1 INFO-VAX	Fri, 21 Mar 2003	Volume 2003 : Issue 158       Contents:3 RE: An opportunity for HP? (wes: Micromedex on VMS) 3 Re: An opportunity for HP? (wes: Micromedex on VMS) 3 Re: An opportunity for HP? (wes: Micromedex on VMS) 3 RE: An opportunity for HP? (wes: Micromedex on VMS)  Re: Carly's feedback link  Re: Carly's feedback link * Re: Cluster, half in the DMZ, half outside* Re: Cluster, half in the DMZ, half outside* Re: Cluster, half in the DMZ, half outside* Re: Cluster, half in the DMZ, half outside* RE: Cluster, half in the DMZ, half outside* Re: Cluster, half in the DMZ, half outside* Re: Cluster, half in the DMZ, half outside* Re: Cluster, half in the DMZ, half outside* Re: Cluster, half in the DMZ, half outside* Re: Cluster, half in the DMZ, half outside! Re: DECserver 90TL docs and image  Re: Demagnetize Disk Drives  Re: Demagnetize Disk Drives = Re: EFI Specifications (was: I Have Seen The IA64 Future !!!) = Re: EFI Specifications (was: I Have Seen The IA64 Future !!!) = Re: EFI Specifications (was: I Have Seen The IA64 Future !!!) = Re: EFI Specifications (was: I Have Seen The IA64 Future !!!) = Re: EFI Specifications (was: I Have Seen The IA64 Future !!!) 7 Re: How do you Detect Ethernet failure from inside code < How to search for last login, but not on disusered accounts.@ Re: How to search for last login, but not on disusered accounts.# Re: I Have Seen The IA64 Future !!! # Re: I Have Seen The IA64 Future !!! # Re: I Have Seen The IA64 Future !!! # Re: I Have Seen The IA64 Future !!! ! Re: IBM starts to work on Power-5 ! Re: IBM starts to work on Power-5 - Re: LOGINOUT.EXE locked by other user problem - Re: LOGINOUT.EXE locked by other user problem - Re: LOGINOUT.EXE locked by other user problem ! Mark Gorham was Re: Rich Marcello  Re: Micromedex on VMS  Re: Micromedex on VMS 1 Re: OT Re: National Moratorium to Appease Tyrants 1 Re: OT Re: National Moratorium to Appease Tyrants  Re: Problems with SSH EAK  Re: Problems with SSH EAK  Request to Sue Re: Rich Marcello  RE: Rich Marcello  Re: Rich Marcello  Re: Rich Marcello  Re: Rich Marcello  Re: Rich Marcello & Re: Sun unix/linux new nasty bug cert!& Re: Sun unix/linux new nasty bug cert!
 www.decus.org  Re: www.decus.org  Re: www.decus.org  X-window and Xlib question X-window and Xlib question- Re: [OT] After two months of SAP training :-(   F ----------------------------------------------------------------------  % Date: Fri, 21 Mar 2003 09:20:03 -0500 ' From: "Main, Kerry" <Kerry.Main@hp.com> < Subject: RE: An opportunity for HP? (wes: Micromedex on VMS)T Message-ID: <BE56C50EA024184DAF48F0B9A47F5CF402660E7A@kaoexc01.americas.cpqcorp.net>   Brian,  = Re: Healthcare ... IDX and Cerner are growing with OpenVMS ..   G Btw - remember, Cerner only runs on AIX or OpenVMS i.e.. not Windows or  HP-UX or Solaris.   E Reference recent (now HP format) OpenVMS Health testimonial brochures  published in 2003:   Cerner= http://h71000.www7.hp.com/openvms/brochures/cerner/cerner.pdf 3 http://h71000.www7.hp.com/openvms/brochures/cerner/    IDX G http://h71000.www7.hp.com/openvms/brochures/dartmouth/css/dartmouth.htm C http://h71000.www7.hp.com/openvms/brochures/dartmouth/dartmouth.pdf @ http://h71000.www7.hp.com/openvms/brochures/idx/ (Compaq format)   Regards   
 Kerry Main Senior Consultant  Hewlett-Packard (Canada) Co.! Consulting & Integration Services  Voice: 613-592-4660  Fax   : 613-591-4477 Email: kerryDOTmain@hpDOTcom-     (remove the DOT's and replace with "."'s)  OpenVMS DCL - the original .COM      -----Original Message-----C From: Brian Tillman [mailto:Brian.Tillman@notnoone.notnohow.com]=20  Sent: March 20, 2003 2:27 PM To: Info-VAX@Mvb.Saic.Com 8 Subject: An opportunity for HP? (wes: Micromedex on VMS)    I >For those using Micromedex on VMS,  today I received information that=20 @ >this product will soon be no longer available on Alpha OpenVMS.  E Sue, can you get anyone to do something about this?  Right before our E eyes, infrastructure for VMS is crumbling, and in one of the area VMS # has been so strong in: health care.  --=20 E Brian Tillman         Internet: Brian.Tillman at smiths-aerospace dot  com 5 Smiths Aerospace  Addresses modified to prevent SPAM. F 3290 Patterson Ave. SE, MS Replace "at" with "@", "dot" with "." Grand Rapids, MI 49512-1991 8        This opinion doesn't represent that of my company   ------------------------------  % Date: Fri, 21 Mar 2003 09:31:22 -0500 1 From: "Lonnie Blevins" <lblevins@regenstrief.org> < Subject: Re: An opportunity for HP? (wes: Micromedex on VMS)0 Message-ID: <b5f7nu$quc$1@hood.uits.indiana.edu>  G Both IDX and CERNER use (where I have seent them) workstations that are L WINDOWS machines.  Only the servers are AIX or OpenVMS.  Users still see the6 WINDOWS interface and think of that as "their system".  2 "Main, Kerry" <Kerry.Main@hp.com> wrote in messageL news:BE56C50EA024184DAF48F0B9A47F5CF402660E7A@kaoexc01.americas.cpqcorp.net. .. Brian,  = Re: Healthcare ... IDX and Cerner are growing with OpenVMS ..   G Btw - remember, Cerner only runs on AIX or OpenVMS i.e.. not Windows or  HP-UX or Solaris.   E Reference recent (now HP format) OpenVMS Health testimonial brochures  published in 2003:   Cerner= http://h71000.www7.hp.com/openvms/brochures/cerner/cerner.pdf 3 http://h71000.www7.hp.com/openvms/brochures/cerner/    IDX G http://h71000.www7.hp.com/openvms/brochures/dartmouth/css/dartmouth.htm C http://h71000.www7.hp.com/openvms/brochures/dartmouth/dartmouth.pdf @ http://h71000.www7.hp.com/openvms/brochures/idx/ (Compaq format)   Regards   
 Kerry Main Senior Consultant  Hewlett-Packard (Canada) Co.! Consulting & Integration Services  Voice: 613-592-4660  Fax   : 613-591-4477 Email: kerryDOTmain@hpDOTcom-     (remove the DOT's and replace with "."'s)  OpenVMS DCL - the original .COM      -----Original Message-----@ From: Brian Tillman [mailto:Brian.Tillman@notnoone.notnohow.com] Sent: March 20, 2003 2:27 PM To: Info-VAX@Mvb.Saic.Com 8 Subject: An opportunity for HP? (wes: Micromedex on VMS)    F >For those using Micromedex on VMS,  today I received information that@ >this product will soon be no longer available on Alpha OpenVMS.  E Sue, can you get anyone to do something about this?  Right before our E eyes, infrastructure for VMS is crumbling, and in one of the area VMS # has been so strong in: health care.  --E Brian Tillman         Internet: Brian.Tillman at smiths-aerospace dot  com 5 Smiths Aerospace  Addresses modified to prevent SPAM. F 3290 Patterson Ave. SE, MS Replace "at" with "@", "dot" with "." Grand Rapids, MI 49512-1991 8        This opinion doesn't represent that of my company   ------------------------------  # Date: Fri, 21 Mar 2003 15:15:21 GMT + From: LESLIE@JRLVAX.HOUSTON.RR.COM (leslie) < Subject: Re: An opportunity for HP? (wes: Micromedex on VMS): Message-ID: <d0Gea.66591$eb.1971681@twister.austin.rr.com>  & Main, Kerry (Kerry.Main@hp.com) wrote: : Brian, : ? : Re: Healthcare ... IDX and Cerner are growing with OpenVMS ..  :   7 I'm the one who brought up IDX -- thanks for the update   2 --Jerry Leslie   (my opinions are strictly my own)9   Note: leslie@jrlvax.houston.rr.com is invalid for email    ------------------------------  % Date: Fri, 21 Mar 2003 18:37:59 -0000 - From: "Steve Spires" <Steve.Spires@torex.com> < Subject: RE: An opportunity for HP? (wes: Micromedex on VMS)E Message-ID: <91947A84607D9D48B8E674A5FAB54DA6854470@tahiti.tinuk.com>   B Well, we sell health systems on VMS and tru64, and most people useD either VTs or PC emulators. Unless you were talking about emulators.  , Perhaps an opportunity for our company then?   ;^D    Steve Spires Technical Consultant Torex Health [T] +44 (0) 1295 274200  [F] +44 (0) 1295 275131 
 www.torex.com    >>-----Original Message-----; >>From: Lonnie Blevins [mailto:lblevins@regenstrief.org]=20  >>Sent: 21 March 2003 14:31  >>To: Info-VAX@Mvb.Saic.Com > >>Subject: Re: An opportunity for HP? (wes: Micromedex on VMS) >> >>6 >>Both IDX and CERNER use (where I have seent them)=20B >>workstations that are WINDOWS machines.  Only the servers are=20? >>AIX or OpenVMS.  Users still see the WINDOWS interface and=20 " >>think of that as "their system". >>7 >>"Main, Kerry" <Kerry.Main@hp.com> wrote in message=20 @ >>news:BE56C50EA024184DAF48F0B9A47F5CF402660E7A@kaoexc01.america >>s.cpqcorp.net. >>.. >>Brian, >>? >>Re: Healthcare ... IDX and Cerner are growing with OpenVMS ..  >>A >>Btw - remember, Cerner only runs on AIX or OpenVMS i.e.. not=20  >>Windows or HP-UX or Solaris. >>@ >>Reference recent (now HP format) OpenVMS Health testimonial=20 >>brochures published in 2003: >>F >>Cerner http://h71000.www7.hp.com/openvms/brochures/cerner/cerner.pdf5 >>http://h71000.www7.hp.com/openvms/brochures/cerner/  >> >>IDX=20@ >>http://h71000.www7.hp.com/openvms/brochures/dartmouth/css/dart	 mouth.htm C http://h71000.www7.hp.com/openvms/brochures/dartmouth/dartmouth.pdf @ http://h71000.www7.hp.com/openvms/brochures/idx/ (Compaq format)   Regards   
 Kerry Main Senior Consultant  Hewlett-Packard (Canada) Co.! Consulting & Integration Services  Voice: 613-592-4660  Fax   : 613-591-4477 Email: kerryDOTmain@hpDOTcom-     (remove the DOT's and replace with "."'s)  OpenVMS DCL - the original .COM      -----Original Message-----@ From: Brian Tillman [mailto:Brian.Tillman@notnoone.notnohow.com] Sent: March 20, 2003 2:27 PM To: Info-VAX@Mvb.Saic.Com 8 Subject: An opportunity for HP? (wes: Micromedex on VMS)    I >For those using Micromedex on VMS,  today I received information that=20 @ >this product will soon be no longer available on Alpha OpenVMS.  E Sue, can you get anyone to do something about this?  Right before our E eyes, infrastructure for VMS is crumbling, and in one of the area VMS # has been so strong in: health care.  --E Brian Tillman         Internet: Brian.Tillman at smiths-aerospace dot  com 5 Smiths Aerospace  Addresses modified to prevent SPAM. F 3290 Patterson Ave. SE, MS Replace "at" with "@", "dot" with "." Grand Rapids, MI 49512-1991 8        This opinion doesn't represent that of my company   ------------------------------   Date: 21 Mar 03 16:17:48 +0100) From: p_sture@elias.decus.ch (Paul Sture) " Subject: Re: Carly's feedback link) Message-ID: <FgGgjStEohg0@elias.decus.ch>   [ In article <3E7A7CF3.6FCD9FA6@fsi.net>, "David J. Dachtera" <djesys.nospam@fsi.net> writes:  > Phillip Helbig wrote:  >>  I >> > Funny how when anyone else has them, they're WMD's, yet when America 0 >> > has them they are simply 'nuclear devices'. >>  J >> Close, but no cigar.  I believe "nuculer" is the correct pronunciation. >  > What's a nuculus?  > ; > (Hint: It's nucleus; hence, nuclear as in "NEW-Clee-er".)  >   ; Be prepared for an unpleasant surprise. Tap "nuclear" into  D http://m-w.com and then click on the third sound button to the right of "Main Entry".   :-(   * Sorry, it wants Javascript enabled too :-(  E Which reminds me. For those interested, I have at last found a decent H online dictionary site, which does both US and UK spellings and meanings  (and works without Javascript) :    http://dictionary.cambridge.org/   --  
 Paul Sture   ------------------------------  # Date: Fri, 21 Mar 2003 18:31:11 GMT 7 From: brad@.homeportal.2wire.net (Bradford J. Hamilton) " Subject: Re: Carly's feedback link? Message-ID: <PTIea.191760$6b3.522447@rwcrnsc51.ops.asp.att.net>   U In article <FgGgjStEohg0@elias.decus.ch>, p_sture@elias.decus.ch (Paul Sture) writes:  > < >Be prepared for an unpleasant surprise. Tap "nuclear" into E >http://m-w.com and then click on the third sound button to the right  >of "Main Entry".  >  >:-(  M So does that mean that if the president jumps off a cliff, that we should all L follow him???  Just because "educated people" pronounce it this way, doesn't make it right.   :-)   N Thanks to my mom for teaching me to do the right thing, no matter what someone else may tell me!    <snip> >--  >Paul Sture   A _________________________________________________________________ 0 Bradford J. Hamilton			"All opinions are my own"/ bMradAhamiPltSon@atMtAbi.cPoSm		"Lose the MAPS"    ------------------------------  + Date: Fri, 21 Mar 2003 12:06:45 +0100 (MET) 9 From: Phillip Helbig <HELBPHI@sysdev.deutsche-boerse.com> 3 Subject: Re: Cluster, half in the DMZ, half outside ; Message-ID: <01KTSAJI11OY9S5MQI@sysdev.deutsche-boerse.com>   H > What exactly is the DMZ ? Is that an official networking term, or justI > some terminology used for one consumer grade home router to the default F > host to receice call requests to ports that are not specified in the > router's SUA table ?    H That's what it appears to be in my "consumer-grade" routers.  According I to another post here, it actually has a stricter definition.  Perhaps my  H routers have a real DMZ, or perhaps they are using the term incorrectly.  I > If you have a NAT/PAT router (such as home routers), then you just need J > to route all incoming calls to the one host on your LAN that you want toE > be "public". Folks from the outside will not be able to reach other 
 > hosts.     Right.    - > Of course, you cannot use a cluster alias.    I What do you mean here?  The "host" to which one forwards things could be  G the cluster alias!  Of course, with a cluster alias, the cluster-alias  F address is used only for INCOMING connections.  However, such routers 1 will allow OUTGOING connections from any address.   H > If you have a real internet connection with fixed IP adresses, one wayI > to insulate part of your cluster is to give "internet" IP adresses only I > to the node you want to be reacheable, and then give the other nodes in I > the cluster local, non-routable IP adresses (such as 10.*). There is no H > way for someone on the internet to reach a node on your lan that has aG > non-routable IP address if you have a real connection to the internet  > without a NAT router.    ------------------------------  + Date: Fri, 21 Mar 2003 12:09:13 +0100 (MET) 9 From: Phillip Helbig <HELBPHI@sysdev.deutsche-boerse.com> 3 Subject: Re: Cluster, half in the DMZ, half outside ; Message-ID: <01KTSAPN6FCO9S5MQI@sysdev.deutsche-boerse.com>   E > > What exactly is the DMZ ? Is that an official networking term, or H > > just some terminology used for one consumer grade home router to tehI > > default host to receice call requests to ports that are not specified  > > in the router's SUA table ?  > H > DeMilitarized Zone- just like in Korea. It's a network segment that's  > physically isolated   B With my router, one can enter the address for the DMZ host in the G configuration menu; since the DMZ host is set in software, it can't be  ) on a PHYSICALLY isolated network segment.   0 > from both the outside world and your internal E > network, subject to access from either direction by a firewall and  6 > rules. This is not lightweight consumer-grade stuff.  F It could be that some lightweight consumer-grade routers (mis)use the   term to indicate a default host.  J > Webservers, mail servers, anything that you would want to be accessible H > from the world at large should go here, where it's isolated from your H > internal systems in case the box gets cracked, and only protocols you ; > want to and from your servers get in/out, as appropriate.    ------------------------------  % Date: Fri, 21 Mar 2003 12:39:18 +0100 ) From: Bart Zorn <B.Zorn@xs4all.nospam.nl> 3 Subject: Re: Cluster, half in the DMZ, half outside 6 Message-ID: <3e7af9e7$0$49105$e4fe514c@news.xs4all.nl>  I DMZ is purely a TCP/IP concept. OpenVMS clusters do not use (TCP/)IP for  I cluster communications. If you can configure your router and/or firewall  C to just bridge non-IP packets, a cluster should be able to use the   network as interconnect.  	 Bart Zorn    Phillip Helbig wrote: H >>What exactly is the DMZ ? Is that an official networking term, or justI >>some terminology used for one consumer grade home router to the default F >>host to receice call requests to ports that are not specified in the >>router's SUA table ?   >  > J > That's what it appears to be in my "consumer-grade" routers.  According K > to another post here, it actually has a stricter definition.  Perhaps my  J > routers have a real DMZ, or perhaps they are using the term incorrectly. >  > I >>If you have a NAT/PAT router (such as home routers), then you just need J >>to route all incoming calls to the one host on your LAN that you want toE >>be "public". Folks from the outside will not be able to reach other 
 >>hosts.   >  > 
 > Right.   >  > - >>Of course, you cannot use a cluster alias.   >  > K > What do you mean here?  The "host" to which one forwards things could be  I > the cluster alias!  Of course, with a cluster alias, the cluster-alias  H > address is used only for INCOMING connections.  However, such routers 3 > will allow OUTGOING connections from any address.  >  > H >>If you have a real internet connection with fixed IP adresses, one wayI >>to insulate part of your cluster is to give "internet" IP adresses only I >>to the node you want to be reacheable, and then give the other nodes in I >>the cluster local, non-routable IP adresses (such as 10.*). There is no H >>way for someone on the internet to reach a node on your lan that has aG >>non-routable IP address if you have a real connection to the internet  >>without a NAT router.  >    ------------------------------  + Date: Fri, 21 Mar 2003 16:15:08 +0000 (UTC) + From: david20@alpha2.mdx.ac.uk (David Webb) 3 Subject: Re: Cluster, half in the DMZ, half outside + Message-ID: <b5fdqc$dbn$1@aquila.mdx.ac.uk>   a In article <5L62oNiGh8uA@eisner.encompasserve.org>, young_r@encompasserve.org (Rob Young) writes: Z >In article <b5cdt1$8va$1@aquila.mdx.ac.uk>, david20@alpha2.mdx.ac.uk (David Webb) writes:l >> In article <863f19d6.0303200410.43294e8f@posting.google.com>, jbrankin@ntlworld.com (Jim Brankin) writes:H >>>Has anyone built a cluster with some of the nodes in the DMZ and some >>>outside of it?  >>> G >>>The nodes in the DMZ would be web servers, obviously, and the others H >>>back end database engines. All the machines would run a cluster-aware@ >>>database product and there would be at least one shared disk. >>>  >>  6 >> There is your problem - "at least one shared disk".L >> Pretty pointless building a cluster if you don't share any disks but that( >> shared disk is a major security hole.O >> Someone who hacked into one of the external VMS cluster members can then put O >> files onto that shared disk. He then just needs to get a commandfile he puts O >> on that shared disk run on one of the internal systems. According to how the O >> cluster is setup this might be as easy as submitting a batch job to a queue  O >> running on one of the internal systems. Also although the shared disk is the P >> only one mounted will any of the other disks be visible ie can a hacker mount >> any of the other disks ?  >>   > I >	I wonder if it would be as simple as all web server accounts and others G >	in use in the DMZ have no INTERACTIVE access, nor BATCH.  It appears  - >	from the description that it may be doable.  > G I was rather assuming that his security people would be looking at what D exposure there was for the internal backend systems if the front endK webservers suffered a (unix terminology) root compromise ie someone taking  7 over the system completely and gaining full privileges. N Since they are VMS systems this is probably much less likely that with Unix or. Windows systems but it is far from impossible.  K I would think it would be very difficult to secure other cluster nodes from D compromise from  a fully privileged user on one node of the cluster.O I would also think it would be impossible to prove that such a setup was really 1 secure to suspicious security people or auditors.   
 David Webb VMS and Unix team leader CCSS Middlesex University      7 >	Sprinkle via IDENTIFIERS to cut down on access (even  A >	accidental access, i.e. you attempt to log into SYSTEM on a DMZ : >	box.  However, your IDENTIFIER for the SYSTEM account is? >	INTERNAL_ACCESS_ONLY and prevents you from going into the DMZ 	 >	boxes).  >  >		EXTERNAL_ACCESS_ONLY  >		BOTH_ACCESS >		INTERNAL_ACCESS_ONLY  > = >	via checks in SYLOGIN.COM, RESTRICTED and CAPTIVE flags set  >	where appropriate, etc.  >  >	I would do a few things.   > & >			1)  SSH access to privved accounts > 3 >			2)  Fine grained auditing with automated alarms  > 6 >			3)  Turning off TCP/IP services not in use (normal3 >			    stuff) and further restricting access.  For 9 >			    instance, you could help a great deal by allowing 3 >			    telnet from only a handful of IP addresses: 
 >Multinet: >  >	Addresses below are fake.  >  >SERVER-CONFIG>sho telnet/full >Service "TELNET":2 >        TCP socket (AF_INET,SOCK_STREAM), Port 23& >        Socket Options = SO_KEEPALIVE >        INIT() = Merge_Image  >        LISTEN() = TCP_Listen$ >        CONNECTED() = TCP_Connected$ >        SERVICE() = Internal_Telnet6 >        Program = "MULTINET:LOADABLE_KERBEROS_TELNET"M >        Accept Hosts = IP-127.0.0.1 , IP-199.224.101.98 , IP-192.227.245.135  >        Accept Nets = "" 3 >        Reject by default all other hosts and nets  > > >	Especially if 1) is impractical (i.e. using TCPIP Services). > = >	I would have a high level of confidence this could be done  9 >	well especially if it is easy to identify all accesses.  >  >				Rob >    ------------------------------  % Date: Fri, 21 Mar 2003 11:39:51 -0500 5 From: "Bochnik, William J" <William_Bochnik@acml.com> 3 Subject: RE: Cluster, half in the DMZ, half outside O Message-ID: <2D75787AAF09C64481BDFD89113BE6D5B36547@ac2kama0102.ac.lp.acml.com>   C In light of all the issues, how about these not being in a cluster?    -----Original Message-----; From: jbrankin@ntlworld.com [mailto:jbrankin@ntlworld.com]  & Sent: Thursday, March 20, 2003 7:10 AM To: Info-VAX@Mvb.Saic.Com / Subject: Cluster, half in the DMZ, half outside     E Has anyone built a cluster with some of the nodes in the DMZ and some  outside of it?  I The nodes in the DMZ would be web servers, obviously, and the others back I end database engines. All the machines would run a cluster-aware database 4 product and there would be at least one shared disk.  H Since the cluster interconnect would connect machines in the DMZ and theA internal network without going through the inner firewall router  J there would be problems convincing the security people it was OK. However,< if you ensure that the cluster interconnect is not carrying H IP or DECnet traffic then I do not see that there is a security problem.  J Has anyone done this? Does anyone see problems with it? What do you think?   - Jim     I The information contained in this transmission may contain privileged and J confidential information and is intended only for the use of the person(s)L named above.  If you are not the intended recipient, or an employee or agentF responsible for delivering this message to the intended recipient, anyK review, dissemination, distribution or duplication of this communication is J strictly prohibited. If you are not the intended recipient, please contactD the sender immediately by reply e-mail and destroy all copies of theI original message. Please note that we do not accept account orders and/or J instructions by e-mail, and therefore will not be responsible for carrying$ out such orders and/or instructions.   ------------------------------  % Date: Fri, 21 Mar 2003 11:56:13 -0400 0 From: JF Mezei <jfmezei.spamnot@vl.videotron.ca>3 Subject: Re: Cluster, half in the DMZ, half outside / Message-ID: <3E7B361D.F30A1E20@vl.videotron.ca>    Phillip Helbig wrote: J > What do you mean here?  The "host" to which one forwards things could be > the cluster alias!    0 Cluster alias is a DNS "trick" from what I read.  N Routers, at least the consumer grade ones, route to a specific IP address, notN a host name. So your inbound call going through the consumer grade router will go to a specific IP.  J Now, if you have a real internet connection with your own DNS server, thenK when "foreign" folks want to connect to www.chocolate.com , then it is your G DNS which will decide to which host in the cluster this will translate.   N But when you are in a consumer grade internet connection (read: crippled) withJ a consumer grade router (NAT), then  the cluster alias is of not much use.   ------------------------------    Date: 21 Mar 2003 11:14:25 -0600 From: briggs@encompasserve.org3 Subject: Re: Cluster, half in the DMZ, half outside 3 Message-ID: <fhnI60uaqHc3@eisner.encompasserve.org>   b In article <3E7B361D.F30A1E20@vl.videotron.ca>, JF Mezei <jfmezei.spamnot@vl.videotron.ca> writes: > Phillip Helbig wrote: K >> What do you mean here?  The "host" to which one forwards things could be  >> the cluster alias!  > 2 > Cluster alias is a DNS "trick" from what I read.  . No.  It's not a DNS trick.  It's an ARP trick.  P > Routers, at least the consumer grade ones, route to a specific IP address, notP > a host name. So your inbound call going through the consumer grade router will > go to a specific IP.  H But if you engage in trickery at the ARP level, one IP can point to more than one machine.    	John Briggs   ------------------------------  + Date: Fri, 21 Mar 2003 17:16:03 +0000 (UTC) + From: david20@alpha2.mdx.ac.uk (David Webb) 3 Subject: Re: Cluster, half in the DMZ, half outside + Message-ID: <b5fhcj$dne$1@aquila.mdx.ac.uk>   b In article <3e7af9e7$0$49105$e4fe514c@news.xs4all.nl>, Bart Zorn <B.Zorn@xs4all.nospam.nl> writes:J >DMZ is purely a TCP/IP concept. OpenVMS clusters do not use (TCP/)IP for J >cluster communications. If you can configure your router and/or firewall D >to just bridge non-IP packets, a cluster should be able to use the  >network as interconnect.  > F However by doing so you will probably defeat the whole point of havingK a DMZ. A DMZ is used to protect vulnerable systems which must be visible to L the outside world such as webservers  whilst also strictly controlling theirN access to internal backend servers such as databases. The idea is that even ifJ the systems in the DMZ are totally compromised the paths by which they canN attack the internal systems are severely curtailed by the restrictions imposed9 by the firewall between the DMZ and the internal systems.   M If you can prove that you can configure a cluster such that a privileged user O on a node in the DMZ cannot leverage those privileges to gain privileged access O on cluster members on the internal network then OK but I would think that would  be almost impossible to do.   
 David Webb VMS and Unix team leader CCSS Middlesex University          
 >Bart Zorn >  >Phillip Helbig wrote:I >>>What exactly is the DMZ ? Is that an official networking term, or just J >>>some terminology used for one consumer grade home router to the defaultG >>>host to receice call requests to ports that are not specified in the  >>>router's SUA table ?  >>   >>  K >> That's what it appears to be in my "consumer-grade" routers.  According  L >> to another post here, it actually has a stricter definition.  Perhaps my K >> routers have a real DMZ, or perhaps they are using the term incorrectly.  >>   >>  J >>>If you have a NAT/PAT router (such as home routers), then you just needK >>>to route all incoming calls to the one host on your LAN that you want to F >>>be "public". Folks from the outside will not be able to reach other >>>hosts.    >>   >>   >> Right.    >>   >>  . >>>Of course, you cannot use a cluster alias.  >>   >>  L >> What do you mean here?  The "host" to which one forwards things could be J >> the cluster alias!  Of course, with a cluster alias, the cluster-alias I >> address is used only for INCOMING connections.  However, such routers  4 >> will allow OUTGOING connections from any address. >>   >>  I >>>If you have a real internet connection with fixed IP adresses, one way J >>>to insulate part of your cluster is to give "internet" IP adresses onlyJ >>>to the node you want to be reacheable, and then give the other nodes inJ >>>the cluster local, non-routable IP adresses (such as 10.*). There is noI >>>way for someone on the internet to reach a node on your lan that has a H >>>non-routable IP address if you have a real connection to the internet >>>without a NAT router.   >>   >    ------------------------------  + Date: Fri, 21 Mar 2003 17:44:56 +0000 (UTC) + From: david20@alpha2.mdx.ac.uk (David Webb) 3 Subject: Re: Cluster, half in the DMZ, half outside + Message-ID: <b5fj2o$dt6$1@aquila.mdx.ac.uk>   b In article <3E7B361D.F30A1E20@vl.videotron.ca>, JF Mezei <jfmezei.spamnot@vl.videotron.ca> writes: >Phillip Helbig wrote:K >> What do you mean here?  The "host" to which one forwards things could be  >> the cluster alias!  > 1 >Cluster alias is a DNS "trick" from what I read.  > 7 There are a number of different types of Cluster alias.        1) DNS round-robin  M Uses DNS and provides load sharing but will still try and connect to a system  which is down ie no failover.   O 2) Load balancing. Uses DNS and dynamic updates to the DNS to provide true load  balancing and failover.    and    3) IP address Cluster alias.  M This provides no load balancing just failover and does NOT involve the use of 
 a DNS server.   N The systems in the cluster are given an extra IP address which is bound to oneL of the systems. All connections to that address are accepted by that system.J When that system is shutdown or crashes the extra IP address automaticallyL moves to one of the other systems in the cluster which then begins accepting connections on this address.      O >Routers, at least the consumer grade ones, route to a specific IP address, not O >a host name. So your inbound call going through the consumer grade router will  >go to a specific IP.  >   = Which can be the third type of Cluster Alias mentioned above.     
 David Webb VMS and Unix team leader CCSS Middlesex University    K >Now, if you have a real internet connection with your own DNS server, then L >when "foreign" folks want to connect to www.chocolate.com , then it is yourH >DNS which will decide to which host in the cluster this will translate. > O >But when you are in a consumer grade internet connection (read: crippled) with K >a consumer grade router (NAT), then  the cluster alias is of not much use.    ------------------------------    Date: 21 Mar 2003 12:31:04 -0600+ From: young_r@encompasserve.org (Rob Young) 3 Subject: Re: Cluster, half in the DMZ, half outside 3 Message-ID: <Pb3V4DRV3806@eisner.encompasserve.org>   Y In article <b5fj2o$dt6$1@aquila.mdx.ac.uk>, david20@alpha2.mdx.ac.uk (David Webb) writes: d > In article <3E7B361D.F30A1E20@vl.videotron.ca>, JF Mezei <jfmezei.spamnot@vl.videotron.ca> writes: >>Phillip Helbig wrote: L >>> What do you mean here?  The "host" to which one forwards things could be >>> the cluster alias!   >>2 >>Cluster alias is a DNS "trick" from what I read. >>9 > There are a number of different types of Cluster alias.  >  >  >  > 1) DNS round-robin > O > Uses DNS and provides load sharing but will still try and connect to a system  > which is down ie no failover.  > Q > 2) Load balancing. Uses DNS and dynamic updates to the DNS to provide true load  > balancing and failover.  >  > and  >  > 3) IP address Cluster alias. > O > This provides no load balancing just failover and does NOT involve the use of  > a DNS server.  > P > The systems in the cluster are given an extra IP address which is bound to oneN > of the systems. All connections to that address are accepted by that system.L > When that system is shutdown or crashes the extra IP address automaticallyN > moves to one of the other systems in the cluster which then begins accepting > connections on this address. >   8 	There is enough noise in this newsgroup.  But sometimes? 	praise has to be given too.  Those were very good descriptions 
 	and answers.    				Rob    ------------------------------  # Date: Fri, 21 Mar 2003 08:21:39 GMT A From: "Colin Butcher" <colin_DOT.butcher_AT@xdelta_DOT.co_DOT.uk> * Subject: Re: DECserver 90TL docs and image= Message-ID: <nYzea.1620$Zt6.13573190@news-text.cableinet.net>   K http://www.dnpg.com/dr/hubs/ is a good starting point for info. No idea how . to get there from elsewhere on their web site!  H The operational images live in the various DECserver kits on the layeredH product CDs, eg: NA7015. They're licensed software, so not available forI free download. Go back to the layered product CDs around the late 90s and  work forwards.   -- Hope this helps. Cheers, Colin.(' (colinDOT.butcherAT@xdeltaDOT.coDOT.uk)i   ------------------------------  % Date: Fri, 21 Mar 2003 09:23:19 +0000o( From: Nic Clews <sendspamhere@127.0.0.1>$ Subject: Re: Demagnetize Disk Drives( Message-ID: <3E7ADA07.C24EEFA@127.0.0.1>   Marc Van Dyck wrote: >   ; > Or you can pay H.P. a fee for not returning broken disks. = > They call that the "defective materials retention" service.n7 > That's what we are doing. When a disk is broken, H.P.f: > replaces it but we keep the broken one, and drill a hole2 > through it, before sending them to the dumpster.  C We do that. Faulty item is replaced, but allows customer controlledeF destruction of the faulty unit. (Due to the overall reliability, it is not an excessive uplift).i   -- e? Regards, Nic Clews a.k.a. Mr. CP Charges, CSC Computer SciencesH nclews at csc dot come   ------------------------------  % Date: Fri, 21 Mar 2003 09:52:37 +0000 ( From: Nic Clews <sendspamhere@127.0.0.1>$ Subject: Re: Demagnetize Disk Drives) Message-ID: <3E7AE0E5.E3320732@127.0.0.1>    Paul Repacholi wrote:a >   E > As long as they stay within your Co, a simple init/erase is all youf@ > really need, unless it is data that they *really * care about. >   B Well no not quite actually. The device holding the data receives aG classification, and a device may pass *up* the chain of classification,rH from "nothing" (sic) through "restricted" to "secret high" (which is NOTH an exhaustive list which describes intermediate and also higher levels),A but a device may never pass down the chain, destruction is not anT! alternative, it is a requirement.   C We deal with clients that have similar security classifications andhF perhaps very obviously, cross usage of the same classification is alsoF utterly prohibited, from a security standpoint. (It should be obvious, but just in case!)  H The transportation requirements of data and devices (or media) that holdG data is a very involved process, even if the data has been 'informally' F made inaccessible. It's something that in general is avoided. It's not= much fun even going to the lavatory doing this sort of thing.   D Interestingly, as you might expect you have to check in with certainA type of police during the journey, and I never did get asked more G questions about the start and end time of the journey, and the distance-B between the locations involved. (Escortation not required for thisH level). Still, I'm more relaxed and sensible now, and it was a few years ago...  C (I no longer officially recall details and the driving was shared).a   -- r? Regards, Nic Clews a.k.a. Mr. CP Charges, CSC Computer Sciences  nclews at csc dot com    ------------------------------  % Date: Fri, 21 Mar 2003 09:11:44 -0500 A From: "Fred Kleinsorge" <kleinsorge@star-dot-zko-dot-dec-dot-com>tF Subject: Re: EFI Specifications (was: I Have Seen The IA64 Future !!!). Message-ID: <3e7b1da0$1_2@hpb10302.boi.hp.com>  = "JF Mezei" <jfmezei.spamnot@vl.videotron.ca> wrote in message1) news:3E79EEF3.88DCB053@vl.videotron.ca.../ > Fred Kleinsorge wrote: > >0L > > Maybe, I wasn't part of the discussion.  But they contributed the ILP toI > > allow it's use.  FAT is sufficiently simple to make it attractive fori this > > specific use.a >r > A few questions: >iF > Is it correct to state that you could have VMS without that FAT file	 stored onyF > the system disk and do basic system stuff such as boot with just the vanillah > EFI stuff available in ROM ? >    Sure, from the network.i  H The OS primary bootstrap lives in the FAT partition.  It is a file basedF boot, not a boot-block based boot.  This, and the partitioning record,J allows multiple-OS disks.  While *we* won't support that (for a while), it
 is useful.  H > Will EFI programs (either the vanilla ROM-based, and the extended ones storedJ > on disk) ever write some files to that FAT partition ? (for instance, to stores > configuration information ?) >i   Yes.  K > When the vanilla (rom based) EFI accesses the FAT partition, does it have K > enough smarts to underestand the FAT file format, with file names etc, or  justL > enough to lookup a few pointers to find the address of code to branch to ? >   + The former.  EFI knows all about FAT files.   G > In other words, if, from the vanilla >>> prompt, you want to boot the-K > extended, smarter EFI program from the FAT, how will you specify that you" want3 > to boot the extended EFI instead of booting VMS ?o  H EFI, by default, gives you a boot menu that allows you to choose what toD load.  You can load the "shell" application, which is a command lineI interface for EFI.  Or you can load an OS_LOADER, like the VMS_LOADER.EFIs  K > Will you specify a FAT file name in the boot command, or just some numberK@ > (similar to specifying 0 or 1 to boot from SYS0 or SYS1 in VMS
 terminology).0 >0  L There is no "boot command", think of it like a NT boot menu.  You can get toK something (the shell) that you can boot from if you want.  The devices haverL the (horrible) but familiar pathname format.  VMS is considering providing aH VMS-friendly "boot manager" which will give you a command line interfaceH that will convert the pathnames (for example) to VMS-friendly names like DKA0..  D > Will VMS have utilities to access that FAT container file ? (get a
 directory,L > read/write files to it ?) If so, will it be made general enough so that we> > could use this to write FAT diskettes or access DOS drives ?  I Not clear.  Yes, we will have a utility to update the FAT container.  But G that does not mean that this will be a general utility for handling any6H generic FAT formatted disk.  It is *just* for handling the FAT container file.    ------------------------------  % Date: Fri, 21 Mar 2003 12:05:51 -0400O0 From: JF Mezei <jfmezei.spamnot@vl.videotron.ca>F Subject: Re: EFI Specifications (was: I Have Seen The IA64 Future !!!)/ Message-ID: <3E7B385E.2ADBBA17@vl.videotron.ca>o   Fred Kleinsorge wrote:N > There is no "boot command", think of it like a NT boot menu.  You can get toM > something (the shell) that you can boot from if you want.  The devices haveyN > the (horrible) but familiar pathname format.  VMS is considering providing aJ > VMS-friendly "boot manager" which will give you a command line interfaceJ > that will convert the pathnames (for example) to VMS-friendly names like > DKA0.   H If you use DOS disk names, does this mean that VMS will be limited to 24N drives when running on IA64 ? What happens when you want to boot from the 25thE disk drive ?  Will A: and D: be assigned to diskette and cdrom drives-L respectively ? How will the hardware determine which drive is the C: drive ?  K Once booted, will VMS have to continue to translate (at very low level) theiK VMS device names into the intel device names ? Or does VMS get to fully map J the names to the real device, at which point there would no longer be a 24 disk drive limit ?   ------------------------------  % Date: Fri, 21 Mar 2003 18:50:05 +0100r$ From: Michael Unger <unger@decus.de>F Subject: Re: EFI Specifications (was: I Have Seen The IA64 Future !!!)+ Message-ID: <00A1D363.26CA9A1F.19@decus.de>-  B "Fred Kleinsorge" <kleinsorge@star-dot-zko-dot-dec-dot-com> wrote:  ? > "JF Mezei" <jfmezei.spamnot@vl.videotron.ca> wrote in message + > news:3E79EEF3.88DCB053@vl.videotron.ca...9 >  > [...]B > J > The OS primary bootstrap lives in the FAT partition.  It is a file basedH > boot, not a boot-block based boot.  This, and the partitioning record,L > allows multiple-OS disks.  While *we* won't support that (for a while), it > is useful.  I To whom? I don't want VMS and Winwoes to co-reside on the very same disk 2 device.w   > [...]  > M > > Will you specify a FAT file name in the boot command, or just some numbereB > > (similar to specifying 0 or 1 to boot from SYS0 or SYS1 in VMS > terminology).: > >p > N > There is no "boot command", think of it like a NT boot menu.  You can get toM > something (the shell) that you can boot from if you want.  The devices havevN > the (horrible) but familiar pathname format.  VMS is considering providing aJ > VMS-friendly "boot manager" which will give you a command line interfaceJ > that will convert the pathnames (for example) to VMS-friendly names like > DKA0.  >  > [...]e  I Considering all the possible combinations of boot flags and devices this OF "menu" has to be displayed on a "multi-headed" graphics screen ... :-)  @ And only "considering providing a VMS-friendly boot manager" ???   Michaela   ------------------------------  + Date: Fri, 21 Mar 2003 17:48:51 +0000 (UTC)t+ From: david20@alpha2.mdx.ac.uk (David Webb)eF Subject: Re: EFI Specifications (was: I Have Seen The IA64 Future !!!)+ Message-ID: <b5fja3$dt7$1@aquila.mdx.ac.uk>m  b In article <3E7B385E.2ADBBA17@vl.videotron.ca>, JF Mezei <jfmezei.spamnot@vl.videotron.ca> writes: >Fred Kleinsorge wrote:.O >> There is no "boot command", think of it like a NT boot menu.  You can get toiN >> something (the shell) that you can boot from if you want.  The devices haveO >> the (horrible) but familiar pathname format.  VMS is considering providing arK >> VMS-friendly "boot manager" which will give you a command line interface K >> that will convert the pathnames (for example) to VMS-friendly names likeo >> DKA0. >oI >If you use DOS disk names, does this mean that VMS will be limited to 24pO >drives when running on IA64 ? What happens when you want to boot from the 25thSF >disk drive ?  Will A: and D: be assigned to diskette and cdrom drivesM >respectively ? How will the hardware determine which drive is the C: drive ?b > L >Once booted, will VMS have to continue to translate (at very low level) theL >VMS device names into the intel device names ? Or does VMS get to fully mapK >the names to the real device, at which point there would no longer be a 24e >disk drive limit ?t  N And even more importantly if a disk fails or is removed will all the disknamesH change (and thereby stop VMS mounting them because of a label mismatch).  
 David Webb VMS and Unix team leader CCSS Middlesex University   ------------------------------    Date: 21 Mar 2003 12:31:41 -0600; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler)uF Subject: Re: EFI Specifications (was: I Have Seen The IA64 Future !!!)3 Message-ID: <Omm4$5k$Kgyb@eisner.encompasserve.org>e  b In article <3E7B385E.2ADBBA17@vl.videotron.ca>, JF Mezei <jfmezei.spamnot@vl.videotron.ca> writes: > Fred Kleinsorge wrote:O >> There is no "boot command", think of it like a NT boot menu.  You can get toiN >> something (the shell) that you can boot from if you want.  The devices haveO >> the (horrible) but familiar pathname format.  VMS is considering providing anK >> VMS-friendly "boot manager" which will give you a command line interfacefK >> that will convert the pathnames (for example) to VMS-friendly names like  >> DKA0. > J > If you use DOS disk names, does this mean that VMS will be limited to 24P > drives when running on IA64 ? What happens when you want to boot from the 25thG > disk drive ?  Will A: and D: be assigned to diskette and cdrom drives-N > respectively ? How will the hardware determine which drive is the C: drive ? >   G    You're confusing Intel device names with DOS device names.  Intel !=-D    DOS.  I suspect "(horrible) but familiar pathname" is much closerF    to the adapter/bus/path/node/unit naes that one typically sees when-    booting Tru64, Solaris, and other eunichs.s   ------------------------------  % Date: Fri, 21 Mar 2003 01:22:12 -0600u- From: Hunter Goatley <goathunter@goatley.com>(@ Subject: Re: How do you Detect Ethernet failure from inside code; Message-ID: <lXyea.34411$3D2.20027@fe02.atl2.webusenet.com>e   Peter Flunger wrote:/ > "Paul Hallam" <paul_hallam@hotmail.com> wrotem > ? >>Does anyone know how to detect the network failure using QIO.e >> > " > I am not sure if there is a way.@ > The folks at HP, namely the TCP/IP group, are doing some quite: > interesting and impressive stuff around detecting TCP/IP; > problems ( including interface failure ) and failing over-% > to other interfaces in such a case.05 > I am not sure if one of them is reading this group.@+ > Maybe somebody from HP can jump in here ?u > Peterf >   > At the risk of sounding like a salesman, I'd like to point outA that MultiNet and TCPware have had interface failover for severala
 years now....r   Hunter ------9 Hunter Goatley, Process Software, http://www.process.com/  goathunter@goatley.com   ------------------------------    Date: 21 Mar 2003 10:26:32 -0800+ From: c00per11242001@yahoo.ca (Vic Mendham)iE Subject: How to search for last login, but not on disusered accounts.t= Message-ID: <f7a73cb1.0303211026.230c391d@posting.google.com>r  C I have a script which searches for last logins older than x, anyonev? have any idea how to modify it so that it disregards an account  flagged as disuser?   @ Is it as simple as putting in an exclude in my search statement?  B $!This command procedure lists the last login date of each user on this system.N $!---------------------------------------------------------------------------- $! $ days = 30o $!D $! Figure out the first day of this month, the last and first day of thetC $! previous month and thirty days prior to the first of this month.o $!-------------------- $ x = 0O $FIRST_LOOP:3 $ first_date = f$cvtime("-''x'-","Absolute","Date")rB $ if f$extract(0,2,first_date) .eqs. "1-" then goto first_of_month $ x = x + 1r $ goto first_loope $! $FIRST_OF_MONTH: $ x = x + 1 6 $ last_of_prev  = f$cvtime("-''x'-","Absolute","Date") $ first_of_prev = "1" +,I f$extract(f$locate("-",last_of_prev),f$length(last_of_prev),last_of_prev)n $ x = x + days5 $ thirty_prior = f$cvtime("-''x'-","Absolute","Date")  $! $ node = f$getsyi("nodename")o> $ write sys$output "Users that have not accessed ''node' since ''thirty_prior'."a: $ write sys$output "(''days' days prior to ''first_date')" $!; $ set message/notext/noseverity/nofacility/noidentification * $ define/user sysuaf sys$system:sysuaf.dat $ mcr authorize list /full *3 $ set message/text/severity/facility/identifications$ $ searc sysuaf.lis "Username:","Last0 Login:"/out=sys$scratch:oldaccounts_searched.tmp $ delete/nolog sysuaf.lis;7 $ open/read infile sys$scratch:oldaccounts_searched.tmp  $! $ oldies = 0B $ write sys$output "Username         Last Login     Mode         "B $ write sys$output "---------------------------------------------" $LOOP: $ read infile data/end=nomores $ user = f$extract(10,16,data) $ read infile data $ inter = f$extract(12,11,data), $ batch = f$extract(45,11,data)s $!A $ if f$edit(inter,"collapse") .eqs. "" then inter = "17-Apr-1963"rA $ if f$edit(batch,"collapse") .eqs. "" then batch = "17-Apr-1963"a3 $! write sys$output "Inter=''inter':Batch=''batch'"r $!4 $ if f$cvtime("''inter'") .gts. f$cvtime("''batch'") $   then $     last_login = inter  $     last_mode  = "Interactive" $   else $     last_login = batch $     last_mode  = "Batch"	 $   endifm $!# $ if last_login .eqs. "17-Apr-1963"  $   then6 $     write sys$output f$fao("!(17AS)",user) + "never" $   else $!D $     if f$cvtime("''last_login'") .lts. f$cvtime("''thirty_prior'") $       then2 $         if f$extract(2,1,last_login) .eqs. "-" -?             then last_login = "0" + f$extract (1,10,last_login) 4 $         write sys$output f$fao("!(17AS)",user) + -+             f$fao("!(15AS)",last_login) + -s&             f$fao("!(12AS)",last_mode) $         oldies = oldies + 1 
 $       endif 	 $   endifm $ goto loopf $! $NOMORE:    B $ write sys$output "---------------------------------------------"F $ write sys$output "There are ''oldies' inactive accounts on ''node'." $ close infile5 $ delete/nolog sys$scratch:oldaccounts_searched.tmp;*c $ exit   ------------------------------  % Date: Fri, 21 Mar 2003 12:55:18 -0600o From: brandon@dalsemi.comtI Subject: Re: How to search for last login, but not on disusered accounts.s1 Message-ID: <03032112551871@dscis6-0.dalsemi.com>   . You will need to modify your search statement:  7 $ searc sysuaf.lis "Username:","Last Login:","Flags:" -.1         /out=sys$scratch:oldaccounts_searched.tmpp   And modify the code:   $LOOP: $ read infile data/end=nomoreA, $ if (f$extract(0,9,data) .eqs. "Username:") $ then  $   user = f$extract(10,16,data)
 $   goto loop  $ endifA $!( $ if (f$extract(0,9,data) .eqs. "Flags:) $ then< $   if (f$search("DisUser",data) .lt. f$length(data)) then -     flag_disuser = "TRUE",
 $   goto loop  $ endif  $!" $ if (flag_disuser) then goto loop $! $ inter = f$extract(12,11,data)t $ batch = f$extract(45,11,data)l    2 This should work - logic wise, may be some bugs...   John Brandon VMS Systems Administratorn Dallas Semiconductor john.brandon@dalsemi.com 972.371.4172 wkn 972.371.4003 fxs   ------------------------------  % Date: Fri, 21 Mar 2003 09:16:55 -0500,A From: "Fred Kleinsorge" <kleinsorge@star-dot-zko-dot-dec-dot-com>I, Subject: Re: I Have Seen The IA64 Future !!!. Message-ID: <3e7b1ed8$1_3@hpb10302.boi.hp.com>  K Is it possible, sure I suppose - if the platform designers chose to do it -a and they don't.t  F You need to do a purge of your expectations based on how the Alpha SRM works.  J The IPF firmware stack in ROM is capable of finding attached bootable diskJ and CD/DVD media - and loading EFI applications from the disks - including: the OS_LOADER, it is also capable of doing a network boot.  K The FAT partition contains diagnostics, additional non-critical (for simpleaK booting) console applications, and the OS_LOADER (think: Primary Bootstrap)sI for the OS.  Trying to find ways of doing away with the FAT partition, is,F like trying to find a way to put APB in ROM.  It is not a useful idea.    = "JF Mezei" <jfmezei.spamnot@vl.videotron.ca> wrote in message ) news:3E79F1CA.67820486@vl.videotron.ca...s > Fred Kleinsorge wrote:K > > On Alpha, all the firmware is in Flash ROM.  So there are CD's that can  beH > > "booted" by the SRM that allow the flash ROM to be reloaded with new, > > versions.  The same is true for Itanium. >sK > Would it be possible to flash the IA64 ROM with the EFI that is stored oni diskJ > so that you'd have the full fledged EFI upon power up by default instead ofK > having to load it manually ? Or will the on-disk EFI be incomplete and in. a>0 > format incompatible with the flashed version ?   ------------------------------  % Date: Fri, 21 Mar 2003 09:21:10 -0500nA From: "Fred Kleinsorge" <kleinsorge@star-dot-zko-dot-dec-dot-com>n, Subject: Re: I Have Seen The IA64 Future !!!. Message-ID: <3e7b1fd7$1_1@hpb10302.boi.hp.com>  1 "Michael Unger" <unger@decus.de> wrote in messagen% news:00A1D297.40346530.21@decus.de...h7 > > Why?  Is there something inherently "bad" about it?  > G > Not neccessarily "bad" -- I just prefer the "keep it simple" approach  > for several reasons: > (a) clear design$ > (b) straightforward implementation > (c) less error prone > (d) simpler bug chasings > (e) easier maintenance >	  L The EFI approach is a far better design that the SRM in many, many ways.  IfL we had it to do over again from scratch, the SRM would have either been doneL using OpenBoot (a nod to Sun), or in a manner closer to how EFI is evolving.K If EFI existed for IA32 when Alpha was done, I would have lobbied for it asl
 the solution.   @ The SRM may look "simple" to you, but it is anything but simple.  J The above isn't a slam of the SRM, the guy most responsible for it I countL as a friend.  They did what they had to do, given the requirements dumped on them... and that was a *lot*.    ------------------------------  % Date: Fri, 21 Mar 2003 09:29:03 -0500aA From: "Fred Kleinsorge" <kleinsorge@star-dot-zko-dot-dec-dot-com> , Subject: Re: I Have Seen The IA64 Future !!!. Message-ID: <3e7b21b0$1_2@hpb10302.boi.hp.com>  1 "Michael Unger" <unger@decus.de> wrote in messagec% news:00A1D297.75A36653.23@decus.de...g >tE > So the "requirements" will be determined by the "Windows" standard?o0 > That's far less than VMS users tend to expect. >g  J Welcome to the real world.  No.  The requirements can be VMS requirements,I or HP-UX requirements, or LINUX requirements, or Windows-64 requirements.AJ But this *isn't* an HP standard, it's a industry standard that is providedH by Intel and a number of other companies that that work with them on it.I So, things that are done are really looked at from a overall perspective.   I If VMS had come back and said "We *must* have MOP support" - working with I the HP firmware people, we would have it put into the ROM for HP systems.yK But that also then limits us to only HP systems.  On the other hand, VMS is L the only OS that uses MOP anymore - so should VMS adapt to the industry norm of a BOOTP like remote boot?  H > Remembering the previous discussion "diagnostics" specific to VMS haveF > to be put into "extended EFI", i.e. "FAT partition" -- checking SCSIJ > commands required by VMS (and not used by Windows) for example. Correct?  B Don't understand what you mean here?  What SCSI checking commands?  I > Will the same be true for special functions of the POST and PCI adapterl > tests? >a   Please be more specific.  J > > > This would avoid the trouble with the FAT "partition" to be embedded > > > into the ODS file system.o > > >	 > >e   Huh?  L > > There is no "trouble".  It's just forcing us to get beyond simple-stupid > > boot-block booting.r >2: > Which makes it more complicated -- see my other posting. >_  L And which I disagree.  And I'm pretty close to being an expert in this area.  G > Perhaps it is time for HP to put some serious explanations on EFI and H > booting VMS on Itanic on the VMS web pages instead of just refering to > the "Intel developer pages". >b  L When the time approaches that you can get VMS for the Itanium, there will be lots of documentation.   ------------------------------  % Date: Fri, 21 Mar 2003 18:51:32 +0100 $ From: Michael Unger <unger@decus.de>, Subject: Re: I Have Seen The IA64 Future !!!+ Message-ID: <00A1D363.5A970F45.21@decus.de>o  B "Fred Kleinsorge" <kleinsorge@star-dot-zko-dot-dec-dot-com> wrote:  3 > "Michael Unger" <unger@decus.de> wrote in message ' > news:00A1D297.75A36653.23@decus.de...c > > J > > Remembering the previous discussion "diagnostics" specific to VMS haveH > > to be put into "extended EFI", i.e. "FAT partition" -- checking SCSIL > > commands required by VMS (and not used by Windows) for example. Correct? > D > Don't understand what you mean here?  What SCSI checking commands?  F If I remember correctly a few years ago SCSI controllers (and disks?) F approved by DEC/Q for usage with VMS contained special ("customized") G firmware due to problems with the "standard" firmware; now it is being iI checked in the "initialization" phase if the controller and disk is "VMS v compliant".e  H The specific issue here is "tagged command queuing" which isn't used by " Winwoes; I don't know about Linux.  H As far as I know disks not being "compliant" will simply be set offline  to avoid data corruption.o  K > > Will the same be true for special functions of the POST and PCI adapterd
 > > tests? > >- >  > Please be more specific.  & - 33 or 66 MHz operation of PCI boards& - 32 or 64 bit operation of PCI borads<    (a Winwoes driver may not be able to support 64 bit mode)( - 3.3 or 5 volts operation of PCI boards: - auto-termination (SCSI adaptors, without external disks)  I > > Perhaps it is time for HP to put some serious explanations on EFI andjJ > > booting VMS on Itanic on the VMS web pages instead of just refering to  > > the "Intel developer pages". > >g > N > When the time approaches that you can get VMS for the Itanium, there will be > lots of documentation.  G It may be valuable for customers to get such information in advance --  F it may be of some influence with respect to "migration" decisions (to $ VMS on Itanic or to another vendor).   Michaela   ------------------------------  % Date: Fri, 21 Mar 2003 12:47:39 +0000n' From: Andrew Harrison SUNUK Consultancyh* Subject: Re: IBM starts to work on Power-5. Message-ID: <3E7B09EB.6050401@nospamn.sun.com>   Greg Cagle wrote:n* > Andrew Harrison SUNUK Consultancy wrote: >  >>? >> Yup and there is also some FUD about Sun in the same article ; >> which your favourite vendor is also touting around. Bothn* >> points seem hardly worth responding to. >  > 9 > Of course, Sun never spreads FUD about the competition.o >   7 You are mistaking fact with fiction. If you can uncoverg4 the facts behind the article Bob posted then you are7 welcome to publish them but the article in nothing moree than an unsubstantiated rumour.e  6 On the other hand say that the GS320/160 don't perform6 as advertised and arn't competitive from a performance3 standpoint is a substantiated fact wel supported by"' collateral published by HPQ themselves.    Regardss Andrew Harrison    ------------------------------  % Date: Fri, 21 Mar 2003 07:32:59 -0800,& From: Greg Cagle <gregc@gregcagle.com>* Subject: Re: IBM starts to work on Power-5/ Message-ID: <v7mc503r2tgla8@corp.supernews.com>n  ( Andrew Harrison SUNUK Consultancy wrote: >  >  > Greg Cagle wrote:  > + >> Andrew Harrison SUNUK Consultancy wrote:  >> >>>r@ >>> Yup and there is also some FUD about Sun in the same article< >>> which your favourite vendor is also touting around. Both+ >>> points seem hardly worth responding to.x >> >> >>: >> Of course, Sun never spreads FUD about the competition. >> > 9 > You are mistaking fact with fiction. If you can uncoverc6 > the facts behind the article Bob posted then you are9 > welcome to publish them but the article in nothing morer! > than an unsubstantiated rumour.e > 8 > On the other hand say that the GS320/160 don't perform8 > as advertised and arn't competitive from a performance5 > standpoint is a substantiated fact wel supported byc) > collateral published by HPQ themselves.t   Hook, line, and sinker.a   -- r
 Greg Cagle gregc at gregcagle dot com   ------------------------------    Date: 20 Mar 2003 23:36:10 -0800# From: dooleys@snowy.net.au (dooley)Z6 Subject: Re: LOGINOUT.EXE locked by other user problem= Message-ID: <1ca82fc6.0303202336.68fd13d9@posting.google.com>t  h Kilgallen@SpamCop.net (Larry Kilgallen) wrote in message news:<R0DHMWyWC3Di@eisner.encompasserve.org>...l > In article <3e7a50e4.0@cfanews.cfa.harvard.edu>, "Gareth V. Williams" <graff@cfa0.cfa.harvard.edu> writes:J > > Has anyone any thoughts on this problem that has afflicted our VMS webN > > server in recent days?  The system is V7.2-1, with Purveyor 2.1 and TCP/IP
 > > 5.0-A. >  CO > > Image name:               SCULLY$DKB0:[SYS0.SYSCOMMON.][SYSEXE]LOGINOUT.EXEaO > > Status:                   %RMS-E-FLK, file currently locked by another userm > H > Effective at least 23 years ago, the most frequent cause of this errorH > is a locally written program that opens SYSUAF.DAT without proper fileE > sharing.  A close second might be an add-on product making the same-
 > mistake.K No - the second most frequent cause of this error is a user written programi6 that opens RIGHTSLIST.DAT without proper file sharing!M Once you have identified the locked file, you may have to use SDA or freewareu1 utilities (FINDLOCK?) to find the offending user.t Phil   ------------------------------    Date: 21 Mar 2003 06:03:07 -0600- From: Kilgallen@SpamCop.net (Larry Kilgallen)e6 Subject: Re: LOGINOUT.EXE locked by other user problem3 Message-ID: <3EAEIaVTzr3H@eisner.encompasserve.org>t  c In article <1ca82fc6.0303202336.68fd13d9@posting.google.com>, dooleys@snowy.net.au (dooley) writes:>  M > No - the second most frequent cause of this error is a user written programe8 > that opens RIGHTSLIST.DAT without proper file sharing!  K I believe the proper description of that would be a program that improperlyIH opens RIGHTSLIST.DAT directly rather than calling the appropriate system	 services.e   ------------------------------    Date: 21 Mar 2003 07:12:56 -0600 From: briggs@encompasserve.org6 Subject: Re: LOGINOUT.EXE locked by other user problem3 Message-ID: <FAlodYZ1buxs@eisner.encompasserve.org>o  ] In article <0jtea.19735$M7.501334@twister.tampabay.rr.com>, "McEagle" <spam@spam.com> writes:aN > I too have seen this problem (infrequently) on my 7.2-1 systems.   Any help?  G Since the status is "%RMS-E-FLK, file currently locked by another user"4= it follows that you're seeing a file lock, not a record lock.   F If we assume that the relevant file is SYSUAF or RIGHTSLIST (extremely, likely) then there is an obvious workaround:  F Open both files yourself for READ/WRITE access with READ/WRITE sharing and hold them open forever.:  C Now, whoever is gumming up the works by opening up those files withuC exclusive access will have problems in _HIS_ application.  You willrE be unaffected because LOGINOUT.EXE is using proper shared file accessa techniques.   C Do you know whether the errors are correlated with the times during C which BACKUP is running?  I believe that BACKUP will open up SYSUAFaF and RIGHTSLIST for exclusive access, even if you use /IGNORE=INTERLOCK5 if it finds them idle when it's time to back them up.l   	John Briggs   ------------------------------  % Date: Fri, 21 Mar 2003 09:19:23 +0000o( From: Nic Clews <sendspamhere@127.0.0.1>* Subject: Mark Gorham was Re: Rich Marcello) Message-ID: <3E7AD91B.FF76413A@127.0.0.1>i   Larry Kilgallen wrote: > t > In article <3e79d2e4$1_1@hpb10302.boi.hp.com>, "Fred Kleinsorge" <kleinsorge@star-dot-zko-dot-dec-dot-com> writes:M > > Don't know exactly what you are asking.  Rich came from the semiconductorrP > > group at Digital Equipment, and today is the Vice President for the HP AlphaP > > Server group.  The OpenVMS group reports to him.  He reports to the Business > > Critical Server group. > D > On the other hand, Mark Gorham, who succeeded Rich Marcello as theG > head of VMS, formerly worked in the VMS group and then went elsewhere G > in the company where he gained additional perspective.  The fact that,H > he is now back in VMS means that he has a broader view of the business' > than if he had been in VMS all along.c > E > Although I have never had any complaints about Mark lacking a broaduD > understanding, I think it is useful to have managers who have doneF > something other than VMS, so his time away is to all of our benefit.C > When I heard Mark had left VMS I was sad, but now that he is backn* > I am happy he was (temporarily) away :-)  * Ee's a good fella and not at all stuck up.  G If you ever get the chance to attend an event where he is, go, and talkd to him.l   -- c? Regards, Nic Clews a.k.a. Mr. CP Charges, CSC Computer Sciencese nclews at csc dot comv   ------------------------------    Date: 21 Mar 2003 07:25:12 -0800( From: bob@instantwhip.com (Bob Ceculski) Subject: Re: Micromedex on VMS= Message-ID: <d7791aa1.0303210725.7ecf66a1@posting.google.com>e  U Lee Y T Mah <lytmah@cha.ab.ca> wrote in message news:<3E7A3DBC.868F5EA0@cha.ab.ca>...f@ > Micromedex is already running on our intranet, and it is being? > accessed at some of our sites.  The text-based version on VMSs= > is available 24x365 and is being accessed over VT terminals.D > in critical areas of our healthcare institution.  We can't migrateB > all these users to PC's that quickly.  Compare the time it takes= > to reset a terminal server port versus rebooting a hung PC.N >  >  > Bob Koehler wrote: > T > > In article <3E79C81F.A89A9FF7@cha.ab.ca>, Lee Y T Mah <lytmah@cha.ab.ca> writes:< > > > Micromedex on VMS (VAX/Alpha) is a text-based databaseI > > > providing toxicology, drug, disease... information to practitionersrA > > > in the healthcare industry.  The vendor wants to move theirt? > > > VMS/AIX/Sun... customers to their internet, intranet, andr > > > Windows platforms. > >iG > >    Remind the vendor where Microsoft states that Windows is not for F > >    critical systems and ask him if he's taking on the liability orA > >    responsibility for making it useable as a critical system.C  > I would say Lee to find another similar product from Cerner or: someone that uses VMS as the database backend at least ...> if these people were smart, the would provide a windows (html,; com or synergy dbl) pretty front ends with VMS still on theg- backend, but they will learn the hard way ...h   ------------------------------  % Date: Fri, 21 Mar 2003 10:09:54 -0700d$ From: Lee Y T Mah <lytmah@cha.ab.ca> Subject: Re: Micromedex on VMS) Message-ID: <3E7B4762.34621A6F@cha.ab.ca>c  @ We do run a Cerner app on our Alpha's, and users can come in viaA Windows or VT terminals.  Micromedex is just a small facet of our-> user applications.  VMS Alpha has a very strong presence here.> I didn't mean to give the impression of doom and gloom with my post on Micromedex.      Bob Ceculski wrote:E  W > Lee Y T Mah <lytmah@cha.ab.ca> wrote in message news:<3E7A3DBC.868F5EA0@cha.ab.ca>... B > > Micromedex is already running on our intranet, and it is beingA > > accessed at some of our sites.  The text-based version on VMSf? > > is available 24x365 and is being accessed over VT terminalshF > > in critical areas of our healthcare institution.  We can't migrateD > > all these users to PC's that quickly.  Compare the time it takes? > > to reset a terminal server port versus rebooting a hung PC.e > >f > >  > > Bob Koehler wrote: > >kV > > > In article <3E79C81F.A89A9FF7@cha.ab.ca>, Lee Y T Mah <lytmah@cha.ab.ca> writes:> > > > > Micromedex on VMS (VAX/Alpha) is a text-based databaseK > > > > providing toxicology, drug, disease... information to practitionersgC > > > > in the healthcare industry.  The vendor wants to move their A > > > > VMS/AIX/Sun... customers to their internet, intranet, ando > > > > Windows platforms. > > >aI > > >    Remind the vendor where Microsoft states that Windows is not foriH > > >    critical systems and ask him if he's taking on the liability orC > > >    responsibility for making it useable as a critical system.t > @ > I would say Lee to find another similar product from Cerner or< > someone that uses VMS as the database backend at least ...@ > if these people were smart, the would provide a windows (html,= > com or synergy dbl) pretty front ends with VMS still on thet/ > backend, but they will learn the hard way ...r   -- Leef  	 L Y T Mahs   ------------------------------   Date: 21 Mar 03 10:09:54 +0100) From: p_sture@elias.decus.ch (Paul Sture) : Subject: Re: OT Re: National Moratorium to Appease Tyrants) Message-ID: <aGRBuVkh$VfK@elias.decus.ch>c  o In article <A88ea.177549$em1.74401@news04.bloor.is.net.cable.rogers.com>, "John Smith" <a@nonymous.com> writes:t >  >  > Paul,o > H > I supply no more political views than anyone else who has participatedA > in this thread. Sure I provide my share of opinions, but I also F > provide a great deal of verifiable facts too. I would point out thatB > Bill Todd began the recent discussion on the Mideast - many haveG > participated in it, and the thread was clearly marked so anyone coulde3 > choose not to delve into it if they chose not to.  > A > If 'editorializing' posts with some reference to politics is an 0 > offence, then many of us here have are guilty. >  > < > I bash HP as I see they deserve it. Just as other here do.D > I was not a fan of the decision to kill Alpha - a belief that many > here also shared.rF > I was not a fan of the merger - a belief that many here also shared.E > I've never criticized VMS Engineering, only Compaq/HP's lack of VMSo > marketing and advertising.F > Many of us are skeptical of exactly what will happen to VMS when theB > Intel porting money is used up, and would like to have a clearerC > statement, either a simple "Read my lips. No new features.", or auE > "Send your prioritized laundry list of new features to us and we'llp; > add the top 30 requests each year for the next 30 years."s > D > I write Stallard, Carly, Marcello, et al. to offer opinions - someE > soft and gentle, some not so gentle; to tell of what customers tell H > me, tell sales reps of opportunities where I think VMS can be sold. OrG > do you want me to post that here so Andrew can tell the local Sun repoG > to call on the potential customer with a bunch of 'Why VMS is no goodl > for your company' literature?  > E > I offer many ideas to improve marketing and visibility of VMS, somehA > serious, some tongue-in-cheek, just as other here do. You might C > consider that to be 'sniping', but maybe it will give a seemingly < > comatose HP some prodding to the point where they actuallyF > market/advertise VMS. I have no idea what actually 'motivates' HP toG > do what it does with respect to not marketing/advertise VMS - do you? B > It's a serious question. Name one other company that has taken a6 > multi-billion dollar asset and failed to promote it. > H > To a large extent we are in maintenance mode with our VMS code becauseG > we find it very hard to push on a rope to market VMS for HP. It is HP F > who should be pulling us and the customer base along. Our use of VMSE > has lessened over the years, as it is harder and harder to convinceoF > our customers that a VMS solution is better for them, even though itB > technically is, and even though we would much prefer to use VMS. > E > You want me to jump in with technical advice? My view is that therenA > are others here who are pushing the envelope with VMS and largenH > installations more than we are and encounter more problems that we do.B > Our problems tend to be ones of skimming the doc set and missing@ > something we should have read rather than finding real bugs orG > limitations in VMS. Again, we are mostly in maintenance mode with ourmG > VMS apps, adding new application features as needed but not ones thathH > stress the o/s - hopefully that will soon change with a new app we are. > pushing a customer to let us build for them. > G > I have conversed in e-mail and by telephone with some who participatenE > in C.o.v.. If you want to communicate off-line, ask. Forgive me forsG > saying so, but as to your point about 'the right to reply off-line' -AH > think of it as meeting a attractive woman and asking for her telephoneA > number - it's her right to give it to you or not, solely at hera
 > discretion.c >  >   3 Thanks for the serious response. It is appreciated.,   --  
 Paul Sture   ------------------------------  # Date: Fri, 21 Mar 2003 16:57:42 GMTE# From: "John Smith" <a@nonymous.com>h: Subject: Re: OT Re: National Moratorium to Appease TyrantsG Message-ID: <awHea.22286$Dt1.9213@news02.bloor.is.net.cable.rogers.com>f  6 "Paul Sture" <p_sture@elias.decus.ch> wrote in message# news:aGRBuVkh$VfK@elias.decus.ch...  > In articleD <A88ea.177549$em1.74401@news04.bloor.is.net.cable.rogers.com>, "John Smith" <a@nonymous.com> writes:d > >o > >n	 > > Paul,  > >g= > > I supply no more political views than anyone else who hasa participatedC > > in this thread. Sure I provide my share of opinions, but I also,C > > provide a great deal of verifiable facts too. I would point outn thatD > > Bill Todd began the recent discussion on the Mideast - many haveC > > participated in it, and the thread was clearly marked so anyoneH could 5 > > choose not to delve into it if they chose not to.  > >bC > > If 'editorializing' posts with some reference to politics is ani2 > > offence, then many of us here have are guilty. > >V > >t> > > I bash HP as I see they deserve it. Just as other here do.F > > I was not a fan of the decision to kill Alpha - a belief that many > > here also shared.2@ > > I was not a fan of the merger - a belief that many here also shared.RC > > I've never criticized VMS Engineering, only Compaq/HP's lack of- VMS0 > > marketing and advertising.D > > Many of us are skeptical of exactly what will happen to VMS when theJD > > Intel porting money is used up, and would like to have a clearerE > > statement, either a simple "Read my lips. No new features.", or ayA > > "Send your prioritized laundry list of new features to us and  we'llo= > > add the top 30 requests each year for the next 30 years."o > >lF > > I write Stallard, Carly, Marcello, et al. to offer opinions - someB > > soft and gentle, some not so gentle; to tell of what customers tellA > > me, tell sales reps of opportunities where I think VMS can bel sold. OrE > > do you want me to post that here so Andrew can tell the local Sun  repbD > > to call on the potential customer with a bunch of 'Why VMS is no good! > > for your company' literature?o > >nB > > I offer many ideas to improve marketing and visibility of VMS, someC > > serious, some tongue-in-cheek, just as other here do. You mightaE > > consider that to be 'sniping', but maybe it will give a seemingly > > > comatose HP some prodding to the point where they actuallyE > > market/advertise VMS. I have no idea what actually 'motivates' HP. toD > > do what it does with respect to not marketing/advertise VMS - do you?D > > It's a serious question. Name one other company that has taken a8 > > multi-billion dollar asset and failed to promote it. > >eB > > To a large extent we are in maintenance mode with our VMS code because3F > > we find it very hard to push on a rope to market VMS for HP. It is HPD > > who should be pulling us and the customer base along. Our use of VMSD> > > has lessened over the years, as it is harder and harder to convinceE > > our customers that a VMS solution is better for them, even thoughl itD > > technically is, and even though we would much prefer to use VMS. > >uA > > You want me to jump in with technical advice? My view is that  therewC > > are others here who are pushing the envelope with VMS and largetF > > installations more than we are and encounter more problems that we do. D > > Our problems tend to be ones of skimming the doc set and missingB > > something we should have read rather than finding real bugs orE > > limitations in VMS. Again, we are mostly in maintenance mode withe ournD > > VMS apps, adding new application features as needed but not ones thatF > > stress the o/s - hopefully that will soon change with a new app we arel0 > > pushing a customer to let us build for them. > >C= > > I have conversed in e-mail and by telephone with some whos participate C > > in C.o.v.. If you want to communicate off-line, ask. Forgive mee for = > > saying so, but as to your point about 'the right to reply  off-line' -o@ > > think of it as meeting a attractive woman and asking for her	 telephone C > > number - it's her right to give it to you or not, solely at herV > > discretion.b > >e > >  >e5 > Thanks for the serious response. It is appreciated.C     You're welcome.i   ------------------------------  % Date: Fri, 21 Mar 2003 07:42:28 +0100r From: Dirk Munk <munk@home.nl>" Subject: Re: Problems with SSH EAK2 Message-ID: <b5ec92$ap7$1@news1.tilbu1.nb.home.nl>   Mike Mabey wrote:D > Hello,G > I have installed the EAK for SSH for OpenVMS and am running into someo > problems.ME > I can open a simple SSH terminal session, but if I attempt to run a:J > command via SSH instead of starting a terminal session it does not work.@ > SFTP is not working either, but I intend to tackle that later.  O We were testing with SCP/SFTP, found several problems, and reported those back dP to the developers. It seems an improved version of the SSH kit will be produced  shortly.   > ( > Any help would be greatly appreciated.	 > Thanks,M > Mike Mabey   ------------------------------  % Date: Fri, 21 Mar 2003 07:59:46 -0500S$ From: Mike Mabey <mmabey@cenhud.com>" Subject: Re: Problems with SSH EAK) Message-ID: <3E7B0CC2.4040501@cenhud.com>e   Thanks Dirk,- I've turned this over to Engineering as well.    Dirk Munk wrote:   > Mike Mabey wrote:i >wH >> I have installed the EAK for SSH for OpenVMS and am running into some >> problems.F >> I can open a simple SSH terminal session, but if I attempt to run aK >> command via SSH instead of starting a terminal session it does not work.lA >> SFTP is not working either, but I intend to tackle that later.- >- >-F > We were testing with SCP/SFTP, found several problems, and reported H > those back to the developers. It seems an improved version of the SSH  > kit will be produced shortly.s >-   ------------------------------  # Date: Fri, 21 Mar 2003 17:10:28 GMTt# From: "John Smith" <a@nonymous.com>a Subject: Request to SuesG Message-ID: <8IHea.22424$Dt1.2626@news02.bloor.is.net.cable.rogers.com>m  @ If you and/or your colleagues will be making the rounds of ISV's9 encouraging them to stay with VMS, then I have a request:u  B Call on Sun and, in particular, the group that build their UnifiedD Development Server (UDS, formerly known as Forte from Forte SoftwareA in Oakland, CA) and do what you can to encourage them to keep VMS D parity with all the other o/s they support. There is still an active customer base on VMS.p  ; Ken McKenzie is the product manager, and Tom Staaden is the E development manager - or at least they were last time I checked about 
 6 months ago./  D You may recall this product from the days when Digital owned a chunkB of the company - Mike Kelley was the person on-point from Digital.E It's a great product - it's what J2EE is still trying to be. And wheniD coupled with VMS clusters, you can effectively get disaster tolerant  applications, o/s, and hardware.   ------------------------------  % Date: Fri, 21 Mar 2003 08:44:57 +0100d4 From: Didier Morandi <Didier.Morandi.nospam@Free.fr> Subject: Re: Rich Marcello& Message-ID: <3E7AC2F9.3000706@Free.fr>   Main, Kerry wrote:   >Didier, >lI ><<< (this is why, today, the last users of VMS in Europe are so numerous-F >to ask for studies on the future of VMS (two requests this week)).<<< >oH >And did you provide a pointer and/or send them the official HP strategy >and roadmaps presentations? >r >Reference: I >http://h18003.www1.hp.com/hps/ipf-enterprise/download/ASDUPDATEDSTRATEGY: >JAN03.pdf e >fF >(Jan 2003 updated - look at slide 21 - OpenVMS on Superdome, mid tier >and low end IPF systems)m >hI >And if they are wondering about OpenVMS compares to other vendor clustera5 >and disaster tolerant products, have them check out:iI >http://www.tru64unix.compaq.com/unix/illuminata_dt_unix_research_note.pd_ >f >aA Kerry, it's nice of you but, again, everyone KNOWS on earth that bG marketing is not synonym of truth. The ONLY way to DEMONSTRATE that HP "B wants to push VMS SALES (not engineering, we all KNOW that) is to 
 ADVERTISE it."   Period.y   D.' http://www.sector7.com, your VMS futureo   ------------------------------  % Date: Fri, 21 Mar 2003 09:09:59 -0500e' From: "Main, Kerry" <Kerry.Main@hp.com>* Subject: RE: Rich MarcelloT Message-ID: <BE56C50EA024184DAF48F0B9A47F5CF402660E79@kaoexc01.americas.cpqcorp.net>   Didier,u  D You mentioned OpenVMS Customers were looking for answers to specific
 questions.  B As an ISV interested in growing Customer interest in OpenVMS, as a> minimum I would expect that when you get these questions, thatH regardless of your personal opinions, you would at least say ".. Here is HP's official position .."  C And yes, better advertising would be good, but that should not stop H ISV's interested in growing OpenVMS (not ones interested in porting fromH OpenVMS) from at least providing Customers with the official HP position on this question.=20   Regardst  
 Kerry Main Senior Consultant' Hewlett-Packard (Canada) Co.! Consulting & Integration Services  Voice: 613-592-4660n Fax   : 613-591-4477 Email: kerryDOTmain@hpDOTcom-     (remove the DOT's and replace with "."'s)" OpenVMS DCL - the original .COM_     -----Original Message-----> From: Didier Morandi [mailto:Didier.Morandi.nospam@Free.fr]=20 Sent: March 21, 2003 2:45 AM To: Info-VAX@Mvb.Saic.Comt Subject: Re: Rich Marcello     Main, Kerry wrote:   >Didier, >sC ><<< (this is why, today, the last users of VMS in Europe are so=201G >numerous to ask for studies on the future of VMS (two requests this=20  >week)).<<<a >!H >And did you provide a pointer and/or send them the official HP strategy   >and roadmaps presentations? >i >Reference:=20H >http://h18003.www1.hp.com/hps/ipf-enterprise/download/ASDUPDATEDSTRATEG >Y
 >JAN03.pdf=20- >"I >(Jan 2003 updated - look at slide 21 - OpenVMS on Superdome, mid tier=20  >and low end IPF systems)g >sD >And if they are wondering about OpenVMS compares to other vendor=20@ >cluster and disaster tolerant products, have them check out:=20H >http://www.tru64unix.compaq.com/unix/illuminata_dt_unix_research_note.p >d >f >sC Kerry, it's nice of you but, again, everyone KNOWS on earth that=20eI marketing is not synonym of truth. The ONLY way to DEMONSTRATE that HP=20LD wants to push VMS SALES (not engineering, we all KNOW that) is to=20
 ADVERTISE it.t   Period.    D.' http://www.sector7.com, your VMS future$   ------------------------------  + Date: Fri, 21 Mar 2003 15:31:42 +0100 (MET) 9 From: Phillip Helbig <HELBPHI@sysdev.deutsche-boerse.com>  Subject: Re: Rich Marcello; Message-ID: <01KTSHDD82U49H1MPS@sysdev.deutsche-boerse.com>d  D > As an ISV interested in growing Customer interest in OpenVMS, as a@ > minimum I would expect that when you get these questions, thatJ > regardless of your personal opinions, you would at least say ".. Here is > HP's official position .." i  F Yes, but what is HP's "official position" with regard to VMS?  We willH help you port to HPUX?  That might not be wise to state.  Or do we have ! a choice of "official positions"?l  F What if the customer says "ah yes, DEC's official position used to be E "NT on ALPHA is the future", and I invested millions in that and got > burned"?  @ (I suddenly started hearing Nazareth's "Love Hurts" in my head.)   ------------------------------  # Date: Fri, 21 Mar 2003 16:49:55 GMT # From: "John Smith" <a@nonymous.com>P Subject: Re: Rich MarcelloG Message-ID: <ToHea.22206$Dt1.2169@news02.bloor.is.net.cable.rogers.com>E  2 "Main, Kerry" <Kerry.Main@hp.com> wrote in messageF news:BE56C50EA024184DAF48F0B9A47F5CF402660E79@kaoexc01.americas.cpqcor p.net... Didier,   D You mentioned OpenVMS Customers were looking for answers to specific
 questions.  B As an ISV interested in growing Customer interest in OpenVMS, as a> minimum I would expect that when you get these questions, thatE regardless of your personal opinions, you would at least say ".. Herep is HP's official position .."  C And yes, better advertising would be good, but that should not stop C ISV's interested in growing OpenVMS (not ones interested in porting  from? OpenVMS) from at least providing Customers with the official HP  position on this question.s     Kerry,  D Sorry to point this out but it is the vendor of VMS, HP, that has to4 be showing the interesting growing VMS, not the ISV.  F The ISV is interested in selling their software. If the ISV can't sellE it on VMS because customers don't think that HP is truly committed to ? VMS, then the ISV and the customer will pursue other platforms.-  @ As an ISV I can talk myself blue telling the customer that HP is@ committed to VMS, but when they see that HP has done zero, nada,D zilch, nichts in the way of marketing and advertising VMS, that is aD HUGE statement of disinterest in the future of VMS by HP - your .pdfC reference excluded as it is a year out of date and not in customerst2 faces like print, TV, magazine, and billboard ads.    F Over the past 2 days there have been more full-page ads from HP in theF local papers - four pages in total - same ads as before except reducedB to 1 page each - two were pitching Linux, and two were pitching HP@ servers generally but no mention of any specific product. No VMS advertising.     Tell me:C a) what is HP's total advertising and marketing budget for the paste year? < b) what percentage of that was allocated to VMS advertising?; c) what percentage of b) was actually spent on advertising?dE d) what percentage of b) was charged as 'overhead' to the VMS budget?o  7 My guesses to the answers of the 4 questions above are:l a) tons  b) zerog c) less than zero  d) lots more than zero  C Someday, in the not too far distant future, the students at HarvardhB Business School are going to do a case study on how HP killed VMS.   ------------------------------  % Date: Fri, 21 Mar 2003 12:11:09 -0400t0 From: JF Mezei <jfmezei.spamnot@vl.videotron.ca> Subject: Re: Rich Marcello/ Message-ID: <3E7B399B.B71127B9@vl.videotron.ca>M   "Main, Kerry" wrote:D > As an ISV interested in growing Customer interest in OpenVMS, as a@ > minimum I would expect that when you get these questions, thatJ > regardless of your personal opinions, you would at least say ".. Here is > HP's official position .."  L NO ! Au contraire. You,d explain that HP is busy integrating people and thatM their position on VMS is starting to change as they realise the value of VMS.aM (bull shit, of course). You wouldn't give the HP "official" position since it 5 says that all new customers will be steered to HP-UX.o  M Herin lies the problem. Many would love to do the VMS marketing if HP doesn'thJ want to. But we need a minimum of support from HP. Remember that the alphaL murder killed a lot of trust/confidence in the so called commitments. It wilP take real concrete and positive actions on the part of HP to rebuild this trust.   ------------------------------  % Date: Fri, 21 Mar 2003 19:11:37 +0100 4 From: Didier Morandi <Didier.Morandi.nospam@Free.fr> Subject: Re: Rich Marcello& Message-ID: <3E7B55D9.2060001@Free.fr>   Main, Kerry wrote:   >Didier, >(E >You mentioned OpenVMS Customers were looking for answers to specific  >questions.? >hC >As an ISV interested in growing Customer interest in OpenVMS, as ac? >minimum I would expect that when you get these questions, thateI >regardless of your personal opinions, you would at least say ".. Here iso >HP's official position .."O >f   Course, I did.  I I even created a yahoo newsgroup for the SNECMA project to migrate their eG workshop tools from VAX/VMS to Itanium, (or Sun or HP or I do not know oF who, the Customer decide). I had a request for quotation from them to D tell them where they should go. I asked HP France, I got some help, G except from the VMS ambassador who NEVER answered my email messages... aD Then, I prepared my study, then I learned that the study was won by H another consultant. So I deleted the Yahoo group. Until the end of last F year, I was still hoping that HP Corporate would do something to keep F French VMS Customers. They OBVIOUSLY were not interested, so I "went" B SAP (my last day of 11 weeks training was today, actually), but I G learned that there are no more new R/3 projects in France anymore, so, gH as I know only two products in my professional life, VMS and APPLE, and G as I need to work to make some money to live, and as there is not inuf  F work on APPLE and no more work on VMS expect to migrate the Customers  elsewere, I joined Sector7.E  < And you can tell Rich and Mark that THEY pushed me to do so.   Regards,   D.   ------------------------------  % Date: Fri, 21 Mar 2003 14:48:38 +0000n' From: Andrew Harrison SUNUK Consultancyf/ Subject: Re: Sun unix/linux new nasty bug cert!l, Message-ID: <3E7B2646.70809@nospamn.sun.com>   Bob Ceculski wrote:e- > the inquirer has the details, and Andrew, Iw, > guess certs aren't important, esp. if they > point to your os ... > * > http://www.theinquirer.net/?article=8436   Well to be more clear.  * Solaris, HP-UX, Tru64, OpenVMS, AIX, Linux  IRIX are all probably vunerable.   Regards  Andrew Harrisons   ------------------------------  + Date: Fri, 21 Mar 2003 17:08:54 +0000 (UTC)d  From: Cthulhu <spambox@rlyeh.it>/ Subject: Re: Sun unix/linux new nasty bug cert! ) Message-ID: <b5fgv6$11u$2@kadath.deep.it>e  P Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com> wrote:  , > Solaris, HP-UX, Tru64, OpenVMS, AIX, Linux" > IRIX are all probably vunerable.  9 So, finally Sun managed to get something that is _really_I multi-platform! ;)   	congratulatamente,e 	    Cthulhu   -- H   Vado dove nessuno giungeB Per strade deserte abitate da spettri         http://www.rlyeh.it/G Vado nel Nulla, per cogliere l'Essenza.       <cthulhu(at)rlyeh(dot)it>    ------------------------------    Date: 20 Mar 2003 23:11:48 -0800/ From: stuie_norris@yahoo.com.au (Stuart Norris)u Subject: www.decus.org= Message-ID: <51262235.0303202311.3d5b416b@posting.google.com>    Hello,   I have been trying to down loadME ftp://www.decus.org/pub/lib/v00454/latapp.zip.  But I cannot get this-E file to download.  Could someone email me the file if they have time.o  9 Is there something wrong with server or is it my broswer?r   Thanks   Stuart.0   ------------------------------  # Date: Fri, 21 Mar 2003 08:30:46 GMTU+ From: LESLIE@JRLVAX.HOUSTON.RR.COM (leslie)e Subject: Re: www.decus.org: Message-ID: <W4Aea.62039$eb.1909245@twister.austin.rr.com>  0 Stuart Norris (stuie_norris@yahoo.com.au) wrote: : Hello, : ! : I have been trying to down loadrG : ftp://www.decus.org/pub/lib/v00454/latapp.zip.  But I cannot get thisoG : file to download.  Could someone email me the file if they have time.a : ; : Is there something wrong with server or is it my broswer?n :   ; I get "page could not be displayed" from Internet Explorer.0  F Try anonymous ftp to ftp.decus.org, and then set the default directory4 to anonymous:[lib.v00454], where latapp.zip resides.  E I've also emailed it, but sometimes Road Runner's email servers think E they're the Pony Express and take three days to send a message across0 Houston.    2 --Jerry Leslie   (my opinions are strictly my own)9   Note: leslie@jrlvax.houston.rr.com is invalid for emailr   ------------------------------  % Date: Fri, 21 Mar 2003 13:10:34 -0500e From: norm.raphael@metso.com Subject: Re: www.decus.org? Message-ID: <OFB13115D6.1CCADFCE-ON85256CF0.00639B3B@metso.com>a   tryc  / ftp://ftp.encompassus.org/lib/v00454/latapp.zip   I [Only three things wrong with the url - www s/b ftp ans /pub s/b deleted, > and in the States it's no longer DECUS, but now Encompass US.]    G From:  stuie_norris@yahoo.com.au (Stuart Norris) on 03/21/2003 02:11 AMt  ; Please respond to stuie_norris@yahoo.com.au (Stuart Norris)u   To:    Info-VAX@Mvb.Saic.Com cc:n   Subject:    www.decus.org      Hello,   I have been trying to down loaddE ftp://www.decus.org/pub/lib/v00454/latapp.zip.  But I cannot get this E file to download.  Could someone email me the file if they have time.b  9 Is there something wrong with server or is it my broswer?r   Thanks   Stuart.t   ------------------------------  % Date: Fri, 21 Mar 2003 10:40:35 -05003( From: Reza Tehrani <mt2151@columbia.edu># Subject: X-window and Xlib questiono, Message-ID: <3E7B3273.C5452469@columbia.edu>   Hi,3  H We have a  AlphaServer running OpenVMS V7.1-2 with TCP/IP V4.2. There isF no problem with telneting and ftping to it from anywhere. However whenE we try to export the display and X-windows library within our subnet,g; the network connection stops working (the error message is t   user> nedit test.c1 XIO:  fatal IO error 65535  on X server "_WSA19:"y>       after 6709 requests (6424 known processed) with 0 events
 remaining. %XLIB-F-IOERROR, xlib io errorC -SYSTEM-F-CONNECFAIL, connect to network object timed-out or failed / %TRACE-F-TRACEBACK, symbolic stack dump followsh    D Now if we do the same thing from outside the subnet everything works fine.   F I heard that it might the problem with our gateway and router settingsE for x-lib port but the fact is we don't have any problem of exportings display for Sun workstaion.   E I have also added x-server port 6000 and font-server port 7001 to theeF OpenVMS network ports but the problem still persists (I wasn't able to" enable them; is that a problem?).   5 X-SERVER    6000  TCP  not defined  0.0.0.0 Disabled e3 FONT-SERVER 7100  TCP  not defined 0.0.0.0 Disabled    Any suggestion is appreciated.   Thanks Reza   ------------------------------  + Date: Fri, 21 Mar 2003 17:21:21 +0100 (MET)p9 From: Phillip Helbig <HELBPHI@sysdev.deutsche-boerse.com>t# Subject: X-window and Xlib question8; Message-ID: <01KTSLH2YJQK9H1MPS@sysdev.deutsche-boerse.com>t  J > We have a  AlphaServer running OpenVMS V7.1-2 with TCP/IP V4.2. There is< > no problem with telneting and ftping to it from anywhere.     # Can you telnet FROM it to anywhere?o   > However whenG > we try to export the display and X-windows library within our subnet,n= > the network connection stops working (the error message is -  F > Now if we do the same thing from outside the subnet everything works > fine.   ; Can you display X11 stuff FROM somewhere else ON the ALPHA?l  F Note that when you display a window FROM the ALPHA TO somewhere else, G this is more akin to telnetting FROM the ALPHA, so let us know if that e works first.  H > I heard that it might the problem with our gateway and router settingsG > for x-lib port but the fact is we don't have any problem of exporting' > display for Sun workstaion.   G > I have also added x-server port 6000 and font-server port 7001 to theeH > OpenVMS network ports but the problem still persists (I wasn't able to$ > enable them; is that a problem?).  > 7 > X-SERVER    6000  TCP  not defined  0.0.0.0 Disabled r5 > FONT-SERVER 7100  TCP  not defined 0.0.0.0 Disabledu  % What are the "OpenVMS network ports"?_  % Note that X can use ports 6000--6063.e  F I had a similar problem on my home system which was probably due to a  wrong MTU value on the router.  D It sounds to me like a) the port is being blocked or b) there is no  route to the host.   ------------------------------  % Date: Fri, 21 Mar 2003 18:48:35 +0100O$ From: Michael Unger <unger@decus.de>6 Subject: Re: [OT] After two months of SAP training :-(* Message-ID: <00A1D362.F11328DA.7@decus.de>  7 "Didier Morandi" <Didier.Morandi.nospam@Free.fr> wrote:.  J > I tell you the truth, I have joined Sector7. No time to waste with SAP. H > Now, I need to find at least one SAP implementation in France to have  > some ROI :-) >  > D. >  > [...]<  : That company seems to be rather experienced. According to:  B http://www.sector7.com/openvms/openvms-porting-and-migration.htm#5  
 <start quote>l VAX VMS to VAX/Alpha OpenVMSI For one reason or another, you may still be running your applications on sG VAX VMS, and the time has come to upgrade to OpenVMS. Only problem is, nF you don't know if it will be a simple "drag-and-drop" exercise, or if F your application code will require modification or enhancement to run I under OpenVMS (on VAX or Alpha). We can help you with both your planning i and porting needs. <end quote>a  C May I please get a quotation for a porting project from VAX VMS to g OpenVMS on VAX?h   Michaele   ------------------------------   End of INFO-VAX 2003.158 ************************ have managers who have doneF > something other than VMS, so his time away is to all of our benefit.C > When I heard Mark had left VMS I was sad, but now that he is backn* > I am happy he was (temporarily) aw                                     	    
            
                                                                                 !    "    #    $    %    &    '    (    )    *    +    ,    -    .    /    0    1    2    3    4    5    6    7    8    9    :    ;    <    =    >    ?    @    A    B    C    D    E    F    G    H    I    J    K    L    M    N    O    P    Q    R    S    T    U    V    W    X    Y    Z    [    \    ]    ^    _    `    a    b    c    d    e    f    g    h    i    j    k    l    m    n    o    p    q    r    s    t    u    v    w    x    y    z    {    |    }    ~                                                                                                                                                                                                                                                                                    Í    č    ō    ƍ    Ǎ    ȍ    ɍ    ʍ    ˍ    ̍    ͍    ΍    ύ    Ѝ    э    ҍ    Ӎ    ԍ    Ս    ֍    ׍    ؍    ٍ    ڍ    ۍ    ܍    ݍ    ލ    ߍ        ፊ    ⍊    ㍊    䍊    半    捊    獊    荊    鍊    ꍊ    덊    썊    퍊                                                                            