the ctrl-W does not work.  4 Do you have any suggestions to avoid this problem or+ do you know of a simple utility to use for a( displaying a list of options to users ?    Robert    F ______________________________________________________________________ Do You Yahoo!?F Il tuo indirizzo gratis e per sempre @yahoo.it su http://mail.yahoo.it   ------------------------------  # Date: Fri, 27 Apr 2001 15:50:52 GMTb2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)@ Subject: Re: Broadcast messages while waiting on INQUIRE command0 Message-ID: <wzgG6.300$5I.5757@news.cpqcorp.net>   In article <20010427151229.87701.qmail@web14307.mail.yahoo.com>, =?iso-8859-1?q?Sandro=20Polato?= <robertpiatto@yahoo.it> writes:i6 :I build a simple menu system based on a DCL procedure5 : it displays a list of option and then wait the userB! :choice with the INQUIRE command.a  G   Please do not use INQUIRE, it is a strange and powerful command, withtG   its full and mysterious capabilities often known only to DCL Wizards.mG   Due to the power of the INQUIRE command, the command is deliberately CJ   disabled within a captive environment.  In most cases, the READ command F   is far more appropriate, and should be substituted for INQUIRE.  AndD   the READ command will function within a captive command procedure.  3 :It works well but when a broadcast message arrives 5 :(i.e. "you have a new message") the top lines scroola+ :up and the choice list appear disordered. c  E   This is normal.  Most commonly, you will simply want to disable thevF   reception of broadcasts.  You may/will also want to provide a means D   of refreshing the display within the menu system itself, often via    a "refresh the screen" option.  E   DCL (and/or images) that provide a CGI interface avoids most of theMD   problems you are seeing -- and you can use a web browser from any @   platform to access a CGI-based DCL (and/or image) menu system.  ( :In this case the ctrl-W does not work.   F   CTRL/R and CTRL/W are used within text editors and within utilities I   that use SMG and similar tools for a screen refresh.  INQUIRE and your iG   DCL environment do not have a handler for CTRL/R or CTRL/W, and thereiD   is no particularly easy way to establish one.  (A /REFRESH handlerE   might be an interesting extension to the current READ command, but fE   this support is not presently available and not presently planned.)e  5 :Do you have any suggestions to avoid this problem or , :do you know of a simple utility to use for ) :displaying a list of options to users ?    I   Menufinder is referenced in the FAQ.  I expect there are other menuing nG   packages around...  And, of course, the CGI stuff referenced earlier.     N  ---------------------------- #include <rtfaq.h> -----------------------------N       For additional, please see the OpenVMS FAQ -- www.openvms.compaq.com    N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------  % Date: Fri, 27 Apr 2001 11:51:14 -0600g4 From: "Michael D. Ober" <mdo.@.wakeassoc.com.nospam>( Subject: Changing UIC of running process2 Message-ID: <okiG6.114$JP1.198606@news.uswest.net>  8 Is there anyway for a running process to change its UIC?   -- Thanks,h
 Mike Ober.   ------------------------------  % Date: Fri, 27 Apr 2001 12:09:38 -0400d- From: "Mitchell, David R." <mitchell@wpi.edu>m Subject: Compaq and SSH H Message-ID: <FE1835D68492D311BF7900508B5BEB0D2CA3C2@petra.admin.wpi.edu>  E It looks like Compaq has licensed secure shell for integration with =n
 Tru64. =20  4 http://www.ssh.com/about/press/2001/2001-04-10B.html   <snip>F "Under the terms of the agreement, Compaq will integrate SSHs Secure = Shello4 with its enterprise-leading server operating system" <snip>  I So, they obviously recognize the importance of including ssh support in =t their8B premier operating system(s).  I guess the announcement about ssh = integrationaH with OpenVMS must have been delayed for some reason.  As a loyal OpenVMS. customer, when can I expect this announcement?   David=   ------------------------------  % Date: Fri, 27 Apr 2001 10:18:32 -0600"% From: Dan O'Reilly <dano@process.com>a Subject: Re: Compaq and SSHnA Message-ID: <5.0.2.1.2.20010427101805.00af1158@ntbsod.psccos.com>1  0 At 10:09 AM 4/27/2001, Mitchell, David R. wrote:K >It looks like Compaq has licensed secure shell for integration with Tru64.  >55 >http://www.ssh.com/about/press/2001/2001-04-10B.html. >> ><snip>-K >"Under the terms of the agreement, Compaq will integrate SSHs Secure Shell 5 >with its enterprise-leading server operating system"e ><snip>d >yN >So, they obviously recognize the importance of including ssh support in theirM >premier operating system(s).  I guess the announcement about ssh integration I >with OpenVMS must have been delayed for some reason.  As a loyal OpenVMSl/ >customer, when can I expect this announcement?   > If you are a Multinet or TCPware customer, you have SSH today.     ------I +-------------------------------+---------------------------------------+oI | Dan O'Reilly                  |                                       | I | Principal Engineer            |  "Why should I care about posterity?  |SI | Process Software              |   What's posterity ever done for me?" | I | http://www.process.com        |                    -- Groucho Marx    | I +-------------------------------+---------------------------------------+    ------------------------------  # Date: Fri, 27 Apr 2001 16:21:32 GMT-/ From: "Richard L. Dyson" <rick-dyson@uiowa.edu>y Subject: Re: Compaq and SSH ) Message-ID: <3AE9563C.613DE6A6@uiowa.edu>e   Mitchell, David R. wrote:n > L > It looks like Compaq has licensed secure shell for integration with Tru64. > 6 > http://www.ssh.com/about/press/2001/2001-04-10B.html >  > <snip>L > "Under the terms of the agreement, Compaq will integrate SSHs Secure Shell6 > with its enterprise-leading server operating system" > <snip> > O > So, they obviously recognize the importance of including ssh support in theirrN > premier operating system(s).  I guess the announcement about ssh integrationJ > with OpenVMS must have been delayed for some reason.  As a loyal OpenVMS0 > customer, when can I expect this announcement?  ? 	Durning a recent teleconference sponsored by Compaq/OpenVMS (afE playback is available from the OpenVMS web page) I asked specifically  aboutlI SSH at the server and client level.  IMO, their answer was that they did  C not think it was important to customers, but would be interested in G hearing from customers if they were interested in getting SSH availabler) with the network products (DECnet/TCPIP).e   	Let them know. :)   Rick -- TH Richard L. Dyson                                    rick-dyson@uiowa.eduH  _   _      _____                http://www-pi.physics.uiowa.edu/~dyson/H | | | |    |_   _|   Systems Analyst                     O: 319/335-1879H | | | | of   | |     The University of Iowa            FAX: 319/335-17536 | \_/ |     _| |_    Department of Physics & Astronomy-  \___/     |_____|   Iowa City, IA 52242-1479    ------------------------------  % Date: Fri, 27 Apr 2001 10:46:25 -0600n% From: Dan O'Reilly <dano@process.com>  Subject: Re: Compaq and SSHeA Message-ID: <5.0.2.1.2.20010427104539.00a92ae8@ntbsod.psccos.com>   . At 10:21 AM 4/27/2001, Richard L. Dyson wrote: >Mitchell, David R. wrote: > >eN > > It looks like Compaq has licensed secure shell for integration with Tru64. > > 8 > > http://www.ssh.com/about/press/2001/2001-04-10B.html > >r
 > > <snip>N > > "Under the terms of the agreement, Compaq will integrate SSHs Secure Shell8 > > with its enterprise-leading server operating system"
 > > <snip> > > L > > So, they obviously recognize the importance of including ssh support in  > theireE > > premier operating system(s).  I guess the announcement about ssh y
 > integrationrL > > with OpenVMS must have been delayed for some reason.  As a loyal OpenVMS2 > > customer, when can I expect this announcement? >iH >         Durning a recent teleconference sponsored by Compaq/OpenVMS (aF >playback is available from the OpenVMS web page) I asked specifically >aboutI >SSH at the server and client level.  IMO, their answer was that they did D >not think it was important to customers, but would be interested inH >hearing from customers if they were interested in getting SSH available* >with the network products (DECnet/TCPIP). >y >         Let them know. :)i  I ..and if you REALLY want it now, contact us at Process Software!  It's ins both of our products.m   ------I +-------------------------------+---------------------------------------+"I | Dan O'Reilly                  |                                       | I | Principal Engineer            |  "Why should I care about posterity?  |uI | Process Software              |   What's posterity ever done for me?" | I | http://www.process.com        |                    -- Groucho Marx    |nI +-------------------------------+---------------------------------------+o   ------------------------------  # Date: Fri, 27 Apr 2001 11:00:20 GMTo= From: system@SendSpamHere.ORG (Brian Schenkenberger, VAXman-)-8 Subject: Re: DCL To Compute IP Address From Hex Number ?0 Message-ID: <009FB270.2FECD495@SendSpamHere.ORG>  Y In article <1010426212951.60771B-100000@Ives.egh.com>, John Santos <JOHN@egh.com> writes:u* >On Thu, 26 Apr 2001, Tim Llewellyn wrote: >i >> n >> e$ >> norm.raphael@jamesbury.com wrote: >> tO >> > This seems to work better.  (The octets appear to need to be reversed fromb >> > below,e% >> > and the value needs conversion.)  >> >P >> > I note with admiration how much cleaner this is than the other submissions. >> >" >> > $ ip = f$trnlnm("sys$rem_id") >> > $ a[0,32]= %x'IP'	 >> > $ a=n\ >> > "''F$fao("!UB.!UB.!UB.!UB",F$cvui(24,8,a),F$cvui(16,8,a),F$cvui(8,8,a),F$cvui(0,8,a))'" >> > >> g\ >>  sure, $FAO looks the neatest way, but why not upgrade to TCPIP Services V5.0A or higher, >> which >> does the hard work for you? > F >In my previous followup, I forgot to note that the F$FAO technique isG >not only the neatest way (basically one or two lines, depending if younF >already have it in a numeric symbol) but also WORKS!  I.E. it doesn'tC >have problems with the sign bit if your IP address is greater thane >128.255.255.255, as mine is.m   :)   --O VAXman- OpenVMS APE certification number: AAA-0001     VAXman(at)TMESIS(dot)COM=            mO city, n., 1. a place where trees are cut down and streets are named after them.y   ------------------------------  + Date: Fri, 27 Apr 2001 10:55:19 +0000 (UTC) ' From: david20@alpha2.mdx.ac.uk (D.Webb)t8 Subject: Re: DCL To Compute IP Address From Hex Number ?+ Message-ID: <9cbj6n$2rk$1@aquila.mdx.ac.uk>P  ` In article <93ZF6.1802$%L5.31949@insync>, LESLIE@209-16-45-102.insync.net (Jerry Leslie) writes:/ >Tim Llewellyn (tim.llewellyn@bbc.co.uk) wrote:a >: o >: uK >: sure, $FAO looks the neatest way, but why not upgrade to TCPIP Services 85 >: V5.0A or higher, which does the hard work for you?E >: e >: (LNM$JOB_811EF380)f >: s8 >:   "SYS$REM_NODE" = ""10.72.8.60       Port: 0    "::" >: w >:   $ ucx sh verh >: o< >:   DIGITAL TCP/IP Services for OpenVMS Alpha Version V5.0A6 >:   on a AlphaServer 1000A 5/300 running OpenVMS V7.1 >: E >nC >That works as long as the address isn't in the hosts database, or u  >isn't resolved by a DNS server: >y >   $ tcpip show version >e< >      Compaq TCP/IP Services for OpenVMS Alpha Version V5.1@ >      on a Digital Personal WorkStation  running OpenVMS V7.2-1 >  >   $ show logical sys$rem_node 4 >     "SYS$REM_NODE" = "OLYSC3::" (LNM$JOB_80C6ED00) >n >   $ show logical sys$rem_idj9 >     "SYS$REM_ID" = "TELNET_95B71B35" (LNM$JOB_80C6ED00)i >u >   $ tcpip show host olysc3 >t >        LOCAL databasee >i >   Host address    Host name  >  >   149.183.27.53   OLYSC3 > 2 >For this application, the IP address is required. >2  - UCX> set service TELNET/LOG_OPTIONS=NOADDRESS2  
 Alpha2:ucx TCPIP> help set service/log    SETn  	   SERVICE        /LOG_OPTIONS             /LOG_OPTIONS=o  d .i .l .f  /        The options have the following meanings:s          Option          Functionb  D        [NO]ACCEPT      Message is logged when a request is accepted.G        [NO]ACTIVATE    Message is logged when the service is activated.2@        [NO]ADDRESS     For auxiliary server messages and OpenVMSC                        security events, the message displays the IP6D                        address as a host name. If host names are notC                        relevant, Compaq recommends that you specify #                        [NO]ADDRESS.n   .t .t .     + This also affects the value of sys$rem_noder   eg     Alpha2:sh log sys$rem*   (LNM$PROCESS_TABLE)    (LNM$JOB_815A2F00)  "   "SYS$REM_ID" = "TELNET_9E5E0205"!   "SYS$REM_NODE" = "158.94.2.5::"t*   "SYS$REM_NODE_FULLNAME" = "158.94.2.5::"     ucx sh host 158.94.2.5        BIND database  ! Server:   158.94.2.9       alpha2    Host address    Host nameP  ! 158.94.2.5      hub.axp.mdx.ac.ukt      
 David Webb VMS and Unix team leader CCSS Middlesex University   ------------------------------  % Date: Fri, 27 Apr 2001 10:11:36 -0400c  From: norm.raphael@jamesbury.com8 Subject: Re: DCL To Compute IP Address From Hex Number ?4 Message-ID: <C2256A3B.004D1B8B.00@jklh21.valmet.com>  E Well, we're technically off-topic here.  The original post called foryA a method to change the HEX IP address into dotted-decimal format.nB VAXMAN did that admirably.  The rest is implementation-specific as@ to how to get that HEX IP address - not trivial as it turns out.  G "If we didn't have standards, we'd have nothing to extend." - Anonymousn          2 david20@alpha2.mdx.ac.uk on 04/27/2001 06:55:19 AM  * Please respond to david20@alpha2.mdx.ac.uk   To:   Info-VAX@mvb.saic.com  cc:H9 Subject:  Re: DCL To Compute IP Address From Hex Number ?t        P In article <93ZF6.1802$%L5.31949@insync>, LESLIE@209-16-45-102.insync.net (Jerry Leslie) writes:f/ >Tim Llewellyn (tim.llewellyn@bbc.co.uk) wrote:i >: >:J >: sure, $FAO looks the neatest way, but why not upgrade to TCPIP Services5 >: V5.0A or higher, which does the hard work for you?u >: >: (LNM$JOB_811EF380)O >:8 >:   "SYS$REM_NODE" = ""10.72.8.60       Port: 0    "::" >: >:   $ ucx sh verm >:< >:   DIGITAL TCP/IP Services for OpenVMS Alpha Version V5.0A6 >:   on a AlphaServer 1000A 5/300 running OpenVMS V7.1 >: >-B >That works as long as the address isn't in the hosts database, or  >isn't resolved by a DNS server: >n >   $ tcpip show version >i< >      Compaq TCP/IP Services for OpenVMS Alpha Version V5.1@ >      on a Digital Personal WorkStation  running OpenVMS V7.2-1 >s >   $ show logical sys$rem_node 4 >     "SYS$REM_NODE" = "OLYSC3::" (LNM$JOB_80C6ED00) >D >   $ show logical sys$rem_idm9 >     "SYS$REM_ID" = "TELNET_95B71B35" (LNM$JOB_80C6ED00)e >k >   $ tcpip show host olysc3 >m >        LOCAL databaseo >e >   Host address    Host namea >n >   149.183.27.53   OLYSC3 > 2 >For this application, the IP address is required. >i  - UCX> set service TELNET/LOG_OPTIONS=NOADDRESS   
 Alpha2:ucx TCPIP> help set service/logm   SET/  	   SERVICE/       /LOG_OPTIONS             /LOG_OPTIONS=g   .  .  .s  /        The options have the following meanings:           Option          Functionu  D        [NO]ACCEPT      Message is logged when a request is accepted.G        [NO]ACTIVATE    Message is logged when the service is activated.t@        [NO]ADDRESS     For auxiliary server messages and OpenVMSC                        security events, the message displays the IP D                        address as a host name. If host names are notC                        relevant, Compaq recommends that you specify2#                        [NO]ADDRESS.i   .o .o .     + This also affects the value of sys$rem_nodel   eg     Alpha2:sh log sys$rem*   (LNM$PROCESS_TABLE)    (LNM$JOB_815A2F00)  "   "SYS$REM_ID" = "TELNET_9E5E0205"!   "SYS$REM_NODE" = "158.94.2.5::"m*   "SYS$REM_NODE_FULLNAME" = "158.94.2.5::"     ucx sh host 158.94.2.5        BIND database  ! Server:   158.94.2.9       alpha2i   Host address    Host namei  ! 158.94.2.5      hub.axp.mdx.ac.uko      
 David Webb VMS and Unix team leader CCSS Middlesex University   ------------------------------  # Date: Fri, 27 Apr 2001 12:39:08 GMTeB From: Simon Clubley <simon_clubley@remove_me.excite.com-Earth.UFP>2 Subject: Duplicate procedures in EVE source code ?4 Message-ID: <MLdG6.607$SZ5.49016@www.newsranger.com>  F I've been looking through the EVE source code and I've noticed that inJ some places the same procedure is defined more than once. In case it makes$ any difference, this is on an Alpha.  = For example, in both EVE$CORE and EVE$WILDCARD, the procedure1F eve$$build_pattern is defined. The EVE source code builds ok, you justI get an informational message that eve$$build_pattern has been superseded.aA If you remove the procedure from EVE$CORE, EVE still compiles ok.t  I Is there something going on that I don't understand or has (as I suspect),2 somebody just not removed the code from EVE$CORE ?  I BTW, looking at the Ultrix code in EVE, does the current VMS (ie 7.2) EVEtF still work in a Ultrix environment, or is that code long out of date ?   Thanks for any replies,m   Simon.     -- ,; Simon Clubley, simon_clubley@remove_me.excite.com-Earth.UFPsJ 'The statement that "it can never happen" is not an acceptable programmingH approach. You must assume it can happen and be in control when it does.'=           -- Ada 95 Quality and Style Guidelines, US DoD AJPOu   ------------------------------  # Date: Fri, 27 Apr 2001 11:49:16 GMTm1 From: CSABA  HARANGOZO   <csabah@zipworld.com.au>t% Subject: Re: Fall of the Roman Empirea9 Message-ID: <01dG6.1527$CN.289121@nostril.pacific.net.au>t  * fabio_compaq@ep-bc.petrobras.com.br wrote:K > UNIX ? It is not a roman name ..... It is Gaelic .. like AsteriX, ObeliX,o > IdotiX , eXpensiviX ! :-)))uG > It should be Unicus if roman ... (it remembers me a hungarian drink -e > Unicum - horrible)  D 	Careful Fabio ! Unicum is a stomach bitter, not just some alcoholic? 	drink you quaff down... Its ingredients are carefully guarded.tD 	It is more like medicine, actually. Sold in nice, dark glass bottle< 	shaped like a ball. On its label it even has a red cross...   	Ah, nostalgia...i 					Cheers,       Csaba  I    ----------------------------------------------------------------------sE    * Csaba I. Harangozo     |    'To err is human', said the hedgehog E    * csabah@zipworld.com.au |           as he dismounted a wirebrush.tI    ----------------------------------------------------------------------s;    EARTH::AUSTRALIA:[SYDNEY]HARANGOZO.CSABA;1, delete? [N]:e   ------------------------------  % Date: Fri, 27 Apr 2001 13:01:44 +0100.  From: steven.reece@quintiles.com% Subject: Re: Fall of the Roman Empire-H Message-ID: <OFDFE6319B.DA17085D-ON80256A3B.0041E3B1@qedi.quintiles.com>  I But I thought most alcohol was purely "for medicinal purposes"...   :-)))-D I mean, it's definitely good as pain relief if taken in large enough/ quantities (sometimes terminal pain relief....)  Steve.  : Csaba Harangozo (csabah at zipworld dot com dot au) wrote: >>>,H      Careful Fabio ! Unicum is a stomach bitter, not just some alcoholicC      drink you quaff down... Its ingredients are carefully guarded.6H      It is more like medicine, actually. Sold in nice, dark glass bottle@      shaped like a ball. On its label it even has a red cross... <<<-   ------------------------------  % Date: Fri, 27 Apr 2001 12:00:00 +0100R0 From: andrew harrison <andrew.nospam@uk.sun.com>P Subject: Re: Hackers crack challenge within 24 hours, and Solaris gets the blame* Message-ID: <3AE95130.56165A1A@uk.sun.com>   David Mathog wrote:r > _ > In article <3AE83385.5583F8C2@uk.sun.com>, andrew harrison <andrew.nospam@uk.sun.com> writes:g > >n$ > >OpenVMS is also not secure unless, > >patched. Timely patching and notification. > >of patches is just as important for OpenVMS > >as it is for any other OS.u > >s > J > A classic piece of FUD - containing a tiny nugget of truth on the top of > huge steaming pile of BS.i >   , I won't asnwer the long rant except to point) out that the hack was a DOS attack not a p) security violation. Since you seem happy -& to make the distinction between a DOS * attack and a security violation in OpenVMS, it is only fair to do the same with Solaris.  1 And if this is the case then OpenVMS if connectedn- to the net using its bundled IP stack has to t- be patched to ensure that it isn't vunerable a to a number of DOS attacks.    regardsiJ > Yes there are occasionally security holes in VMS that must be addressed,I > but there is one every bloody WEEK in Solaris, and the vast majority ofcK > them allow root access to any pimply faced script kiddy with a few rubles K > and a telephone.  I'm migrating to Solaris in spite of its security holestL > because I need some software that runs there and not on VMS (and for otherL > political/economic reasons which have long since been beaten to death). SoK > I spent most of yesterday shutting Solaris security holes and am far fromeK > sure that I've closed them all.  I was particularly displeased to see all-J > of the SNMP pieces which were running by default, half of which had CERTI > warnings on them and no solutions, and which I didn't need in the first H > place.  And when I went to restrict access to rpcbind was horrified toM > realize that Sun still (as of 1/01) ships a version that allows connectionsmI > from *anywhere*.  The fellow who wrote TCP Wrappers had a variant which-D > would reject connections by IP number, so I installed that.  But IH > shouldn't have had to, the version in Solaris should have been able to > apply this filter already. > J > Whatever other failings VMS and Compaq may suffer, inattention to system > security is not one of them. > - > >If you havn't bothered following the patchi, > >paper chase for OpenVMS then your systems, > >are as vunerable as the Solaris for Intel > >box was.r > J > Ridiculous.  I do not recall the last "attack from the other side of theK > planet over the internet with a port scanner and gain SYSTEM control on asJ > VMS system" security hole.  Conversely, there have been so many securityL > holes of that type on Solaris that it would take an hour to list them all. >  > > ( > >Not a good effort on your part Larry. > >O >  > Nor yours Andrew.c >  > David Mathog > mathog@caltech.edu@ > Manager, sequence analysis facility, biology division, Caltech   -- i Andrew Harrisone Enterprise IT Architecte   ------------------------------    Date: 27 Apr 2001 13:31:45 +0200G From: Jan Vorbrueggen <jan@mailhost.neuroinformatik.ruhr-uni-bochum.de>uP Subject: Re: Hackers crack challenge within 24 hours, and Solaris gets the blameH Message-ID: <y41yqeh93y.fsf@mailhost.neuroinformatik.ruhr-uni-bochum.de>  2 andrew harrison <andrew.nospam@uk.sun.com> writes:  F > > Since there's apparently a fix, would you care to tell us what the > > vulnerability was? > Its a DOS attack.   F A DOS attack gives you access to the machine? Explain how that can be!   	Jan   ------------------------------  % Date: Fri, 27 Apr 2001 12:49:24 +0200|= From: Arne =?iso-8859-1?Q?Vajh=F8j?= <arne.vajhoej@gtech.com>wP Subject: Re: Hackers crack challenge within 24 hours, and Solaris gets the blame) Message-ID: <3AE94EB4.414D46A9@gtech.com>p   andrew harrison wrote:# > OpenVMS is also not secure unless1+ > patched. Timely patching and notification - > of patches is just as important for OpenVMSt > as it is for any other OS. > , > If you havn't bothered following the patch+ > paper chase for OpenVMS then your systems + > are as vunerable as the Solaris for Intel-
 > box was.   ????  - Would you care to post the following 3 lists:m,   - security related patches for Solaris 2.60   - security related patches for Solaris 2.7 / 70   - security related patches for Solaris 2.8 / 8   And then we create the 3 lists:p'   -security related patches for VMS 6.2 '   -security related patches for VMS 7.1t'   -security related patches for VMS 7.2m  1 Then we have some solid information for comparinge vulnerability.  0 If you do not post those lists, then we wil just3 have to conclude that you share our conviction thata VMS is much safer than Solaris.b   Arne   ------------------------------  % Date: Fri, 27 Apr 2001 12:56:18 +0100c0 From: andrew harrison <andrew.nospam@uk.sun.com>P Subject: Re: Hackers crack challenge within 24 hours, and Solaris gets the blame* Message-ID: <3AE95E62.D841F2CE@uk.sun.com>   Jan Vorbrueggen wrote: > 4 > andrew harrison <andrew.nospam@uk.sun.com> writes: > H > > > Since there's apparently a fix, would you care to tell us what the > > > vulnerability was? > > Its a DOS attack.- > H > A DOS attack gives you access to the machine? Explain how that can be! > 
 >         Jan   - It didn't as the report says it allowed them   to bring the Solaris box down.   Regardsh -- I Andrew Harrisonr Enterprise IT Architecta   ------------------------------  + Date: Thu, 26 Apr 2001 23:23:39 +0000 (UTC)c' From: david20@alpha2.mdx.ac.uk (D.Webb) P Subject: Re: Hackers crack challenge within 24 hours, and Solaris gets the blame+ Message-ID: <9caalr$nlh$1@aquila.mdx.ac.uk>e  ] In article <3AE8654E.9D8D5726@uk.sun.com>, andrew harrison <andrew.nospam@uk.sun.com> writes: # >Shane.F.Smith@Healthnet.com wrote:s >>  E >> Since there's apparently a fix, would you care to tell us what thet >> vulnerability was?7 >> 0 >@ >Its a DOS attack. >t  9 Not according to the folks on the Argus stand at Infosec.yI They say it's a major security hole (details of which they cannot releaseTJ until Sun has had a chance to provide a patch). It does though only affect Solaris on Intel.aJ Now it's possible the people on the stand don't know what they are talkingH about. However it is also true that a Denial of Service attack would notN have won the prize !! In order to win the prize you had to have enough control& of the system to deface some webpages.    
 David Webb VMS and Unix team leader CCSS Middlesex University   ------------------------------  % Date: Fri, 27 Apr 2001 08:42:35 -0400e5 From: David Beatty <David.Beatty@qwertysasasdfgh.com> P Subject: Re: Hackers crack challenge within 24 hours, and Solaris gets the blame2 Message-ID: <OmbpOhEA2gGye0QaPGLJOYFj3mab@4ax.com>  3 On Fri, 27 Apr 2001 12:00:00 +0100, andrew harrisont! <andrew.nospam@uk.sun.com> wrote:i   >David Mathog wrote: >> t` >> In article <3AE83385.5583F8C2@uk.sun.com>, andrew harrison <andrew.nospam@uk.sun.com> writes: >> >% >> >OpenVMS is also not secure unlessm- >> >patched. Timely patching and notification / >> >of patches is just as important for OpenVMS- >> >as it is for any other OS. >> > >> 1K >> A classic piece of FUD - containing a tiny nugget of truth on the top of: >> huge steaming pile of BS. >>   >b- >I won't asnwer the long rant except to pointk* >out that the hack was a DOS attack not a * >security violation. Since you seem happy ' >to make the distinction between a DOS l+ >attack and a security violation in OpenVMS$- >it is only fair to do the same with Solaris.t >C2 >And if this is the case then OpenVMS if connected. >to the net using its bundled IP stack has to . >be patched to ensure that it isn't vunerable  >to a number of DOS attacks. u >s  ; Andrew, do you realize the DOS Ping-of-Death attack appliesr; to UCX V4.0 and V4.1 only, which are no longer supported by < Compaq (with the exception noted below)?  Supported releases/ are V5.1, V5.0 with ECO 2, and V4.2 with ECO 4.s  ; The only exception is for VAX/VMS V5.5-2, which falls underS; Prior Version Support (it's eight years old) and should rune; UCX 4.0 ECO 5.  It can run V3.3, which was never vulnerablegC to the Ping-of-Death DOS attach, it's just not currently supported.0   David R. Beattyl   >regardsK >> Yes there are occasionally security holes in VMS that must be addressed, J >> but there is one every bloody WEEK in Solaris, and the vast majority ofL >> them allow root access to any pimply faced script kiddy with a few rublesL >> and a telephone.  I'm migrating to Solaris in spite of its security holesM >> because I need some software that runs there and not on VMS (and for otherTM >> political/economic reasons which have long since been beaten to death). So L >> I spent most of yesterday shutting Solaris security holes and am far fromL >> sure that I've closed them all.  I was particularly displeased to see allK >> of the SNMP pieces which were running by default, half of which had CERTxJ >> warnings on them and no solutions, and which I didn't need in the firstI >> place.  And when I went to restrict access to rpcbind was horrified to.N >> realize that Sun still (as of 1/01) ships a version that allows connectionsJ >> from *anywhere*.  The fellow who wrote TCP Wrappers had a variant whichE >> would reject connections by IP number, so I installed that.  But I*I >> shouldn't have had to, the version in Solaris should have been able to9 >> apply this filter already.5 >> 2K >> Whatever other failings VMS and Compaq may suffer, inattention to system5 >> security is not one of them.  >> .. >> >If you havn't bothered following the patch- >> >paper chase for OpenVMS then your systems - >> >are as vunerable as the Solaris for Inteld >> >box was. >>  K >> Ridiculous.  I do not recall the last "attack from the other side of themL >> planet over the internet with a port scanner and gain SYSTEM control on aK >> VMS system" security hole.  Conversely, there have been so many security M >> holes of that type on Solaris that it would take an hour to list them all.t >> X >> >) >> >Not a good effort on your part Larry.d >> > >> i >> Nor yours Andrew. >> n >> David Mathogs >> mathog@caltech.eduXA >> Manager, sequence analysis facility, biology division, Caltecha   ------------------------------  % Date: Fri, 27 Apr 2001 14:59:05 +010020 From: andrew harrison <andrew.nospam@uk.sun.com>P Subject: Re: Hackers crack challenge within 24 hours, and Solaris gets the blame* Message-ID: <3AE97B29.E48526BA@uk.sun.com>   "D.Webb" wrote:t > _ > In article <3AE8654E.9D8D5726@uk.sun.com>, andrew harrison <andrew.nospam@uk.sun.com> writes:e% > >Shane.F.Smith@Healthnet.com wrote:i > >>G > >> Since there's apparently a fix, would you care to tell us what thev > >> vulnerability was?  > >> > >t > >Its a DOS attack. > >  > ; > Not according to the folks on the Argus stand at Infosec. K > They say it's a major security hole (details of which they cannot releasefL > until Sun has had a chance to provide a patch). It does though only affect > Solaris on Intel.aL > Now it's possible the people on the stand don't know what they are talkingJ > about. However it is also true that a Denial of Service attack would notP > have won the prize !! In order to win the prize you had to have enough control( > of the system to deface some webpages. >  > David Webb > VMS and Unix team leader > CCSS > Middlesex University    2 Well thats very very odd since the attack allready3 has a patch for it and the patch has been availableh! since the beginning of the year. d   regardse Andrew Harrisons Enterprise IT Architect4   ------------------------------  % Date: Fri, 27 Apr 2001 14:54:52 +0100.0 From: andrew harrison <andrew.nospam@uk.sun.com>P Subject: Re: Hackers crack challenge within 24 hours, and Solaris gets the blame* Message-ID: <3AE97A2C.E5F70A30@uk.sun.com>   David Beatty wrote:t > 5 > On Fri, 27 Apr 2001 12:00:00 +0100, andrew harrison # > <andrew.nospam@uk.sun.com> wrote:M >  > >David Mathog wrote: > >>b > >> In article <3AE83385.5583F8C2@uk.sun.com>, andrew harrison <andrew.nospam@uk.sun.com> writes: > >> >' > >> >OpenVMS is also not secure unlessS/ > >> >patched. Timely patching and notifications1 > >> >of patches is just as important for OpenVMSs  > >> >as it is for any other OS. > >> > > >>M > >> A classic piece of FUD - containing a tiny nugget of truth on the top of  > >> huge steaming pile of BS. > >> > >a/ > >I won't asnwer the long rant except to point + > >out that the hack was a DOS attack not a + > >security violation. Since you seem happyE( > >to make the distinction between a DOS- > >attack and a security violation in OpenVMS./ > >it is only fair to do the same with Solaris.a > > 4 > >And if this is the case then OpenVMS if connected/ > >to the net using its bundled IP stack has too/ > >be patched to ensure that it isn't vunerableS > >to a number of DOS attacks. > >- > = > Andrew, do you realize the DOS Ping-of-Death attack appliesi= > to UCX V4.0 and V4.1 only, which are no longer supported byj> > Compaq (with the exception noted below)?  Supported releases1 > are V5.1, V5.0 with ECO 2, and V4.2 with ECO 4.o >   2 Do you realise that POD is only one of a number of6 denial of service attacks that UCX has been vunerable.  2 POD was the first, Land/Teardrop/Smurf/Naptha all  followed on afterwards.o  2 And Solaris 7 while not out of maintenance is not  the current Solaris release.   Regardse Andrew Harrisonu Enterprise IT Architectm   ------------------------------  % Date: Fri, 27 Apr 2001 11:23:34 -0400c5 From: David Beatty <David.Beatty@qwertysasasdfgh.com>eP Subject: Re: Hackers crack challenge within 24 hours, and Solaris gets the blame2 Message-ID: <vovpOkybS7AWi9c1t+N25ZrigyxZ@4ax.com>  3 On Fri, 27 Apr 2001 14:54:52 +0100, andrew harrison2! <andrew.nospam@uk.sun.com> wrote:r   >David Beatty wrote: >> o6 >> On Fri, 27 Apr 2001 12:00:00 +0100, andrew harrison$ >> <andrew.nospam@uk.sun.com> wrote: >> e >> >David Mathog wrote:T >> >>ec >> >> In article <3AE83385.5583F8C2@uk.sun.com>, andrew harrison <andrew.nospam@uk.sun.com> writes:p >> >> >e( >> >> >OpenVMS is also not secure unless0 >> >> >patched. Timely patching and notification2 >> >> >of patches is just as important for OpenVMS! >> >> >as it is for any other OS.H >> >> >  >> >>hN >> >> A classic piece of FUD - containing a tiny nugget of truth on the top of >> >> huge steaming pile of BS.. >> >>  >> >0 >> >I won't asnwer the long rant except to point, >> >out that the hack was a DOS attack not a, >> >security violation. Since you seem happy) >> >to make the distinction between a DOSt. >> >attack and a security violation in OpenVMS0 >> >it is only fair to do the same with Solaris. >> >5 >> >And if this is the case then OpenVMS if connectedi0 >> >to the net using its bundled IP stack has to0 >> >be patched to ensure that it isn't vunerable >> >to a number of DOS attacks.  >> > >> e> >> Andrew, do you realize the DOS Ping-of-Death attack applies> >> to UCX V4.0 and V4.1 only, which are no longer supported by? >> Compaq (with the exception noted below)?  Supported releasesu2 >> are V5.1, V5.0 with ECO 2, and V4.2 with ECO 4. >>   > 3 >Do you realise that POD is only one of a number of-7 >denial of service attacks that UCX has been vunerable.  >E3 >POD was the first, Land/Teardrop/Smurf/Naptha all ] >followed on afterwards. >-       Teardrop/Land2  ;     According to the CERT advisory, Digital TCP/IP Servicesu# is not affected by this DOS attack.   	     SmurfO  >     While Digital TCP/IP Services is affected by this DOS, theE correction can be (and should be) implemented at the router level --   see the CISCO entry for this.   
     Naptha  <     As we have discussed before, just because Digital TCP/IP< Services is affected on Tru64 does not necessarily mean that OpenVMS is affected.  3 >And Solaris 7 while not out of maintenance is not j >the current Solaris release.y  < I never said anything about Solaris 7, although others have.$ I know it's not the current release.   >n >Regards >Andrew Harrison >Enterprise IT Architect   ------------------------------  + Date: Fri, 27 Apr 2001 15:23:52 +0000 (UTC)e' From: david20@alpha2.mdx.ac.uk (D.Webb) P Subject: Re: Hackers crack challenge within 24 hours, and Solaris gets the blame+ Message-ID: <9cc2u8$6kv$1@aquila.mdx.ac.uk>e  ] In article <3AE95130.56165A1A@uk.sun.com>, andrew harrison <andrew.nospam@uk.sun.com> writes:F >David Mathog wrote: >> r` >> In article <3AE83385.5583F8C2@uk.sun.com>, andrew harrison <andrew.nospam@uk.sun.com> writes: >> > >t- >I won't asnwer the long rant except to points* >out that the hack was a DOS attack not a * >security violation. Since you seem happy ' >to make the distinction between a DOS  + >attack and a security violation in OpenVMS - >it is only fair to do the same with Solaris.  >    Andrew,c  & What is it about you and DOS attacks ?  < Where did you get the idea that this hack was a DOS attack ?G Since my last post on this subject pointing out that this was denied byhF the people on the Argus stand at Infosec I have had a chance to read a report on the Argus web site  , http://www.argus-systems.com/events/infosec/   Please read it in particular :-f   "a The Exploite  < The LSD hack was not an exploit of the PitBull software. LSD> became aware of a vulnerability in Intel x86 operating systemsD through a posting to the NetBSD advisory (that vulnerability was not discovered by LSD).i9 LSD was able to use that vulnerability to create a kernelrA level vulnerability in the base Solaris x86 operating system thattA was running on the system that Argus had deployed for the hackingrE challenge. The vulnerability exploited by LSD relates specifically tooJ operating system implementations supporting the Intel x86 architecture. In? addition to Solaris for x86, the vulnerability may affect other ; operating systems that support the Intel x86 architecture. c% This vulnerability had not previouslyo8 been posted on Solaris bug tracking web sites or mailingN lists, and to the best of our knowledge no patch was or is presently available to correct this vulnerability. s "    andl   "o Why Was the System Vulnerable?  ; Two factors contributed to the success of the hack. For the F hack to be successful both factors had to be operative. Elimination ofI either factor would have completely foiled the attack. Those factors are:s  : 1.The underlying base operating system had a vulnerability1   that was not identified through code analysis. o7 2.The particular system implementation that was exposedhG   during this event allowed users to log in and create shell accounts. l   "t                       L You are a mighty strange Enterprise IT architect if you think this describes a Denial of Service attack.h  M I have just received a mail message from Glenn Everhart explaining more aboutiM the hack. I apologise to Glenn and the group if he has already posted this :-   K " And you are right. The hole has been documented; it was an issue on LinuxfK or maybe openBSD in the past, on X86. Has to do with entry to other "rings" F on x86 and the fact that if users can create these inter-ring gatewaysG they can wind up being able to run in kernel. The structures are called-I LDTs. Getting into random kernel locations is about as serious a securityo hole as you get. "s  I Unfortunately Glenn neglected to post where he obtained this information.w     
 David Webb VMS and Unix team leader CCSS Middlesex University   ------------------------------  % Date: Fri, 27 Apr 2001 17:30:34 +0100p  From: steven.reece@quintiles.comP Subject: Re: Hackers crack challenge within 24 hours, and Solaris gets the blameH Message-ID: <OFD3901F94.6BD59F74-ON80256A3B.0059FE6C@qedi.quintiles.com>  H So, given David's information and the input from Glenn, what do we think that Andrew will go with next :|  E a) It's Intel's fault for developing a flawed platform that they then: allowed Sun to program on ?w  E b) It's Argus's fault for telling people about it and responding to a9, security vulnerability in the way they did ?  I c) Sun knew about it and were just about to release a patch, although the K patch had been available to contract customers for many months because this G is old news and every commercial customer of this software knows of theM problem (except Argus) ?   :-))   David Webb wrote:   
 >>>Andrew,  & What is it about you and DOS attacks ?  < Where did you get the idea that this hack was a DOS attack ?G Since my last post on this subject pointing out that this was denied bylF the people on the Argus stand at Infosec I have had a chance to read a report on the Argus web site <trim> <<<s   ------------------------------  % Date: Fri, 27 Apr 2001 10:24:21 -0700.! From: Shane.F.Smith@Healthnet.commP Subject: Re: Hackers crack challenge within 24 hours, and Solaris gets the blameD Message-ID: <OFB6F148A6.B107087E-ON88256A3B.005F7FC2@foundation.com>  G You know, Andrew, I still haven't seen you explain what the attack was, K only that it was a DOS. That covers a very wide range of attacks. Which DOS. was it?j   ShaneH          D andrew harrison <andrew.nospam@uk.sun.com> on 04/27/2001 04:00:00 AM  < Please respond to andrew harrison <andrew.nospam@uk.sun.com>   To:   Info-VAX@Mvb.Saic.Comu cc:.  K Subject:  Re: Hackers crack challenge within 24 hours, and Solaris gets theu       blame      David Mathog wrote:  >t< > In article <3AE83385.5583F8C2@uk.sun.com>, andrew harrison" <andrew.nospam@uk.sun.com> writes: > > $ > >OpenVMS is also not secure unless, > >patched. Timely patching and notification. > >of patches is just as important for OpenVMS > >as it is for any other OS.- > >- >-J > A classic piece of FUD - containing a tiny nugget of truth on the top of > huge steaming pile of BS.  >j  , I won't asnwer the long rant except to point( out that the hack was a DOS attack not a( security violation. Since you seem happy% to make the distinction between a DOSd* attack and a security violation in OpenVMS, it is only fair to do the same with Solaris.  1 And if this is the case then OpenVMS if connected , to the net using its bundled IP stack has to, be patched to ensure that it isn't vunerable to a number of DOS attacks.    regardsfJ > Yes there are occasionally security holes in VMS that must be addressed,I > but there is one every bloody WEEK in Solaris, and the vast majority of K > them allow root access to any pimply faced script kiddy with a few rublesnK > and a telephone.  I'm migrating to Solaris in spite of its security holesvF > because I need some software that runs there and not on VMS (and for otheroI > political/economic reasons which have long since been beaten to death).b SoK > I spent most of yesterday shutting Solaris security holes and am far from-K > sure that I've closed them all.  I was particularly displeased to see alleJ > of the SNMP pieces which were running by default, half of which had CERTI > warnings on them and no solutions, and which I didn't need in the first@H > place.  And when I went to restrict access to rpcbind was horrified toA > realize that Sun still (as of 1/01) ships a version that allowsi connections.I > from *anywhere*.  The fellow who wrote TCP Wrappers had a variant which>D > would reject connections by IP number, so I installed that.  But IH > shouldn't have had to, the version in Solaris should have been able to > apply this filter already. >>J > Whatever other failings VMS and Compaq may suffer, inattention to system > security is not one of them. >B- > >If you havn't bothered following the patche, > >paper chase for OpenVMS then your systems, > >are as vunerable as the Solaris for Intel > >box was.  >eJ > Ridiculous.  I do not recall the last "attack from the other side of theK > planet over the internet with a port scanner and gain SYSTEM control on aaJ > VMS system" security hole.  Conversely, there have been so many securityG > holes of that type on Solaris that it would take an hour to list thema all. >s > >n( > >Not a good effort on your part Larry. > >c >p > Nor yours Andrew.n >4 > David Mathog > mathog@caltech.edu@ > Manager, sequence analysis facility, biology division, Caltech   -- Andrew HarrisonC Enterprise IT Architects   ------------------------------  % Date: Fri, 27 Apr 2001 10:36:12 -0700s! From: Shane.F.Smith@Healthnet.comaP Subject: Re: Hackers crack challenge within 24 hours, and Solaris gets the blameD Message-ID: <OF6D3892F7.1900C59C-ON88256A3B.005FE212@foundation.com>  J Could it be that (gasp) Andrew didn't do his homework? :-) You /know/ theyK wouldn't have given that much money away if the hacker didn't genuinely win  it.h   Shanej          ; david20@alpha2.mdx.ac.uk (D.Webb) on 04/26/2001 04:23:39 PMt  3 Please respond to david20@alpha2.mdx.ac.uk (D.Webb)o   To:   Info-VAX@Mvb.Saic.Comh cc:   K Subject:  Re: Hackers crack challenge within 24 hours, and Solaris gets the        blamee    : In article <3AE8654E.9D8D5726@uk.sun.com>, andrew harrison" <andrew.nospam@uk.sun.com> writes:# >Shane.F.Smith@Healthnet.com wrote:  >>E >> Since there's apparently a fix, would you care to tell us what thes >> vulnerability was?p >> >h >Its a DOS attack. >a  9 Not according to the folks on the Argus stand at Infosec.tI They say it's a major security hole (details of which they cannot releaseeJ until Sun has had a chance to provide a patch). It does though only affect Solaris on Intel.iJ Now it's possible the people on the stand don't know what they are talkingH about. However it is also true that a Denial of Service attack would notF have won the prize !! In order to win the prize you had to have enough control & of the system to deface some webpages.    
 David Webb VMS and Unix team leader CCSS Middlesex University   ------------------------------  % Date: Fri, 27 Apr 2001 18:09:52 +0100u  From: steven.reece@quintiles.com  Subject: Re: IBM bought InformixH Message-ID: <OF8F028A84.74938307-ON80256A3B.005E1E66@qedi.quintiles.com>  G Might make some sense for some folks out there for Compaq to buy Ingres K from the one who's name I shall not utter (and I don't mean Macbeth!!!).  A G number of sites have been burned by the increases in prices for Ingres.s Steve.   Fabio wrote:J >>>So,  is time to Compaq re-buy RDB from Oracle  . . .   or Sybase ... or
 Ingres ...<<<o   ------------------------------  % Date: Fri, 27 Apr 2001 14:21:26 -0300n) From: fabio_compaq@ep-bc.petrobras.com.br   Subject: Re: IBM bought InformixL Message-ID: <OFBA7D6548.3492327F-ON03256A3B.005F4232@ep-bc.petrobras.com.br>  @ The company is planning the migration of all Ingres databases to* Oracle in a near future ( until  DEC 2002)   Regardsi   FC        1 steven.reece@quintiles.com em 27/04/2001 14:09:525  , Favor responder a steven.reece@quintiles.com             Info-VAX@Mvb.Saic.Comm        Assunto: Re: IBM bought Informix        G Might make some sense for some folks out there for Compaq to buy IngresaK from the one who's name I shall not utter (and I don't mean Macbeth!!!).  A G number of sites have been burned by the increases in prices for Ingres.s Steve.   Fabio wrote:J >>>So,  is time to Compaq re-buy RDB from Oracle  . . .   or Sybase ... or
 Ingres ...<<<n   ------------------------------  # Date: Fri, 27 Apr 2001 15:59:03 GMT 2 From: Piyush Avichal<pa@it.singer-friedlander.com>$ Subject: Installing OpenVMS - HELP!!4 Message-ID: <bHgG6.902$SZ5.72489@www.newsranger.com>  J I have a blank MicroVAX 3100-80 and OpenVMS 7.1 CD. I am trying to installP OpenVMS on this machine. I have booted off the CD, but all I get is a standaloneH backup prompt. I was expecting a menu with the option to do upgrades and& installations like you do with Alphas.  F Can anyone tell me how I can install the operating system from the CD?   Thanks In Advance,   Piyush.    ------------------------------  % Date: Fri, 27 Apr 2001 17:18:11 +0100-  From: steven.reece@quintiles.com( Subject: Re: Installing OpenVMS - HELP!!H Message-ID: <OF3A8A8C22.0CBBC551-ON80256A3B.0057BE38@qedi.quintiles.com>   Piyush, E If you follow the instructions in the Upgrade and Installation Manuala you'll see := $ BACKUP/IMAGE/VERIFY/NOALIAS sourcedrive:VMS071.B/SAVE_SET -       targetdevice:  I There is (contrary to popular belief and as of v7.1) a menu on the VAX CD8I if you boot from SYS1 instead of SYS0, although it has fewer functions on ) it than the Alpha menu.  The options are:l  & 1) Execute DCL commands and procedures 2) Shut down this system  H Selecting option 1 means that you can restore VMS071.B to the new system disk using the command above.e  F I'd really recommend looking at a copy of the upgrade and installationI manual for the version of VMS that you are installing.  It will make yourn life much easier.  Steve.   Piyush Avachal wrote:sE >>>I have a blank MicroVAX 3100-80 and OpenVMS 7.1 CD. I am trying tot installSE OpenVMS on this machine. I have booted off the CD, but all I get is a 
 standaloneH backup prompt. I was expecting a menu with the option to do upgrades and& installations like you do with Alphas.  F Can anyone tell me how I can install the operating system from the CD? <<<    ------------------------------  % Date: Fri, 27 Apr 2001 12:35:41 -0400c2 From: "Kent, Philip  JW1811" <kent@jwfc.jfcom.mil>( Subject: RE: Installing OpenVMS - HELP!!H Message-ID: <5B57189920E7D41190B500606D210686011A69F5@mailsvr.jfcom.mil>  K >I have a blank MicroVAX 3100-80 and OpenVMS 7.1 CD. I am trying to installfF >OpenVMS on this machine. I have booted off the CD, but all I get is a
 standaloneI >backup prompt. I was expecting a menu with the option to do upgrades and ' >installations like you do with Alphas.n >oG >Can anyone tell me how I can install the operating system from the CD?  >- >Thanks In Advance,- >- >Piyush.  G YOU MAY BE INTERESTED IN REVIEWING THE VMS 7.1 INSTALLATION MANUALS AT:s7 http://www.openvms.compaq.com:8000/ssb71/6487/6487p.htm   L Once you have standalone backup TO, you can transfer VMS071.B to your system disk:>4 BACKUP/IMAGE/VERIFY CD:VMS071.B/SAV NEW_SYSTEM_DISK:   THEN BOOT THE NEW SYSTEM DISKo   ------------------------------  # Date: Fri, 27 Apr 2001 16:38:34 GMT,2 From: Piyush Avichal<pa@it.singer-friedlander.com>( Subject: Re: Installing OpenVMS - HELP!!4 Message-ID: <eghG6.954$SZ5.76814@www.newsranger.com>  0 Thanks for your help. I have now figured it out!  K In article <bHgG6.902$SZ5.72489@www.newsranger.com>, Piyush Avichal says...n >nK >I have a blank MicroVAX 3100-80 and OpenVMS 7.1 CD. I am trying to install Q >OpenVMS on this machine. I have booted off the CD, but all I get is a standalonenI >backup prompt. I was expecting a menu with the option to do upgrades and>' >installations like you do with Alphas.a >eG >Can anyone tell me how I can install the operating system from the CD?y >o >Thanks In Advance,m >p >Piyush. >k >h   ------------------------------  % Date: Fri, 27 Apr 2001 19:13:22 +0200S+ From: Arne Bergseth <Arne.Bergseth@dnv.com>a( Subject: Re: Installing OpenVMS - HELP!!' Message-ID: <3AE9A8B2.CA7DFCFF@dnv.com>u   Pyiush,n  E On the DECUS website of the OpenVMS Hobbyist program, there is both a> simplified tutorial,* and links to COMPAQ OpenVMS documentation:+ http://www.montagar.com/hobbyist/index.htmlt  4 Also the OpenVMS Frequentlu Asked Questions page at:5 http://www.openvms.compaq.com/wizard/openvms_faq.htmlt( contains information that may be useful.  
 Arne BergsethC   ------------------------------  # Date: Fri, 27 Apr 2001 17:36:10 GMTn8 From: hammond@not@peek.ppb.cpqcorp.net (Charlie Hammond)( Subject: Re: Installing OpenVMS - HELP!!0 Message-ID: <e6iG6.304$5I.5594@news.cpqcorp.net>  5 In article <bHgG6.902$SZ5.72489@www.newsranger.com>, a4 Piyush Avichal<pa@it.singer-friedlander.com> writes:  K >I have a blank MicroVAX 3100-80 and OpenVMS 7.1 CD. I am trying to installtQ >OpenVMS on this machine. I have booted off the CD, but all I get is a standalone I >backup prompt. I was expecting a menu with the option to do upgrades ands' >installations like you do with Alphas.e >dG >Can anyone tell me how I can install the operating system from the CD?l  F If you need to ask this question, then you should probably obtaina and5 read the OpenVMS VAX Upgrade and Installation manual.t   That said...  J (1) Use standalone backup to restore saveset B to your target system disk.H For example, if your CD is DKA400: and your target system disk is DKA0:,1 you might use a command similar to the following:.  1     $ BACKUP/IMAGE DKA400:VMS071.B/SAVE_SET DKA0:t  < (2) Boot the target system disk and follow the instructions.A (Leave the OpenVMS VAX operating system CD-ROM in the CD drive --,- it will be used to install additional files.)t  E NOTE: Unless you have specific reasons/needs, select ALL the options..   -- nK     Charlie Hammond -- Compaq Computer Corporation -- Pompano Beach  FL USAhH        (hammond@not@peek.ppb.cpqcorp.net -- remove "@not" when replying)J       All opinions expressed are my own and not necessarily my employer's.   ------------------------------  % Date: Fri, 27 Apr 2001 17:53:24 +0930 % From: Jeremy Begg <jeremy@vsm.com.au>w@ Subject: Re: Mounting and Management station: comments please...) Message-ID: <3AE92C7C.2090409@vsm.com.au>    Robert Deininger wrote:s  L > In article <Pine.LNX.4.21.0104231529490.21828-100000@irys.stanpol.com.pl>, > gs@stanpol.com.pl wrote: >  >  >>  Can anyone comment:  >>  B >> http://www.openvms.compaq.com/openvms/products/argus/index.html> >> "And, OpenVMS Management Station mounts volumes much faster; >>  than is possible with DCL or command procedures, [...]"b >>       ********************s >>   >>  WHAT ? Why ?> >>  Can't see a way why MOUNT/NOREBUILD followed with /REBUILDA >> in subprocesses (one for disk) *must* be slower than any otherr< >> way. Ok, "specialised" .EXE where does the same will save< >> image activactions, but IMHO this is *nothing* worth with& >> typical time of the device mount...8 >>  "Marketing" ? "With Windows you are 'the best'" ? ;> >...H > I thought I was joking, but reading further, I see that they really DO6 > expect this billy-box to be involved in VMS reboots. > J > I assume there were hordes of VMS system managers demanding this productL > over the last few years.  After all, we know that Compaq always listens to' > its customers where VMS is concerned.t  I To be fair to Compaq, I think the Argus project originated in the mid-90slH when Digital was trying to wash its hands of VMS and VMS engineering was; trying to demonstrate that you could make VMS look like NT.o  H While I never used it much, I can well imagine that there would be sitesJ which like its GUI for managing queues and users.  But I agree that having& it run my disk mounts is rather scary!   	Jeremy Begg   ------------------------------  # Date: Fri, 27 Apr 2001 15:34:12 GMTo* From: "AConway" <aconway@optushome.com.au> Subject: network card on win2kG Message-ID: <UjgG6.50768$Xx3.287315@news1.eburwd1.vic.optushome.com.au>a  . ok lets hope ive got the right place... anyway  = my network card is a ' PCI Ethernet DEC 21041 Based Adapter 'nF im looking for the install drivers for win2k as my isp (Optus home, in= Australia) says this network card isnt compatible with win2k..= (even though on their website it says they are 2k compliant!)yJ  ive only had cable for just over a year, and arent really looking forwarf% to having to buy a new network card..e  I anyway, i rang up my isp and they said its not compatible but people HAVE B got it to work ( i guessed by getting different drivers or editing
 something)   Any help would be appreciatede   thanks   ------------------------------  # Date: Fri, 27 Apr 2001 16:14:14 GMTw2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)" Subject: Re: network card on win2k0 Message-ID: <qVgG6.301$5I.5714@news.cpqcorp.net>  t In article <UjgG6.50768$Xx3.287315@news1.eburwd1.vic.optushome.com.au>, "AConway" <aconway@optushome.com.au> writes:/ :ok lets hope ive got the right place... anyway3  J   Uh, no.  This is the OpenVMS newsgroup, and is not known for Windows NT H   nor generic DEC hardware discussions.  Nor are many of the folks here 9   particularly familiar with Microsoft operating systems.e  > :my network card is a ' PCI Ethernet DEC 21041 Based Adapter '0 :im looking for the install drivers for win2k...  B   There are various adapters that were available, this may well be=   a DE450 series controller.  Check the W2K HCL as a start...a  J :anyway, i rang up my isp and they said its not compatible but people HAVEC :got it to work ( i guessed by getting different drivers or editingp :something)   F   Ask your ISP, or check the Microsoft databases -- there are a coupleC   of references to the DE450 at the Microsoft support website.  eg:k  B     http://support.microsoft.com/support/kb/articles/q156/3/26.asp  (   The following might or might not help:  9     ftp://ftp.digital.com/pub/Digital/info/semiconductor/i?       networks-and-communications/software/dsc-software-nc.html2  B   I have no idea if the above drivers will work on W2K, nor if the@   problem referenced in the Microsoft article has been resolved.  G :and arent really looking forwarf to having to buy a new network card..o     Um, faster NICs are cheap.    H   (Whether appreciated or not, new software can mean new hardware.  ThisG   model is common in the PC space, and is becoming more common in other H   product spaces as the prices of hardware drop, as the capabilities of J   the new hardware increase, as the requirements of new software increase,I   and as the cost of sustaining and support engineering for the existing  G   hardware increases.  This is definitely disconcerting to those of us  @   that have been in the business for a while.  But I digress...)  G   Note: I have no idea if DE450 is tested or supported on Windows 2000.   K   Note: comp.sys.dec would often be suggested as a better spot for generic t/   DIGITAL hardware questions and discussions...-  N  ---------------------------- #include <rtfaq.h> -----------------------------N       For additional, please see the OpenVMS FAQ -- www.openvms.compaq.com    N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------    Date: 27 Apr 2001 10:56:15 -05009 From: Kilgallen@eisner.decus.org.nospam (Larry Kilgallen)iJ Subject: newsgroup manners (was: 80% is Good Enough for M$. Now it's 100%)3 Message-ID: <6PqDt$gwSrt8@eisner.encompasserve.org>r  n In article <3GJZVTREJtgi@tachxxsoftxxconsult>, wayne@tachysoft.xxx.065234.killspam.015d (Wayne Sewell) writes:] > In article <3AE87870.EE6B82F7@fsi.net>, "David J. Dachtera" <djesys.nospam@fsi.net> writes:b >> John Eisenschmidt wrote:  >>> H >>> Christ, I don't know which is sadder: The software or the people who >>> make it.	 >> [snip]n >>> For god sakes  >> oF >> Just my opinion, mind you, but I do take umbrage to such use of theK >> Lord's name(s) in vain. Other folks react badly to obscenities and otherf  >> profanities. Watcha gonna do? >> a > P > Personally, I don't think religion belongs in a technical forum in *any* form,O > pro or con.  I find religious quotes in sigs, such as those found in posts byiK > that speakeasy guy, to be every bit as offensive as you find the above.  t   I agree with Wayne.    ------------------------------  % Date: Fri, 27 Apr 2001 10:41:02 -0700n! From: Shane.F.Smith@Healthnet.comeN Subject: Re: newsgroup manners (was: 80% is Good Enough for M$. Now it's 100%)D Message-ID: <OF8B07EBB4.B0DF9653-ON88256A3B.00611B8D@foundation.com>   Ditto. Totally.h          J Kilgallen@eisner.decus.org.nospam (Larry Kilgallen) on 04/27/2001 08:56:15 AM  E Please respond to Kilgallen@eisner.decus.org.nospam (Larry Kilgallen).   To:   Info-VAX@Mvb.Saic.Com  cc:w  K Subject:  newsgroup manners (was: 80% is Good Enough for M$. Now it's 100%)l    . In article <3GJZVTREJtgi@tachxxsoftxxconsult>,? wayne@tachysoft.xxx.065234.killspam.015d (Wayne Sewell) writes:t= > In article <3AE87870.EE6B82F7@fsi.net>, "David J. Dachtera"T <djesys.nospam@fsi.net> writes:e >> John Eisenschmidt wrote:r >>>eH >>> Christ, I don't know which is sadder: The software or the people who >>> make it.	 >> [snip]l >>> For god sakest >>F >> Just my opinion, mind you, but I do take umbrage to such use of theK >> Lord's name(s) in vain. Other folks react badly to obscenities and othero  >> profanities. Watcha gonna do? >> >oJ > Personally, I don't think religion belongs in a technical forum in *any* form, F > pro or con.  I find religious quotes in sigs, such as those found in posts byI > that speakeasy guy, to be every bit as offensive as you find the above.3   I agree with Wayne.3   ------------------------------  % Date: Fri, 27 Apr 2001 11:26:10 +0100h0 From: andrew harrison <andrew.nospam@uk.sun.com>8 Subject: Re: OpenVMS article - please explain last line!* Message-ID: <3AE94942.AAC9E225@uk.sun.com>   Paul Repacholi wrote:2 > 4 > andrew harrison <andrew.nospam@uk.sun.com> writes: > E > > You still don't seem to get it. The old result was run in what isr9 > > the current ES40 memory subsystem, it hasn't changed.  > A > > Changing the CPU's to 833 Mhz will only change the single CPUsF > > result, but won't change the 4 CPU result which is bottlenecked on > > the memory subsystem.a > C > But it is NOT limited by the switch. THAT, was Kerrie's statment,s > the D-chip bandwidth.t > $ > I quote below so you are reminded. > C > = Current ES40 and DS20E systems use a 5.2Gb/sec cross bar switchsF > = technology.  I am sure folks here realize the difference between a2 > = cross bar switch and standard bus technology." >   4 What part of this discussion did you not understand.  1 Kerry claimed that tsmurphys STREAMS posting was a5 false citing the marketing cross bar switch bandwidthi0 number as proof of this. Neither tsmurphy or my 5 posts refer to the cross bar as being the bottleneck  3 we simply point out that the actual data throughpute for the ES40 is not 5.2 GB/s.   4 I corrected him and again he reiterated this claim.   2 If you don't understand the difference between the7 actual memory bandwidth of a system, which is in effect6< what STREAMS measures and a marketing bullet point refering 0 to the fastest point in the ES40 while ignoring - any other considerations when why not say so.d    3 > Now, what part of the above don't you understand.n >   1 Remarkable, you accuse me of incompehension when i5 you are arguing in support of someone who is claimingc2 that the memory bandwidth numbers published by his own company are false. -  G > BTW, when are we going to see the no-prefetch Sun SPEC result Andrew?r >   . Ahh there you go again, how many attempts are . you going to make to change the subject before you admit that Kerry was BSing.n   Regardsn Andrew Harrisonf Enterprise IT Architect    ------------------------------  % Date: Fri, 27 Apr 2001 11:50:18 +0100n0 From: andrew harrison <andrew.nospam@uk.sun.com>8 Subject: Re: OpenVMS article - please explain last line!* Message-ID: <3AE94EEA.7CA46066@uk.sun.com>  " Shane.F.Smith@Healthnet.com wrote: > L > Are you completely incapable of understanding that the subject had changed > before I got involved? > I > As I posted earlier, I have no direct knowledge of the bandwidth issueseJ > here, but based on past experience of the pair of you I consider Kerry aK > considerably more reliable source than you are. I expect if I had time or-M > inclination to do the research, I'd find Kerry right again, and you being ahH > total spin merchant as usual. You have no credibility in my eyes. From& > others' posts, I know I'm not alone. >   8 Sorry this is total BS. Kerrys previous discussions with4 me have be characterised by an almost total lack of 4 beleivable content from Kerry unless you are someone who needs to beleive Kerry.    And this is no different.   6 Kerry included in his response to my suggestions that 7 5.2 GB/s was BS a section cut from a Compaq white paperr$ explaining what memory bandwidth is.  ? http://www.compaq.com/alphaserver/technology/mem-bandwidth.html1  5 This is the best example I have ever seen from Kerry :3 of the selective way he parcels out information to 82 the OpenVMS choir. If Kerry had been a little less/ selective he would have included the following h! paragraph from the same document.   
 Which says   "Rule of Thumb? What if a vendor publishes "raw" bandwidth and does not publish @ programmer-available (Streams) bandwidth? Is there a conversion  factor or rule of thumb? m  > Yes. Take the vendor's claimed bandwidth and multiply by 0.5. @ If the vendor protests that this is unfair, challenge the vendor? to prove it by running the Streams benchmark and by submitting s
 the results. e  = Note: A factor of 0.10 may be more appropriate if the vendor  0 marketing literature lacks precise information."  @ Laugh I almost wet myself. Compaqs own technical people advocate9 exactly what tsmurphy and myself have said, use STREAMS.    ; Incedentally there is a STREAMS result for the ES40 and itsd= almost exactly 50% of the RAW number Kerry published which ash2 the Compaq white paper points out is marketing BS.  8 The fact that Kerry chose not to include this snippet in7 his response is hardly supprising since it destroys hisA7 argument, the fact that he chose to quote from the samei white paper is less explicable.-  7 Were this an isolated incident then I would give Kerry 59 the benefit of the doubt, sadly it is one of a long list.i   regards: > Shanea > F > andrew harrison <andrew.nospam@uk.sun.com> on 04/26/2001 03:51:21 AM > > > Please respond to andrew harrison <andrew.nospam@uk.sun.com> >  > To:   Info-VAX@Mvb.Saic.ComO > cc:o > ; > Subject:  Re: OpenVMS article - please explain last line!n > $ > Shane.F.Smith@Healthnet.com wrote: > >wL > > Here we go again, Andrew. You responded to Paul by ignoring what he saidJ > > and trying to make one of your own points on another subject. Again. IM > > explained what you'd done (although it was obvious), and you do it again.TF > > My post, and indeed Paul's, were not directly connected to Kerry's	 > content/K > > - both were on the subject of your usual tactics. So, as you can see, Io > allyL > > did the homework set for the subject, and aced it. You, however, flunkedB > > badly by turning geography homework in to the english teacher. > >  > 3 > Shane you are far too clever for me, there was me-0 > thinking that Paul had attempted to change the6 > subject because Kerry was BSing badly and obviously. > 0 > When in fact you were both trying to give us a > geography lesson.H > M > > Now, if you want to change the subject back to memory bandwidth, that's a  > > different topic. > >S > 5 > OH good, so are you finally after all the attempted - > avoidance offering to answer the question ?a >  > Let me refresh you.  > 4 > Do you support Kerrys claims of 5.2 GB/s bandwidth0 > and by default agree with his allegations that- > tsmurphys STREAMS posting is false ?  Its a  > simple yes/no. > 0 > Perhaps as you have offered to get back to the0 > bandwidth question after your most instructive3 > geography lecture you could answer this question.g > 4 > Of course if you don't actually have a clue either2 > way then perhaps you shouldn't have got involved > in the first place.e > 	 > > Shane  > >iH > > andrew harrison <andrew.nospam@uk.sun.com> on 04/25/2001 06:39:09 AM > > @ > > Please respond to andrew harrison <andrew.nospam@uk.sun.com> > >e > > To:   Info-VAX@Mvb.Saic.Comn > > cc:: > >p= > > Subject:  Re: OpenVMS article - please explain last line!f > >i& > > Shane.F.Smith@Healthnet.com wrote: > > >-M > > > Gee, Andrew, you obviously never studied English particularly well, didnH > > > you? Still, I gave up expecting well constructed language from you > yearsPM > > > ago. I was expressing no position on Kerry's post, because I don't knowZ > > thewJ > > > text of it. I'd lost track (and largely interest) by that point, and > it'dI > > > been snipped. I was simply pointing out your syntactic trap for thee
 > > rather > > > feeble attempt it was. > > >uK > > > As for 5+GB, well, in the absence of direct knowledge if Kerry posted  > it,t > > IgL > > > have a lot more faith in its accuracy than if you posted it, let's put > it2 > > > that way. There again, that's not difficult. > > >e > >a< > > So you responded to my post without reading the previous7 > > posts in the same thread. How remarkable, don't yous5 > > think you would be better off reading Kerrys postB% > > and the one he was responding to.l > > 6 > > Oh no lets think that would be inconvenient if you= > > read Kerrys thread you will then have to make a judgementA+ > > as to its veracity. Better not read it.  > >o9 > > Of course if you don't read it then you should not bea! > > posting responses should you.  > >M4 > > Seems to be another rock and hard place decision > > for you. > >O6 > > To help you out the following lines are from Kerry	 > > post.. > >- > > Kerrys posting > >-C > > "Yep, a real mystery how people can post such false numbers :-)n > >@C > > Current ES40 and DS20E systems use a 5.2Gb/sec cross bar switcho > > technology.mJ > > I am sure folks here realize the difference between a cross bar switch > > andr > > standard bus technology."  > >  > > This was in response to. > >uD > > "I don't understand how somebody can post such a blatantly false
 > > statement.; > > when stream is a click away. Here are the real numbers:c > >i' > > Intel Pentium 4 1.4 GHz - 1574 MB/sy > > Compaq ES40 - 1338 MB/sp > > Compaq DS20 - 1323 MB/s  > > HP B2000 - 960 MB/s ! > > AMD Athlon 800 MHz - 586 MB/so) > > Intel Pentium III 733 MHz - 544 MB/s"O > >SA > > All you need to do is decide if the 5.2 GB/s claim is true ore@ > > if the STREAMS results published by Compaq, which the poster: > > Kerry is accusing of falsehood refered to are correct. > >E; > > Perhaps instead of accusing me of lack of understandingk< > > of the english language you would be better off actually7 > > reading a bit of english. At the moment your school,7 > > report for english reads "does not do his homework"f > >d > > regards. > > > Shanei > > >WJ > > > andrew harrison <andrew.nospam@uk.sun.com> on 04/24/2001 05:25:07 AM > > >uB > > > Please respond to andrew harrison <andrew.nospam@uk.sun.com> > > > ! > > > To:   Info-VAX@Mvb.Saic.Comw	 > > > cc:o > > >y? > > > Subject:  Re: OpenVMS article - please explain last line!  > > >l( > > > Shane.F.Smith@Healthnet.com wrote: > > > >tL > > > > Unless I'm misreading him, he's doing neither. He's using sarcasm to > > take > > > ahJ > > > > potshot at Sun. And in response, you're doing a crude variation of > thesF > > > > "When did you stop beating your wife?" verbal trap. Run out of > > > imagination, > > > > eh Andrew? > > > >r > > > ) > > > Actually you are mistaken its not au4 > > > "when did you stop beating your wife" question1 > > > but it is "between a rock and a hard place"- > > > question.- > > > 2 > > > Both you a Paul know that it is matketing BS* > > > but to admit this would be terrible. > > >r# > > > Hence the attempt at sarcasm.X > > >15 > > > What do you think of Kerrys claim, do you thinki3 > > > the 5+ GB/s is marketing BS or the truth ????s > > >I
 > > > regards1
 > > > > Shanee > > > >lL > > > > andrew harrison <andrew.nospam@uk.sun.com> on 04/23/2001 08:58:12 AM > > > >sD > > > > Please respond to andrew harrison <andrew.nospam@uk.sun.com> > > > >y# > > > > To:   Info-VAX@Mvb.Saic.Com  > > > > cc:  > > > > A > > > > Subject:  Re: OpenVMS article - please explain last line!t > > > >. > > > > Paul Repacholi wrote:a	 > > > > >d< > > > > > andrew harrison <andrew.nospam@uk.sun.com> writes:	 > > > > >nI > > > > > > "Main, Kerry" wrote: As is making claims for existing systemse > whichsG > > > > > > are not supported by the data you youselves have published.w	 > > > > >f& > > > > > OK Kerry, naughty boy! Smak.	 > > > > >RH > > > > > You should know that it is only aceptable to enrich the mental > > gardenseH > > > > > of you poor punters by spreading infomation based on NDA data.	 > > > > >s > > > >a1 > > > > So was Kerrys posting bullshit or are you. > > > > defending it.T > > > >e > > > > Regardsp > > > > Andrew Harrisonn > > > > Enterprise IT Architecta > > >u > > > -- > > > Andrew Harrisons > > > Enterprise IT Architect  > >M > > -- > > Andrew Harrison  > > Enterprise IT Architect  >  > -- > Andrew Harrisont > Enterprise IT Architectn   -- r Andrew Harrison. Enterprise IT Architectd   ------------------------------  % Date: Fri, 27 Apr 2001 12:44:18 +0100 0 From: andrew harrison <andrew.nospam@uk.sun.com>8 Subject: Re: OpenVMS article - please explain last line!* Message-ID: <3AE95B92.94158763@uk.sun.com>   Paul Repacholi wrote:n > 4 > andrew harrison <andrew.nospam@uk.sun.com> writes: > E > > You still don't seem to get it. The old result was run in what iss9 > > the current ES40 memory subsystem, it hasn't changed.c > A > > Changing the CPU's to 833 Mhz will only change the single CPUnF > > result, but won't change the 4 CPU result which is bottlenecked on > > the memory subsystem.a > C > But it is NOT limited by the switch. THAT, was Kerrie's statment,v > the D-chip bandwidth.  > $ > I quote below so you are reminded. > C > = Current ES40 and DS20E systems use a 5.2Gb/sec cross bar switchvF > = technology.  I am sure folks here realize the difference between a2 > = cross bar switch and standard bus technology." > 3 > Now, what part of the above don't you understand.d >   . read the white paper Kerry selected his memory bandwidth definitions from.,  ? http://www.compaq.com/alphaserver/technology/mem-bandwidth.htmlr  1 Read down to the Rules of Thumb section once you f3 have done that I would suggest that you can do one   of two things.  1 A. Admit total defeat, you should get an appology 3    from Kerry for missleading you which may offset B!    the pain this will cause you. n  0 B. Keep on arguing. You will see from the white +    paper that this isn't an option but you u    are welcome to keep it up.n  . If there is anything in the Compaq white paper% you don't understand then please ask.   c2 If you are having difficulty with 50% of 5.2 GB/s  then let me know.   1 If you need the STREAMS result for the ES40 then   let me know.  1 If you want to know how close to 50 % of 5.2 GB/s-, the ES40 STREAMS result is then let me know.  n, If you want to know what to do when you are - in a hole and you are still digging then let   me know.   regards  Andrew Harrisond Enterprise IT Architectl   ------------------------------  % Date: Fri, 27 Apr 2001 13:17:05 +0000t  From: Steve.Spires@yellgroup.com8 Subject: Re: OpenVMS article - please explain last line!/ Message-ID: <00256A3B.0048FD9D.00@quegw01.btyp>i  L Contact:   Tel: 3063  -  IS - Infrastructure, 1st Floor, Bridge Street Plaza     Andrew H said;  	 _________f  + If you want to know what to do when you are , in a hole and you are still digging then let me know.  	 _________e  N An admission on your part that you have more experience of this than others on this board?    ;^Dk   Steve So   ------------------------------  % Date: Fri, 27 Apr 2001 14:40:07 +0100c0 From: andrew harrison <andrew.nospam@uk.sun.com>8 Subject: Re: OpenVMS article - please explain last line!* Message-ID: <3AE976B7.DDF89A30@uk.sun.com>  ! Steve.Spires@yellgroup.com wrote:i > N > Contact:   Tel: 3063  -  IS - Infrastructure, 1st Floor, Bridge Street Plaza >  > Andrew H said; >  > _________e > - > If you want to know what to do when you are . > in a hole and you are still digging then let
 > me know. >  > _________u > P > An admission on your part that you have more experience of this than others on
 > this board?. >   5 Now why would I admit to something that is obviously y not true. :)  2 But since you have jumped into this thread as well' do you agree with Kerry or disagree ???d     regards@ Andrew Harrisonn Enterprise IT Architect    ------------------------------  % Date: Fri, 27 Apr 2001 10:22:36 -07000! From: Shane.F.Smith@Healthnet.com08 Subject: Re: OpenVMS article - please explain last line!D Message-ID: <OFED2ED987.F665BAA8-ON88256A3B.005F5EF8@foundation.com>  F Read what I said again, Andrew. As usual, you're attacking a different point from the one I'm making.   Shanew          D andrew harrison <andrew.nospam@uk.sun.com> on 04/27/2001 03:50:18 AM  < Please respond to andrew harrison <andrew.nospam@uk.sun.com>   To:   Info-VAX@Mvb.Saic.Como cc:   9 Subject:  Re: OpenVMS article - please explain last line!f    " Shane.F.Smith@Healthnet.com wrote: > D > Are you completely incapable of understanding that the subject had changedb > before I got involved? >nI > As I posted earlier, I have no direct knowledge of the bandwidth issuessJ > here, but based on past experience of the pair of you I consider Kerry aK > considerably more reliable source than you are. I expect if I had time or-K > inclination to do the research, I'd find Kerry right again, and you beingp auH > total spin merchant as usual. You have no credibility in my eyes. From& > others' posts, I know I'm not alone. >a  8 Sorry this is total BS. Kerrys previous discussions with3 me have be characterised by an almost total lack ofe4 beleivable content from Kerry unless you are someone who needs to beleive Kerry.    And this is no different.n  5 Kerry included in his response to my suggestions thats7 5.2 GB/s was BS a section cut from a Compaq white paperb$ explaining what memory bandwidth is.  ? http://www.compaq.com/alphaserver/technology/mem-bandwidth.html   4 This is the best example I have ever seen from Kerry2 of the selective way he parcels out information to2 the OpenVMS choir. If Kerry had been a little less. selective he would have included the following! paragraph from the same document.t  
 Which says   "Rule of Thumb? What if a vendor publishes "raw" bandwidth and does not publishe? programmer-available (Streams) bandwidth? Is there a conversione factor or rule of thumb?  = Yes. Take the vendor's claimed bandwidth and multiply by 0.5.o@ If the vendor protests that this is unfair, challenge the vendor> to prove it by running the Streams benchmark and by submitting the results.  < Note: A factor of 0.10 may be more appropriate if the vendor0 marketing literature lacks precise information."  @ Laugh I almost wet myself. Compaqs own technical people advocate8 exactly what tsmurphy and myself have said, use STREAMS.  ; Incedentally there is a STREAMS result for the ES40 and itse= almost exactly 50% of the RAW number Kerry published which as 2 the Compaq white paper points out is marketing BS.  8 The fact that Kerry chose not to include this snippet in7 his response is hardly supprising since it destroys hise7 argument, the fact that he chose to quote from the sameg white paper is less explicable.P  6 Were this an isolated incident then I would give Kerry9 the benefit of the doubt, sadly it is one of a long list.o   regardsi > Shaner >wF > andrew harrison <andrew.nospam@uk.sun.com> on 04/26/2001 03:51:21 AM >-> > Please respond to andrew harrison <andrew.nospam@uk.sun.com> >  > To:   Info-VAX@Mvb.Saic.Como > cc:. >.; > Subject:  Re: OpenVMS article - please explain last line!! >e$ > Shane.F.Smith@Healthnet.com wrote: > > G > > Here we go again, Andrew. You responded to Paul by ignoring what he  saidJ > > and trying to make one of your own points on another subject. Again. IF > > explained what you'd done (although it was obvious), and you do it again.F > > My post, and indeed Paul's, were not directly connected to Kerry's	 > content K > > - both were on the subject of your usual tactics. So, as you can see, I  > alleD > > did the homework set for the subject, and aced it. You, however, flunkeduB > > badly by turning geography homework in to the english teacher. > >  > 3 > Shane you are far too clever for me, there was me 0 > thinking that Paul had attempted to change the6 > subject because Kerry was BSing badly and obviously. >y0 > When in fact you were both trying to give us a > geography lesson.t > K > > Now, if you want to change the subject back to memory bandwidth, that'se ao > > different topic. > >a >r5 > OH good, so are you finally after all the attemptedm- > avoidance offering to answer the question ?m >  > Let me refresh you.p >I4 > Do you support Kerrys claims of 5.2 GB/s bandwidth0 > and by default agree with his allegations that- > tsmurphys STREAMS posting is false ?  Its au > simple yes/no. >f0 > Perhaps as you have offered to get back to the0 > bandwidth question after your most instructive3 > geography lecture you could answer this question.m >a4 > Of course if you don't actually have a clue either2 > way then perhaps you shouldn't have got involved > in the first place.m >i	 > > Shane  > >tH > > andrew harrison <andrew.nospam@uk.sun.com> on 04/25/2001 06:39:09 AM > >n@ > > Please respond to andrew harrison <andrew.nospam@uk.sun.com> > >  > > To:   Info-VAX@Mvb.Saic.Como > > cc:  > >e= > > Subject:  Re: OpenVMS article - please explain last line!g > >l& > > Shane.F.Smith@Healthnet.com wrote: > > >dI > > > Gee, Andrew, you obviously never studied English particularly well,e didyH > > > you? Still, I gave up expecting well constructed language from you > yearsbH > > > ago. I was expressing no position on Kerry's post, because I don't know > > theaJ > > > text of it. I'd lost track (and largely interest) by that point, and > it'dI > > > been snipped. I was simply pointing out your syntactic trap for thee
 > > rather > > > feeble attempt it was. > > >tK > > > As for 5+GB, well, in the absence of direct knowledge if Kerry posted  > it,t > > I H > > > have a lot more faith in its accuracy than if you posted it, let's put. > it2 > > > that way. There again, that's not difficult. > > >w > >i< > > So you responded to my post without reading the previous7 > > posts in the same thread. How remarkable, don't youe5 > > think you would be better off reading Kerrys post % > > and the one he was responding to.r > >o6 > > Oh no lets think that would be inconvenient if you= > > read Kerrys thread you will then have to make a judgementI+ > > as to its veracity. Better not read it.w > >v9 > > Of course if you don't read it then you should not bes! > > posting responses should you.r > > 4 > > Seems to be another rock and hard place decision > > for you. > >y6 > > To help you out the following lines are from Kerry	 > > post.n > >d > > Kerrys posting > >dC > > "Yep, a real mystery how people can post such false numbers :-)  > > C > > Current ES40 and DS20E systems use a 5.2Gb/sec cross bar switcht > > technology.mJ > > I am sure folks here realize the difference between a cross bar switch > > andn > > standard bus technology."o > >  > > This was in response to. > > D > > "I don't understand how somebody can post such a blatantly false
 > > statementr; > > when stream is a click away. Here are the real numbers:  > >t' > > Intel Pentium 4 1.4 GHz - 1574 MB/st > > Compaq ES40 - 1338 MB/sr > > Compaq DS20 - 1323 MB/sp > > HP B2000 - 960 MB/s ! > > AMD Athlon 800 MHz - 586 MB/sr) > > Intel Pentium III 733 MHz - 544 MB/s"T > > A > > All you need to do is decide if the 5.2 GB/s claim is true orh@ > > if the STREAMS results published by Compaq, which the poster: > > Kerry is accusing of falsehood refered to are correct. > >n; > > Perhaps instead of accusing me of lack of understanding < > > of the english language you would be better off actually7 > > reading a bit of english. At the moment your schoola7 > > report for english reads "does not do his homework"  > >l > > regardst > > > Shanei > > >dJ > > > andrew harrison <andrew.nospam@uk.sun.com> on 04/24/2001 05:25:07 AM > > > B > > > Please respond to andrew harrison <andrew.nospam@uk.sun.com> > > >.! > > > To:   Info-VAX@Mvb.Saic.Com-	 > > > cc:, > > >,? > > > Subject:  Re: OpenVMS article - please explain last line!  > > >S( > > > Shane.F.Smith@Healthnet.com wrote: > > > > I > > > > Unless I'm misreading him, he's doing neither. He's using sarcasm, to > > take > > > a J > > > > potshot at Sun. And in response, you're doing a crude variation of > thedF > > > > "When did you stop beating your wife?" verbal trap. Run out of > > > imagination, > > > > eh Andrew? > > > >  > > > ) > > > Actually you are mistaken its not ac4 > > > "when did you stop beating your wife" question1 > > > but it is "between a rock and a hard place"i > > > question.l > > >t2 > > > Both you a Paul know that it is matketing BS* > > > but to admit this would be terrible. > > >s# > > > Hence the attempt at sarcasm.o > > >t5 > > > What do you think of Kerrys claim, do you thinko3 > > > the 5+ GB/s is marketing BS or the truth ????h > > >e
 > > > regards 
 > > > > Shane  > > > >sI > > > > andrew harrison <andrew.nospam@uk.sun.com> on 04/23/2001 08:58:12h AM > > > >kD > > > > Please respond to andrew harrison <andrew.nospam@uk.sun.com> > > > >s# > > > > To:   Info-VAX@Mvb.Saic.Comd > > > > cc:  > > > >eA > > > > Subject:  Re: OpenVMS article - please explain last line!l > > > >i > > > > Paul Repacholi wrote:h	 > > > > > < > > > > > andrew harrison <andrew.nospam@uk.sun.com> writes:	 > > > > > I > > > > > > "Main, Kerry" wrote: As is making claims for existing systems  > which G > > > > > > are not supported by the data you youselves have published.h	 > > > > > & > > > > > OK Kerry, naughty boy! Smak.	 > > > > >eH > > > > > You should know that it is only aceptable to enrich the mental > > gardenseH > > > > > of you poor punters by spreading infomation based on NDA data.	 > > > > >e > > > >t1 > > > > So was Kerrys posting bullshit or are youh > > > > defending it.  > > > >  > > > > Regards  > > > > Andrew HarrisonE > > > > Enterprise IT Architect  > > >M > > > -- > > > Andrew Harrison  > > > Enterprise IT Architecto > >  > > -- > > Andrew Harrison4 > > Enterprise IT Architecte >t > -- > Andrew Harrison  > Enterprise IT Architect    -- Andrew Harrisonp Enterprise IT Architect    ------------------------------  + Date: Wed, 25 Apr 2001 15:30:56 +0000 (UTC) ' From: david20@alpha2.mdx.ac.uk (D.Webb) % Subject: Re: OpenVMS Hobbyist versionu+ Message-ID: <9c6qjg$seh$1@aquila.mdx.ac.uk>d  Z In article <3AE6916A.6557D7FD@vmsone.com>, John Wisniewski <wisniewski@vmsone.com> writes: >sms@antinode.org wrote: >d0 >> From: John Wisniewski <wisniewski@vmsone.com> >>' >> > VAX and Alpha 7.3   (Coming Soon!)0 >> >>    I'm ready. >>O >> > We have tried to put a good selection of layered products (C, Fortran, andV- >> > such - see the webpage-- on the HobbyistjP >> > distributions.   Perhaps when VMS goes over to DVD we can get all the SW on >> > one disk;-) >>K >>    Perhaps if someone develops a standard scheme whereby the kits can be J >> compressed, a good deal more could be packed in.  Am I the only one whoJ >> thinks that the capability to install directly from a compressed kit is >> overdue?r >>  M Why not also put all the hobbiest licensed products on an ftp/web server and   allow users to download them ?  1 They can't run them unless they have the license.s  
 David Webb VMS and Unix team leader CCSS Middlesex University      J >>    Survey question:  How many people would object to compressed layeredI >> product kits (using, say, ZIP), which would need to be expanded onto a @ >> local read-write disk before installation (using the current," >> handicapped PCSI or VMSINSTAL)? >>G >>    Just as an impractical example, the VAX VMS072.% save sets occupyrH >> about 203000 blocks.  After ZIP, about 103000.  I assume that typicalD >> product kits would see similar shrinkage.  The VAX TCPIP V5.0 kitH >> (including docs) drops from about 37500 blocks to about 12500 blocks.H >> With twice the effective space available, it should be easy enough toK >> cram in a few more things, yes?  A convenient procedure to expand onto adG >> disk or tape might make it a little more convenient.  (How many kitst! >> can't fit onto a TK50 or two?)  >>K >> ------------------------------------------------------------------------p >>F >>    Steven M. Schweda               (+1) 651-699-9818  (voice, home)F >>    382 South Warwick Street        (+1) 763-781-0308  (voice, work)J >>    Saint Paul  MN  55105-2547      (+1) 763-781-0309  (facsimile, work)< >>    sms@antinode.org                sms@provis.com  (work) > F >Then someone would complain that they don't have enough disk space;-) > K >we'll see if we can zip the layered products in the 7.3 Hobbyist cdroms... + >(and Unzip them when we build a DVD kit;-)h >c >John Wisniewski >wisniewski@vmsone.com >n   ------------------------------  % Date: Fri, 27 Apr 2001 09:55:30 -0500t" From: Earl Lakia <lakia@ipact.com>) Subject: Re: PCs Need to Query RMS files.-) Message-ID: <3AE98862.3762DB2B@ipact.com>1  + Experience with the following two products:u  
 Product 1-D We also wrote an ODBC ourselves using  SYWARE's tool kit (Dr. DeeB).D It severed process variables from Process Data (lots of process dataI saved to Block I/O files, global sections, and  keyed files).   A typical  query we served were:w  * Select Flow,Pressure,Temperature from xxxx'   where Time > yesterday and time<today   7 We wrote our own because in reality the process historyP6 are in a single table that has over 7000 columns.   In6 effect, the xxxx became views against the 7000 columns: for particular tags the user wanted.  The kit was designed4 for UNIX heads, but we are able to migrate without a lot of trouble..  	 Product 2r7 Have also used the EasySoft product and it worked fine.c5 They provide some tools to convert and define the RMSi record layout.    C In both situations, our PC clients were Microsoft Access and Excel.a   -earl  -------'
 Earl Lakia
 IPACT Inc.     Lock Horsburgh wrote:y  D > I need to provide access from PCs to the data in a big application" > running on a VAX, under VMS 7.1. >h> > This is mainly a management information and ad hoc reportingD > issue at this time, though there is a desire to Windowise the userH > interface sometime, if only to stop people asking why they have to useJ > this ancient stuff. ("Because it works very well" doesn't seem to be the > answer they want). >eG > Some of the data is in Oracle, so it should be fairly straightforwardlB > to use ODBC drivers that will allow pc applications to query the > database.  >eJ > However, older modules of this application use RMS files. There are 200+J > RMS files, millions of records, and I don't know of any simple tool that& > will let PC users query these files. > B > One option would be to load all the data from the RMS files intoG > Oracle every night, but that is a lot of processing and a lot of disk  > space. >gE > There was a proposal to copy data into a PC database - like Access,rC > Paradox, or dBase - and let users query it there. This would seem16 > feasible for one or two smaller subsystems at least.D > A lot of the interesting stuff is in floating point formats, so it9 > would take more than just a straight copy of the files.I > D > Obviously I'm not the first to have this problem, so I hoped maybe/ > somebody could suggest a better way to do it.  > A > Does anybody even have an opinion about whether one PC database_I > is a better target than another, due to import conversions or whatever?h >oC > I'm not very familiar with what is available on VAX/VMS, so there-A > may well be something blindingly obvious to you that I have noti
 > thought of.p >u@ > The budget is as near zero as makes no difference, by the way,H > so cheap solutions preferred, but all suggestions gratefully received. >w > Lock.    ------------------------------  % Date: Fri, 27 Apr 2001 17:14:42 +0100PC From: "Lock Horsburgh" <lock@lhorsburgh.NO.RUBBISH.freeserve.co.uk>o) Subject: Re: PCs Need to Query RMS files. / Message-ID: <9cc60i$vu1$1@newsg4.svr.pol.co.uk>   , Thanks for all the replies, on ng and email.   So far we have:r  E Easysoft (excellent example of a website that actually tells you what 5 products they have, what they do, and what they cost)i  L SolutionsIQ CONNX - gives you a demo, says it can do Oracle, RDB and RMS allH in the same Query, but nothing about price. Site is not clear if it is a. reporting tool or just an ODBC driver or what.  J WRQ Verastream - visited, asked for info. I have used WRQ Reflection for > 15 years, and like it a lot.  A Synergex: web site is the worst I have ever seen for impenetrable.J gobbledygook. Not sure I want to talk to these people, I have no idea whatK they do for a living. Search on RMS + ODBC produces over 200 hits, but theneL you can't look at them without a password. Good idea, somebody might want to& mess up their day by buying something.  L Hummingbird: gobbledygook not as bad, but couldn't find any sensible product info.i  1 Do it Yourself - too much to do, too little time.r    E I'm now looking for prices and demos, but still grateful for any more K comments (especially "that works great" or "that's a dog") and suggestions.g     Lock.n   ------------------------------  % Date: Fri, 27 Apr 2001 09:35:47 +0200o= From: Arne =?iso-8859-1?Q?Vajh=F8j?= <arne.vajhoej@gtech.com>e; Subject: Re: R.I.P. FTP.WKU.EDU.  Long live FTP.PROCESS.COMh( Message-ID: <3AE92153.BEF8C6B@gtech.com>   Roy Omond wrote: > Hunter Goatley wrote:a( > > I no longer have anything to do with > > MadGoat Software.o >  > *Yikes* !  The end of an era.S
 > How sad.  9 I think it is quite some time since Matt went commercial.    Arne   ------------------------------   Date: 26 Apr 2001 22:02:04 CDT= From: wayne@tachysoft.xxx.065234.killspam.015d (Wayne Sewell)n5 Subject: Re: Random access to 2 bytes in a large files. Message-ID: <yErvsROSZTh0@tachxxsoftxxconsult>  \ In article <3AE844AF.9F374B1D@videotron.ca>, JF Mezei <jfmezei.spamnot@videotron.ca> writes:* > Thanks for the various responses so far. > O > I checked the sectiosn documentation again, and it is true that for a privateMK > section file, it doesn't consume GBLPAGES (sysgen) and PGFLQUO (process).e > O > However, if I map a 100 meg file to my process' virtual memory,  Do I need toi< > ensure that the system has enough virtual address space ?    yesr   >(remember, this is aw< > small memory system for which 100 meg is a huge quantity).  H this determines how much *physical* memory you can access, not *virtual*   >  > K > Also, wouldn't the process' Working set limits have to be large enough ton! > accomodate the whole 100 megs ?t >   N No.  The size of the working set and of the process virtual address space haveM little to do with each other.  The working set consists of the pages that arecM currently in physical memory.  Pages in the virtual address space don't exist L until you access that address and bring them into the working set.  Only theL pages you actually access reside in physical memory (for the purpose of thisM discussion we will ignore page fault clusters).  Once you fill up the workingnL set, the pager starts throwing old pages out of physical memory to make roomG for the new ones.  How much paging occurs depends on how many pages you 1 actually access and how randomly you access them.K       -- eO =============================================================================== M Wayne Sewell, Tachyon Software Consulting  (281)812-0738  wayne@tachysoft.xxxa: http://www.tachysoft.xxx/www/tachyon.html and wayne.html  K change .xxx to .com in addresses above, assuming you are not a spambot  :-) O =============================================================================== B Jed Clampett, checking into hotel: "This place got a cement pond?"+ Ellie May: "And do yuh let critters in it?"n   ------------------------------  % Date: Fri, 27 Apr 2001 10:37:26 -0400g- From: JF Mezei <jfmezei.spamnot@videotron.ca>y5 Subject: Re: Random access to 2 bytes in a large fileo, Message-ID: <3AE98424.C5418EF8@videotron.ca>   Wayne Sewell wrote:cP > No.  The size of the working set and of the process virtual address space have > little to do with each other.u  M Correct, I was thinking of the "virtual pages" in the SHOW PROC/CONT display.   J On a MAC, this is quite simple, you have physical memory size, and virtualK memory size. But I never figured out what the virtual memory size parameter.K was for VMS (for the system as a whole and for an individual process). (For 	 VAX, 7.2),    M Also, in the case of section files, If I request a byte that is not currentlyrM in physical memory, will the system fetch a single 512 byte block to map that < request byte ? Or would it page multiple blocks are a time ?  J For my first application, there would be locality of the data with perhapsK about 2 or 3 non consecvutive blocks being accessed many times. (increasinge6 longitude, and slightly varying latitude by +- 1 row).   ------------------------------  % Date: Fri, 27 Apr 2001 16:44:56 +0100e4 From: John Laird <john@laird-towers.freeserve.co.uk>5 Subject: Re: Random access to 2 bytes in a large fileh8 Message-ID: <qfqietg3tpg8791dsjpung8a6d6t92okhe@4ax.com>  , On Wed, 25 Apr 2001 02:51:04 -0400, JF Mezei% <jfmezei.spamnot@videotron.ca> wrote:o  : >OK, let me start by saying that I am very memory limited. >hO >I have two flat files of about 50 megs each which provide elevation data for acK >continent, each sample is a short (2 bytes) that contains the elevation inlO >metres for that location. The data is arranged in rows. The file has 4800 rowsy >each with 4800 columns. >uN >My wish is to be able to access the data randomly (what is the altitude for a >particular point ?).  > N >If I use a process section to map the file to memory,  is it correct to stateK >that I would have to increase not only that process's pgflquo but also thea/ >system GBLPAGFIL parameters by 100 megabytes ?tM >Would that consume too many resources to map 2 50meg files to memory on a 16 
 >meg system ?s >rM >How would performance of a section file compare to a home grown routine that N >would read the appropriate block in the file if the requested point is not in >a cache of a few blocks ?  G See the other responses for discussion of efficiency, ease of use, etc.sB Using the paging mechanism to automatically perform the random I/OG simply by referencing HEIGHT(x,y) is a no-brainer, once you have mappedsH the section.  However, you have traded away effective control over whichB areas of the file are retained for locality of reference, since (ID think) there is no concept of "age" when pages are selected to go toE make room for others.  The modified list mechanism won't help because G your data is presumably readonly.  To avoid your code disappearing fromhF memory, you may want to investigate $LCKPAG or $LKWSET (which is which has paged from my memory...).t  ? Is your requirement to truly randomly access data points, on anoD as-needed basis, in which case you can presumably afford the 10-20msD required for each likely I/O, or are you more likely to be accessingF continuously within a locality, in which case the I/O is going to biteG really badly because of the ordering of your data ?  In other words, if F you need to look up 1000 data points at a time, can you afford to wait! 10-20 seconds in the worst case ?   H I/O is so slow you probably ought to consider a user routine anyway, and? make your HEIGHT reference a function call rather than an array D reference.  The call overhead will be very small, of the order of 10F instructions or so, and you now have complete control of what is goingD on.  If you want to stick with the (user) I/O, you can now implement= your own FIFO lists if that helps, or consider read-ahead (orsB read-sideways) .  You can test buffering strategies against likelyD access patterns.  Is your row, column ordering the best way - if youE need locality of data, this might not be optimal.  What about storingmH small rectangles instead ?  This will cut down on overall storage at the? expense of some start-x overhead - I've not seen many perfectlyoA rectangular continents, and why waste space on sea level ?  Is it F possible you could extend this further and store a reference elevationF per rectangle, and cut the individual point data down to bytes, if theH delta values can be restricted to 255 metres ?  (You can infer from thisG that we don't have much mountainous terrain in the UK...)  Much of your'* data could halve in size if this was done.  H In a previous existence, I used to work with multi-gigabyte seismic dataH sets and multi-megabyte mapping surfaces, on 3-vup Vaxstations with slow disks.  I feel your pain :-)     	John, --  
 John Laird   ------------------------------    Date: 27 Apr 2001 17:43:59 +0200G From: Jan Vorbrueggen <jan@mailhost.neuroinformatik.ruhr-uni-bochum.de>n5 Subject: Re: Random access to 2 bytes in a large file H Message-ID: <y4g0eus5z4.fsf@mailhost.neuroinformatik.ruhr-uni-bochum.de>  / JF Mezei <jfmezei.spamnot@videotron.ca> writes:   O > Also, in the case of section files, If I request a byte that is not currently O > in physical memory, will the system fetch a single 512 byte block to map thath> > request byte ? Or would it page multiple blocks are a time ?  N I think you can set the page fault cluster size in the map request. Otherwise,D SYSGEN parameter PFCDEFAULT applies, which usually is 64 (pagelets).   	Jan   ------------------------------  , Date: Fri, 27 Apr 2001 16:02:57 +0200 (CEST): From: "Gotfryd Smolik, VMS lists" <gotfryd@stanpol.com.pl>F Subject: Re: Reply before or after? (was: seeing escapes is believing)J Message-ID: <Pine.LNX.4.21.0104271542430.11626-100000@irys.stanpol.com.pl>  * On Mon, 23 Apr 2001, Didier Morandi wrote:  ? +Well, Gotfryd, you remark raises actually another question :-)K    -:)  < +Should answers be posted before or after the original text? +sL +My preference is _before_ because when I ask for the next unread, I can see& +immediately the answer to my question  A  Yes. But the assumption that *anyone reader* reads all and wholeo# threads is not neccessarily true :) >  Especially that is important when the answer can be separated% to separate parts of original text...   ( +(question text that I do know by heart)  A  But think of one, who reads the post/mail as a separate message,h< *or* reads are sorted not directly after the one you replies> (supose more answers, where can be related to differrent parts  of original post)  1 +without having to scroll until I find the reply.r  >  Och, sure. Probably more (that "readable position") important: is the "cut unnecessary part of original" point as you say	 at end :)e   [...]AO +If there is a better reason or an habit or any other etiquette-based reason ton7 +post replies *after* the original text, I will change.v  $ AFAIR that was a standard in past :)F  Really, I must agree: *some* points changes thru time. The connectionB in most cases are faster (and anyone who gets GBytes of movies canA not think, that b.ex. the unnecessary citate may be important fornA someone else who uses dial-up), the interface changes to graphics>A (even, if I am SURE that I am not the only dinosaur who uses pinei? in terminal mode :] and here is problem "how to display a splitA in news threads" etc.>@  But, IMHO, the  answer after original in *most* cases (suposing: the "cut unneccessary") is hm... more elegant, at least :)   +Advices, everybody?     I am only one with opinion ? ;)   +D.lE +(as far as cutting unnecessary text, I always do so since more than i3 +10 years, which is not the case for everyone here.v    Really, agree !    Best regards - Gotfrydp   -- oE =====================================================================iF $ ON F$ERROR("LANGUAGE","ENGLISH","IN_MESSAGE").GT.F$ERROR("NORMAL") - 		THEN EXCUSE/OBJECT=MEr. $!                        GS@stanpol.zabrze.plE =====================================================================    ------------------------------  % Date: Fri, 27 Apr 2001 09:31:27 +0200>= From: Arne =?iso-8859-1?Q?Vajh=F8j?= <arne.vajhoej@gtech.com>k Subject: Re: SAMBA) Message-ID: <3AE9204F.33DAE83E@gtech.com>1   "Jackson, Louise" wrote:A > What is SAMBA, and how dose it differ from PathWork's software?   A The basic difference is that Samba is freeware while PathWorks is  a commercial product.a  G There are also differences in functionality and performance, but you ifo@ that is what you are interested in, then I suggest you ask a bit more specific.   Arne   ------------------------------  % Date: Fri, 27 Apr 2001 11:17:44 +0200)( From: "Bruin, J.M. de" <Bruin@WT.TNO.NL> Subject: RE: SAMBAC Message-ID: <2ABC8BB85FE5D411AC100008C7F37BC227D94B@wt15.wt.tno.nl>(  F You can always take a look at www.samba.org to see what Samba means a=
 nd can do.   Mark   > -----Original Message-----5 > From: Arne Vajh=F8j [mailto:arne.vajhoej@gtech.com]z$ > Sent: Friday, April 27, 2001 09:31 > To: Info-VAX@Mvb.Saic.Com  > Subject: Re: SAMBA >=20 >=20 > "Jackson, Louise" wrote:C > > What is SAMBA, and how dose it differ from PathWork's software?2 >=20C > The basic difference is that Samba is freeware while PathWorks is: > a commercial product.l >=20A > There are also differences in functionality and performance,=20o > but you ifB > that is what you are interested in, then I suggest you ask a bit > more specific. >=20 > Arne >=20   ------------------------------  % Date: Fri, 27 Apr 2001 11:39:09 +0100O/ From: Nigel Arnot <sysmgr@maxwell.ph.kcl.ac.uk>y Subject: re: SAMBA6 Message-ID: <009FB297.2353E401.2@maxwell.ph.kcl.ac.uk>   >  >  > What is SAMBA, d  D An open-source inplementation of a CIFS (aka SMB aka Microsoft) file server. See www.samba.org.  2 > and how dose it differ from PathWork's software?  D Both are solutions to the same basic problem: allowing Microsoft PCs* access to files on a non-Microsoft server.  4 SAMBA is open source and the license costs nothing. L Pathworks (I think now known as Compaq Advanced Server) is closed and costs.  ? In practice on VMS, pathworks is  far more  integrated into thetG operating system, and is cluster-aware. It's also backed by the Compaq  D service organisation. These features make it suitable for Enterprise usage.  G Samba/VMS probably offers lower performance (I'm guessing - anyone have J benchmarks?) and if you find a problem, you may have to debug it yourself.F That said, it works pretty well, and is ideal if your requirements are? for something cheap, and if your usage isn't mission-critical. y  > Samba is extensively used on Linux and other Unixes. And thereG is some serious support behind samba.org in that HP and IBM and severalnK NAS vendors use SAMBA as the basis of solutions that they sell. If you wantwG a server for MS systems but don't want to pay oodles to MS, I recommendrM it wholeheartedly - but unless the requirement is to share files between PCs nI and VMS, running it on a Unix box is more "mainstream" and finds greater   community support.   	Yours,p
 		Nigel Arnoto- 		NRA@MAXWELL.PH.KCL.AC.UK                   r  7 		"In the beginning there was nothing, which exploded."    ------------------------------    Date: 27 Apr 2001 09:16:30 -05003 From: malmberg@encompasserve.org (John E. Malmberg)e Subject: Re: SAMBA3 Message-ID: <6nVP4usxz$Mn@eisner.encompasserve.org>e  L > "Jackson, Louise" <louise.jackson@lmco.company> 04/26/2001 11:29:25 AM >>> >tA > What is SAMBA, and how dose it differ from PathWork's software?I  5 For information on the Compaq Pathworks software see:   3 http://www.openvms.compaq.com/pathworks32/spds.html   A In quick summary, there are two products in the Pathworks family,   @ The PC client software that supplies LAT, DECNET, eXcursion, and? PowerTerm VT Emulation.  Use of it requires a software license.a< The license to use this software is included in some but not  all Alphaserver OpenVMS systems.  = The X-11 emulator, from my personal experience, is one of thef; best available commercially, and at the last time I checked > (a few years back) it was the lowest cost one on the market by a considerable ammount.   A Since AFAIK, all of the X-11 and terminal emulation vendors offeru? free time-bombed demo's, you can take the time to form your ownR opinion.    H The other is the server software.  Pathworks now known as Advance Server provides the following:o  8 File and print services (requires client access license)G External Authentication (I do not think requires client access license)tE OpenVMS DCL administration for an Microsoft Windows NT server domain.l@ ( I do not think this function requires a client access license)< NT management tools can also manage most of Advanced Server.  B Support and handholding is available from Compaq for this product.  D What Compaq Pathworks or Advanced Server does not supply is a method9 for OpenVMS to access data on a Microsoft Windows system.o     What is SAMBA?  @ SAMBA is an open source client and server software that providesI the LANMAN protocl on TCP/IP.  SAMBA only provides the TCP/IP protocol inh3 contrast to Pathworks providing DECNET and NETBeui.   : Information on SAMBA is available at HTTP://www.samba.org/  B The latest version of SAMBA known to me to be ported to OpenVMS is@ 2.0.6. and is expected to be on the OpenVMS FREEWARE 5.0. CDROM.  0 Or see http://encompasserve.org/~malmberg/samba/  > I am not aware of any commercial support for SAMBA on OpenVMS.  C The SAMBA client portions can work on the same system as is hosting  Pathworks or Advanced Server.   @ SAMBA versions older than 2.0.7 can have problems with Microsoft
 Windows 2000.:  @ For various reasons, the performance of SAMBA on OpenVMS will be: slower than Advanced Server, once both are properly tuned.  ! I am not aware of any benchmarks.n  F At this time, there does not seem to be anyone admitting that they are9 actively working on a newer release of SAMBA for OpenVMS.n  ' But with opensource, you never know :-)-   -John- wb8tyw@qsl.network Personal Opinion Only    ------------------------------  # Date: Fri, 27 Apr 2001 12:55:07 GMTn$ From: Lee Y T Mah <lytmah@cha.ab.ca> Subject: Samba vs NFSZ) Message-ID: <3AE96C2C.9ECB9801@cha.ab.ca>8  G I would like to implement TCPWARE's NFS module to transfer data between:G several Windows 98 machines and a VMS/Alpha production cluster.  It hasnA been suggested that SAMBA (E.g., V2.0.3 or higher) may be able to G perform the same function.  Has anyone had experience doing this?  What ! are the con's and pro's?  Thanks.    -- Lee    ------------------------------  % Date: Fri, 27 Apr 2001 10:04:47 -0300 ) From: fabio_compaq@ep-bc.petrobras.com.brs Subject: Re: Samba vs NFSeL Message-ID: <OFFD853003.21DF0702-ON03256A3B.00479A1D@ep-bc.petrobras.com.br>  J Even SAMBA is free, and a lot of people tell me it is a good product, I am aware to put "open source code"aI in my productions systems. You can call me "delayed", but I dont agree tot" put "Open Software" in my systems.H It is a question of support, availability and security. You dont imagine! the "back holes" an open softwarer	 can have.y  " I am turning to the mainframe ....   Regards    FC        5 Lee Y T Mah <lytmah@cha.ab.ca> em 27/04/2001 09:55:07o  0 Favor responder a Lee Y T Mah <lytmah@cha.ab.ca>             Info-VAX@Mvb.Saic.Como       Assunto: Samba vs NFSl    G I would like to implement TCPWARE's NFS module to transfer data betweensG several Windows 98 machines and a VMS/Alpha production cluster.  It has A been suggested that SAMBA (E.g., V2.0.3 or higher) may be able tolG perform the same function.  Has anyone had experience doing this?  What=! are the con's and pro's?  Thanks.=   -- Lee=   ------------------------------  % Date: Fri, 27 Apr 2001 16:02:58 +0200 + From: "Peter Kille" <peter.kille@volvo.com>x Subject: Re: Samba vs NFS., Message-ID: <9cbu70$f80$1@vg170.it.volvo.se>  I I dont agree with you, i have been using "open source code" in production= for some time now=L and have not had any problems worth mentioning. I think you have to consider
 what softwarelI you are using for what purpose and so on. And of course do proper testing- and have+ proper routines before going to production.z  K I have 100+ Linux clients NFS mounting an Alphaserver in one of our plants,l i have xx numberF of Window clients accessing a Linux fileserver through Samba, all user administration is made through the NT-domain.  G Note, im running Samba on Linux, not VMS, the 2.0.3 version on VMS is as beta,nH (see http://www.ifn.ing.tu-bs.de/ifn/sonst/samba-vms.html) there i agreeJ with you, "do not use for production". I have downloaded and tested it but% it seems to me that all functions are  not implemented/debugged yet.    BR Peter6 <fabio_compaq@ep-bc.petrobras.com.br> wrote in messageF news:OFFD853003.21DF0702-ON03256A3B.00479A1D@ep-bc.petrobras.com.br...L > Even SAMBA is free, and a lot of people tell me it is a good product, I am! > aware to put "open source code" K > in my productions systems. You can call me "delayed", but I dont agree toi$ > put "Open Software" in my systems.J > It is a question of support, availability and security. You dont imagine# > the "back holes" an open softwareh > can have.a >u$ > I am turning to the mainframe .... >-	 > Regards  >e > FC >  >  >1 >:7 > Lee Y T Mah <lytmah@cha.ab.ca> em 27/04/2001 09:55:07v >o2 > Favor responder a Lee Y T Mah <lytmah@cha.ab.ca> >f >h >e >       Info-VAX@Mvb.Saic.Comn >6 >9 >h > Assunto: Samba vs NFSA >2 > I > I would like to implement TCPWARE's NFS module to transfer data betweentI > several Windows 98 machines and a VMS/Alpha production cluster.  It haslC > been suggested that SAMBA (E.g., V2.0.3 or higher) may be able toeI > perform the same function.  Has anyone had experience doing this?  Whate# > are the con's and pro's?  Thanks.n >  > -- > Leef >  >  >0 >w >  >c >i   ------------------------------  # Date: Fri, 27 Apr 2001 14:32:23 GMT(+ From: psherwin@strayduck.com (Paul Sherwin)  Subject: Re: Samba vs NFSs2 Message-ID: <3ae98114.14036605@news.onetel.net.uk>  @ On Fri, 27 Apr 2001 12:55:07 GMT, Lee Y T Mah <lytmah@cha.ab.ca> wrote:  H >I would like to implement TCPWARE's NFS module to transfer data betweenH >several Windows 98 machines and a VMS/Alpha production cluster.  It hasB >been suggested that SAMBA (E.g., V2.0.3 or higher) may be able toH >perform the same function.  Has anyone had experience doing this?  What" >are the con's and pro's?  Thanks. >oE Samba works well, but I've never used it seriously under VMS. TCPwareyD NFS is better if you want to store VMS files on non-VMS systems, butD if you just want to store non-VMS files on VMS systems Samba will beA fine. It can be a little tricky to set up and is a port of a Unix C program, so if you have no Unix experience you may struggle for thel4 first few hours. It all depends what you want to do.  E I find SMB networking over TCP/IP (which is what Samba implements) isd much faster than NFS.n  F If you are nervous about putting Samba on your production systems, youE could build a little Linux gateway box running Samba and DECnet - anyTE old Pentium box will do. DECnet support is included in the V2.4 Linux  kernels.  	 HTH, Pauls  @ Paul Sherwin Consulting     22 Monmouth Road, Oxford OX1 4TD, UK= Phone  +44 (0)1865 721438   http://www.psherwin.strayduck.comu9 Mobile +44 (0)7931 578334   mailto:psherwin@strayduck.comy   ------------------------------  % Date: Fri, 27 Apr 2001 14:59:11 +0000i  From: Steve.Spires@yellgroup.com( Subject: Set RADIX and OCTAL arithmetic./ Message-ID: <00256A3B.005256F6.00@quegw01.btyp>    cc:w bcc:L Contact:   Tel: 3063  -  IS - Infrastructure, 1st Floor, Bridge Street Plaza   Set RADIX and OCTAL arithmetic.f    3 Did I imagine there used to be a SET RADIX command?o  P I need to do some arithmetic operations in OCTAL, and thought SET RADIX would be the easiest way.  P I've been trying to do this for a while, and eventually cooked up some very crapO DCL to do it, but it ended up relatively complicated [having to check when each)P part of the number reached 7 etc] so is there an easy way? Is there a simple wayM to perhaps use F$FAO in some way to achieve this? I have been looking at this ( for so long I have become [DCL] blind...   Steve Spires     [Information] -- PostMaster:D This transmission is intended solely for the addressee(s) and may beL confidential. If you are not the named addressee, or if the message has beenP addressed to you in error, you must not read, disclose, reproduce, distribute or use this transmission.  L Delivery of this message to any person other than the named addressee is notH intended in any way to waive confidentiality.  If you have received thisF transmission in error please contact the sender or delete the message.  
 Thank you.   ------------------------------  # Date: Fri, 27 Apr 2001 14:13:44 GMTa2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman), Subject: Re: Set RADIX and OCTAL arithmetic.0 Message-ID: <s8fG6.293$5I.5673@news.cpqcorp.net>  R In article <00256A3B.005256F6.00@quegw01.btyp>, Steve.Spires@yellgroup.com writes:4 :Did I imagine there used to be a SET RADIX command?  @   In utilities such as the debugger, yes, but not in DCL itself.  I :I need to do some arithmetic operations in OCTAL, and thought SET RADIX e :would be the easiest way.  -   I presume you are aware of the %o notation.t  N  ---------------------------- #include <rtfaq.h> -----------------------------N       For additional, please see the OpenVMS FAQ -- www.openvms.compaq.com    N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------  % Date: Fri, 27 Apr 2001 15:25:49 +00002  From: Steve.Spires@yellgroup.com, Subject: Re: Set RADIX and OCTAL arithmetic./ Message-ID: <00256A3B.0054C6A8.00@quegw01.btyp>c  L Contact:   Tel: 3063  -  IS - Infrastructure, 1st Floor, Bridge Street Plaza    5 >:Did I imagine there used to be a SET RADIX command?:  A >  In utilities such as the debugger, yes, but not in DCL itself.   N Ah! It's amazing how you misremember when you're struggling to find an answer!  I >:I need to do some arithmetic operations in OCTAL, and thought SET RADIXd >:would be the easiest way.i  . >  I presume you are aware of the %o notation.   Yes, I guess I'm just weary...  M I thought to myself "I know, I'll make use of the invaluable resource that issN the second edition of the Writing Real Programs in DCL tome" only to find that it specifically excludes OCTAL.,   ;^Da   Steve Spires   ------------------------------  % Date: Fri, 27 Apr 2001 08:28:08 -0300y) From: fabio_compaq@ep-bc.petrobras.com.br  Subject: Sync Comm - PCI Card L Message-ID: <OF04362B1C.1E69CBB6-ON03256A3B.003EC700@ep-bc.petrobras.com.br>  B I remember DEC used to have a Synchronous Communications PCI card.< At www.dnpg.com there is no information about.t his product.  0 Do you know where can I get informations about ?   Regardse   FC   ------------------------------   Date: 27 Apr 2001 11:56:10 GMT0 From: "Alan Boyles" <alan.boyles@mindspring.com>! Subject: Re: Sync Comm - PCI Cardo0 Message-ID: <01c0cf10$e004a100$529356d1@boylesa>  I There are a couple of Sync boards available depending on the machine typexF and the number of ports,  the PBXDP-AA or AB.  These are Emulex boardsF which I believe will EOL this year and then be replaced by a couple of others.    Alan  4 fabio_compaq@ep-bc.petrobras.com.br wrote in articleC <OF04362B1C.1E69CBB6-ON03256A3B.003EC700@ep-bc.petrobras.com.br>...hD > I remember DEC used to have a Synchronous Communications PCI card.> > At www.dnpg.com there is no information about.t his product. > 2 > Do you know where can I get informations about ? > 	 > Regardse >  > FC >  >    ------------------------------  % Date: Fri, 27 Apr 2001 09:13:32 -0300e) From: fabio_compaq@ep-bc.petrobras.com.br ! Subject: Re: Sync Comm - PCI Card=L Message-ID: <OFAC3B441B.C60A7362-ON03256A3B.00422EF7@ep-bc.petrobras.com.br>  J I discovered that the DS10L doesnt support theses boards for OpenVMS. Just the DS10 standard . . .s  X http://www.compaq.com/alphaserver/options/asds10l/asds10l_19_48_options.html#OptListHead   Regards-     FC        A "Alan Boyles" <alan.boyles@mindspring.com> em 27/04/2001 08:56:10-  < Favor responder a "Alan Boyles" <alan.boyles@mindspring.com>             Info-VAX@Mvb.Saic.Comc      ! Assunto: Re: Sync Comm - PCI Cardi    I There are a couple of Sync boards available depending on the machine typenF and the number of ports,  the PBXDP-AA or AB.  These are Emulex boardsF which I believe will EOL this year and then be replaced by a couple of others.t   Alan  4 fabio_compaq@ep-bc.petrobras.com.br wrote in articleC <OF04362B1C.1E69CBB6-ON03256A3B.003EC700@ep-bc.petrobras.com.br>...2D > I remember DEC used to have a Synchronous Communications PCI card.> > At www.dnpg.com there is no information about.t his product. >D2 > Do you know where can I get informations about ? >w	 > Regards  >s > FC >a >    ------------------------------  # Date: Fri, 27 Apr 2001 14:31:12 GMTe& From: "john nixon" <jnixon@cfl.rr.com>! Subject: Re: Sync Comm - PCI Card0? Message-ID: <QofG6.105786$fs3.17399446@typhoon.tampabay.rr.com>=  J I am glad you pointed that out.  I will be putting together a proposal forJ an Alpha solution to replace our old MicroVAXes.  We have four MicroVAXes,K each with six or 7  Q-Bus synch devices.  Since the softwaer for Alpha onlyeH supports one synch PCIcard each, and each card only has up to 4 ports, IF will need 7 DS10s to replace 4 MVAX 3600s.  (The hardware supports twoL cards, but the software only supports one).  I was planning on using DS10Ls,L but now I see that I will need to go with the somewhat more expensive DS10s.E This is going to be expensive considering all the licenses (including G cluster) etc... just to support  Bisynch.  (Cheaper than maintenance onB( microvaxes, but expensive all the same).  I I heard that they will have an 8 or 16 port card, but it will only be fortL Unix.  I  would    * * *  R E A L L Y  * * *  like to see this larger optionE qualified on VMS.  I (and Compaq)  may lose this simple VAX-Alpha VMSiF migration to a 3rd party solution if I have to put in 6 or 7 alphas to replace 4 MVAXes.a  I Does anyone know, or can anyone find out, if there are plans to ever qual " the 8 or 16 port (PBXDD-*) on VMS?    6 <fabio_compaq@ep-bc.petrobras.com.br> wrote in messageF news:OFAC3B441B.C60A7362-ON03256A3B.00422EF7@ep-bc.petrobras.com.br...L > I discovered that the DS10L doesnt support theses boards for OpenVMS. Just > the DS10 standard . . .t >u >lL http://www.compaq.com/alphaserver/options/asds10l/asds10l_19_48_options.html #OptListHead >,	 > Regardsr >a >i > FC >i >c >i >.C > "Alan Boyles" <alan.boyles@mindspring.com> em 27/04/2001 08:56:10  >s> > Favor responder a "Alan Boyles" <alan.boyles@mindspring.com> >n >A >e >       Info-VAX@Mvb.Saic.Com  >u >n >h# > Assunto: Re: Sync Comm - PCI Cardf >M >yK > There are a couple of Sync boards available depending on the machine typeuH > and the number of ports,  the PBXDP-AA or AB.  These are Emulex boardsH > which I believe will EOL this year and then be replaced by a couple of	 > others.u >p > Alan >r6 > fabio_compaq@ep-bc.petrobras.com.br wrote in articleE > <OF04362B1C.1E69CBB6-ON03256A3B.003EC700@ep-bc.petrobras.com.br>...-F > > I remember DEC used to have a Synchronous Communications PCI card.@ > > At www.dnpg.com there is no information about.t his product. > >64 > > Do you know where can I get informations about ? > >e > > Regardsk > >m > > FC > >4 > >0 >1 >9 >  >> >  >  >h   ------------------------------  # Date: Fri, 27 Apr 2001 14:55:00 GMTo From: Dirk Munk <munk@home.nl>! Subject: Re: Sync Comm - PCI Cards' Message-ID: <3AE98844.9AD4760A@home.nl>q   Alan Boyles wrote:  K > There are a couple of Sync boards available depending on the machine typesH > and the number of ports,  the PBXDP-AA or AB.  These are Emulex boardsH > which I believe will EOL this year and then be replaced by a couple of	 > others.l  L These are Emulex XP boards. The AA model has 2 ports on the card itself, butH can no longer be ordered AFAIK (status: maintenance). The AB model has 4M ports and the AC model has 8 ports, but this is done with a kind of break-outrJ box that is connected to the actual boards by means of a cable & big plug.J I would like to know something more about the new sync cards, is there any information available ?s   Regards,   Dirk       >r >n > Alan > 6 > fabio_compaq@ep-bc.petrobras.com.br wrote in articleE > <OF04362B1C.1E69CBB6-ON03256A3B.003EC700@ep-bc.petrobras.com.br>...IF > > I remember DEC used to have a Synchronous Communications PCI card.@ > > At www.dnpg.com there is no information about.t his product. > >u4 > > Do you know where can I get informations about ? > >l > > Regardsn > >o > > FC > >d > >e   ------------------------------  # Date: Fri, 27 Apr 2001 14:57:41 GMT  From: Dirk Munk <munk@home.nl>! Subject: Re: Sync Comm - PCI Cardh' Message-ID: <3AE988E5.ABB19B7C@home.nl>s  X I see no reason why they should not work. Maybe no one tested them with VMS in a DS10L ?    * fabio_compaq@ep-bc.petrobras.com.br wrote:  L > I discovered that the DS10L doesnt support theses boards for OpenVMS. Just > the DS10 standard . . .m >oZ > http://www.compaq.com/alphaserver/options/asds10l/asds10l_19_48_options.html#OptListHead >b	 > Regards  >n > FC >.C > "Alan Boyles" <alan.boyles@mindspring.com> em 27/04/2001 08:56:10s >h> > Favor responder a "Alan Boyles" <alan.boyles@mindspring.com> >o >       Info-VAX@Mvb.Saic.Comp >M# > Assunto: Re: Sync Comm - PCI Cardo >oK > There are a couple of Sync boards available depending on the machine typesH > and the number of ports,  the PBXDP-AA or AB.  These are Emulex boardsH > which I believe will EOL this year and then be replaced by a couple of	 > others.d >r > Alan >o6 > fabio_compaq@ep-bc.petrobras.com.br wrote in articleE > <OF04362B1C.1E69CBB6-ON03256A3B.003EC700@ep-bc.petrobras.com.br>...hF > > I remember DEC used to have a Synchronous Communications PCI card.@ > > At www.dnpg.com there is no information about.t his product. > >e4 > > Do you know where can I get informations about ? > >a > > Regardsp > >1 > > FC > >n > >m   ------------------------------  # Date: Fri, 27 Apr 2001 15:21:43 GMTE From: Dirk Munk <munk@home.nl>! Subject: Re: Sync Comm - PCI CardF' Message-ID: <3AE98E86.5E75B36F@home.nl>a  K I think you're mistaken. The DS10 supports at least 2 PBXDP-AB cards with 4t< ports each with VMS according to the supported options list.V http://www.compaq.com/alphaserver/options/asds10/asds10_19_48_options.html#OptListHead  " So that would be 8 ports per DS10.O I can't imagine why it is stated that 2 cards are supported in VMS but only onep can be used ?1?s  O But there is also the AC model with 8 ports. It is not mentioned in any options0O list, but is it supported. So that would  make 16 ports. And I'm quite sure you P could even put 3 of those boards in a DS10. So with a bit of luck you can get 24O sync ports on one DS10 ... Ask Compaq to make a statement, but as far as I know  it can be done.    john nixon wrote:a  L > I am glad you pointed that out.  I will be putting together a proposal forL > an Alpha solution to replace our old MicroVAXes.  We have four MicroVAXes,M > each with six or 7  Q-Bus synch devices.  Since the softwaer for Alpha only J > supports one synch PCIcard each, and each card only has up to 4 ports, IH > will need 7 DS10s to replace 4 MVAX 3600s.  (The hardware supports twoN > cards, but the software only supports one).  I was planning on using DS10Ls,N > but now I see that I will need to go with the somewhat more expensive DS10s.G > This is going to be expensive considering all the licenses (including=I > cluster) etc... just to support  Bisynch.  (Cheaper than maintenance on-* > microvaxes, but expensive all the same). > K > I heard that they will have an 8 or 16 port card, but it will only be foreN > Unix.  I  would    * * *  R E A L L Y  * * *  like to see this larger optionG > qualified on VMS.  I (and Compaq)  may lose this simple VAX-Alpha VMStH > migration to a 3rd party solution if I have to put in 6 or 7 alphas to > replace 4 MVAXes.  > K > Does anyone know, or can anyone find out, if there are plans to ever qualu$ > the 8 or 16 port (PBXDD-*) on VMS? >a8 > <fabio_compaq@ep-bc.petrobras.com.br> wrote in messageH > news:OFAC3B441B.C60A7362-ON03256A3B.00422EF7@ep-bc.petrobras.com.br...N > > I discovered that the DS10L doesnt support theses boards for OpenVMS. Just > > the DS10 standard . . .i > >n > >SN > http://www.compaq.com/alphaserver/options/asds10l/asds10l_19_48_options.html > #OptListHead > >t > > Regards  > >y > >d > > FC > >  > >c > >. > >vE > > "Alan Boyles" <alan.boyles@mindspring.com> em 27/04/2001 08:56:10  > >m@ > > Favor responder a "Alan Boyles" <alan.boyles@mindspring.com> > >. > >r > >n > >       Info-VAX@Mvb.Saic.Com2 > >2 > >. > >A% > > Assunto: Re: Sync Comm - PCI CardM > >  > > M > > There are a couple of Sync boards available depending on the machine typeuJ > > and the number of ports,  the PBXDP-AA or AB.  These are Emulex boardsJ > > which I believe will EOL this year and then be replaced by a couple of > > others.b > >  > > Alan > >o8 > > fabio_compaq@ep-bc.petrobras.com.br wrote in articleG > > <OF04362B1C.1E69CBB6-ON03256A3B.003EC700@ep-bc.petrobras.com.br>... H > > > I remember DEC used to have a Synchronous Communications PCI card.B > > > At www.dnpg.com there is no information about.t his product. > > > 6 > > > Do you know where can I get informations about ? > > >A
 > > > Regardso > > >e > > > FC > > >W > > >F > >u > >  > >f > >t > >w > >  > >e   ------------------------------  # Date: Fri, 27 Apr 2001 15:56:48 GMTe& From: "john nixon" <jnixon@cfl.rr.com>! Subject: Re: Sync Comm - PCI Card ? Message-ID: <4FgG6.105905$fs3.17459666@typhoon.tampabay.rr.com>   K I should have been a little more specific.  When I said the "software" only D supports one card, I should have said that the SPD for SNA 2780/3780K software  (which is not the only product I use, but is the primary one thatFJ uses the bisynch cards) specifies a maximum of 1 interface (2-8 ports) per system.7  . http://www.compaq.com/info/SP6457/SP6457PF.PDF  H The only PBXDP variants available are the -AA (2 port) and the -AB (fourK port).   There was at one time  -CA variant available, but it was listed asuF "ADP UNIQUE".  I don't know for sure, but I suspect that was a special& device delivered to a single customer.  L Even if something "might" work, or "should" work,  I cannot buy it unless itF is "guaranteed by Compaq" to work.  It is bad enough when Compaq drops. support for something we already have working.  @ It is my understanding that the new boards will be OEMed by DIGIK International  (www.digi.com) , and from what I have heard, there will onlygL be the 2 and 4 port variants (or at least only those two to be qualified for VMS)  K If anyone has more up-to-date or contradictory information, I would love toi hear from you.  + "Dirk Munk" <munk@home.nl> wrote in messagea! news:3AE98E86.5E75B36F@home.nl...oK > I think you're mistaken. The DS10 supports at least 2 PBXDP-AB cards withm 4a> > ports each with VMS according to the supported options list. >eL http://www.compaq.com/alphaserver/options/asds10/asds10_19_48_options.html#O
 ptListHead > $ > So that would be 8 ports per DS10.H > I can't imagine why it is stated that 2 cards are supported in VMS but only one > can be used ?1?y > I > But there is also the AC model with 8 ports. It is not mentioned in anyr options.H > list, but is it supported. So that would  make 16 ports. And I'm quite sure youK > could even put 3 of those boards in a DS10. So with a bit of luck you canr get 24L > sync ports on one DS10 ... Ask Compaq to make a statement, but as far as I know > it can be done.  >a > john nixon wrote:  > J > > I am glad you pointed that out.  I will be putting together a proposal forhB > > an Alpha solution to replace our old MicroVAXes.  We have four MicroVAXes,uJ > > each with six or 7  Q-Bus synch devices.  Since the softwaer for Alpha onlyL > > supports one synch PCIcard each, and each card only has up to 4 ports, IJ > > will need 7 DS10s to replace 4 MVAX 3600s.  (The hardware supports twoH > > cards, but the software only supports one).  I was planning on using DS10Ls, I > > but now I see that I will need to go with the somewhat more expensivet DS10s.I > > This is going to be expensive considering all the licenses (including K > > cluster) etc... just to support  Bisynch.  (Cheaper than maintenance on , > > microvaxes, but expensive all the same). > > I > > I heard that they will have an 8 or 16 port card, but it will only bet foreI > > Unix.  I  would    * * *  R E A L L Y  * * *  like to see this largerm optionI > > qualified on VMS.  I (and Compaq)  may lose this simple VAX-Alpha VMS-J > > migration to a 3rd party solution if I have to put in 6 or 7 alphas to > > replace 4 MVAXes.n > > H > > Does anyone know, or can anyone find out, if there are plans to ever qual& > > the 8 or 16 port (PBXDD-*) on VMS? > ><: > > <fabio_compaq@ep-bc.petrobras.com.br> wrote in messageJ > > news:OFAC3B441B.C60A7362-ON03256A3B.00422EF7@ep-bc.petrobras.com.br...K > > > I discovered that the DS10L doesnt support theses boards for OpenVMS.e Just > > > the DS10 standard . . .t > > >S > > >X > >wL http://www.compaq.com/alphaserver/options/asds10l/asds10l_19_48_options.html > > #OptListHead > > >-
 > > > Regardst > > >- > > >- > > > FC > > >- > > >  > > >d > > >lG > > > "Alan Boyles" <alan.boyles@mindspring.com> em 27/04/2001 08:56:10- > > >-B > > > Favor responder a "Alan Boyles" <alan.boyles@mindspring.com> > > >  > > >  > > >M! > > >       Info-VAX@Mvb.Saic.Comd > > >  > > >- > > >-' > > > Assunto: Re: Sync Comm - PCI Card1 > > >4 > > >2J > > > There are a couple of Sync boards available depending on the machine typeL > > > and the number of ports,  the PBXDP-AA or AB.  These are Emulex boardsL > > > which I believe will EOL this year and then be replaced by a couple of
 > > > others.T > > >c
 > > > Alan > > >t: > > > fabio_compaq@ep-bc.petrobras.com.br wrote in articleI > > > <OF04362B1C.1E69CBB6-ON03256A3B.003EC700@ep-bc.petrobras.com.br>...sJ > > > > I remember DEC used to have a Synchronous Communications PCI card.D > > > > At www.dnpg.com there is no information about.t his product. > > > >t8 > > > > Do you know where can I get informations about ? > > > >k > > > > Regardse > > > >n
 > > > > FC > > > >t > > > >  > > >e > > >o > > >r > > >e > > >r > > >D > > >" >l >o   ------------------------------  % Date: Fri, 27 Apr 2001 16:08:04 +0200-+ From: "Peter Kille" <peter.kille@volvo.com>p0 Subject: Re: Tabular Timeline Tracking Tool (T4), Message-ID: <9cbugi$ihs$1@vg170.it.volvo.se>  K I have also tried to get info about this, please let me know if you get any> answers.   BR Peter1 _________________________________________________t Peter Kille  Systems Managerm Volvo Information Technology	 Dept 8740  SE-54187 Skvde, Swedeno Telephone: +46 500 475491C Mobile Phone: +46 703487163- E-mail: Peter.Kille@volvo.com:      : "Richard L. Dyson" <rick-dyson@uiowa.edu> wrote in message# news:3AE82118.3E2B9AEF@uiowa.edu...DI > I have read about T4 a couple times now, but have not been able to find F > it anywhere on the OpenVMS web pages or software rollout lists, etc. >DI > Does anyone know the status of this product or where there is more infod > about it?a > 
 > Regards, > Rick > --J > Richard L. Dyson                                    rick-dyson@uiowa.eduJ >  _   _      _____                http://www-pi.physics.uiowa.edu/~dyson/J > | | | |    |_   _|   Systems Analyst                     O: 319/335-1879J > | | | | of   | |     The University of Iowa            FAX: 319/335-17538 > | \_/ |     _| |_    Department of Physics & Astronomy/ >  \___/     |_____|   Iowa City, IA 52242-1479    ------------------------------  # Date: Fri, 27 Apr 2001 17:37:16 GMT:2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)0 Subject: Re: Tabular Timeline Tracking Tool (T4)0 Message-ID: <g7iG6.305$5I.5859@news.cpqcorp.net>  [ In article <3AE82118.3E2B9AEF@uiowa.edu>, "Richard L. Dyson" <rick-dyson@uiowa.edu> writes:pH :I have read about T4 a couple times now, but have not been able to findE :it anywhere on the OpenVMS web pages or software rollout lists, etc.1 : H :Does anyone know the status of this product or where there is more info
 :about it?  J   The reason you are having trouble getting information involves the levelL   of difficulty folks are encountering locating information you are seeking.H   Having checked all the usual spots, I can not locate any such product,L   package, server, widget or item available within Compaq.  If this T4 does J   exist, the associated information is difficult to locate, or the productI   name that was cited does not match the product name that was used.  (Orc   it is not a Compaq product?)  J   Without more details -- such as a source or a webpage referencing this, H   or a contact or similar source such, I doubt that I'd be able to find !   additional information on this.-  L   [Please, don't cross-post stuff over to the Ask The Wizard area.  Thanks!]  N  ---------------------------- #include <rtfaq.h> -----------------------------N       For additional, please see the OpenVMS FAQ -- www.openvms.compaq.com    N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------  # Date: Fri, 27 Apr 2001 16:10:15 GMTp4 From: LESLIE@209-16-45-102.insync.net (Jerry Leslie)1 Subject: Re: TCP/IP 5.1 Allow Host Table To Wipedf) Message-ID: <HRgG6.1824$%L5.32250@insync>n  0 David J. Dachtera (djesys.nospam@fsi.net) wrote: : Jerry Leslie wrote:  : > B : > TCP/IP Services for OpenVMS Alpha Version V5.1 will delete all7 : > local hosts when the following command is executed:  : > # : >   $ tcpip set nohost /noconform  : [snip] : I : I seem to remember that being a "feature" all the way back to UCX V4.x.w   Yes. b   : - : BTW - isn't that supposed to be "/CONFIRM"?  :   I Yup, but "/CONFORM" was accepted by the system when I performed the test.s  % --Jerry Leslie   leslie@clio.rice.edup/                  leslie@209-16-45-97.insync.neto;                  leslie@209-16-45-102.insync.net is invalido2                  (my opinions are strictly my own)   ------------------------------  # Date: Fri, 27 Apr 2001 07:24:20 GMTm% From: "Joseph Gill" <gilljb@home.com>6S Subject: Trying to connect HSG80 console port to Bay Networks Annex terminal servert< Message-ID: <E89G6.34381$B22.7922755@news1.rdc2.pa.home.com>  L Has anyone connected the HSG80 console port to a Bay Networks Annex terminalL server?  I've checked the VMS FAQ for the pinouts but that didn't help and IA can't find the pinouts for the Bay Network Annex terminal server.    Thanks,    Joeo   ------------------------------  + Date: Mon, 23 Apr 2001 16:19:56 +0000 (UTC)u' From: david20@alpha1.mdx.ac.uk (D.Webb)CP Subject: Re: using a pipe command into a CP creates DCL$OUTPUT_xxxxxxx.LOG files+ Message-ID: <9c1knc$o3n$1@aquila.mdx.ac.uk>D   In article <rdeininger-2304011016160001@user-2ive774.dialup.mindspring.com>, rdeininger@mindspring.com (Robert Deininger) writes:n8 >In article <9c17tk$jda@news.or.intel.com>, "Iris Green" ><iris.green@intel.com> wrote: >3N >> We use the pipe command extensively inside DCL procedures, in the following >> way:m* >> $ pipe sho system | sear sys$input text > K >In this particular case, you might want to look at SHOW SYSTEM/SEARCH, and J >avoid the pipe command.  You'll save the cost of creating the subprocess,- >and probably get the functionality you need.- >    Rather different functionality.   &  pipe sho system | sear sys$input text  -  returns just those lines containing the text8    H   SHOW SYSTEM/SEARCH won't actually work on my VMS 7.2-1 systems without   also specifying /PAGE    eg   Alpha2:sh sys/search=dispbN %DCL-W-CONFLICT, illegal combination of command elements - check documentation	  \NOPAGE\   D And when /PAGE is specified it  pages through the SHOW SYSTEM output' highlighting lines containing the text.h    H I don't know of any way of getting SHOW SYSTEM/SEARCH to only return theI lines containing a piece of text rather than just highlighting that text.     
 David Webb VMS and Unix team leader CCSS Middlesex University   ------------------------------  + Date: Fri, 27 Apr 2001 02:16:41 -0700 (PDT)c$ From: nclews <nclews at csc dot com>5 Subject: Re: Re: V7.3 in clusters w/earlier versions? + Message-ID: <313.988233636504@iw1.etin.com>a   In reply to:M http://www.etin.com/article/Article.jsp?messageID=11962077&folder=comp.os.vmsD    X I have to say, you'll _want_ that XFC, after some tests I ran, I was suitably impressed!  e And just to comment on wider support of mixed versions, quite recently Colorado CSC dialled into and ne diagnosed a cluster running 5.5-2, 6.1 and 6.2 plus oddball hardware, and identified the problem. It r) wasn't the operating system either.......S  ` Forget your preconceptions with other operating systems, imagine trying to get W3.11, W95, W98, 9 WNT3 and WNT4 co-existing in the same way. No, maybe not.    Regards, Nic ----I Posted via http://www.etin.com - the FREE public USENET portal on the Web F Complete SEARCHING, BROWSING, and POSTING of text and BINARY messages!   ------------------------------    Date: 27 Apr 2001 09:11:37 -0500+ From: young_r@encompasserve.org (Rob Young) 5 Subject: Re: Re: V7.3 in clusters w/earlier versions?63 Message-ID: <2zaGuKX6glka@eisner.encompasserve.org>l  R In article <313.988233636504@iw1.etin.com>, nclews <nclews at csc dot com> writes: >  > In reply to:O > http://www.etin.com/article/Article.jsp?messageID=11962077&folder=comp.os.vmsh >  > G > I have to say, you'll _want_ that XFC, after some tests I ran, I was v > suitably impressed!d >   ! 	But it isnt' as fast as Linux.  s  D 	Long running tongue-in-cheek troll for the caching bashers.  For anE 	excellent counter-point to the caching bashers, this is a good read:.  O http://www.tmk.com/ftp/decus-sig-tapes/vlt00b/vmslt00b/net/vms-cache-future.txtw  < 	(Getting the above in early as you aren't allowed to be tooC 	excited about XFC.... you see it isn't as good as Linux.  So don'td 	get too happy!)   				Rob    *-
 RIP Novell *-   ------------------------------  % Date: Fri, 27 Apr 2001 10:26:32 -0400o  From: norm.raphael@jamesbury.com5 Subject: Re: Re: V7.3 in clusters w/earlier versions? 4 Message-ID: <C2256A3B.004E78A2.00@jklh21.valmet.com>   The question is:  = [When] will the XFC be patched back into V7.2,V7.1,V6.2,etc.?t andwB Will mixed [version/architecture] clusters run safely before that?     Then: $ How about VAX, is it even an option?  $ (Okay, that more than one question.)  4 [Disclaimer:  I am not up to speed on the XFC, yet.]          3 young_r@encompasserve.org on 04/27/2001 10:11:37 AMn  + Please respond to young_r@encompasserve.orge   To:   Info-VAX@mvb.saic.com. cc:I6 Subject:  Re: Re: V7.3 in clusters w/earlier versions?          J In article <313.988233636504@iw1.etin.com>, nclews <nclews at csc dot com> writes:t >m > In reply to:O > http://www.etin.com/article/Article.jsp?messageID=11962077&folder=comp.os.vmst >t >IF > I have to say, you'll _want_ that XFC, after some tests I ran, I was > suitably impressed!e >v  #      But it isnt' as fast as Linux.7  H      Long running tongue-in-cheek troll for the caching bashers.  For anI      excellent counter-point to the caching bashers, this is a good read:S  O http://www.tmk.com/ftp/decus-sig-tapes/vlt00b/vmslt00b/net/vms-cache-future.txt   @      (Getting the above in early as you aren't allowed to be tooG      excited about XFC.... you see it isn't as good as Linux.  So don'ti      get too happy!)                       Robs   *n
 RIP Novell *s   ------------------------------    Date: 27 Apr 2001 12:04:14 -0500+ From: young_r@encompasserve.org (Rob Young) 5 Subject: Re: Re: V7.3 in clusters w/earlier versions?u3 Message-ID: <TCxb2hJXuf8C@eisner.encompasserve.org>   W In article <C2256A3B.004E78A2.00@jklh21.valmet.com>, norm.raphael@jamesbury.com writes:l  7 	Disclaimer:  I don't work for Compaq and am relying ons7 	memory.  If I discover something different in the next 9 	24 or so hours, I'll follow up to my own post.  Not goodf' 	to leave dangling wrongness out there.r   >  >  > The question is: > ? > [When] will the XFC be patched back into V7.2,V7.1,V6.2,etc.?A > and.D > Will mixed [version/architecture] clusters run safely before that? >   < 	Don't get excited about going back further than the current? 	version.  In other words, it *may* (can't recall) make it intol; 	7.2-1 but doubt it goes further back (after all those are iH 	unsupported versions).  This is something I am really unsure about, so 8 	maybe I dig a bit at lunch or Hoff or Kerry pop in :-).   >  > Then:m& > How about VAX, is it even an option? > 3 	Nope.  32-bit.  XFC is 64-bit.  XFC is Alpha only:n  . http://www.openvms.compaq.com/openvms/sdk.html    H            Extended File Cache (XFC) Version 1.0 support (Alpha only) --I            Replacement for VIOC. Extends the capabilities of cluster fileh	 		cache. o  & > (Okay, that more than one question.) > 6 > [Disclaimer:  I am not up to speed on the XFC, yet.] >    	Read this:   8 http://www.decus.gr.jp/decus99/sessioncd/NOTES/OV166.PDF   				Roba   ------------------------------    Date: 27 Apr 2001 13:39:35 +0200G From: Jan Vorbrueggen <jan@mailhost.neuroinformatik.ruhr-uni-bochum.de>s' Subject: Re: VMS and Fast Etherchannel?sH Message-ID: <y4y9smfu6g.fsf@mailhost.neuroinformatik.ruhr-uni-bochum.de>  - John Eisenschmidt <jeisensc@aaas.org> writes:o  L > If we take a piece of PC hardware (which I was crucifying earlier, make noK > mistake) which has two onboard NICs, whether I put NT or OpenBSD on it, I I > can configure those two NICs to act as 1, bound to the same IP address.g  J Dunno about IP (I always thought you need to distinguish each interface by0 IP?), but DECnet Phase V will do this for you...   	Jan   ------------------------------  # Date: Fri, 27 Apr 2001 11:55:06 GMT  From: danco@pebble.org () ' Subject: Re: VMS and Fast Etherchannel?e- Message-ID: <slrn9eikc4.2in.danco@pebble.org>i  . On 27 Apr 2001 13:39:35 +0200, Jan Vorbrueggen8 <jan@mailhost.neuroinformatik.ruhr-uni-bochum.de> wrote:  K >Dunno about IP (I always thought you need to distinguish each interface byw >IP?)...  H That's probably operating system specific.  I have a Redhat Linux systemJ here that has the same address on four interfaces (ppp0, ppp1, eql, eth1).F The Redhat Linux route command lets you specify which interface you're, refering to, so that makes them unambiguous.   - Dan    ------------------------------    Date: 27 Apr 2001 16:18:17 +0200G From: Jan Vorbrueggen <jan@mailhost.neuroinformatik.ruhr-uni-bochum.de>b' Subject: Re: VMS and Fast Etherchannel?nH Message-ID: <y44rvajuja.fsf@mailhost.neuroinformatik.ruhr-uni-bochum.de>   danco@pebble.org () writes:e  M > >Dunno about IP (I always thought you need to distinguish each interface byn
 > >IP?)...J > That's probably operating system specific.  I have a Redhat Linux systemL > here that has the same address on four interfaces (ppp0, ppp1, eql, eth1).H > The Redhat Linux route command lets you specify which interface you're. > refering to, so that makes them unambiguous.  C Won't that confuse ARP if those interfaces are on the same network?    	Jan   ------------------------------  % Date: Fri, 27 Apr 2001 11:11:52 -0400 + From: John Eisenschmidt <jeisensc@aaas.org>a' Subject: Re: VMS and Fast Etherchannel?h# Message-ID: <sae9541a.051@aaas.org>n  G Presumably, all four have different MAC addresses. ARP is a layer two =1? issue, binding them to multiple address is a layer three issue.r  G Unless the network drivers clone the MAC address across all the bound =SK interfaces (which is not what is being said here) and then an ARP against =eI the IP address would be fine. It's then the drivers problem of managing =l those packets.=20e  D >>> Jan Vorbrueggen <jan@fsnif.neuroinformatik.ruhr-uni-bochum.de> = 04/27/2001 10:18:17 AM >>> danco@pebble.org () writes:u  L > >Dunno about IP (I always thought you need to distinguish each interface = by
 > >IP?)...J > That's probably operating system specific.  I have a Redhat Linux systemG > here that has the same address on four interfaces (ppp0, ppp1, eql, =o eth1).H > The Redhat Linux route command lets you specify which interface you're. > refering to, so that makes them unambiguous.  C Won't that confuse ARP if those interfaces are on the same network?n   	Jan   ------------------------------    Date: 27 Apr 2001 10:42:31 -05001 From: Graham Allan <allan@physics.umn.edu.nospam> ' Subject: Re: VMS and Fast Etherchannel?o2 Message-ID: <w534rva1h94.fsf@pitcairn.spa.umn.edu>  - Ryan Moore <rmoore@rmoore.dyndns.org> writes:   / > On Thu, 26 Apr 2001, John Eisenschmidt wrote:(N > > My question was simply is that possible under OpenVMS? Can I treat the twoJ > > onboard NICs in the DS10 as 1 with a single IP address? And if so, how  > > would I go about doing that. > K > I believe Multinet 4.3 can do that.  Don't know about TCP/IP Services forMC > OpenVMS (UCX)... probably not in v5.0.  Don't know about new 5.1.   I Well, Multinet has something called "Paired Network Interface on a Common H Ethernet Cable", but it sounds a bit like each interface probably has toJ have a distinct IP address. Having read the Multinet docs I'm not sure howJ it's supposed to work. I'm pretty certain it's a different thing from Fast3 Etherchannel, which I think works at a lower level.C   Graham --  I -------------------------------------------------------------------------/, Graham Allan - I.T. Manager - (612) 624-50409 School of Physics and Astronomy - University of Minnesota I -------------------------------------------------------------------------    ------------------------------  % Date: Fri, 27 Apr 2001 08:26:29 -0400e5 From: David Beatty <David.Beatty@qwertysasasdfgh.com>eP Subject: Re: VMS gets positive mention in Compaq Q1 financial analystsconference2 Message-ID: <OmXpOjyQRKXps1kzf6riIhF3pNvT@4ax.com>  , On Thu, 26 Apr 2001 19:05:59 -0400, JF Mezei% <jfmezei.spamnot@videotron.ca> wrote:e   >"Richard D. Piccard" wrote:O >> Since VAX is a CISC architecture, it is axiomatic that a 1 GHz VAX should bes( >> expected to outperform a 1 GHz Alpha. >r >uO >This generally makes sense. However, when you factor in architectural features.I >such as pipelining, and branch prediction etc whcih are made possible byaI >having a simple fixed length instruction set, I don't think that you can E >automatically say that a 1ghz VAX would be faster than a 1ghz Alpha.- >-N >However, I wouldn't mind having an upgrade to a 1ghz VAX chip for my Microvax >II :-) :-) :-) :-)l >cL >> The fact that there are 1 GHz chips in the IA-32 space tells us that fastN >> CISC is possible.  What it would have taken to make a 1 GHz VAX chip is far >> beyond my knowledge.n >rM >It would be interesting to see how much money Intel had to invest to upgrade O >its 8086 game controller into the Pentium enterprise computer, versus how muche7 >money Digital invested to develop the brand new Alpha.7 >- >-L >I personally maintain that there was nothing wrong with the VAX, except theL >fact that Digital refused to consider that it was competing against cheaperG >systems from Intel and Sun etc. Digital seemed to insist that its onlyeM >competitor was the IBM mainframe at one point, which allowed Sun to start ups >and break into the market.e >e  B The huge irony here is that the guys who created Sun pitched their. idea to Digital and were turned down cold!!!!   O >Had Digital kept VAX prices low to compete against HP/Apollo and Sun, I am note# >sure sun would have survived long.s   ------------------------------  % Date: Fri, 27 Apr 2001 08:39:10 -0500e+ From: Christopher Smith <csmith@amdocs.com>7P Subject: RE: VMS gets positive mention in Compaq Q1 financial analystsconferenceL Message-ID: <3B55D7F383B0D31197D9009027541CBF0D9D1DC3@cmiexch1.cmi.itds.com>   > -----Original Message-----6 > From: JF Mezei [mailto:jfmezei.spamnot@videotron.ca]  = > It would be interesting to see how much money Intel had to . > invest to upgrade>7 > its 8086 game controller into the Pentium enterprise   > computer, versus how much 8 > money Digital invested to develop the brand new Alpha.  L Does anyone else think "enterprise computer" is a strongly optimistic phraseD for pentium? :)  I would have said "accelerated game controller," or something similar.    H I'm not saying you can't do good things with pentiums.  Sequent has been; very successful in the past at taking intel CPUs and makingtD minicomputer-class systems out of them.  They were good systems too,L generally.  It isn't done much, though, and I'd say the number 1 application+ for intel chips is still "game controller.",   Regards,   Chrism  ! Christopher Smith, Perl Developerp Amdocs - Champaign, IL   /usr/bin/perl -e '? print((~"\x95\xc4\xe3"^"Just Another Perl Hacker.")."\x08!\n");t '       ------------------------------  % Date: Fri, 27 Apr 2001 15:17:14 +0100 0 From: andrew harrison <andrew.nospam@uk.sun.com>P Subject: Re: VMS gets positive mention in Compaq Q1 financial analystsconference* Message-ID: <3AE97F6A.A15197F4@uk.sun.com>   Christopher Smith wrote: >  > > -----Original Message-----8 > > From: JF Mezei [mailto:jfmezei.spamnot@videotron.ca] > > > > It would be interesting to see how much money Intel had to > > invest to upgraden8 > > its 8086 game controller into the Pentium enterprise > > computer, versus how much : > > money Digital invested to develop the brand new Alpha. > N > Does anyone else think "enterprise computer" is a strongly optimistic phraseF > for pentium? :)  I would have said "accelerated game controller," or > something similar. > J > I'm not saying you can't do good things with pentiums.  Sequent has been= > very successful in the past at taking intel CPUs and making5F > minicomputer-class systems out of them.  They were good systems too,N > generally.  It isn't done much, though, and I'd say the number 1 application- > for intel chips is still "game controller."  >   ; Your favorite systems vendor disagrees. This is from their F web site  6 "Advantages of the new Compaq ProLiant 8-Way computers  3 In the enterprise space, Compaq has merged the low c5 cost and easy to use advantages of PCs with the high o. reliability, serviceability, manageability and4 availability advantages of the mid range computers.   ? These qualities are needed to address today's customer needs."     OR  < "This server was designed for mission-critical environments : and offers an outstanding combination of high performance 6 and high availability features" From another Proliant 	 brochure.h     regardsa
 > Regards, >  > ChrisG > # > Christopher Smith, Perl Developere > Amdocs - Champaign, IL >  > /usr/bin/perl -e 'A > print((~"\x95\xc4\xe3"^"Just Another Perl Hacker.")."\x08!\n");y > 'o >    -- r Andrew Harrison  Enterprise IT Architect-   ------------------------------  % Date: Fri, 27 Apr 2001 10:30:49 -0500r+ From: Christopher Smith <csmith@amdocs.com>cP Subject: RE: VMS gets positive mention in Compaq Q1 financial analystsconferenceL Message-ID: <3B55D7F383B0D31197D9009027541CBF0D9D1DC5@cmiexch1.cmi.itds.com>   > -----Original Message-----9 > From: andrew harrison [mailto:andrew.nospam@uk.sun.com]s   > Christopher Smith wrote:  < > > generally.  It isn't done much, though, and I'd say the  > number 1 application/ > > for intel chips is still "game controller."f  = > Your favorite systems vendor disagrees. This is from their  
 > web site  8 > "Advantages of the new Compaq ProLiant 8-Way computers  5 > In the enterprise space, Compaq has merged the low  7 > cost and easy to use advantages of PCs with the high l0 > reliability, serviceability, manageability and6 > availability advantages of the mid range computers.   A > These qualities are needed to address today's customer needs." n   > OR  > > "This server was designed for mission-critical environments < > and offers an outstanding combination of high performance 8 > and high availability features" From another Proliant  > brochure.n  H Well, I'm not sure who my "favorite systems vendor" is these days.  It'sJ probably not Compaq, even though they may vend my favorite systems. :)  ItL used to be silicon graphics -- before they started getting their hands dirtyI with those x86 "game controllers."  Even today, the level of support they G offer (even just for free, on their web-page) for an SGI is incredible.,K Compaq might take some lessons in web-design from the SGI people, at least,jI even if SGI now has less innovation and more stupidity than Compaq. (This6L is, IMO, due to both the good things happening at Compaq, and the bad things at SGI.) ;)6  E Compaq could certainly become my favorite vendor if they continue therG genuine effort that I've seen (yeah, I'm really beginning to think it'sd3 genuine) to really sell VMS and the alpha platform.   K On the other hand, I'm starting to like starbridge systems, though I'm sureb1 they won't be in my price range for a long while.n  G Just for reference, what do you think, Andrew -- does the x86 (assumingtL standard architecture, and not heavily modified, a la Sequent) belong in the "enterprise" class?m   Regards,   Chris   ! Christopher Smith, Perl Developert Amdocs - Champaign, IL   /usr/bin/perl -e '? print((~"\x95\xc4\xe3"^"Just Another Perl Hacker.")."\x08!\n");  '    ------------------------------  % Date: Fri, 27 Apr 2001 16:50:48 +01001  From: steven.reece@quintiles.comP Subject: Re: VMS gets positive mention in Compaq Q1 financial analystsconferenceH Message-ID: <OF4D9B9973.77460717-ON80256A3B.0056A43A@qedi.quintiles.com>  E But then they keep trying to re-invent their "operating system" (ha!)rJ offerings so that people like you and I will forget how bad their last one was. Bad news guys, it didn't work. Steve.   Larry Kilgallen wrote :m >>>tI In the interest of leveling the playing field, Microsoft keeps increasingn- the number of operating systems they support.o <<<    ------------------------------  % Date: Fri, 27 Apr 2001 16:56:58 +0100t0 From: andrew harrison <andrew.nospam@uk.sun.com>P Subject: Re: VMS gets positive mention in Compaq Q1 financial analystsconference* Message-ID: <3AE996CA.EDD825A5@uk.sun.com>   Christopher Smith wrote: >   I > Just for reference, what do you think, Andrew -- does the x86 (assuminglN > standard architecture, and not heavily modified, a la Sequent) belong in the > "enterprise" class?m >   A No, it isn't Intels design point and anything that falls outside g< Intels standard chipsets tends to end up being marginalised.  > The Sequent was a good example, they did a good job of working: around the constraints placed on them by Intels processor A architecture (2 bit CPU IDs and the rest) but the effort involved?; in doing this ment that they were unable to keep pace with m1 the increasing clock speeds on offer from Intel. _  : DG who took a much more generic approach were able to pump; their clock speeds up more quickly (then they got bought byo EMC).I  6 This does not mean that it won't happen in the future.  3 Infiniband for example would offer commodity Intel c2 vendors the ability to build more enterprise class2 servers and HP will provide IA64 in the SuperDOme + but this isn't remotely a commodity device.   2 Of course if you believe the MS spiel on "Scaling 0 Out", which is basically lots of small boxes in 4 a cluster with a parallel app on top, which Compaqs 2 brass obviously do then of course Intel boxes can  be enterprise servers. :):):)i   regardsa Andrew Harrisoni Enterprise IT Architectp   ------------------------------  % Date: Fri, 27 Apr 2001 11:38:51 -0500 + From: Christopher Smith <csmith@amdocs.com>rP Subject: RE: VMS gets positive mention in Compaq Q1 financial analystsconferenceL Message-ID: <3B55D7F383B0D31197D9009027541CBF0D9D1DC6@cmiexch1.cmi.itds.com>   > -----Original Message-----9 > From: andrew harrison [mailto:andrew.nospam@uk.sun.com]_  4 > Of course if you believe the MS spiel on "Scaling 2 > Out", which is basically lots of small boxes in 6 > a cluster with a parallel app on top, which Compaqs 4 > brass obviously do then of course Intel boxes can  > be enterprise servers. :):):)   I Well, there's nothing wrong with clusters and parallel apps.  The problem K --as with anything -- is when you have a weak architecture supporting them,g they tend to collapse. :)o  K On the other hand, if you have a cluster done right, you may not notice thepL collapse of a machine here and there.  In assuming that, though, we're faced9 with the question:  When has M$ ever done anything right?c   Regards,   Chris   ! Christopher Smith, Perl DeveloperI Amdocs - Champaign, IL   /usr/bin/perl -e '? print((~"\x95\xc4\xe3"^"Just Another Perl Hacker.")."\x08!\n");e '    ------------------------------  % Date: Fri, 27 Apr 2001 12:46:14 +0200m= From: Arne =?iso-8859-1?Q?Vajh=F8j?= <arne.vajhoej@gtech.com> . Subject: Re: VMS Loyalist dies from starvation) Message-ID: <3AE94DF6.5F8FAFF6@gtech.com>    andrew harrison wrote: > Jan Vorbrueggen wrote:: > > "Terry C. Shannon" <terryshannon@mediaone.net> writes:R > > > As once it apparently was. CPQ is not completely oblivious to this fact, butK > > > those on the inside maintain that the cost of the port would never be M > > > amortized by SAP-on-VMS sales. If such a port was easily and affordably O > > > do-able, and would make good economic sense, I suspect CPQ would go ahead- > > > and do it. > > P > > Given the price of SAP, and the potential market, I wonder what type of crapN > > the SAP code is that porting it should be so difficult. I'd really like toO > > know that, given that SAP supports what - at least several different Unices 0 > > and probably some of the IBM 370 & Co. OSes. > B > Its not the porting thats the issue. Its the testing, OS skills, > market development etc etc.h > ? > The actual process of doing the port is a tiny consideration.c   Very good point.  . (and it is not that often I agree with Andrew)  @ That this indeed is an important factor is easy to verufy by theE fact that Compaq VMS group has choosen to drop commerial support (but  realeaseE as freeware) several products that are actively being used internallyV5 and therefor still need to be technically maintained..   Arne   ------------------------------  % Date: Fri, 27 Apr 2001 12:53:39 +0100.0 From: andrew harrison <andrew.nospam@uk.sun.com>. Subject: Re: VMS Loyalist dies from starvation* Message-ID: <3AE95DC3.29BC330C@uk.sun.com>   Arne Vajh=F8j wrote: > =e   > andrew harrison wrote: > > Jan Vorbrueggen wrote:< > > > "Terry C. Shannon" <terryshannon@mediaone.net> writes:J > > > > As once it apparently was. CPQ is not completely oblivious to thi= s fact, butmJ > > > > those on the inside maintain that the cost of the port would neve= r beJ > > > > amortized by SAP-on-VMS sales. If such a port was easily and affo= rdablyJ > > > > do-able, and would make good economic sense, I suspect CPQ would = go ahead > > > > and do it. > > >nJ > > > Given the price of SAP, and the potential market, I wonder what typ=	 e of crap J > > > the SAP code is that porting it should be so difficult. I'd really = like tomJ > > > know that, given that SAP supports what - at least several differen= t Unices2 > > > and probably some of the IBM 370 & Co. OSes. > >eD > > Its not the porting thats the issue. Its the testing, OS skills, > > market development etc etc.' > >tA > > The actual process of doing the port is a tiny consideration.  > =-   > Very good point. > =   0 > (and it is not that often I agree with Andrew) > =n  B > That this indeed is an important factor is easy to verufy by theG > fact that Compaq VMS group has choosen to drop commerial support (butr
 > realeaseG > as freeware) several products that are actively being used internallya7 > and therefor still need to be technically maintained.t > =C   > Arne    6 It also raises the question of how usefull COE will be to OpenVMS.   1 It will make porting easier but without further =i  - investment by Compaq in the joint marketing =   - progams, test systems, benchmarking and the =n  / selling of OpenVMS (and its available market) =1  - to the ISV's new ports will not necessarely =    happen.o  2 I have been working on a project which requires an1 application which currently does not run on Sun =t  . we have put together a project plan with the =  , apps vendor to get the app available on Sun.  * Only a tiny part of the project budget and- time is allocated to actually doing the port.n  0 The vendor is much more interested in testing, =  . benchmarking, and how thay and Sun are going =  . to make sure that the app is sucessfull from =   a sales standpoint on Sun.  	 Regards =0   Andrew Harrisonr Enterprise IT Architecth   ------------------------------  % Date: Fri, 27 Apr 2001 08:26:02 -0400D. From: "Jerry Alan Braga" <jabraga@flanagan.ca> Subject: Re: VT POP3 client?4 Message-ID: <AAdG6.231821$Z2.2424431@nnrp1.uunet.ca>  @ Visit the following www site and look at the MAILBOX for openvms   http://patrick.lequere.net    9 "Alphaman" <alphaman64@nixspam-home.com> wrote in message,5 news:vXNF6.26505$U4.5981279@news1.rdc1.tn.home.com... K > I've been playing with PINE as I thought it was a POP3 client.  All I canU5 > seem to find reference to is its IMAP capabilities.  > I > Can it be used as a POP3 client?  What setting is required to set that?  Is4 > it the "inbox-path", and if so, what's the syntax? >aL > If not, are there any other terminal based POP3 clients available?  If notJ > that, any tools for getting mail from my ISP's POP3 server to my VMSmailK > account?  I've looked far and wide, but the closest references I've found H > have been Netscape, Mozilla, (both of which don't fit my VT need), and PINE > (IMAP only?) >a > Wishfully, > Aaron  > --@ > Aaron Sakovich  http://members.home.net/sakovich/alphaman.html@ > Make April 15 just another day:        http://www.fairtax.org/> > "Things do not change, people change." (Henry David Thoreau) >0 >e >r   ------------------------------  % Date: Fri, 27 Apr 2001 09:29:27 +0200 = From: Arne =?iso-8859-1?Q?Vajh=F8j?= <arne.vajhoej@gtech.com>s2 Subject: Re: Warning - Charon-VAX Hobbyist Edition) Message-ID: <3AE91FD7.B811DF57@gtech.com>    Timothy Stark wrote:G > Since my Charon-VAX hobbyist edition will expire on May 1, 2001 (NextcI > week), I checked Charon-VAX web site for latest version but I found outeF > that they decided to suspend the hobbyist program for a while (untilC > this summer!).  They mentioned that they rather are interested incD > commerical customers their first priority over hobbyist customers.@ > It looks like discrimination aganist hobbyist customers to me.% > I am not happy with SRI's decision.a   Discrimination ????H  < Are BMW discriminating against me because they will not give me a free BMW 750 ?   B They run a business. They make whatever decisions they want to try@ and maximize their profit. They make think a hobbyist version isD a good way of getting new customers. They may decide that a hobbyistB version is a waste of time. They may decide that they want to have@ a hobbyist version, but that ressource constraints force them to5 concentrate on the commercial version for the moment.d   Arne   ------------------------------  # Date: Fri, 27 Apr 2001 14:37:08 GMTn- From: goathunter@goatley.com (Hunter Goatley)g2 Subject: Re: Warning - Charon-VAX Hobbyist Edition0 Message-ID: <3ae983a3.48673338@swen.process.com>  P On Thu, 26 Apr 2001 22:13:22 GMT, goathunter@goatley.com (Hunter Goatley) wrote:  C >Last I knew, they had 0 commercial interest in a Linux version, so " >it's being back-burnered for now. >eA >My understanding is that the Hobbyist version will be back soon.! > < >(Note: I'm not formally affiliated with SRI, I just run the  >CHARON-VAX-Users mailing list.) > + This is what their Hobbyist page has today:-  A >     The hobbyist versions of CHARON-VAX that were available for E >     download on this site work until the end of April 2001. We haveoI >     temporarily suspended the hobbyist program while we are redesigning2E >     our site, moving it to a faster server and focussing for now on G >     support of our commercial customers. We plan to implement again anG >     CHARON-VAX package for private use on a home computer system, buttI >     we have to fit this in with our other activities before the summer.w  I >     We realize that there will be interruption in the availability of anF >     valid hobbyist version and our aplologies for that. The delay isI >     due to our engineering priorities and the time necessary to design, I >     build and test a specific hobbyist release for the Windows NT4/2000nB >     platform. As we have just decided to develop a full functionC >     CHARON-VAX product for Linux, we will generate a new hobbyist ) >     version for Linux at a later stage.h  H >     However, we do not plan to provide a specific hobbyist W98 versionI >     (although the NT4/2000 version might partly work), because we wouldsC >     have to significantly modify our code for CD/Floppies to workA >     correctly on W98.0   Hunter ------9 Hunter Goatley, Process Software, http://www.process.com/19 goathunter@goatley.com     http://www.goatley.com/hunter/.   ------------------------------  % Date: Fri, 27 Apr 2001 08:54:17 -0700s: From: "Frederick Hoenisch" <Fred.hoenisch@gems9.gov.bc.ca>6 Subject: Warning: EV68/833 and Oracle Rdb incompatible+ Message-ID: <3ae99628$1@obsidian.gov.bc.ca>t  K A warning to system managers that are keen on upgrading to EV68 technology.   K This past weekend we upgraded our AlphaServer 4100 to a dual processor ES40pI with EV68/833s.  The upgrade went relatively smoothly, that is, until the E Rdb (7.0.6.1) startup tried to run and reported the following errors:"  ; %COSI-F-UNSUPP_HW_EVX, unsupported hardware DECchip variantvA %RMU-F-FATALOSI, Fatal error from the Operating System Interface. H %RMU-F-FTL_RMU, Fatal error for RMU operation at 22-APR-2001 10:21:18.44  I Until COMPAQ and ORACLE sort out this hardware certification issue, don'ttK plan on upgrading to the EV68 processors if Rdb is mission critical to youriL organisation.  Worthy of note, is that COMPAQ responded to our issue as soon) as they were made aware of the situation.f     Frederick W. Hoenischg6 OpenVMS Services - ITSD - Province of British ColumbiaB E317 (3GR) 4000 Seymour Place,  Victoria, British Columbia, CANADA V8X 4S8 ) (work) 250 387 5673    (fax) 250 387 5231n   Fred.Hoenisch@Gems9.gov.bc.cai  I Disclaimer: The opinions and statements contained in this posting are thedJ sole responsibility of the author and have not in any way been reviewed or/ approved by my employer or any network service.    ------------------------------  + Date: Mon, 23 Apr 2001 12:48:16 +0000 (UTC)f' From: david20@alpha2.mdx.ac.uk (D.Webb) * Subject: Re: What is better?? VMS or Unix?+ Message-ID: <9c18af$lh3$1@aquila.mdx.ac.uk>p  b In article <003101c0cbf3$a9447020$ec036e64@nepco.com.jo>, abdullah <abdullah@nepco.com.jo> writes:- >This is a multi-part message in MIME format.  > , >------=_NextPart_000_002E_01C0CC04.6BD17B00 >Content-Type: text/plain; >	charset="windows-1256", >Content-Transfer-Encoding: quoted-printable >  >Hi allmK >My comanpy is a big company in Jordan, we have 1 DEC machine (7610), one =a >Alpha (2000) and 2 VAX 8530;D >3? >But to move to new technologies, we need to study it carefully G >so i like some one to tell is it better to stay in VMS or to move to =a% >other OS, and what is the beneifits?r >:
 >Thank you >l  $ Two reasons to move away from VMS :-  K 1) The latest versions of your applications are supported on Platform X but. will not be supported on VMS.e   2) Corporate policy    ie    9 "All new systems will be Windows 2000/Solaris/Platform X"8   or  - "VMS is dying we must move to something else"   J (The latter is not really true but it can be difficult to shift management2 from that opinion once it has become established).       Reasons to stick with VMS   8 1) Migration from Vax to Alpha is usually fairly simple.  4 2) System Manager/Operator familiarity. VMS is VMS. $    No need to train staff in new OS.  2 3) No need to rewrite all your command procedures.  > 4) No need to rewrite all your inhouse utilities/applications.  A 5) Retain current stability. Other platforms can be less stable -eI    especially in some cases if you run more than a single application on mJ    each machine  => With some platforms you may need to deploy more boxes.  " 6) Retain current security levels.I    Unix and Microsoft platforms have vastly more well publicised securitye:    holes => script kiddie heaven (when compared with VMS).:    => Need for increased security monitoring and patching.  1 7) Clustering - VMS clustering is still the best.v?    Tru64 Clustering is improving but has some major limitationsh0    1) Restricted range of cluster interconnects I    2) Too many disks required to setup cluster - a two node Tru64 cluster 3       requires 4 Shared disks to setup the cluster a"       -  Boot disk for each system!       -  Cluster common root disk        -  Quorum disk  nM       You can use a Raid-array and cut these disks out of that storage so youyL       don't always need to dedicate a 4GB drive to just be a quorum disk :) K       (Note. in Tru64 the quorum disk must be dedicated to this function - h*       you cannot put anything else on it).    K    3) Patching a Tru64 cluster is a nightmare - Too many reboots of clustert       members.    J In short sticking with VMS will be the most cost effective solution unless  D 1) The organisation has already invested in the alternative platformN ie If you already have a large number of Solaris systems then the organisationI will already have spent money on Corporate Firewalls, Intrusion detectionOD systems Etc and will have already trained its staff in using the OS.   ande  B 2) There are not a large number of inhouse Utilities/applications  or scripts to be rewritten.l    K These are just a few things off the top of my head. I'm sure others can adda other considerations.     
 David Webb VMS and Unix team leader CCSS Middlesex University    , >------=_NextPart_000_002E_01C0CC04.6BD17B00 >Content-Type: text/html;- >	charset="windows-1256", >Content-Transfer-Encoding: quoted-printable >w? ><!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">i
 ><HTML><HEAD>o6 ><META content=3D"text/html; charset=3Dwindows-1256" = >http-equiv=3DContent-Type>h: ><META content=3D"MSHTML 5.00.2314.1000" name=3DGENERATOR> ><STYLE></STYLE> ></HEAD> ><BODY bgColor=3D#ffffff>d5 ><DIV><FONT face=3DArial size=3D2>Hi all</FONT></DIV>bJ ><DIV><FONT face=3DArial size=3D2>My comanpy is a big company in Jordan, =
 >we have 1=20hB >DEC machine (7610), one Alpha (2000) and 2 VAX 8530;</FONT></DIV> ><DIV>&nbsp;</DIV>G ><DIV><FONT face=3DArial size=3D2>But to move to new technologies, we =n >need to study=20t >it carefully</FONT></DIV>K ><DIV><FONT face=3DArial size=3D2>so i like some one to tell is it better =i >to stay in=20D >VMS or to move to other OS, and what is the beneifits?</FONT></DIV> ><DIV>&nbsp;</DIV>F ><DIV><FONT face=3DArial size=3D2>Thank you</FONT></DIV></BODY></HTML> >s. >------=_NextPart_000_002E_01C0CC04.6BD17B00-- >    ------------------------------  # Date: Fri, 27 Apr 2001 17:59:17 GMTr0 From: sander@vmsbiz.enet.dec.com (Warren Sander); Subject: Re: www.compaq.com/openvms news from Sue Skonetski 0 Message-ID: <VriG6.310$5I.5836@news.cpqcorp.net>  
 $! Doc at end  $ lf[0,7] = %X0A	! line feed! $ digits = -		! X bit map for 0-4uE 	"0xff0xff0xff0xc30x990x990x990x990x990x990x990x990xc30xff0xff0xff"+-eE 	"0xff0xff0xff0xcf0xc70xcf0xcf0xcf0xcf0xcf0xcf0xcf0xcf0xff0xff0xff"+-oE 	"0xff0xff0xff0xc30x990x9f0x9f0xcf0xe70xf30xf90xf90x810xff0xff0xff"+- E 	"0xff0xff0xff0xc30x990x9f0x9f0xc70x9f0x9f0x9f0x990xc30xff0xff0xff"+-,C 	"0xff0xff0xff0xcf0xcf0xc70xc70xcb0xcb0xcd0x810xcf0x870xff0xff0xff"c( $ digits = digits +-	! X bit map for 5-9E 	"0xff0xff0xff0x810xf90xf90xf90xc10x9f0x9f0x9f0x990xc30xff0xff0xff"+-DE 	"0xff0xff0xff0xc70xf30xf90xf90xc10x990x990x990x990xc30xff0xff0xff"+-eE 	"0xff0xff0xff0x810x990x9f0x9f0xcf0xcf0xe70xe70xf30xf30xff0xff0xff"+- E 	"0xff0xff0xff0xc30x990x990x990xc30x990x990x990x990xc30xff0xff0xff"+-sC 	"0xff0xff0xff0xc30x990x990x990x990x830x9f0x9f0xcf0xe30xff0xff0xff"n $! $open:3 $ open/read/write/error=read_error counter 'p1'.txto $ read counter records $ count = f$integer(record) + 1'B $ record = f$extract(f$length(count)-1,7,"000000"+f$string(count)) $ write/update counter recorda $ close counterc5 $ write sys$output "Content-type: image/x-xbitmap"+lfn; $ write sys$output "Expires: Thu, 15 Mar 2001 00:00:00 GMT"t= $ write sys$output "Cache-Control: no-cache, must-revalidate"g% $ write sys$output "Pragma: no-cache"e $ write sys$output lfr. $ write sys$output "#define count_width 56"+lf/ $ write sys$output "#define count_height 16"+lfp4 $ write sys$output "static char count_bits[] = {"+lf $ x = 0o $xloop:  $ hold = ""0 $ y = 1  $yloop:-( $ c = f$integer(f$extract(y-1,1,record))2 $ hold = hold + f$extract(((c*64)+(4*x)),4,digits)% $ if y .lt. 7 then hold = hold + ", "  $ y = y + 1V $ if y .lt. 8 then goto yloopn $ if x .eq. 15 $  then hold = hold + "};" $  else hold = hold + ", " $ endifb $ write sys$output hold+lf $ x = x + 1e $ if x .lt. 16 then goto xloop $ goto exitc $read_error: $ status = $statusC $ if status .eq. %X1001828A then goto open	!file locked, just retrya@ $ if status .eq. %X10018292		!file not found, (try to) create it4 $  then open/write/error=read_error counter 'p1'.txt $	write counter "0000000"u $	close countero $	goto opens $ endifQ- $ write sys$output "Content-type: text/plain"i $ write sys$output ""n$ $ write sys$output f$message(status) $exit: $ exit   $!6 $! COUNTER.COM -- just to see if I could do it in DCL.5 $! call this by <IMG SRC="/cgi-bin/counter.com/page">wB $! Where page is the P1 Argument and is used to make the file nameE $! for the TXT File which holds the "count" IF the file doesn't existtE $! it is created with the "count" starting at 1. The file file can bel  $! edited and the count changed. $!F $! debug it by calling it by itself: http://.../cgi-bin/counter.com...J $! returns an XBM image of the number of hits; or plaintext with the errorJ $! You will need to create the counter yourself if the server doesn't have $! write access.   -- sB ------------------------------------------------------------------6 Warren Sander                        OpenVMS MarketingD Compaq Computer Corporation          Work:  warren.sander@compaq.comE 200 Forest Street MR01-3/J1          Personal: sander@ma.ultranet.comL3 Marlboro, MA 01752                   (508) 467-487596    My opinions are my own and I only speak for myself -          Read http://www.openvms.compaq.com/ :B ------------------------------------------------------------------   ------------------------------  % Date: Fri, 27 Apr 2001 08:55:26 +0200a! From: Didier Morandi <DMo@ims.ch>d1 Subject: [INFO] PCSI for Newbie version x0.0-0.00 & Message-ID: <3AE917DE.3C83B6DD@ims.ch>  +                         =================== +                         POLYCENTER_INFO.TXT +                         ===================     0                    V1.1-0 27-apr-2001 D. Morandi8             (for any questions: Didier.Morandi@wcxn.org)7              or http://groups.yahoo.com/group/vmsinstalt    F The PolyCenter Software Installation utility (aka PCSI) supersedes the$ VMSINSTAL.COM DCL command procedure.  D To build a kit with PCSI, the following actions should be performed:   1. Documentation ================  L Read the OpenVMS System Management Utilities Reference Manual M-Z pages 3-1 K to 3-55 and the Polycenter Software Installation Utility Developer's Guide.     
 2. Check-listt
 =============e  D Fill the questionaire pp. 3-1 to 3-4 of the above Developer's Guide.    ! 3. Product Description File (PDF)V! =================================i  ) Write the .PCSI$DESC file such like this:s (the comment character is "--")   J --[start of TOTO.PCSI$DESC]----------------------------------------------- --! product IMS VMS TOTO v1.0-0 full;t --0 if (<software DEC AXPVMS version below V7.2-1>);     error UNSUPP_VMS_VER abort;  endif; --L execute preconfigure "@PCSI$SOURCE:[TEMP]TOTO$PRE_CONFIGURE.COM" interactive*         uses [TEMP]TOTO$PRE_CONFIGURE.COM; --3 rights identifier TOTO with "/attributes=resource";p --. directory [TOTO] owner TOTO protection private access control3         ("(IDENTIFIER=[TOTO],ACCESS=READ+EXECUTE)",l3          "(IDENTIFIER=[*,*],         ACCESS=NONE)",oD          "(IDENTIFIER=[TOTO],OPTIONS=DEFAULT,ACCESS=READ+EXECUTE)"); -- account TOTO with ( ?  "/uic=[200,010]/account=TOTO/device=MYDISK:/directory=[TOTO]",n  "TOTO_id",r  "/pwdlifetime=""300-"""); --3 file [SYSMGR]TOTO.COM owner TOTO protection privateg access control3         ("(IDENTIFIER=[TOTO],ACCESS=READ+EXECUTE)",p,          "(IDENTIFIER=[*,*], ACCESS=NONE)"); --3 file [SYSEXE]TOTO.EXE owner TOTO protection private- access control.         ("(IDENTIFIER=[TOTO],ACCESS=EXECUTE)",,          "(IDENTIFIER=[*,*], ACCESS=NONE)"); --3 file [SYSHLP]TOTO.HLP owner TOTO protection privateu access control+         ("(IDENTIFIER=[TOTO],ACCESS=READ)",B,          "(IDENTIFIER=[*,*], ACCESS=NONE)"); --J execute postinstall "@PCSI$SOURCE:[TEMP]TOTO$POST_INSTALL.COM" interactive)         uses [TEMP]TOTO$POST_INSTALL.COM;' -- end product; --H --[end of TOTO.PCSI$DESC]-----------------------------------------------      4. Product Text File (PTF) ==========================  / Create the help and errors text file like this:   J --[start of TOTO.PCSI$TEXT]-----------------------------------------------  " =product IMS VMS TOTO v1.0-0 full; 1 UNSUPP_VMS_VER =prompt OpenVMS 7.2-1 required.nD The version of TOTO you are installing runs only under OpenVMS 7.2-1
 or higher.  B Please check whether you are installing this software on the right= system, or do upgrade this system to OpenVMS 7.2-1 or higher.n  H --[end of TOTO.PCSI$TEXT]-----------------------------------------------     5. Pre and Post installation ============================  O Write the DCL pre and post-installation procedures which should (eventually) beiC executed before and after the installation. A standard name for the,G pre-configuration procedure could be TOTO$PRE_CONFIGURE.COM and for thev( post-installation TOTO$POST_INSTALL.COM      6. Packaging ============  . Build a DCL procedure to create the .PCSI kit. Example:   $ product package TOTO - o      /base=VMS -  C      /producer=IMS -                     ! the name of your companyfE      /source=MYDISK:[TOTO.PDF] -         ! where to find PDF and PTF hC      /destination=MYDISK:[KITS] -        ! where to put .PCSI file tJ      /material=MYDISK:[TOTO.DEV] -       ! where to find product material       /format=sequential   ; This will create a file named IMS-VMS-TOTO-V0100-0-1.PCSI;1o     7. Testn =======w  ( Test the kit with the following command:   $ PRODUCT INSTALL TOTO  L Please note that the kit file name is IMS-VMS-TOTO-V0100-0-1.PCSI;1 but the 4 command to install it is only $ PRODUCT INSTALL TOTO  P To install the software on another disk than the default sys$sysdevice disk, use the /DESTINATION qualifier.    Example:  2 $ PRODUCT INSTALL TOTO/DEST=ANOTHER_DISK:[000000.]  0 Note that the "[000000.]" syntax is *mandatory*.    0 To log the installation, add the /LOG qualifier.   Example:  6 $ PRODUCT INSTALL TOTO/DEST=ANOTHER_DISK:[000000.]/log    N To watch what happens in terms of accounts creation and modifications, enable O AUTHORIZATION and FILE ACCESS FAILURE auditing before the installation with thee following commands:   & $ set audit/alarm/enable=authorization0 $ set audit/alarm/enable=file_access=failure=all $ reply/enable     Enjoy. ---m  8 (also posted in http://groups.yahoo.com/group/vmsinstal)   ------------------------------  % Date: Fri, 27 Apr 2001 10:09:48 -0400e2 From: norm lastovica <norman.lastovica@oracle.com>C Subject: Re: [INFO] SYSTEM-F-INSFSPTS, insufficient SPTEs availables* Message-ID: <3AE97DAC.AA77E634@oracle.com>   Didier Morandi wrote:n >  > Remember?h > I > %SQL-F-ERRATTDEF, Could not use database file specified by SQL$DATABASE 8 > -RDB-F-SYS_REQUEST, error from system services requestH > -RDMS-F-CANTCREGBL, error creating and mapping database global section2 > -SYSTEM-F-INSFSPTS, insufficient SPTEs available 	mH This is a well understood condition.  Please contact Oracle Rdb support F directly.  The issue is that there is not enough contiguous 32-bit S0 G address space to use the SHARED MEMORY IS SYSTEM feature that you have t specified for your database.    F What can you do to increase this space?  First, make sure that you areH using at least VMS V7.2 since a bunch of stuff was moved out of S0 spaceE into S2.  Second, don't make it a habit of constantly opening/closingaI databases (especially if you're making parameter changes to the database)v3 as this can increase fragmentation of system space.n   > Q > We have had a few weeks ago a discussion in here about these messages, the fact R > that INSFSPTS is a VAX only message displayed on an Alpha system running OpenVMS5 > 7.2-1, and the parameters to change to suppress it.l > R > At the Customer's place, I succeeded to raise the number of Rdb databases opened@ > at the same time in (by?) increasing GBLPAGES and GBLSECTIONS. > 0 > Many of you said that I'm exhausting S0 space. > Q > A (real) Rdb expert from DEC^H^H^H COMPAQ Switzerland came to take care of this1O > problem. He said that S0 space was exhausted and that the next version of Rdbj4 > should allow S2 access. ORACLE didn't confirm yet. >  > Fyi. >  > D. > --F > The fastest loading over-the-pond WEB page ever: http://www.wcxn.org   -- 6> norman lastovica / oracle rdb engineering / usa / 610.696.4685   ------------------------------  % Date: Fri, 27 Apr 2001 10:11:02 -0400y2 From: norm lastovica <norman.lastovica@oracle.com>C Subject: Re: [INFO] SYSTEM-F-INSFSPTS, insufficient SPTEs availableH* Message-ID: <3AE97DF6.D3052AD7@oracle.com>  3 And, for more focused attention on Rdb, I'd suggestm4 comp.databases.rdb and/or the Oracle Rdb list server via www.jcc.com.   Didier Morandi wrote:O >  > Remember?o > I > %SQL-F-ERRATTDEF, Could not use database file specified by SQL$DATABASE 8 > -RDB-F-SYS_REQUEST, error from system services requestH > -RDMS-F-CANTCREGBL, error creating and mapping database global section2 > -SYSTEM-F-INSFSPTS, insufficient SPTEs available > Q > We have had a few weeks ago a discussion in here about these messages, the facteR > that INSFSPTS is a VAX only message displayed on an Alpha system running OpenVMS5 > 7.2-1, and the parameters to change to suppress it.  > R > At the Customer's place, I succeeded to raise the number of Rdb databases opened@ > at the same time in (by?) increasing GBLPAGES and GBLSECTIONS. > 0 > Many of you said that I'm exhausting S0 space. > Q > A (real) Rdb expert from DEC^H^H^H COMPAQ Switzerland came to take care of this2O > problem. He said that S0 space was exhausted and that the next version of Rdb-4 > should allow S2 access. ORACLE didn't confirm yet. >  > Fyi. >  > D. > --F > The fastest loading over-the-pond WEB page ever: http://www.wcxn.org   --  > norman lastovica / oracle rdb engineering / usa / 610.696.4685   ------------------------------   End of INFO-VAX 2001.235 ************************