1 INFO-VAX	Wed, 12 Apr 2000	Volume 2000 : Issue 205       Contents:$ Re: "%BACKUP-F-WAITIDLEBCB" question	 ...I hope + Re: 36 Gb disk drive question (supported ?) + Re: 36 Gb disk drive question (supported ?)  Re: Access limiting on DECnet  Re: Access limiting on DECnet  Re: Access limiting on DECnet ) Re: Advice Needed on New AlphaServer 2100 ) Re: Advice Needed on New AlphaServer 2100 ) Re: Advice Needed on New AlphaServer 2100 ) Re: Advice Needed on New AlphaServer 2100 ) Re: Advice Needed on New AlphaServer 2100 ) Re: Advice Needed on New AlphaServer 2100 ) Re: Advice Needed on New AlphaServer 2100  AST sanity check.  Re: CDE equivalent of xhost ! Re: CPU factor for Alpha systems? ! Re: CPU factor for Alpha systems?  Re: DE500 problem/docs Re: DE500 problem/docs re:  Re: DE500 problem/docs  DE500 problem/docs Re: DECnet access  Re: DECnet access , Re: DECW$DEVICE-I-NODEVICE and no decwindows( DECW$DEVICE-I-NODEVICE and no decwindows Denying DECnet access  Re: Denying DECnet access  Re: Denying DECnet access  Re: Denying DECnet access . Denying DECnet access from one node to another Re: Double broadcast messages  Re: Double broadcast messages  Re: ebay to drop Sun?  Re: ebay to drop Sun?  Re: ebay to drop Sun?  faster/smarter FTP clients?  Re: faster/smarter FTP clients? 2 Re: Galaxy (or perhaps Wildfire) and shared memory2 Re: Galaxy (or perhaps Wildfire) and shared memoryP Re: how to communicate with CICS directly nor through CICS Transaction Server fo HSJ80 experience?  Re: HSJ80 experience? " Re: Is sys$dclexh ever not called?" Re: Is sys$dclexh ever not called? Re: Manuals for DEC 3000-800 Re: netscape 6 Re: netscape 6 Re: netscape 6 Re: netscape 6 Re: netscape 6 Re: netscape 6 Re: netscape 6 Re: netscape 6 New VMS books coming out Re: New VMS books coming out Re: New VMS books coming out  Re: Portable unformatted files 2 Re: RMS errors Senility Re: signature file in MAIL Re: signature file in MAIL Re: signature file in MAIL& Re: Sun considers VMS a "mainframe" OS That's got it !  Vax Equipment for Sale Web Standards (was: netscape 6) # Re: Web Standards (was: netscape 6) : Re: Where to find DECserver Network Access Software (DNAS): Re: Where to find DECserver Network Access Software (DNAS): Re: Where to find DECserver Network Access Software (DNAS)6 Re: Why are there no French people in this newsgroup ?6 Re: Why are there no French people in this newsgroup ?6 Re: Why are there no French people in this newsgroup ?6 Re: Why are there no French people in this newsgroup ?% Re: Wierd Elsa Gloria Synergy Problem  Yamaha CRW8424 on VMS  Re: Yamaha CRW8424 on VMS  Re: Yamaha CRW8424 on VMS   F ----------------------------------------------------------------------  % Date: Wed, 12 Apr 2000 10:35:54 +0100 * From: "Richard Brodie" <R.Brodie@rl.ac.uk>- Subject: Re: "%BACKUP-F-WAITIDLEBCB" question , Message-ID: <8d1g1s$1b5e@newton.cc.rl.ac.uk>  ` "Michael Fields" <mwfields@hotmail.com> wrote in message news:8d0mn8$8k11@chnews.ch.intel.com...  M >Hi, all.  I'm new to the world of VMS mail and monitoring and have come upon  >a question.  = VMS version and other relevant information is always helpful.   I > What I can't figure out though is why the backup failed.  I know what a J > "Baddir" is and how to fix it.  However, I don't know what "WAITIDLEBCB" > means.  ; Try HELP/MESSAGE WAITIDLEBCB. This reveals that WAITIDLEBCB D is a "can't happen" type of error that indicates a bug in the BACKUP utility.   > Can anyone offer me a hand?   F Yes. Either log a call with your support people or pull the patch fromF the web site. It will be ALPBACK??? or similar. Sounds like a bug fromE a while back though - you might care to check how current your system  is.    ------------------------------  % Date: Wed, 12 Apr 2000 14:09:41 +0100 - From: Andy Williams <Andy.Williams@liffe.com>  Subject: ...I hopeM Message-ID: <78E5B8E274DBD1118D6800805FE60E77207EAE@ntprdex4.admin.liffe.com>   E ... just to close the loop, the point about OPER privilege overriding A the NCP SET NODE setting is documented in the 'DECnet for OpenVMS B Network Management Utilities' manual (for VMS V6.0) on page 3-106.  C One last thing I have to check: If a privileged process on system A A opens a network link to another privileged process on system B...    -Andy    -----Original Message----- From: Andy Williams , Posted At: Wednesday, April 12, 2000 1:32 PM Posted To: vms# Conversation: Denying DECnet access  Subject: That's got it !     Thanks Larry !  D Don't even ask me why I didn't think of doing the same on the remote system. senility I expect...   Cheers,    -Andy    -----Original Message-----2 From: kilgallen@eisner.decus.org (Larry Kilgallen)# [mailto:kilgallen@eisner.decus.org] - Posted At: Wednesday, April 12, 2000 12:29 PM  Posted To: vms# Conversation: Denying DECnet access " Subject: Re: Denying DECnet access    F In article <8d1hc8$261$1@nnrp1.deja.com>, Andy <andyw9804@my-deja.com> writes:   H > After a long search I unearthed some old (V6.0) Phase IV documentationH > but can't really find anything relating to this except for the NCP SETG > NODE xxx ACCESS NONE command. This works as documented but the access 6 > restriction is overridden if the user has OPER priv.  D I did not recall the OPER nuance, but certainly that cannot have anyG effect on the target node, since that node has no information regarding , privilege of the originating process.  Thus:   	If node A has:     		MCR NCP SET NODE B ACCESS NONE   	and node B has:    		MCR NCP SET NODE A ACCESS NONE  * the isolation you seek should be provided.  A The lack of a Phase V equivalent, seems to be the final matter in A which DECnet-Plus still has not caught up to regular DECnet.  The ? US DECUS Security SIG raised this concern to DEC before Phase V 
 was released.    ------------------------------  % Date: Wed, 12 Apr 2000 11:12:40 -0400 " From: Dan Sugalski <dan@sidhe.org>4 Subject: Re: 36 Gb disk drive question (supported ?)6 Message-ID: <4.3.0.20000412111014.01d33860@24.8.96.48>  2 At 12:57 AM 4/12/00 -0400, Robert Deininger wrote:H >On Tue, Apr 11, 2000 1:09 PM, Jerome LAURET <jlauret@?.chem.sunysb.edu> >wrote:  > >  >  > H >I can't speak for VMS 7.2, but I am currently trying to figure out whatF >to do about a 36 GB Quantum drive that won't work on a DPW 600au withC >VMS 7.1-1H2.  The console sees it, but VMS complains about invalid  >inquiry data and gives up.   K IIRC the 36G drives are large enough that their sector count overflows the  H 24-bit integer SCSI uses to hold that info and the remainder gets stuck K someplace else reasonably standard. (I don't remember the details, alas) I  J think there's an ECO out for 7.1 and 6.2 to handle this--check the latest > SCSI and system patches and see if applying them fixes things.   					Dan  L ----------------------------------------------------------------------------L Dan Sugalski                          General and VMS-specific perl training
 dan@sidhe.org >                                       Mail me for more details   ------------------------------    Date: 12 Apr 2000 13:36:44 -04004 From: "Robert Deininger" <rdeininger@mindspring.com>4 Subject: Re: 36 Gb disk drive question (supported ?), Message-ID: <B51A2C6E-2B4A3F@165.247.26.214>  B On Wed, Apr 12, 2000 11:12 AM, Dan Sugalski <dan@sidhe.org> wrote:3 >At 12:57 AM 4/12/00 -0400, Robert Deininger wrote: I >>On Tue, Apr 11, 2000 1:09 PM, Jerome LAURET <jlauret@?.chem.sunysb.edu>  >>wrote: >> > >> >>I >>I can't speak for VMS 7.2, but I am currently trying to figure out what G >>to do about a 36 GB Quantum drive that won't work on a DPW 600au with D >>VMS 7.1-1H2.  The console sees it, but VMS complains about invalid >>inquiry data and gives up. > K >IIRC the 36G drives are large enough that their sector count overflows the   I >24-bit integer SCSI uses to hold that info and the remainder gets stuck  K >someplace else reasonably standard. (I don't remember the details, alas) I   K >think there's an ECO out for 7.1 and 6.2 to handle this--check the latest  ? >SCSI and system patches and see if applying them fixes things.   E Thanks.  I think I've got the latest SCSI patch, but I may be missing A some related system patches.  Searching for patches seems to have : gotten harder lately; the directory tree is getting weird.  F Compaq service screamed "3rd party disk!" and declined to even commentG on whether the system is missing a crucial patch.  (I know it must be,  $ since it *should* be running 7.1-2.)  B Do SCSI-3 devices have a wider integer for the sector count?  ThisB disk reports itself as SCSI-3, and I suspected that might make theC driver unhappy.  Couldn't get an offical answer about that, either.      ---------------------------  Robert Deininger rdeininger@mindspring.com    ------------------------------  % Date: Wed, 12 Apr 2000 10:06:30 -0400  From: Everhart <ge@gce.com> & Subject: Re: Access limiting on DECnet' Message-ID: <38F482E6.A3EEDE1C@gce.com>   C You can use "NCP set access none" to block access to nodes. However E it is more useful to remember that if you use instead ncp set default  accessD none, and then set send and receive passwords for all nodes, you canG set up a DECnet so that nodes communicate only with nodes they know the G passwords for and that know their passwords. You set a receive password D on each node and set send password on each other node to the receive password for the other node.  D Bingo. For an interloper node to access anything, it has to know theF other nodes' passwords, and the other nodes have to know its password.= This gives a network that is rather resistant to interlopers.    ------------------------------  % Date: Wed, 12 Apr 2000 16:59:44 +0200 5 From: Oswald Knoppers <Oswald.Knoppers@whitehouse.nl> & Subject: Re: Access limiting on DECnet- Message-ID: <38F48F60.CD862F9F@whitehouse.nl>    Everhart wrote:   F > none, and then set send and receive passwords for all nodes, you canI > set up a DECnet so that nodes communicate only with nodes they know the I > passwords for and that know their passwords. You set a receive password F > on each node and set send password on each other node to the receive > password for the other node.  H IIRC the send and receive passwords are only for point-to-point circuits and not for ethernet.    Oswald   ------------------------------  # Date: Wed, 12 Apr 2000 14:46:17 GMT 2 From: kilgallen@eisner.decus.org (Larry Kilgallen)& Subject: Re: Access limiting on DECnet' Message-ID: <2000Apr12.104617.1@eisner>   E In article <38F482E6.A3EEDE1C@gce.com>, Everhart <ge@gce.com> writes: E > You can use "NCP set access none" to block access to nodes. However G > it is more useful to remember that if you use instead ncp set default  > accessF > none, and then set send and receive passwords for all nodes, you canI > set up a DECnet so that nodes communicate only with nodes they know the I > passwords for and that know their passwords. You set a receive password F > on each node and set send password on each other node to the receive > password for the other node. > F > Bingo. For an interloper node to access anything, it has to know theH > other nodes' passwords, and the other nodes have to know its password.? > This gives a network that is rather resistant to interlopers.   > Providing you don't use Ethernet, where those send and receive= passwords are not supported (presumably because eavesdropping  is so easy).   ------------------------------  % Date: Wed, 12 Apr 2000 00:11:08 -0400 * From: David A Froble <davef@tsoft-inc.com>2 Subject: Re: Advice Needed on New AlphaServer 2100- Message-ID: <38F3F75C.2DCFE55F@tsoft-inc.com>    Brian Oster wrote: >   ? Looks like you're getting lots of advice, and less answers. :-)   H > We are looking to upgrade our VMS Server with a used AlphaServer 2100.G > Our old system is a Dec 2000-300 axp.  Following are the basic specs:   + And your new system will also be quite old.   8 > AlphaServer 2100 4/275  -  2 CPUs  - 4MB onboard cache > 512MB RAM " > 1 RZ29B-VW 4.3GB Wide SCSI Drive4 > 3 RZ29B-VW 4.3GB Wide SCSI Drive (Setup as Raid 5) > 3 Channel Raid Controller  > Fast SCSI-2 controller$ > 8-Bay Internal Storage Works Shelf > 10/100 NIC Card E > CD Rom Drive, Floppy, Pedistal Enclosure, Embedded SVGA Video, etc.  > Open VMS v7.1-2 Base License > NAS 200 License , > Guaranteed eligible for Compaq maintenance >  > Price:  $9,500  J Some questions.  Is all of the above coming with the 2100?  Are you movingG anything from your DEC 2000-300?  What is currently on your old system?   L Another suggested a new DS10 system.  As David indicated, advantages are theL 3-year warranty, thus saving 3 years of basic maintenance, and a much fasterP system.  If the RZ29 drives listed are from your old system, they will move over) to the DS10, just as well as to the 2100.   O If you need new disks, get new 9, 18, or 36 GB drives and forget about the RAID J stuff unless you specifically need it.  The best idea is to keep things asK simple as you can.  I don't see any tape device, guess you're handling that  seperately.   B > Being relativly new to the Digital/VMS world, I have a couple of > questions: > F > 1.  Is this priced about right considering the hardware and the factH > that it is a used system?  If it is way over priced, can you recommend2 > any other vendors that deal with used equipment?  N I really hate to second guess a dealer who's in business to sell used systems,K but I wouldn't pay that much for a 2100, a first generation Alpha, when new * systems are using third generation Alphas.  < > 2.  We have 18 user licenses on the old system.  Are theseG > transferable to the new system?  If not, is there a transfer fee?  Do B > we have to buy all new licenses?  Can the license PAKs simply be > installed on the new system?  M Depends on the type of user licenses.  Since they are on alpha, it's possible L that you have the license(s) that are NOT tied to the CPU, and then they canP just be put on the new system.  If your license begins with QL-MT3 then they canN be freely moved from system to system.  Post the license part numbers, and you can get a specific answer.  D > 3.  Will the NAS license cover all of our networking needs such asA > LAT, TCP/IP?  With our old system, I believe we had to purchase F > licenses for these separately but the licensing seem to be different > for this system.  P The major component of NAS is TCP/IP, and it also includes end-node DECnet, LAT, etc.  H > 4.  Is there anything else to consider, or any other hidden costs that: > we might not have anticipated with regards to licensing?  E Don't know, you haven't posted as much information as you could have.   4 > Brian Oster  Remove NOSPAM from my email to reply.   Dave   --  4 David Froble                       Tel: 724-529-04504 Dave Froble Enterprises, Inc.      Fax: 724-529-0596; 170 Grimplin Road               E-Mail: davef@tsoft-inc.com  Vanderbilt, PA  15486    ------------------------------  # Date: Wed, 12 Apr 2000 14:57:23 GMT , From: bgoster@flash.net.NOSPAM (Brian Oster)2 Subject: Re: Advice Needed on New AlphaServer 2100/ Message-ID: <38f48717.580176069@news.flash.net>   2 On Wed, 12 Apr 2000 00:11:08 -0400, David A Froble <davef@tsoft-inc.com> wrote:   >Brian Oster wrote:  >>   > @ >Looks like you're getting lots of advice, and less answers. :-)   Thats ok, I need both. :)    <my original message snipped>   K >Some questions.  Is all of the above coming with the 2100?  Are you moving H >anything from your DEC 2000-300?  What is currently on your old system?  A The only thing we are moving is the tape drive (TLZ09 I believe).    > M >Another suggested a new DS10 system.  As David indicated, advantages are the M >3-year warranty, thus saving 3 years of basic maintenance, and a much faster Q >system.  If the RZ29 drives listed are from your old system, they will move over * >to the DS10, just as well as to the 2100.  F We also got a quote on a new DS10 (apx $13k - $15k).  We are basicallyC trying to get adequate performance for as little money as possible. C There is also the possibility that we may be required to move to NT E (due to our ERP software) and we wanted to be able to convert our new C server over to an NT box if needed.  It is my understanding that NTuF will run on the 2100, but not on the DS10.  The 2100 should be a worldE of difference from our Dec 2000-300 (64MB RAM, Data Drive=1 4GB).  Wee@ are mainly memory (RAM) bound on our current server with 16 - 184 concurrent users and 512MB will hopefully cure that.     >eP >If you need new disks, get new 9, 18, or 36 GB drives and forget about the RAIDK >stuff unless you specifically need it.  The best idea is to keep things as2L >simple as you can.  I don't see any tape device, guess you're handling that >seperately.  F Disk space is not really a problem, we will be doubling our disk spaceA for our data.  I did think about going to 3 9GB drives simply foraB performance reasons, since disk performance is probably the second? area where we need the most improvment when moving from our oldtD system.  Would 3 9GB drives be significantly faster than the 3 4.3GBD RZ29B-VW drives?  I would like to keep the RAID for both performance and fault tollerance reasons.e   >yC >> Being relativly new to the Digital/VMS world, I have a couple of 
 >> questions:o >> hG >> 1.  Is this priced about right considering the hardware and the factbI >> that it is a used system?  If it is way over priced, can you recommenda3 >> any other vendors that deal with used equipment?  > O >I really hate to second guess a dealer who's in business to sell used systems,aL >but I wouldn't pay that much for a 2100, a first generation Alpha, when new+ >systems are using third generation Alphas.   E Is $9500, in your opinion, way overpriced?  I mean are we paying 1 or / 2 thousand to much or 5 or 6 thousand too much?e     >a= >> 2.  We have 18 user licenses on the old system.  Are theseaH >> transferable to the new system?  If not, is there a transfer fee?  DoC >> we have to buy all new licenses?  Can the license PAKs simply be  >> installed on the new system?  > N >Depends on the type of user licenses.  Since they are on alpha, it's possibleM >that you have the license(s) that are NOT tied to the CPU, and then they can Q >just be put on the new system.  If your license begins with QL-MT3 then they caneO >be freely moved from system to system.  Post the license part numbers, and you  >can get a specific answer.e  F I finally got a straight answer from Compaq on this one.  They told meB they should work, since we are not trying to use them in a clusterE environment.  I just wanted to get confirmation on this, because thatF2 could be a siginificant, and costly error to make.   >eE >> 3.  Will the NAS license cover all of our networking needs such as B >> LAT, TCP/IP?  With our old system, I believe we had to purchaseG >> licenses for these separately but the licensing seem to be different- >> for this system.e >1Q >The major component of NAS is TCP/IP, and it also includes end-node DECnet, LAT,: >etc.U  F Great!  We are currently running LAT, but would like to switch over to IP.c   >FI >> 4.  Is there anything else to consider, or any other hidden costs thatn; >> we might not have anticipated with regards to licensing?  >iF >Don't know, you haven't posted as much information as you could have. >i5 >> Brian Oster  Remove NOSPAM from my email to reply.  >  >Davei >h >-- 5 >David Froble                       Tel: 724-529-0450o5 >Dave Froble Enterprises, Inc.      Fax: 724-529-0596e< >170 Grimplin Road               E-Mail: davef@tsoft-inc.com >Vanderbilt, PA  15486  2 Brian Oster  Remove NOSPAM from my email to reply.   ------------------------------  # Date: Wed, 12 Apr 2000 15:19:42 GMT , From: bgoster@flash.net.NOSPAM (Brian Oster)2 Subject: Re: Advice Needed on New AlphaServer 2100/ Message-ID: <38f48ed8.582161123@news.flash.net>   C On Tue, 11 Apr 2000 19:38:11 -0500, JM <vmswiz@geonospamcities.com>- wrote:   >Brian Oster wrote:0   <snip>   >>  I >> 4.  Is there anything else to consider, or any other hidden costs that-; >> we might not have anticipated with regards to licensing?f >cG >Make sure this 2100 is NOT an EISA-based machine. Ours has been a paindA >for the past 2 years and was finally shut down today. SWXCR aka nB >Raid-array 210's are annoying. These don't appear to age well. WeF >were running NT 4.0, so VMS may be entirely different on this system. >  >You want PCI! >iC >Another potential problem is the raid controller wants narrow scsiiD >drives. All the old RZ26-VA and RZ28-VA's are reaching end of theirF >useful lives. Even 4gb replacement RZ29's are getting hard to find inG >reasonably new condition. Go with a newer controller in a pci slot andiF >new drives. I'd sacrifice raid for non-raid and new drives if cost isH >the objective. New drives tend to last 3-5 years without failure if youB >make an effort to never power them off and always keep them cool. >o  B The system comes with 8 EISA slots and 3 PCI.  As I mentioned in aE later post disk drive performance is a concern, but I would also likenC RAID 5 on the new system for fault tollerance.  The quote says that A the RZ29B-VW drives are wide SCSI.  Is this correct?  I suppose I C could forget about the additional fault tollerance of RAID 5 if onexF 9GB drive would provide better performance than 3 4GB drives in a RAID configuration.        2 Brian Oster  Remove NOSPAM from my email to reply.   ------------------------------  # Date: Wed, 12 Apr 2000 15:23:27 GMTy, From: bgoster@flash.net.NOSPAM (Brian Oster)2 Subject: Re: Advice Needed on New AlphaServer 2100/ Message-ID: <38f4942c.583525505@news.flash.net>C  A On 11 Apr 2000 23:38:07 GMT, mathog@seqaxp.bio.caltech.edu (DavidE Mathog) wrote:  _ >In article <38f39c06.519999389@news.flash.net>, bgoster@flash.net.NOSPAM (Brian Oster) writes:iH >>We are looking to upgrade our VMS Server with a used AlphaServer 2100.G >>Our old system is a Dec 2000-300 axp.  Following are the basic specs:  >aF >If I were you I'd go with a DS10 instead.  You'll get 3 years supportK >and it will be much faster than the 2100 (just be careful when working on  L >the console - the X11 server can use up all the CPU cycles.)  It will also L >be much smaller physically - put it on a shelf instead of using up 6 sq. ftL >of floor space.  Get an U2W controller directly from Intraserver and two orE >three 18Gb disks.  Buy the disks and memory third party.  That will  I >probably cost about the same as the 2100, but will be less overall sincehJ >you won't have to pay for service (unless you need better than warranty.)  B We are also looking at the DS10, but there is a possiblity that weC might be moving to NT in the future and it is my understanding thattD the DS10 does not support NT while the 2100 does.  If I new for sureB that we would be running VMS for the next 3 years, than a new DS10B probably would be the best answer since we would save a lot on our maintenance contract.a  2 Brian Oster  Remove NOSPAM from my email to reply.   ------------------------------  % Date: Wed, 12 Apr 2000 17:26:06 +0100s- From: Tim Llewellyn <tim.llewellyn@bbc.co.uk>c2 Subject: Re: Advice Needed on New AlphaServer 2100) Message-ID: <38F4A39E.F5944DD6@bbc.co.uk>i   Brian Oster wrote:   >- >-D > We are also looking at the DS10, but there is a possiblity that weE > might be moving to NT in the future and it is my understanding thatrF > the DS10 does not support NT while the 2100 does.  If I new for sureD > that we would be running VMS for the next 3 years, than a new DS10D > probably would be the best answer since we would save a lot on our > maintenance contract.e   Briane  : You DO know that Compaq will not be supporting NT on alpha after NT4 SP6, don't you?i   Regardsa --6 Tim Llewellyn, OpenVMS Infrastructure, Remarcs Project0 MedAS at the BBC, Whiteladies Road, Bristol, UK.A Email tim.llewellyn@bbc.co.uk. Home tim.llewellyn@cableinet.co.uk   A I speak for myself only and my views in no way represent those ofr MedAS or the BBC.i   ------------------------------   Date: 12 Apr 2000 16:41:59 GMT) From: leslie@clio.rice.edu (Jerry Leslie)o2 Subject: Re: Advice Needed on New AlphaServer 2100' Message-ID: <8d290n$bab$1@joe.rice.edu>s  - Brian Oster (bgoster@flash.net.NOSPAM) wrote:d  D : We are also looking at the DS10, but there is a possiblity that weE : might be moving to NT in the future and it is my understanding thateF : the DS10 does not support NT while the 2100 does.  If I new for sureD : that we would be running VMS for the next 3 years, than a new DS10D : probably would be the best answer since we would save a lot on our : maintenance contract.b  E The 2100 will run NT 4.0, Service Pack 6, but not any NT/Windows 2000d versions after that.  4 --Jerry Leslie     (my opinions are strictly my own)   ------------------------------  # Date: Wed, 12 Apr 2000 17:09:05 GMTs, From: bgoster@flash.net.NOSPAM (Brian Oster)2 Subject: Re: Advice Needed on New AlphaServer 2100/ Message-ID: <38f4aa8a.589251038@news.flash.net>   1 On Wed, 12 Apr 2000 17:26:06 +0100, Tim Llewellyng  <tim.llewellyn@bbc.co.uk> wrote:   >o >E >Brian Oster wrote:F >E >> >>E >> We are also looking at the DS10, but there is a possiblity that wedF >> might be moving to NT in the future and it is my understanding thatG >> the DS10 does not support NT while the 2100 does.  If I new for surelE >> that we would be running VMS for the next 3 years, than a new DS10oE >> probably would be the best answer since we would save a lot on ourt >> maintenance contract. >  >Brian >t; >You DO know that Compaq will not be supporting NT on alphar >after NT4 SP6, don't you? >a  F Yes.  And I am beginning to wonder if we should forget about trying toE buy a system that would support both OS's.  Can anyone tell me how NTtC would perform on the 2100 as opposed to an Intel based machine.  In-B other words would a 2100 perform about the same as a PIII 500, 256F RAM, SCSI Hard drives, etc.  If NT performs poorly on the 2100 than weF probably would be better off buying the DS10 and then selling it if we moved to NT.    2 Brian Oster  Remove NOSPAM from my email to reply.   ------------------------------  % Date: Wed, 12 Apr 2000 13:25:15 -0400k0 From: JF Mezei <jfmezei.spamnot@vl.videotron.ca> Subject: AST sanity check./ Message-ID: <38F4B179.D506A437@vl.videotron.ca>a  M With all the coming wildfires and galaxy explosions and clones of VMS running 6 in the same body... I just want to get a sanity check:  L If is still architecturally correct to expect that in a single process, only) one AST can be running at any one time ? @  J Also, assuming it has got its own RAB stream, are RMS writes to an indexedV file permitted while in AST mode ? (the main program will be doing reads and deletes).U (eg: transactions come in via ASTs, but the main program reads/deletes transactions).    ------------------------------   Date: 12 Apr 2000 08:57:00 GMTC From: cczanj@axp0.ccc.nottingham.ac.uk (Andy.Jack@nottingham.ac.uk)o$ Subject: Re: CDE equivalent of xhost4 Message-ID: <8d1dos$qpt$1@oyez.ccc.nottingham.ac.uk>  [ In article <8cv8mg$53t$1@info.service.rug.nl>, helbig@astro.rug.nl (Phillip Helbig) writes:c5 >In article <8cv6lk$kjg$1@oyez.ccc.nottingham.ac.uk>,rG >cczanj@axp0.ccc.nottingham.ac.uk (Andy.Jack@nottingham.ac.uk) writes: d >oG >> I've been using the "new" CDE-style desktop on my Alpha for ages nowtI >> (running VMS 7.2-1) yet I find myself stumped as to the correct way to  >> modify the security.    [...]d > I >Click on the icon with the mouse, window, colour palette, three T's etc p >and then go to security.  e  E Thanks. I made the mistake of trying to find it in the documentation.1C Thanks also to Paul Sture of Zrich, who also mailed me the answer.>   AndyH ------------------------------------------------------------------------? Academic Computing Services, E-mail: Andy.Jack@nottingham.ac.uk 5 University of Nottingham,    Phone:  +44 115 951 3328a5 Nottingham NG7 2RD, UK       Fax:    +44 115 951 3358e   ------------------------------  % Date: Wed, 12 Apr 2000 11:36:04 +0100c- From: Tim Llewellyn <tim.llewellyn@bbc.co.uk>i* Subject: Re: CPU factor for Alpha systems?) Message-ID: <38F45194.8A3FF708@bbc.co.uk>    Dan Sugalski wrote:I   >eG > And don't forget the mandatory patches, either, unless you're fond ofC, > unprivileged users crashing your system...  @  hmmm, isn't is about time for a "remastered" 7.2-1 CD, with the patch for that bug included.   --6 Tim Llewellyn, OpenVMS Infrastructure, Remarcs Project0 MedAS at the BBC, Whiteladies Road, Bristol, UK.A Email tim.llewellyn@bbc.co.uk. Home tim.llewellyn@cableinet.co.uko  A I speak for myself only and my views in no way represent those of- MedAS or the BBC.n   ------------------------------  % Date: Wed, 12 Apr 2000 11:58:37 -0400i" From: Dan Sugalski <dan@sidhe.org>* Subject: Re: CPU factor for Alpha systems?6 Message-ID: <4.3.0.20000412115313.01d32620@24.8.96.48>  / At 11:36 AM 4/12/00 +0100, Tim Llewellyn wrote:r     >Dan Sugalski wrote: >s > >hI > > And don't forget the mandatory patches, either, unless you're fond oft. > > unprivileged users crashing your system... >dB >  hmmm, isn't is about time for a "remastered" 7.2-1 CD, with the >patch for that bug included.u  J Well, I'd say it warrants 7.2-2, or at least 7.2-1.1, but then I'm not in J VMS Engineering's shoes and I don't have to do a full matrix test to make 5 sure that it doesn't break something horribly either.u   					Dan  L ----------------------------------------------------------------------------L Dan Sugalski                          General and VMS-specific perl training
 dan@sidhe.orgt>                                       Mail me for more details   ------------------------------  / Date: Wed, 12 Apr 2000 09:05:30 +0200 (MET DST)e& From: Rudolf Wingert <win@fom.fgan.de> Subject: Re: DE500 problem/docsm/ Message-ID: <200004120607.IAA26629@fom.fgan.de>v   Hello,  D did you use the netconfig command procedure to configure the networkB under OpenVMS. If you will have the network interface, that is notF all what you need. You must it configure for the different protocolls, such as DECnet IV, TCP/IP, LAT.t   Regards Rudolf Wingert   ------------------------------   Date: 11 Apr 2000 21:48:56 CST& From: Rudolf Wingert <win@fom.fgan.de> Subject: Re: DE500 problem/docs / Message-ID: <200004120616.IAA27066@fom.fgan.de>v   Hello,  B our experience is, that it will be better to set both sides to the@ wished parameter (e.g: DE500 >>>SET EWx0_MODE FASTFD, set switch@ to 100Mb/s and full duplex). We did see the following problem if we used the "autonegotiation":   	o the 100Mb/s LED lights,9 	o the LANCP SHOW DEVICE/PARAMETER shows 100Mb/s and full 	 	  duplexr$ 	o but the real speed was less 1Mb/s  G After setup both side fix, we did see full wire speed (93Mb/s sustant).a   Regards Rudolf Wingert   ------------------------------  % Date: Wed, 12 Apr 2000 11:28:06 +0000e/ From: Nigel Arnot <sysmgr@maxwell.ph.kcl.ac.uk> $ Subject: re:  Re: DE500 problem/docs7 Message-ID: <009E87FA.A2DF5BFB.25@maxwell.ph.kcl.ac.uk>    >  > mumfrd@my-deja.com wrote:a > > I > > I have a de500 thats giving me problems. Anyone know where I can find1 > > any DE500 docs on the web? > >    [snip]- > > This is OpenVMS 6.2-1h3, Decnet phase IV.l > I > The recommendation is to use the DE500 for VMS versions > = 7.1, so youA3 > should look for an earlier type of ethernet card.i > ,m  * There are are multiple revisions of DE500.  K The DE500-XA is OK on VMS 6.2-1H3 (It's what this message is going through)e  5 the -AA (and -BA?) need more recent versions of VMS. u  G One other thing to check is auto-negotiation of line speed and duplex. hI It sometimes doesn't work right between different vendors. You could try nM setting the speed explicitly at both ends (which needs a >>> console command g$ on the Alpha that I can't remember).   	Yours,v
 		Nigel Arnot - 		NRA@MAXWELL.PH.KCL.AC.UK                   c  7 		"In the beginning there was nothing, which exploded."v   ------------------------------  % Date: Wed, 12 Apr 2000 06:58:09 -0400t2 From: "Richard B. Gilbert" <DRAGON@compuserve.com> Subject: DE500 problem/docs 7 Message-ID: <200004120658_MC2-A0E6-E9C3@compuserve.com>a  J         There are at least three flavors of DE500:  DE500-XA, DE500-AA, a= ndJ DE500-BA.  The DE500-BA variant is not supported under VMS V6.2-1H3 until=  J you apply a series of patches.  I believe that ALPLAN05_062 is one of the= m.F  There are several others that are required to fully support the card.  3 Message text written by INTERNET:mumfrd@my-deja.comrF >I have a de500 thats giving me problems. Anyone know where I can find any DE500 docs on the web?  D If not maybe someone can help. We just bought a used AS4100, but the# de500 won't connect to the network.eE The LINK light comes on at the de500 end, but not at the 3com networka$ end. I've tried 2 cables, no change.  G At the >>> level, a SHOW CONFIG shows the de500 OK, but when I boot VMS.# and run netconfig, it tells me thatm  there are "no lines configured".  ) This is OpenVMS 6.2-1h3, Decnet phase IV.    Anybody have any ideas?e   Thanks in advancen   <s   ------------------------------  % Date: Wed, 12 Apr 2000 12:59:53 -0400t From: Everhart <ge@gce.com>y Subject: Re: DECnet access' Message-ID: <38F4AB89.F57C2AFC@gce.com>Z  A Though a couple folks claim set access none and passwords are fornB point to point, they were used in a DECnet of 8000+ nodes for many@ years, worked nicely. Admittedly the passwords are not encryptedC so could be sniffed on an Ethernet, but the connections are treated>E as though point to point closely enough that the link level passwordsa@ are of some value. Give someone long enough on your ethernet andB all bets are off; passwords nevertheless buy you a bit of time and* make accidents or casual intrusion harder.  G Mind I haven't tried this for a few years, but I doubt much that Decneto phase IV has changed either.   ------------------------------  # Date: Wed, 12 Apr 2000 17:25:52 GMTh2 From: kilgallen@eisner.decus.org (Larry Kilgallen) Subject: Re: DECnet access' Message-ID: <2000Apr12.132552.1@eisner>o  E In article <38F4AB89.F57C2AFC@gce.com>, Everhart <ge@gce.com> writes: C > Though a couple folks claim set access none and passwords are forh > point to point,   A I do not recall anyone claiming that SET ACCESS NONE was only for  point ot point.o   ------------------------------  % Date: Wed, 12 Apr 2000 01:48:22 -0500s% From: Chris Scheers <asi@airmail.net>c5 Subject: Re: DECW$DEVICE-I-NODEVICE and no decwindowsdO Message-ID: <9E4B476F517B061F.F0292ECE70556716.D9F3DFC8EE42D067@lp.airnews.net>i   lucverhoelst@my-deja.com wrote:a > 0 > I installed VMS 7.1 on a DEC2000 model 300 AXPH > This was a Fresh install , including format of the disk and during theJ > installation the layered product DECWINDOWS was selected and installed . >  > Rom version 2.2  > SRM FW version 370. > Doing a show config on the boot prompt I see > 1  ISA   VGA OKa5 > The installation completed all on the Blue screen .h7 > The VMS startup terminates correctly (job terminated)l! > but the DECWindows do not startf. > On the end of the startup there is a message3 > DECW$DEVICE-I-Nodevice, no graphics devices foundT > 9 > The License for VMS and DW-Motif are loaded and active.i" > I can login on the OPA0: device. > Rund DECW$Configure I geto > Number of Screens 0 ! > Screen devices and order : NONE  > # > Using MC SYSGEN> Windows_system 1g > ( > So the windows parameter is set to 1 . > " > Any suggestions how to proceed ?    9 Make sure that your video card is a Compaq QVision 1024E.i  E AFAIK, this is the only video card supported on the DEC 2000-300 with  VMS and Motif.  H (Well, I found that you can also use a QVision 1280E, but only in 1024 x
 768 mode.)  
 Good luck!  G ------------------------------------------------------------------------$ Chris Scheers, Applied Synergy, Inc.  G 817-237-3360 (Voice)    817-237-3074 (Fax)    Internet: asi@airmail.net)   ------------------------------  % Date: Wed, 12 Apr 2000 06:58:10 -0400.2 From: "Richard B. Gilbert" <DRAGON@compuserve.com>1 Subject: DECW$DEVICE-I-NODEVICE and no decwindowst7 Message-ID: <200004120658_MC2-A0E6-E9C4@compuserve.com>h  G         Did you install *ALL* of the DECwindows support?  There is Base G Support (or some similar name) installed as part of the O/S.  There are0; also separate DECW* savesets that are installed separately.eB The "savesets" may be PCSI files in V7.1.  The second part must be= installed if the machine is to act as both client and server.r  <         Do you have a *supported* graphics device installed?  9 Message text written by INTERNET:lucverhoelst@my-deja.com / >I installed VMS 7.1 on a DEC2000 model 300 AXPtF This was a Fresh install , including format of the disk and during theH installation the layered product DECWINDOWS was selected and installed .   Rom version 2.2n SRM FW version 370, Doing a show config on the boot prompt I see 1  ISA   VGA OKd3 The installation completed all on the Blue screen .n5 The VMS startup terminates correctly (job terminated)f but the DECWindows do not startl, On the end of the startup there is a message1 DECW$DEVICE-I-Nodevice, no graphics devices founda  7 The License for VMS and DW-Motif are loaded and active.l  I can login on the OPA0: device. Rund DECW$Configure I get  Number of Screens 0  Screen devices and order : NONEs  ! Using MC SYSGEN> Windows_system 1o  & So the windows parameter is set to 1 .    Any suggestions how to proceed ? <v   ------------------------------  # Date: Wed, 12 Apr 2000 09:58:36 GMT " From: Andy <andyw9804@my-deja.com> Subject: Denying DECnet access) Message-ID: <8d1hc8$261$1@nnrp1.deja.com>0  G Well, this was a new one for me... We had a little 'incident' the otherA@ day caused by a live application configuration being copied to aD support cluster & then fired up. As a result of this I've been askedF whether it's possible to deny all forms of DECnet access from one node: to another on the same network. Here's a simplified setup:  ? OpenVMS VAX V6.2 on all systems along with DECnet phase IV. TheeD requirement is to prevent all DECnet access (SET HOST, task-to-task,G copies etc.) from node N1 to node N2. These systems are both DECnet endtD nodes and are on the same physical bit of wire. We have a VAXstationF 4000 acting as the DECnet area router also on the same physical bit of wire.c  F Now, I know this isn't necessarily the best approach, but while I workG to realign the powers-that-be it's also raised an interesting technicaltE challenge. While N1 mustn't be able to communicate with N2 it must bee! allowed to talk to N3 through Nn.oF After a long search I unearthed some old (V6.0) Phase IV documentationF but can't really find anything relating to this except for the NCP SETE NODE xxx ACCESS NONE command. This works as documented but the accesse4 restriction is overridden if the user has OPER priv.  ( Has anybody ever looked at this before ?   -Andyt    & Sent via Deja.com http://www.deja.com/ Before you buy.S   ------------------------------  % Date: Wed, 12 Apr 2000 12:40:54 +0200i5 From: Oswald Knoppers <Oswald.Knoppers@whitehouse.nl> " Subject: Re: Denying DECnet access- Message-ID: <38F452B6.426D13C3@whitehouse.nl>-   Andy wrote:, > H > After a long search I unearthed some old (V6.0) Phase IV documentationH > but can't really find anything relating to this except for the NCP SETG > NODE xxx ACCESS NONE command. This works as documented but the accessn6 > restriction is overridden if the user has OPER priv.  D You can probably do something in sylogin.com and check on the remoteD node. The problem is that this will not prevent access via poor mans! routing (copy a::b::file.ext []).n  H So the question is do you want to prevent accidental access or malicious access?    Regards,   Oswald   ------------------------------  # Date: Wed, 12 Apr 2000 11:29:04 GMTi2 From: kilgallen@eisner.decus.org (Larry Kilgallen)" Subject: Re: Denying DECnet access' Message-ID: <2000Apr12.072904.1@eisner>B  N In article <8d1hc8$261$1@nnrp1.deja.com>, Andy <andyw9804@my-deja.com> writes:  H > After a long search I unearthed some old (V6.0) Phase IV documentationH > but can't really find anything relating to this except for the NCP SETG > NODE xxx ACCESS NONE command. This works as documented but the access 6 > restriction is overridden if the user has OPER priv.  D I did not recall the OPER nuance, but certainly that cannot have anyG effect on the target node, since that node has no information regardingb, privilege of the originating process.  Thus:   	If node A has:u    		MCR NCP SET NODE B ACCESS NONE   	and node B has:    		MCR NCP SET NODE A ACCESS NONE  * the isolation you seek should be provided.  A The lack of a Phase V equivalent, seems to be the final matter inwA which DECnet-Plus still has not caught up to regular DECnet.  Theu? US DECUS Security SIG raised this concern to DEC before Phase Va
 was released.-   ------------------------------  % Date: Wed, 12 Apr 2000 13:12:57 -0400-0 From: JF Mezei <jfmezei.spamnot@vl.videotron.ca>" Subject: Re: Denying DECnet access. Message-ID: <38F4AE97.C46D89F@vl.videotron.ca>   Larry Kilgallen wrote: >         If node A has:0 >                 MCR NCP SET NODE B ACCESS NONE >         and node B has:l0 >                 MCR NCP SET NODE A ACCESS NONE   Actually wouldn't: node A has:i! MC NCP SET NODE B ACCESS OUTGOINGi  M That would prevent node B from accessing node A, but would still allow node A  to access node B.o    K So the commands would need to be placed on all the target nodes, instead ofu the single "insecure" node.s  	 Question:D  3 Would the above still prevent poor man's routing ? l  M if, on node B, I try to access C::A::login.com , will node A realise that the K call is coming from B and block it, or will it only see the call from C and : think it is legal because C has full inbound access to A ?   ------------------------------  % Date: Tue, 11 Apr 2000 12:29:56 +0100h- From: Andy Williams <Andy.Williams@liffe.com> 7 Subject: Denying DECnet access from one node to anotherrM Message-ID: <78E5B8E274DBD1118D6800805FE60E771FFC17@ntprdex4.admin.liffe.com>o  G Well, this was a new one for me... We had a little 'incident' the other0H day caused by a live application configuration being copied to a supportD cluster & then fired up. As a result of this I've been asked whetherA it's possible to deny all forms of DECnet access from one node tod7 another on the same network. Here's a simplified setup:B  ? OpenVMS VAX V6.2 on all systems along with DECnet phase IV. The D requirement is to prevent all DECnet access (SET HOST, task-to-task,G copies etc.) from node N1 to node N2. These systems are both DECnet endeD nodes and are on the same physical bit of wire. We have a VAXstationF 4000 acting as the DECnet area router also on the same physical bit of wire.n  F Now, I know this isn't necessarily the best approach, but while I workG to realign the powers-that-be it's also raised an interesting technicalpE challenge. While N1 mustn't be able to communicate with N2 it must be ! allowed to talk to N3 through Nn.rF After a long search I unearthed some old (V6.0) Phase IV documentationF but can't really find anything relating to this except for the NCP SETE NODE xxx ACCESS NONE command. This works as documented but the access@4 restriction is overridden if the user has OPER priv.  ( Has anybody ever looked at this before ?   -Andy0   ------------------------------  % Date: Wed, 12 Apr 2000 11:50:50 +0100e- From: Tim Llewellyn <tim.llewellyn@bbc.co.uk>a& Subject: Re: Double broadcast messages) Message-ID: <38F4550A.7F6E8478@bbc.co.uk>w   Peter LANGSTOEGER wrote:  c > In article <38F3A93E.33D6995@vl.videotron.ca>, JF Mezei <jfmezei.spamnot@vl.videotron.ca> writes:e
 > >Situation:y > > @ > >Am logged into node BIKE as user JFMEZEI in a DECTERM window.< > >From there, I SET HOST to node VELO and login as JFMEZEI. > > ! > >BIKE and VELO are in a cluster-R > > (each with their own SYSUAF and system diSk - not a homogenous cluster - yet). > >iQ > >whenever JFMEZEI on VELO gets a new mail message, two broadcasts are delivered2N > >to the DECterm window. (I guess one delivered to JFMEZEI on VELO and one to: > >JFMEZEI on BIKE, both being on the same DECterm window) > > Q > >The problem is that doing SET TERM/NOBROADCAST disables the broadcast for bothe7 > >sessions on the same terminal so I get 0 messages...w > >I5 > >Is there a way to eliminate duplicate broadcasts ?r >nH > Yes. Issue a SET BROADCAST=NOMAIL on BIKE before doing a SET HOST VELO  ' Yes, that solves the question JF asked..  R However, I'd suggest setting up a DECNET object in the cluster to start a DECTerm,L then, just start a DECterm on VELO by issuing a command on BIKE. If you haveL default proxy enabled for user JFMEZEI in the cluster you don't even need toM type username and password again. Works great in clusters. I'm sure I've seen-5 an example posted here in the past.  Anyway, try this1     $ open/read mynet sys$nets $ sysnet=F$TRNLNM("SYS$NET") $ node=F$ELEMENT (0,":",sysnet) # $ set display/node="''node'"/create3 $ show display $ create/terminal/detached
 $ close mynetn $ exit  F Setup a DECNET object DECTERM to use this on VELO, then invoke it with   BIKE $ type velo::"0=decterm"d  P I know this isn't quite what you asked for, JF, but you won't get two broadcasts if you don;t lo9g in twice.D   --6 Tim Llewellyn, OpenVMS Infrastructure, Remarcs Project0 MedAS at the BBC, Whiteladies Road, Bristol, UK.A Email tim.llewellyn@bbc.co.uk. Home tim.llewellyn@cableinet.co.ukr  A I speak for myself only and my views in no way represent those ofr MedAS or the BBC.-   ------------------------------  % Date: Wed, 12 Apr 2000 13:05:10 -0400 0 From: JF Mezei <jfmezei.spamnot@vl.videotron.ca>& Subject: Re: Double broadcast messages/ Message-ID: <38F4ACC6.CE29C327@vl.videotron.ca>    Tim Llewellyn wrote: > $ open/read mynet sys$net  > $ sysnet=F$TRNLNM("SYS$NET")! > $ node=F$ELEMENT (0,":",sysnet)e% > $ set display/node="''node'"/create0 > $ show display > $ create/terminal/detached > $ close mynet  > $ exit  K Thanks. I'll have to fiddle with it to get it to work. VELO only has "base"pL decwindows support, so create/terminal doesn't work there. What exactly doesN "base" decwindows install on a node ? I woudl have expected that functionalityN to have existed (run create/terminal on a "base" decwindows to redirect output+ on a workstation that has full decwindows).i  M I'll probably fiddle with copying enough files to get it to work though. (set0 display works).    ------------------------------  % Date: Wed, 12 Apr 2000 14:52:15 +0100tB From: Andrew Harrison SUNUK Consultancy <andrew.nospam@uk.sun.com> Subject: Re: ebay to drop Sun?* Message-ID: <38F47F8E.D05D5A22@uk.sun.com>   Malcolm Dunnett wrote:  ) > In article <2000Apr11.091813.1@eisner>, : >     kilgallen@eisner.decus.org (Larry Kilgallen) writes: >e2 > >> or did they just not make it into the top 13? > >w  > > I know !  I know !  I know !& > > Call on me, teacher.  Call on me ! > >-J > > If you scroll down further on the page you will see that Sun is listedI > > as benchmarking only for the prior version where there are 4 entries,o: > > not for the latest version where there are 14 entries. > J >    I must say that SAP should be commended on the phenomenal performanceP > enhancements they've apparently put into release 4. We've all heard how CompaqD > has no systems which can "play with the big boys", yet the puny 32K > cpu GS320 has roughly twice the performance of the 64 CPU  Sun Enterprise J > 10000. Of course I'm sure this still doesn't make it a suitable platform > for Ebay :-)  - You of course arn't comparing the same fruit.i  A The Sun number is old, using old slower Sun CPU's, an E10000 witheC a slower backplane and longer latency than the current machine, oldi> storage, an old version of Informix and an old version of SAP.  > The GS320 ran a new version of Oracle, SAP etc and hasn't been. released yet, hardly a meaningfull comparison.  D Incedentally the Fujitsu box that is ahead of the GS320 is basically= an E10000 clone, it uses SPARC CPU's, runs Solaris, its CPU'sn@ are almost identical in performance as the E10000's and it has a& similar cross-bar switch architecture.  @ So you conclusions are that a yet to be released Alpha server is> slightly slower than currently available SPARC/Solaris system.  E Do you actually think that Sun does not have anything in the pipelineE) at the high end, or HP or IBM or Fujitsu.5  @ How very embarassing, given the position that Alpha now occupiesC at the bottom of most peoples selection list it needs to be way way.@ faster, you have to attract peoples attention and hold it, being- slightly slower itsn't going to do the trick.r   Regardso Andrew Harrison0 Enterprise IT Architects   ------------------------------  # Date: Wed, 12 Apr 2000 14:49:27 GMTe2 From: kilgallen@eisner.decus.org (Larry Kilgallen) Subject: Re: ebay to drop Sun?' Message-ID: <2000Apr12.104927.1@eisner>e  o In article <38F47F8E.D05D5A22@uk.sun.com>, Andrew Harrison SUNUK Consultancy <andrew.nospam@uk.sun.com> writes:h  B > So you conclusions are that a yet to be released Alpha server is@ > slightly slower than currently available SPARC/Solaris system.  . Where the SPARC system has twice as many CPUs.   ------------------------------    Date: 12 Apr 2000 09:39:47 -0700* From: dunnett@mala.bc.ca (Malcolm Dunnett) Subject: Re: ebay to drop Sun?, Message-ID: <1iEAv$i4TA35@malvm2.mala.bc.ca>  ( In article <2000Apr12.104927.1@eisner>, 8     kilgallen@eisner.decus.org (Larry Kilgallen) writes:  q > In article <38F47F8E.D05D5A22@uk.sun.com>, Andrew Harrison SUNUK Consultancy <andrew.nospam@uk.sun.com> writes:c > C >> So you conclusions are that a yet to be released Alpha server isgA >> slightly slower than currently available SPARC/Solaris system.  > 0 > Where the SPARC system has twice as many CPUs.  ,   Well, if I were a shill for Sun I'd reply:  D   But of course that's no indication that a 64 CPU wildfire would beF faster than a 64 CPU SPARC system, or that Compaq can ( or will ) ever sell such a beast.  K   And of course you shouldn't assume that Sun doesn't have even better gear5J in the pipeline ( whereas you should (apparently) assume that the GS320 asC tested is the penultimate Alpha system and no faster CPUs will ever8 be available )     ;-)    ------------------------------   Date: 12 Apr 2000 16:11:51 GMT2 From: mathog@seqaxp.bio.caltech.edu (David Mathog)$ Subject: faster/smarter FTP clients?, Message-ID: <8d2787$qpq@gap.cco.caltech.edu>  C I think that both the Multinet FTP client and MGFTP need some work,20 the former much more so than the latter.  (LONG)    Consider an operation like this:  : $ ftp/user=anonyymous/password="something" remote.host.gov cd /put/data/directory binary1 get smallbinary datadisk:[somedir]smallbinary.dat 0 get hugebinary  datadisk:[somedir]hugebinary.dat quit  C If I use either the Multinet ftp client or MGFTP in either case thelE transfer is slowed to a great extent because the FTP clients use the  B default extend size for the target disk.  Or at least a very smallG extend size, because I can hear the disk doing a rapid "tick-tick-tick"l which is the hallmark of h   1.  add a few blocks 2.  request an extendt
 3.  goto 1  G If I know ahead of time that I'm only going to transfer a big file thene" I can do BEFORE the ftp command a:   $ set rms/extend=1000   J and it seems to speed things up.  DIR/SIZE from another window while it's I running shows the file extending in 1000 block chunks.  But that isn't aneK appropriate setting for a mixture of small and large files - which is often  what is being transferred. e  K Shouldn't the FTP client be able to do a better job of this?  I mean, when  % a GET starts one sees a message like:I  M <150 Sending "/H/Download/testfile" (3526656 bytes). Mode STREAM Type BINARY..  L The server has told the client that the file will be 3526656 bytes, or 6888 K blocks, so why doesn't the FTP client (either one) allocate a file of that -J size to start with?  That would eliminate all need for extends and should G speed things up a bit.   I would also greatly appreciate it some other -J network optimizations and disk write optimizations were handled by the FTPC clients. The Multinet FTP client, in particular, just reeks in some H instances. In these examples a 3526656 byte testfile is being served viaH WAR-FTPD from a WNT machine.  BINARY get times on some 100baseT machines are: u  / client                    get to        secondsh, LINUX (Intel)             /dev/null     4.33  , WNT   (intel)             C:/TEMP       3.25  - VMS(DS10, Multinet FTP)   nla0:        57.627s-    (version 4.2A-X)       []file.dat   57.646h, VMS(DS10, MGFTP)          nla0:         3.33,    (version V2.2-2)       []file.dat   18.039 VMS(DS10, Netscape 3.03)  nla0:         3.5   (stopwatch)i9                                        12.3   (stopwatch)e  J (because of file caching transfer time to a file was not tested on Linux, C and transfer time on WNT represents only time to RAM, not to disk.).  D This shows that while there's nothing wrong with the Multinet TCP/IPC transport, the Multinet FTP client moves data very poorly from this C particular FTP server.  (If this is a "window" size problem then isnB there some reason the client could not connect at one window size,L detect the mismatch, and reconnect at an appropriate size before attempting K the regular login and data transfer?)  Interestingly, the huge network timeoF apparently completely subsumes the extra disk write operations for theK Multinet FTP client.  (Perhaps it's multithreaded). Conversely, MGFTP moves.K data across the net as well as any other FTP client on any other platform.  I When it comes time to save the data to disk though, the MGFTP client doesuD alright, but 15 seconds to store 3.5 Mb is really too slow. Previous1 timings on this DS10, which uses U2W disks (see: i  \   http://x26.deja.com/[ST_rn=ps]/getdoc.xp?AN=544450275&CONTEXT=955553269.886898705&hitnum=5   )   J showed that a 30720000 byte file could be copied disk to disk on this DS10Y in 9 seconds, so the expected write time here is certainly less than 9 * 3526656/30720000e = 1 second.e  G I don't mean to suggest that the Multinet FTP client is always bad.  IteJ appears to work very well to itself.  (However, I'm not entirely sure whatH the next set of commands is testing since I don't think the packets everJ actually go out through the ethernet port as the light on the card doesn't flash)  A $ ftp/user=anonyymous/password="something" seqaxp.bio.caltech.eduI cd /lasergraphicsc binary
 statistics get testfile nlao: get killme.dat nla0:C <VMS retrieve of ANON_ROOT:[PUB.LASERGRAPHICS]KILLME.DAT;1 started.W4 <Transfer completed.  3526704 (8) bytes transferred.* 3526704 bytes transferred at 22427370 bps.* Run time = 0. ms, Elapsed time = 1258. ms. get killme.dat killme2.datC <VMS retrieve of ANON_ROOT:[PUB.LASERGRAPHICS]KILLME.DAT;1 started. 4 <Transfer completed.  3526704 (8) bytes transferred.* 3526704 bytes transferred at 11016646 bps.* Run time = 0. ms, Elapsed time = 2561. ms. quit $   D That is, it's "network" time was only 1.26 seconds, and file storage? time was only an additional 1.3 seconds, approximately the same1) as a simple COPY operation between disks.o   Regards,   David Mathog mathog@seqaxp.bio.caltech.edug? Manager, sequence analysis facility, biology division, Caltech     ------------------------------  # Date: Wed, 12 Apr 2000 16:26:23 GMT - From: goathunter@PROCESS.COM (Hunter Goatley)i( Subject: Re: faster/smarter FTP clients?, Message-ID: <38f4a2c0.18860299@news.wku.edu>  A On 12 Apr 2000 16:11:51 GMT, mathog@seqaxp.bio.caltech.edu (David  Mathog) wrote:  D >I think that both the Multinet FTP client and MGFTP need some work,1 >the former much more so than the latter.  (LONG)e >v [...] D >If I use either the Multinet ftp client or MGFTP in either case theF >transfer is slowed to a great extent because the FTP clients use the C >default extend size for the target disk.  Or at least a very small:H >extend size, because I can hear the disk doing a rapid "tick-tick-tick" >which is the hallmark of  >gA Funny you should mention.  I recently modified MGFTP (but haven't1A released it yet) and we're working to modify MultiNet FTP so thats= the initial allocation quantity is the size of the file to be C transferred.  No more extends required at all, assuming the initial  size can be determined.a  L >Shouldn't the FTP client be able to do a better job of this?  I mean, when & >a GET starts one sees a message like: >eN ><150 Sending "/H/Download/testfile" (3526656 bytes). Mode STREAM Type BINARY. >tM >The server has told the client that the file will be 3526656 bytes, or 6888 TL >blocks, so why doesn't the FTP client (either one) allocate a file of that K >size to start with?  That would eliminate all need for extends and should o  F MGFTP does that now.  It will also use the SIZE command, if the remote server supports it.   H >speed things up a bit.   I would also greatly appreciate it some other K >network optimizations and disk write optimizations were handled by the FTP-D >clients. The Multinet FTP client, in particular, just reeks in someI >instances. In these examples a 3526656 byte testfile is being served viagI >WAR-FTPD from a WNT machine.  BINARY get times on some 100baseT machines  >are:   D Some of these may be the default window sizes, as has been discussed! here in the not-too-distant past.C  3 I'll release my MGFTP change in the next day or so.e   Hunter       Hunter ------9 Hunter Goatley, Process Software, http://www.process.com/d8 <goathunter@PROCESS.COM>     http://www2.wku.edu/hunter/   ------------------------------  % Date: Wed, 12 Apr 2000 09:04:37 -0400o* From: Clair Grant <grant@evms.zko.dec.com>; Subject: Re: Galaxy (or perhaps Wildfire) and shared memory / Message-ID: <38F43C25.5A79691@evms.zko.dec.com>w   JF Mezei wrote:t > 3 > Can hard partitions be dynamically reconfigured ?o >   E What is your definition of dynamic? (I don't mean to be argumentativecE but every company seems to have their own definition and yet everyone1 uses the same terminology.)l  mN > Also, if you create 2 hard partitions, each running VMS, is it fair to stateK > that these 2 instances will not be able to form a galaxy but will have tof' > cluster with more traditional means ?m > N > (or: does the formation of a galaxy require soft partitions in a wildfire ?)  D Two VMS in different hard partitions can't be in the same Galaxy. IfF they are in the same cluster they will communicate via an external I/O path.s       -- o Clair Grante VMS Exec Group Project Leader  COMPAQ Computer Corporation  110 Spit Brook Rd. Nashua, NH  03062e   ------------------------------  % Date: Wed, 12 Apr 2000 13:17:38 -0400s0 From: JF Mezei <jfmezei.spamnot@vl.videotron.ca>; Subject: Re: Galaxy (or perhaps Wildfire) and shared memory / Message-ID: <38F4AFB0.21A2D90A@vl.videotron.ca>a   Clair Grant wrote:5 > > Can hard partitions be dynamically reconfigured ?o > >k > G > What is your definition of dynamic? (I don't mean to be argumentativeeG > but every company seems to have their own definition and yet everyonei > uses the same terminology.)i  M By dynamic, I meant whether you can re-allocate resources dynamically without & rebooting the whole kit and caboodle.   N I know that with galaxies, that seems to be a big marketing advantage that oneM can "on the fly", switch a cpu from one instance to another when the workloadiN on the other increases. So I was wondering if the same could be accompliseshedG with hard partitions or whether hard partitions are configured when youpJ power-up the beast, and once you've begun to boot an operating system, you0 can't reconfigure the machine. Is that correct ?   ------------------------------  % Date: Tue, 11 Apr 2000 17:18:38 +01008. From: Roger@natron.demon.co.uk (Roger Barnett)Y Subject: Re: how to communicate with CICS directly nor through CICS Transaction Server fou- Message-ID: <155178613wnr@natron.demon.co.uk>,  : Depending on what you actually want to do, you may be able to use a CORBA-based approach.  c Try Iona   http://www.iona.com. or check the OMG web site   http://www.omg.org0 for other companies that support CICS platforms.   -- o
 Roger Barnett    ------------------------------  + Date: Wed, 12 Apr 2000 07:35:31 -0600 (MDT)n) From: John Nebel <nebel@athena.csdco.com>  Subject: HSJ80 experience?G Message-ID: <Pine.OSF.4.21.0004120732320.10021-100000@athena.csdco.com>a   Hello,  G Any HSJ80 experiences out there?  Good/bad/indifferent?  The specs lookh pretty good.  
 John Nebel   ------------------------------  # Date: Wed, 12 Apr 2000 13:55:19 GMTe/ From: "John Nixon" <jorlnixon@worldnet.att.net>e Subject: Re: HSJ80 experience?C Message-ID: <bf%I4.630$WF.45739@bgtnsc04-news.ops.worldnet.att.net>,  E I am considering one myself because I still have three VAX 78xxs with J CIXCDs.  If I had all Alpha, I would certainly be looking at HSGs instead.  G I like the idea of the HSJ80 having two CI ports, but you still have to K assign individual disks/arrays to a particular CI on the HSJ.  I was hopingnI that it could somehow miraculously use both CIs to pump twice the IO to a,L single disk array, but that is not the case.  At any rate, you need multipleI CI adapters on your host to take advantage of that major feature.  If youe6 have CIXCDs, make sure you have the 4k packet updates.      6 "John Nebel" <nebel@athena.csdco.com> wrote in messageA news:Pine.OSF.4.21.0004120732320.10021-100000@athena.csdco.com...  >  > Hello, >,I > Any HSJ80 experiences out there?  Good/bad/indifferent?  The specs lookd > pretty good. >i > John Nebel >b   ------------------------------  % Date: Wed, 12 Apr 2000 09:57:08 -0400 . From: "Kenneth Randell" <kenr@datametrics.com>+ Subject: Re: Is sys$dclexh ever not called? + Message-ID: <8d1v6e$5a2$1@bob.news.rcn.net>m  ! Wayne Sewell wrote in message ...r   <snip>   >:L >Fatal errors in exec rundown handlers would not be *quite* so catastrophic.G >They would kill the process, which of course is already dying when theN routineNL >is called.  :-)  On the other hand, the rundown would not complete whatever it
 >is doing. >y    L If you are going to write/debug EXEC mode handlers, I would strongly suggestF making sure that BUGCHECKFATAL and SYSTEM_CHECK are set to 0.  If not,L then EXEC level exceptions WILL cause a system crash.  This can be useful in? debugging your code, but may not really be what you want to do.    Ken Randell    ------------------------------  # Date: Wed, 12 Apr 2000 14:31:34 GMTs From: briggs@eisner.decus.org)+ Subject: Re: Is sys$dclexh ever not called?h' Message-ID: <2000Apr12.103134.1@eisner>i  l In article <8uTykc22CpeU@tachxxsoftxxconsult>, wayne@tachyon.xxx.301332.killspam.0537 (Wayne Sewell) writes:c > In article <#mAYvWvo$GA.303@cpmsnbbsa04>, "cstranslations" <cstranslations@email.msn.com> writes:.N >> What I'm wondering is if the registered routine is called if someone uses a9 >> "big stick" to hit it over the head - say sys$delprc. l > P > Exit handlers are not called if sys$delprc is called, or if a "stop/id=xxx" is7 > done on the process, which amounts to the same thing.   ' A few other cases of possible interest.o  H In the case of a $ STOP (no PID or process name specified) from the sameE process then image rundown occurs without invoking exit handlers eveniC though process rundown does not occur.  This is often quite useful.e2 Especially with multi-tasking apps written in Ada.  D In the case of an exit handler itself calling $EXIT I seem to recallF that previously declared exit handlers are not invoked.  That's a veryC unreliable recollection, however.  Exit handlers are executed LIFO.r  F In the case of an exit handler going into an infinite loop, previously3 declared exit handlers are, obviously, not invoked.n  @ In the case of an $UNWIND in the context of an exit handler or aA condition handler triggered from an exit handler, I have no clue.e  & 	John Briggs			briggs@eisner.decus.org   ------------------------------  # Date: Wed, 12 Apr 2000 12:29:50 GMTo, From: koehler@eisner.decus.org (Bob Koehler)% Subject: Re: Manuals for DEC 3000-800 ' Message-ID: <2000Apr12.082950.1@eisner>e  Y In article <38F33A9C.9A6E76D1@bbc.co.uk>, Tim Llewellyn <tim.llewellyn@bbc.co.uk> writes:   F > Someone (Bob Koehler?) was asking for 3000-300 manuals. They are at:  8 Nope, not me.  My 3000 is a 600S and I have the manuals.  F ----------------------------------------------------------------------? Bob Koehler                     | Computer Sciences Corporatione= Hubble Space Telescope Payload  | Federal Sector, Civil Group E  Flight Software Team           | please remove ".aspm" when replying    ------------------------------  % Date: Wed, 12 Apr 2000 11:13:25 +0100a* From: "Richard Brodie" <R.Brodie@rl.ac.uk> Subject: Re: netscape 6i, Message-ID: <8d1i8a$18pe@newton.cc.rl.ac.uk>  ; "Michael Austin" <maustin@nc.prestige.net> wrote in message ) news:38F3E00D.C9F7F172@nc.prestige.net...,J > too bad IE is not available for anything other than Wintel.... The majorH > problem with WWW is the inescapable fact that there are sites that can7 > only be read in their entirety with one or the other.a  J Fine. Obviously there are plenty of people out there who like to turn away? business. Why not put up a big banner - "we care far more aboutsD packaging than content". If I happened to be on a Wintel box I mightK deign to fire up IE, if not I would go looking for someone that has a clue.2  D I don't really want to start a long thread on this but now there areC rich standards for the web, there are no excuses for not keeping tocA them. I realise that sometimes you might have to visit such sitesn- but you can always complain to the webmaster.a   ------------------------------  % Date: Wed, 12 Apr 2000 00:32:15 -0400n* From: David A Froble <davef@tsoft-inc.com> Subject: Re: netscape 6a- Message-ID: <38F3FC4F.BFFF8F13@tsoft-inc.com>l   Jerome LAURET wrote: > c > In article <8cvfsf$n5f@gap.cco.caltech.edu>, mathog@seqaxp.bio.caltech.edu (David Mathog) writes:r6 > |>Netscape 6 is out now for Windows, Mac, and Linux.= > |>(At least the "Preview Release" is, whatever that means.)C > |>- > |>When can we expect version 6 for OpenVMS?  > |> > |>Regards, > |> > |>David Mathog! > |>mathog@seqaxp.bio.caltech.edueB > |>Manager, sequence analysis facility, biology division, Caltech > |> > F >         Interresting question but since the version 5 is still under> > beta-release like state, I am wondering if 6 will ever come. > 8 >                   Jerome LAURET S.U.N.Y. @ Stony Brook& >        ,,,,,      Dept. of Chemistry- >       ( o o )     Stony Brook NY 11794-34008= >   ---m---U---m---------------------------------------------t( >   E-mail: jlauret@mail.chem.sunysb.edu> >   URL   : http://nucwww.chem.sunysb.edu/jlauret/jlauret.html  N It looks like even numbered versions are being skipped, so the nex VMS version! should be Netscape Version 7. :-)    Dave   -- a4 David Froble                       Tel: 724-529-04504 Dave Froble Enterprises, Inc.      Fax: 724-529-0596; 170 Grimplin Road               E-Mail: davef@tsoft-inc.comi Vanderbilt, PA  15486e   ------------------------------  # Date: Wed, 12 Apr 2000 11:11:57 GMTd= From: system@SendSpamHere.ORG (Brian Schenkenberger, VAXman-)  Subject: Re: netscape 6i0 Message-ID: <009E87D6.DA86A6A7@SendSpamHere.ORG>  Y In article <8d1i8a$18pe@newton.cc.rl.ac.uk>, "Richard Brodie" <R.Brodie@rl.ac.uk> writes: 
 >{...snip...}eE >I don't really want to start a long thread on this but now there areeD >rich standards for the web, there are no excuses for not keeping toB >them. I realise that sometimes you might have to visit such sites. >but you can always complain to the webmaster.  
 Somewhat O.T.a  M I heard a report on the news about a possible 'remedy' in the monopoly$chlockhM anti-trust case -- open source IE royalty-free!  This, IMHO, is a double edgeo sword!  L The problem with monopoly$chlock bundling IE with their excuse for an O/S isL that it makes the use of non-standard web features/functions pervasive.  GetL IE running and you can get all of the proprietary web served features comingM from monopoly$chlock's IIS.  If the source for IE is made 'open' and 'royaltytM free', IE will become more and more ingrained as the 'standard' making monop-o9 oly$schlock's attempts to monopolize the web even easier!o  L If a site can't adhere to the standards and/or baises itself to non-standard8 based browsers and web features, they'll not get my biz.   --N VAXman- OpenVMS APE certification number: AAA-0001           VAXman@TMESIS.COM  L GNU Freeware -- What does the GNU *really* stand for?  Garbage!  Not Usable!   ------------------------------  # Date: Wed, 12 Apr 2000 11:18:06 GMT 2 From: kilgallen@eisner.decus.org (Larry Kilgallen) Subject: Re: netscape 6e' Message-ID: <2000Apr12.071806.1@eisner>   ` In article <38F3E00D.C9F7F172@nc.prestige.net>, Michael Austin <maustin@nc.prestige.net> writes:  . > This is a multi-part message in MIME format.  & Please don't send MIME to comp.os.vms.  = > too bad IE is not available for anything other than Wintel.   E Internet Explorer is available for Macintosh.  There was a news storytC yesterday with one of the web "standards" groups blasting MicrosoftrA for not supporting all aspects of HTML 4.0 on the latest InternetvD Explorer for Windows (and instead spending the effort on proprietaryD extensions).  At the same time, that group praised Microsoft for theC high degree of HTML 4.0 support in Internet Explorer for Macintosh.n   ------------------------------  # Date: Wed, 12 Apr 2000 16:36:48 GMT?- From: goathunter@PROCESS.COM (Hunter Goatley)e Subject: Re: netscape 6t, Message-ID: <38f4a5f6.19682782@news.wku.edu>  B What's really too bad, IMO, is Netscape 6.  From my playing aroundD with it, it's just more bloatware that runs slower and isn't as easy to use as Netscape 4.7.i   Hunter     Hunter ------9 Hunter Goatley, Process Software, http://www.process.com/o8 <goathunter@PROCESS.COM>     http://www2.wku.edu/hunter/   ------------------------------  % Date: Wed, 12 Apr 2000 13:20:41 -0400.5 From: "John L. Ferguson" <John.L.Ferguson@compaq.com>  Subject: Re: netscape 606 Message-ID: <8d2ba0$qbs$1@mailint03.im.hou.compaq.com>  K Here in OpenVMS, we are planning to provide Netscape Navigator 6 on OpenVMS6G Alpha.  As Navigator 6 will be based upon Mozilla, we have been porting = Mozilla over the past year to make the Navigator port easier..  L As far as I know, Netscape has not released source code so that porting pastJ the reference platforms can begin.  I don't have a date for release of the source code.  K One of the issues (at least from a Compaq viewpoint) is working through the.J porting contract issues resulting from the sale of Netscape to AOL and the Sun/Netscape Alliance.  
 Stay tuned...A   John L. Ferguson Johh.L.Ferguson@compaq.com Product Manager, Compaq   A David Mathog wrote in message <8cvfsf$n5f@gap.cco.caltech.edu>... 3 >Netscape 6 is out now for Windows, Mac, and Linux. : >(At least the "Preview Release" is, whatever that means.) >A* >When can we expect version 6 for OpenVMS? >e	 >Regards,  > 
 >David Mathoge >mathog@seqaxp.bio.caltech.edu? >Manager, sequence analysis facility, biology division, Caltech    ------------------------------  % Date: Wed, 12 Apr 2000 13:42:31 -0400 % From: "Brian Tillman" <tillman_brian>, Subject: Re: netscape 6 $ Message-ID: <38f4b5b5$1@news.si.com>  ? >too bad IE is not available for anything other than Wintel....a  9 I though MSIE was available on Macs and NT/Alpha as well.    -- -  B  Brian Tillman                   Internet: tillman_brian at si.comB  Smiths Industries, Inc.                   tillman at swdev.si.com>  3290 Patterson Ave. SE, MS      Addresses modified to prevent=  Grand Rapids, MI 49512-1991     SPAM.  Replace "at" with "@"(9         This opinion doesn't represent that of my companyT   ------------------------------  % Date: Wed, 12 Apr 2000 13:44:39 -0400r% From: "Brian Tillman" <tillman_brian>r Subject: Re: netscape 6o$ Message-ID: <38f4b635$1@news.si.com>  L >Here in OpenVMS, we are planning to provide Netscape Navigator 6 on OpenVMSH >Alpha.  As Navigator 6 will be based upon Mozilla, we have been porting> >Mozilla over the past year to make the Navigator port easier.  ' And when will there be a VAX version???n -- r  B  Brian Tillman                   Internet: tillman_brian at si.comB  Smiths Industries, Inc.                   tillman at swdev.si.com>  3290 Patterson Ave. SE, MS      Addresses modified to prevent=  Grand Rapids, MI 49512-1991     SPAM.  Replace "at" with "@" 9         This opinion doesn't represent that of my companyh   ------------------------------  # Date: Wed, 12 Apr 2000 13:39:52 GMTd From: jgessling@yahoo.com(! Subject: New VMS books coming outN) Message-ID: <8d1ub5$g27$1@nnrp1.deja.com>h  
 Greetings,  @ I noticed on Amazon just now that they have listed two VMS books due out this summer.  They are:f  : Introduction to OpenVMS - Lesley Ogilvie Rice  (june 2000); OpenVMS Performance Management - Joginder Sethi (june 2000)u  3 Looks like a positive sign for VMS, anyone have any  information on these?b   Jim     & Sent via Deja.com http://www.deja.com/ Before you buy.z   ------------------------------  % Date: Wed, 12 Apr 2000 10:30:22 -0400 + From: Clayton Kroh <ckroh@MAIL.CLARION.EDU>s% Subject: Re: New VMS books coming out + Message-ID: <38F4887E.92A0B8CB@clarion.edu>   C I have had the Introduction to Open VMS by Lesley Ogilvie Rice bookeG ($39.95) on order with Amazon since 12/3/99 when it was stated that thedG book would be published in December. I haven't received any informationnG as to the status of the order. Guess I will wait until June to see if Iy get it.    jgessling@yahoo.com wrote:   > Greetings, >RB > I noticed on Amazon just now that they have listed two VMS books! > due out this summer.  They are:0 >4< > Introduction to OpenVMS - Lesley Ogilvie Rice  (june 2000)= > OpenVMS Performance Management - Joginder Sethi (june 2000)- > 5 > Looks like a positive sign for VMS, anyone have anyt > information on these?n >e > Jimc >y( > Sent via Deja.com http://www.deja.com/ > Before you buy.I   ------------------------------    Date: 12 Apr 2000 11:02:56 -0500. From: rjordan@Mercury.mcs.net (Richard Jordan)% Subject: Re: New VMS books coming oute+ Message-ID: <8d26ng$2rll$1@Mercury.mcs.net>t  H There was also a book called something like "OpenVMS and the Internet"; H I had it on order for a few months but Amazon canceled, stating that theJ publisher had cancelled.  Very disappointing.  I registered my displeasure= with BH (who prints Digital Press books now) but to no avail.f   Rich Jordanh rjordan@mcs.nete   >l >mD >I have had the Introduction to Open VMS by Lesley Ogilvie Rice bookH >($39.95) on order with Amazon since 12/3/99 when it was stated that theH >book would be published in December. I haven't received any informationH >as to the status of the order. Guess I will wait until June to see if I >get it. >G >jgessling@yahoo.com wrote:n >h
 >> Greetings,e >>C >> I noticed on Amazon just now that they have listed two VMS books " >> due out this summer.  They are: >>= >> Introduction to OpenVMS - Lesley Ogilvie Rice  (june 2000)d> >> OpenVMS Performance Management - Joginder Sethi (june 2000) >>6 >> Looks like a positive sign for VMS, anyone have any >> information on these? >> >> Jim   ------------------------------  # Date: Wed, 12 Apr 2000 12:23:51 GMTh, From: koehler@eisner.decus.org (Bob Koehler)) Subject: Re: Portable unformatted files 2a' Message-ID: <2000Apr12.082351.1@eisner>a  [ In article <8cv6ra$4hu$1@info.service.rug.nl>, helbig@astro.rug.nl (Phillip Helbig) writes:gG > In article <2000Apr11.080726.1@eisner>, koehler@eisner.decus.org (Bobd  I > More to the subject of this newsgroup, isn't VMS still alive, well and u: > in heavy use at the Space Telescope Science Institute?     Yes.  J > What about the rumour of the International Space Station being "full of  > VAXen"---any truth to that?r  H I believe none.  There is a VAX/VMS based Ada cross compiler being used H to support code development, I forget the target chip (possibly 8051).  G For this reason there are a large number of VAXen on the space station y= project, but none that I know of on the space station itself./  F ----------------------------------------------------------------------? Bob Koehler                     | Computer Sciences Corporation-= Hubble Space Telescope Payload  | Federal Sector, Civil Group E  Flight Software Team           | please remove ".aspm" when replying0   ------------------------------   Date: 12 Apr 2000 01:44 -0400@ From: hein@eps.zko.dec.c*m Subject: Re: RMS errors & Message-ID: <12APR200001442946@miasys>  b In article <38F3A5A8.46967575@nc.prestige.net>, Michael Austin <maustin@nc.prestige.net> writes...  3 >When trying to read all record sequentially I get:n > 8 >%RMS-F-CHK, bucket format check failed for VBN = 447472  " - Restore a backup! (ya right  :^)  = - Request Support from compaq to revive the file for a price.V  < - Use DUMP/BLOC=(STA=problem-minus-bucket,COUN=3-bucket) andG   ANAL/RMS/INT ... POSI /BUCK to learn about the nature of the problem.p  = Check out my rms_tools directory on the freeware cd/websites. F There is a zap tool there that can help you verify basic bucket formatG and help patch the bucket, to the bucket(s) before it avoid the problem % area (at the cost of some data loss).   0 Standard patch techniques. May need more than 1. 	- make check byte matchE 	- adjust first-free-byte to point just after last known valid recordJA 	- patch next-vbn in prior bucket to point to next bucket, jumpin.; 	   around the problem bucket(s) loosing the record in theme- 	- reconstruct data in bucket (challenging!).o   > Q >I dumped the data for these blocks and it does appear to be corrupt data.  There-  > yeah... so go on, speak to us! Don't give us just teaser info.   Cheers,R 	Hein    ------------------------------  % Date: Wed, 12 Apr 2000 13:34:00 +0100g+ From: Craig Brooks <Craig.Brooks@liffe.com>n Subject: SenilityeM Message-ID: <78E5B8E274DBD1118D6800805FE60E77205956@ntprdex4.admin.liffe.com>   G As a colleague of Andy's (I sit opposite him) I can totally support thea Senility excuse !!!!!!     Craig       -----Original Message-----, From: 	Andy Williams   Sent:	12 April 2000 13:32o Posted To:	vms# Conversation:	Denying DECnet accessd Subject:	That's got it !   Thanks Larry !  D Don't even ask me why I didn't think of doing the same on the remote system. senility I expect...   Cheers,c   -Andyi   -----Original Message-----2 From: kilgallen@eisner.decus.org (Larry Kilgallen)# [mailto:kilgallen@eisner.decus.org]-- Posted At: Wednesday, April 12, 2000 12:29 PM3 Posted To: vms# Conversation: Denying DECnet accessI" Subject: Re: Denying DECnet access    F In article <8d1hc8$261$1@nnrp1.deja.com>, Andy <andyw9804@my-deja.com> writes:,  H > After a long search I unearthed some old (V6.0) Phase IV documentationH > but can't really find anything relating to this except for the NCP SETG > NODE xxx ACCESS NONE command. This works as documented but the accessl6 > restriction is overridden if the user has OPER priv.  D I did not recall the OPER nuance, but certainly that cannot have anyG effect on the target node, since that node has no information regardingn, privilege of the originating process.  Thus:   	If node A has:o    		MCR NCP SET NODE B ACCESS NONE   	and node B has:    		MCR NCP SET NODE A ACCESS NONE  * the isolation you seek should be provided.  A The lack of a Phase V equivalent, seems to be the final matter intA which DECnet-Plus still has not caught up to regular DECnet.  Then? US DECUS Security SIG raised this concern to DEC before Phase Vl
 was released.l   ------------------------------   Date: 12 Apr 2000 10:01:39 GMT* From: helbig@astro.rug.nl (Phillip Helbig)# Subject: Re: signature file in MAILx. Message-ID: <8d1hi3$qgs$1@info.service.rug.nl>  B In article <38f39ecf$1@news.kapsch.co.at>, eplan@kapsch.net (Peter LANGSTOEGER) writes: c  + > 2) MAIL/OLD (eg. via symbol in LOGIN.COM)e  F That's it!  I use MAIL/OLD because the XLATE stuff in EDT doesn't workB with the new MAIL (enables me to SPAWN out of EDT when it has been called from MAIL). M  ' Why was MAIL rewritten?  (Was it in C?)h   When will this be fixed?  G I mean, introducing the /OLD qualifier is rather embarrassing since it 8H admits there is a need to use it.  (OK, it's not documented, but still.)     Phillip Helbig     --M Kapteyn Instituut                    Email .............. helbig@astro.rug.nl M Rijksuniversiteit Groningen          Tel. ................... +31 50 363 4067-M Postbus 800                          Fax .................... +31 50 363 6100pM NL-9700 AV Groningen                 Web ... http://www.astro.rug.nl/~helbig/h  M ******************************** formerly at ********************************-  M this will still work for a while -----> Email ......... p.helbig@jb.man.ac.uk"M University of Manchester                Tel. ... +44 1477 571 321 (ext. 2635).M Jodrell Bank Observatory                Fax ................ +44 1477 571 618eM Macclesfield                            Telex ................ 36149 JODREL GtM UK-Cheshire SK11 9DL                    Web ... http://www.jb.man.ac.uk/~pjh/t  H My opinions are not necessarily those of either of the above institutes.  N <A HREF=" http://gladia.astro.rug.nl:8000/helbig/hire/hire.html ">HIRE ME!</A>   ------------------------------  % Date: Wed, 12 Apr 2000 13:20:26 +0100 - From: Tim Llewellyn <tim.llewellyn@bbc.co.uk> # Subject: Re: signature file in MAILp) Message-ID: <38F46A0A.94DDF35B@bbc.co.uk>t   Phillip Helbig wrote:'   >lH > I mean, introducing the /OLD qualifier is rather embarrassing since itJ > admits there is a need to use it.  (OK, it's not documented, but still.)  6 Ahem, backwards compatibility is one of VMS's forte's.   Regardsl   --6 Tim Llewellyn, OpenVMS Infrastructure, Remarcs Project0 MedAS at the BBC, Whiteladies Road, Bristol, UK.A Email tim.llewellyn@bbc.co.uk. Home tim.llewellyn@cableinet.co.ukD  A I speak for myself only and my views in no way represent those ofn MedAS or the BBC.f   ------------------------------   Date: 12 Apr 2000 13:28:26 GMT* From: helbig@astro.rug.nl (Phillip Helbig)# Subject: Re: signature file in MAILh. Message-ID: <8d1tlq$1pm$1@info.service.rug.nl>  7 In article <38F46A0A.94DDF35B@bbc.co.uk>, Tim Llewellyn " <tim.llewellyn@bbc.co.uk> writes:   J > > I mean, introducing the /OLD qualifier is rather embarrassing since itL > > admits there is a need to use it.  (OK, it's not documented, but still.) > 8 > Ahem, backwards compatibility is one of VMS's forte's.  F What's the smiley for tongue in cheek?  Normally, no extra qualifiers ' are needed for backwards compatibility.    ------------------------------  % Date: Wed, 12 Apr 2000 17:15:54 +0100mB From: Andrew Harrison SUNUK Consultancy <andrew.nospam@uk.sun.com>/ Subject: Re: Sun considers VMS a "mainframe" OSn* Message-ID: <38F4A13A.7875C307@uk.sun.com>   Rob Young wrote:  q > In article <38F32BEC.1E5F3963@uk.sun.com>, Andrew Harrison SUNUK Consultancy <andrew.nospam@uk.sun.com> writes:iH > > Most commercial UNIX's have batch support, batch control systems andI > > batch scheduler classes nowdays, they also have things like fairshare@ > > schedulers.a > >i > F >         Got to learn more about this fairshare scheduling stuff, andG >         gang scheduling.  At a recent site, performance was lousy and-Q >         they had to tweek scheduling (a large multi-processor non-Sun Unix box)nL >         and then things got better.  Wonder why something as sophisticatedM >         as that can be left to the end-user.  By the way, I have no details K >         so probably shouldn't be mentioning it... just thinking out loud.   A Fairshare schedulers do need tweeking though all you are doing isuF deciding what % of a systems resources a particular application, user,3 group of users get when the system is fully loaded.r  > For a description of a commercial Fairshare scheduler look at,   http://www.softway.com.au   C They have a product called Share II which is a Fairshare scheduler,w  @ Sun has a product called Solaris Resource Manager which is based6 on Share II, HP have a product of their own called PRM   Regardst Andrew Harrisoni Enterprise IT Architect    ------------------------------  % Date: Wed, 12 Apr 2000 13:32:17 +0100d- From: Andy Williams <Andy.Williams@liffe.com>  Subject: That's got it !M Message-ID: <78E5B8E274DBD1118D6800805FE60E7720594C@ntprdex4.admin.liffe.com>    Thanks Larry !  D Don't even ask me why I didn't think of doing the same on the remote system. senility I expect...   Cheers,i   -Andy    -----Original Message-----2 From: kilgallen@eisner.decus.org (Larry Kilgallen)# [mailto:kilgallen@eisner.decus.org]a- Posted At: Wednesday, April 12, 2000 12:29 PM- Posted To: vms# Conversation: Denying DECnet access:" Subject: Re: Denying DECnet access    F In article <8d1hc8$261$1@nnrp1.deja.com>, Andy <andyw9804@my-deja.com> writes:@  H > After a long search I unearthed some old (V6.0) Phase IV documentationH > but can't really find anything relating to this except for the NCP SETG > NODE xxx ACCESS NONE command. This works as documented but the access 6 > restriction is overridden if the user has OPER priv.  D I did not recall the OPER nuance, but certainly that cannot have anyG effect on the target node, since that node has no information regardingr, privilege of the originating process.  Thus:   	If node A has:r    		MCR NCP SET NODE B ACCESS NONE   	and node B has:    		MCR NCP SET NODE A ACCESS NONE  * the isolation you seek should be provided.  A The lack of a Phase V equivalent, seems to be the final matter in-A which DECnet-Plus still has not caught up to regular DECnet.  Them? US DECUS Security SIG raised this concern to DEC before Phase V 
 was released.    ------------------------------  # Date: Wed, 12 Apr 2000 14:28:22 GMTp From: antonack@my-deja.com Subject: Vax Equipment for Saler) Message-ID: <8d215l$jck$1@nnrp1.deja.com>e  G I know that this may be off-topic, and if so I apologize, but if anyoneoC is in need of a TA-91 tape unit, or two HSC70's or requestor cards,/3 please send me an email at mark@chathamnetworks.com-   Thanks.e    & Sent via Deja.com http://www.deja.com/ Before you buy.n   ------------------------------  # Date: Wed, 12 Apr 2000 12:40:07 GMTb2 From: kilgallen@eisner.decus.org (Larry Kilgallen)( Subject: Web Standards (was: netscape 6)' Message-ID: <2000Apr12.084007.1@eisner>6  p In article <009E87D6.DA86A6A7@SendSpamHere.ORG>, system@SendSpamHere.ORG (Brian Schenkenberger, VAXman-) writes:  N > The problem with monopoly$chlock bundling IE with their excuse for an O/S isN > that it makes the use of non-standard web features/functions pervasive.  GetN > IE running and you can get all of the proprietary web served features comingO > from monopoly$chlock's IIS.  If the source for IE is made 'open' and 'royaltyeO > free', IE will become more and more ingrained as the 'standard' making monop-o; > oly$schlock's attempts to monopolize the web even easier!e  D I have taken the hint and will not bother asking Brian to contribute( to the Bill Gates Legal Defense Fund :-)  N > If a site can't adhere to the standards and/or baises itself to non-standard: > based browsers and web features, they'll not get my biz.  B The HTML validator at W3.ORG offers a logo you can post indicatingG compliance with HTML V4.0.  If only that were expected and respected...1   ------------------------------  # Date: Wed, 12 Apr 2000 16:10:11 GMTc= From: system@SendSpamHere.ORG (Brian Schenkenberger, VAXman-)n, Subject: Re: Web Standards (was: netscape 6)0 Message-ID: <009E8800.84294078@SendSpamHere.ORG>  \ In article <2000Apr12.084007.1@eisner>, kilgallen@eisner.decus.org (Larry Kilgallen) writes:q >In article <009E87D6.DA86A6A7@SendSpamHere.ORG>, system@SendSpamHere.ORG (Brian Schenkenberger, VAXman-) writes:e >oO >> The problem with monopoly$chlock bundling IE with their excuse for an O/S is O >> that it makes the use of non-standard web features/functions pervasive.  Get-O >> IE running and you can get all of the proprietary web served features comingnP >> from monopoly$chlock's IIS.  If the source for IE is made 'open' and 'royaltyP >> free', IE will become more and more ingrained as the 'standard' making monop-< >> oly$schlock's attempts to monopolize the web even easier! >aE >I have taken the hint and will not bother asking Brian to contribute ) >to the Bill Gates Legal Defense Fund :-)T  J The man has enough money to bribe every justice all the way to the supremeI court so why would he need my meager contribution.  Besides, I've alreadyML pledged to give as much to Bill Gates' Legal Defense Fund as he contributed  to mine.   --N VAXman- OpenVMS APE certification number: AAA-0001           VAXman@TMESIS.COM  L GNU Freeware -- What does the GNU *really* stand for?  Garbage!  Not Usable!   ------------------------------   Date: 12 Apr 2000 07:21:00 GMT3 From: gartmann@immunbio.mpg.de (Christoph Gartmann)aC Subject: Re: Where to find DECserver Network Access Software (DNAS).0 Message-ID: <8d184s$41s$2@n.ruf.uni-freiburg.de>  { In article <022dacf6.eb7e1e3d@usw-ex0110-075.remarq.com>, Mike Price <mike.priceNOmiSPAM@littlewoods.co.uk.invalid> writes:n9 >Our CDs came with the decserver 700s we order - we get aa7 >'free' CD with each one and the software is pre-loaded-  G Are you sure it is the "DECserver Network Access Software" and not justl the "DECserver 700 Software"?E   Regards,    Christoph Gartmannm  H -----------------------------------------------------------------------+H | Max-Planck-Institut fuer      Phone   : +49-761-5108-464   Fax: -452 |H | Immunbiologie                                                        |H | Postfach 1169                 Internet: gartmann@immunbio.mpg.de     |H | D-79011  Freiburg, FRG                                               |H +------------ http://www.immunbio.mpg.de/english/menue.html -----------+   ------------------------------   Date: 12 Apr 2000 07:18:25 GMT3 From: gartmann@immunbio.mpg.de (Christoph Gartmann)(C Subject: Re: Where to find DECserver Network Access Software (DNAS)00 Message-ID: <8d1801$41s$1@n.ruf.uni-freiburg.de>  [ In article <38F2F4D8.61BADEEA@uiowa.edu>, "Richard L. Dyson" <rick-dyson@uiowa.edu> writes:oF >Does anyone know how or where I can find the DECserver Network AccessG >Software (DNAS, #QB-0LWAA-WA) v2.3?  Where would I look for it?  Is itjH >on a normal CD distribution or is it a special product I have to order?E >If so, any idea how much it would cost?  Could a regular VAR get it?I  I It is a separate product. And I think it went over to Cabetron. I remeberhG having a hard time getting it through Digital but I got it (just befores- the merger). I don't remeber the cost though.k   Regards,    Christoph Gartmannr  H -----------------------------------------------------------------------+H | Max-Planck-Institut fuer      Phone   : +49-761-5108-464   Fax: -452 |H | Immunbiologie                                                        |H | Postfach 1169                 Internet: gartmann@immunbio.mpg.de     |H | D-79011  Freiburg, FRG                                               |H +------------ http://www.immunbio.mpg.de/english/menue.html -----------+   ------------------------------  % Date: Wed, 12 Apr 2000 19:11:00 +0200o  From: "Peter" <oudakker@gmx.net>C Subject: Re: Where to find DECserver Network Access Software (DNAS)2< Message-ID: <955559435.24514.0.pluto.d4ee6809@news.demon.nl>  ; > >Our CDs came with the decserver 700s we order - we get a-9 > >'free' CD with each one and the software is pre-loadedo >eI > Are you sure it is the "DECserver Network Access Software" and not justa > the "DECserver 700 Software"?@  K I think he means the wweng2.sys (bootfile for the DECServer 700) what needs(% to but put in the MOM$LOAD directory.<  H In the past you needed a separate license for NAS but later (I'll beliveB from 6.2 and up) it was standard available within the VMS-license.   ------------------------------  % Date: Wed, 12 Apr 2000 08:45:15 +0100t8 From: "benchabane@wissal.dz" <benchabane@mail.wissal.dz>? Subject: Re: Why are there no French people in this newsgroup ? 2 Message-ID: <200004120745.IAA28844@mail.wissal.dz>  G Il n'y a pas que des Fran=E7ais, il y a des francophones aussi ........h   L.Benchabane.....Algiers    G >Mais si, il y a des Francais sur comp.os.vms, mais je rentre juste de=h
  vacances,H >et j'ai m=EAme 2 stations sous VMS =E0 la maison et quelques pages web=
  consacr=E9eso >=E0 ce sujet, par exemple:  >o% >  http://www.multimania.com/pmoreau/a >T >I don't like Mr Bill. > @ >VMS is not dead in France, just near dead ... comme partout :-) >  >Patrick >r   ------------------------------  # Date: Wed, 12 Apr 2000 11:19:37 GMT0) From: "Neil Rieck" <n.rieck@sympatico.ca>:? Subject: Re: Why are there no French people in this newsgroup ?e; Message-ID: <dZYI4.74807$1C2.2213157@news20.bellglobal.com>@  4 David Turner <d_b_turner@yahoo.com> wrote in message) news:seqcfl5mrfg119@corp.supernews.com..., > This is a weird deal.... > K > Netherlands, Germany, US UK, Ozzy's;  pretty much everyone, but no Frenchc >a > Strange huh ?i >f# > Do they still use VMS en France ?p% > Or PC's only  - Vive Monsieur Bill! J > Actually - if one tries to pronouce Vms with a french accent, it doesn't) > sound 'arf as good as Solaris or Unix -e >eL Sysops in Quebec in love Solaris and it's probably because of the good multiK language support. On the flip side, I'm seeing an increase in the number ofbC web sites running or supporting OpenVMS in France and Germany. ManyqJ Europeans disillusioned by the realities of Linux are moving (slowly) backL to English oriented OpenVMS. (If you learned English in school, you've got aF better chance of running an OpenVMS system where the commands and help system are meaningful)   Some OpenVMS support in Europe: K http://www2.cenaath.cena.dgac.fr/ftp/VMS/ (Centre d'Etudes de la Navigationr Aerienne, France).9 http://www.uniud.it/ftp/vms/ (University of Udine, Italy)sG http://www.decus.de:8080/www/vms/sw.html (DECUS in Germany; Darmstadt ,a Germany)H http://vzdmzj.zdv.uni-mainz.de/www/IPS/HTML_PERL/ (Universitt of Mainz, Germany)J Note: http://axp616.gsi.de:8080/www/vms/sw.html (GSI in Darmstadt, Germany has moved to DECUS.de)   Some alternate news groups:e news:de.comp.os.vms (Germany)a news:fr.comp.os.vms (France)  
 Neil Rieck* Kitchener(New Berlin?)/Waterloo/Cambridge, Ontario, Canada.! http://www3.sympatico.ca/n.rieck/o6 http://www3.sympatico.ca/n.rieck/docs/alpha_diary.html   ------------------------------  # Date: Wed, 12 Apr 2000 12:46:19 GMTh2 From: kilgallen@eisner.decus.org (Larry Kilgallen)? Subject: Re: Why are there no French people in this newsgroup ?e' Message-ID: <2000Apr12.084619.1@eisner>i  g In article <dZYI4.74807$1C2.2213157@news20.bellglobal.com>, "Neil Rieck" <n.rieck@sympatico.ca> writes:i  N > Sysops in Quebec in love Solaris and it's probably because of the good multiM > language support. On the flip side, I'm seeing an increase in the number of E > web sites running or supporting OpenVMS in France and Germany. ManyaL > Europeans disillusioned by the realities of Linux are moving (slowly) backN > to English oriented OpenVMS. (If you learned English in school, you've got aH > better chance of running an OpenVMS system where the commands and help > system are meaningful)  : VMS has great capabilities for working in other languages,; but DEC never put the effort in (like Apple does) to timelyT; release of language variants that have all the commands and ( help text translated to other languages.  9 An interesting parallel discussion in comp.lang.ada about 9 the fact that Ada keyword are in English found many whoset; native language is _not_ English saying they even preferredD9 the keywords for Ada being in English since it made quiteD; an obvious distinction that these were keywords rather than-= being just some natural language word in their native tongue.g   ------------------------------  % Date: Wed, 12 Apr 2000 17:02:48 +02001> From: "Jean-Franois Marchal" <jean-francois.marchal@x9000.fr>? Subject: Re: Why are there no French people in this newsgroup ? + Message-ID: <8d234m$i3r$1@minus.oleane.net>   ? "Larry Kilgallen" <kilgallen@eisner.decus.org> wrote in messageo! news:2000Apr12.084619.1@eisner...oJ > In article <dZYI4.74807$1C2.2213157@news20.bellglobal.com>, "Neil Rieck" <n.rieck@sympatico.ca> writes:  < > VMS has great capabilities for working in other languages,= > but DEC never put the effort in (like Apple does) to timely = > release of language variants that have all the commands and * > help text translated to other languages.   Who remembers from LOLA ?a  1 "La petite LOLA me turlupine" (french spoonerism)l   Jean-Franois Marchal  X9000 - LYON   ------------------------------   Date: 12 Apr 2000 06:40:01 GMT+ From: Tom Leitner <tom@radar.tu-graz.ac.at> . Subject: Re: Wierd Elsa Gloria Synergy Problem1 Message-ID: <8d15o1$fr0$1@fstgss02.tu-graz.ac.at>*  4 Fred Kleinsorge <kleinsorge@star.zko.dec.com> wrote:  G > Tom Leitner wrote in message <8ct8n0$kfn$1@fstgss02.tu-graz.ac.at>...t  J >>It doesn't buy much in terms of video memory. But it buys a lot in termsG >>of pixmap memory required in the X server. With 16 bit depth, pixmaps H >>(and CDE, for example needs lots of them) are only half as big as with9 >>24 bit depth (2 bytes per pixel vs. 4 bytes per pixel).c >>  L >  Yeah, but PIXMAPs are done as memory pixmaps, so it's just VM.  You would1 > need a lot of PIXMAPs before you run out of VM.b  H Sure. But shuffling around pixmaps with double size takes a.) longer and1 b.) can get very very slow if paging is involved.y  G Our application (wiis.tu-graz.ac.at) needs to do full screen animations-J of weather images. Guess what happens on a 128MB RAM machine when the userE tries to animate some 40 images or so. 40 images are 1280x1024x4x40 =iE 209715200 bytes which are 200 Megs. Even when there is enough VM, the L animation almost comes to a grinding halt when pixmaps need to be constantlyJ paged in. In contrast with 16bpp, I have only 100 Megs which is manageable with 128MB RAM.o  E I just don't see the point of using 24bpp. 16bpp truecolor has enoughDD colors by far and consumes much less RAM, as I've pointed out above.   Tom  -- eN Dr. Tom Leitner             Dept. of Communications, Graz Univ. of Technology,K tom@radar.tu-graz.ac.at     Inffeldgasse 12, A-8010 Graz / Austria / Europe : PGP public key @ http://wiis.tu-graz.ac.at/people/tom.html   ------------------------------  % Date: Wed, 12 Apr 2000 12:08:32 +0200r% From: "GM" <gmant@supereva.it.NOSPAM>O Subject: Yamaha CRW8424 on VMS9 Message-ID: <002901bfa467$0f951b00$1be8ccc1@esrin.esa.it>t   Hello,  L does somebody successfully used this CD-writer on VMS ? I know that previous models CRW2260/4260 work, . but these old models aren't available anymore.  
 Thank you,   Giorgios   ------------------------------  # Date: Wed, 12 Apr 2000 10:56:01 GMT0= From: system@SendSpamHere.ORG (Brian Schenkenberger, VAXman-)d" Subject: Re: Yamaha CRW8424 on VMS0 Message-ID: <009E87D4.A027C393@SendSpamHere.ORG>  a In article <002901bfa467$0f951b00$1be8ccc1@esrin.esa.it>, "GM" <gmant@supereva.it.NOSPAM> writes:d >Hello,s >SM >does somebody successfully used this CD-writer on VMS ? I know that previouss >models CRW2260/4260 work,/ >but these old models aren't available anymore.o >y >Thank you,h >  >Giorgio >a >h  I What are you using for writer software and to which VMS platform are you s connecting this device?d   --N VAXman- OpenVMS APE certification number: AAA-0001           VAXman@TMESIS.COM  L GNU Freeware -- What does the GNU *really* stand for?  Garbage!  Not Usable!   ------------------------------  % Date: Wed, 12 Apr 2000 15:21:37 +0200 % From: "GM" <gmant@supereva.it.NOSPAM>e" Subject: Re: Yamaha CRW8424 on VMS9 Message-ID: <000d01bfa482$08f52310$1be8ccc1@esrin.esa.it>i  J > What are you using for writer software and to which VMS platform are you > connecting this device?f  I I think to use a dedicated AXP workstation (DEC 3000/300, 255 or DPW 433)  running 6.1 to 7.2-1.d; I heard about a SW called CDRECORD (also available for VAX, # http://www.tmesis.com/cdrom/), then ? I didn't check yet the SWs provided on the OVMS freeware cd v4.d   Giorgiou   ------------------------------   End of INFO-VAX 2000.205 ************************