0 INFO-VAX	Sat, 16 Feb 2002	Volume 2002 : Issue 91      Contents: Re: "Crashless Windows"  Re: "Crashless Windows" 
 Backup query.  Re: Backup query.  Re: Backup query.  RE: Backup query.  Re: Backup query.  RE: Backup query.  Re: Backup query.  Re: Backup query.  Re: Backup query.  Re: Backup query. 8 Re: Batch job termination - OpenVMS v7.3 v. OpenVMS v7.18 Re: Batch job termination - OpenVMS v7.3 v. OpenVMS v7.1 Re: Carly interview on CNNC Re: compaq logo on DECwindows login screen...how to get the old one C Re: compaq logo on DECwindows login screen...how to get the old one  Creating a mailbox...  Re: Creating a mailbox...  Re: DCL procedure's  Re: DCL procedure's - Re: Did Intel Bite Off More Than It Can Chew?  Re: Gordon Bell site Re: Gordon Bell site Re: Gordon Bell site Re: Gordon Bell site Re: Gordon Bell site Re: Gordon Bell site Re: Gordon Bell site Re: Gordon Bell site RE: Gordon Bell site Re: Gordon Bell site Re: Gordon Bell site Re: Gordon Bell site Re: Gordon Bell site? Re: Hello Compaq!  What happens to us VMS users if Itanic dies? ? Re: Hello Compaq!  What happens to us VMS users if Itanic dies? C Re: Re: Hello Compaq!  What happens to us VMS users if Itanic dies? ? Re: Hello Compaq!  What happens to us VMS users if Itanic dies? ? Re: Hello Compaq!  What happens to us VMS users if Itanic dies? ? Re: Hello Compaq!  What happens to us VMS users if Itanic dies? C Re: Re: Hello Compaq!  What happens to us VMS users if Itanic dies? ? Re: Hello Compaq!  What happens to us VMS users if Itanic dies? ? Re: Hello Compaq!  What happens to us VMS users if Itanic dies? ? Re: Hello Compaq!  What happens to us VMS users if Itanic dies? ? Re: Hello Compaq!  What happens to us VMS users if Itanic dies? ? Re: Hello Compaq!  What happens to us VMS users if Itanic dies? ? Re: Hello Compaq!  What happens to us VMS users if Itanic dies? ? Re: Hello Compaq!  What happens to us VMS users if Itanic dies? ? Re: Hello Compaq!  What happens to us VMS users if Itanic dies? ? Re: Hello Compaq!  What happens to us VMS users if Itanic dies? ? Re: Hello Compaq!  What happens to us VMS users if Itanic dies? + Re: How to become a Hobbyist (the hard way) . Re: How to get prices on CPQ OpenVMS products?. Re: In stock: ES40 Model 2 USD7895 VMS Ready ! Re: Itanium troubles6 Re: JDK 1.3.1 past due (feature not found in the beta)/ Re: Load Balancing DNA across 2*2MBit WAN links 8 Re: Longest VMS _cluster_ uptime, was: Re: 1 Year uptime OpenVMS Times -- RIP???  Re: OpenVMS Times -- RIP???  Re: OpenVMS Times -- RIP???  Re: OpenVMS Times -- RIP???  Oracle problem.  Re: OT: "Crashless Windows" & Re: OT: Proliant DL and DS-10 cabinets& Re: OT: Proliant DL and DS-10 cabinets7 re: Reading OPCOM messages (was: Creating a mailbox...) 5 Re: Security for VMS systems visable to the  Internet 5 Re: Security for VMS systems visable to the  Internet 5 Re: Security for VMS systems visable to the  Internet  Re: Software Question $ Re: Ultimate Workstation vs. AS 1200 Very Very Slow Telnet. Re: Very Very Slow Telnet. Re: Very Very Slow Telnet. Re: Very Very Slow Telnet.- Re: What is the best NFS package for OpenVMS?  Re: XP 1000 won't boot  F ----------------------------------------------------------------------  # Date: Sat, 16 Feb 2002 04:05:52 GMT 1 From: "David J. Dachtera" <djesys.nospam@fsi.net>   Subject: Re: "Crashless Windows"' Message-ID: <3C6DDC29.52674F54@fsi.net>    JF Mezei wrote:  >  > "David J. Dachtera" wrote:3 > > How many wrongs *DOES* it take to make a right?  > N > That doesn't matter. When enough wrongs are made in the same direction, thatM > direction becomes industry standard and the harm done by that wrong becomes  > accepted.   G Sort of like society. Violence was abhorrent to us, until it became the F norm in first in "the arts", later in everyday life. Now it's accepted4 as a fact of everyday life. Doesn't make it "right".  D > Windows can be compared to cars that generate exhaust. People haveL > gotten used to accepting the pollution, noise and dammage/deaths caused byF > cars and don't demand cars that don't pollute and don't kill people.  E Well, until the laws of physics get repealed, we'll have to live with H the difference between the momentum of a moving vehicle and the level of$ impact the human body can withstand.  B As to pollution, e-mail me privately (anyone!) and I'll share someE long-lost research (*NOT* mine! let me make that unmistakably clear!) B with you. It's possible today, and it's been possible for going onF twenty(20) years - without tons of batteries and without combustion ofG any kind. One of the necessary elements is radioactive (deuterium), the  other (gallium) is not.   L > > PDP groups about 60's and 70's technology running merrily away while theG > > whiz-bang, latest-and-greatest crowd are rebooting every few hours, - > > fending off viruses and hackers, etc. ...  > P > But in the end, the windows crap still produces more than the 1970s technology* > even with all its blue screen of deaths.  F Dunno 'bout that. Can a Wintel box outperform a S/360 of that vintage?& Well, yeah - suppose the new ones can.  ' > You couldn't run NASDAQ's web site on  > punched cards.   Agreed.    > [snip]N > But HP insists on continuing in that path and there is nothing you and I canM > do to prevent it. And with Compaq out of the way, HP has a better chance of H > survival. Remember that this whole merger deal isn't about combining 22 > companies, it is about eliminating a competitor.  B Hhmmm... Personally, outside of the desktop, I never saw that muchC overlap between the Q and HP. What in the HP line corresponds to or @ competes with the rack-mount Proliant servers we find in so many datacenters in America?    --   David J. Dachtera  dba DJE Systems  http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/    ------------------------------  # Date: Sat, 16 Feb 2002 04:06:32 GMT 1 From: "David J. Dachtera" <djesys.nospam@fsi.net>   Subject: Re: "Crashless Windows"' Message-ID: <3C6DDC59.A7161B66@fsi.net>    Larry Kilgallen wrote: > ] > In article <3C6C8934.A4860D01@fsi.net>, "David J. Dachtera" <djesys.nospam@fsi.net> writes:  > > David Beatty wrote:  > >> [snip]  > >> if you B > >> have a mission critical app. that's running on Windows [snip] > > $ > > ...your stockholders should sue. > C > But in the meantime, you should make sure you are getting paid by B > the hour, with a premium for overtime and a minimum for on-call.   AMEN!    --   David J. Dachtera  dba DJE Systems  http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/    ------------------------------  % Date: Fri, 15 Feb 2002 20:20:36 -0000 ; From: "Leigh G. Bowden" <LGBowden@bowdenfamily.fsnet.co.uk>  Subject: Backup query./ Message-ID: <a4jqbt$6l8$2@newsg4.svr.pol.co.uk>   K I've been asked to improve the backups on a AS2100 with 2 CPU's and 1 GB of J RAM it has 16 x 2.1 GB disks but shadowed as 8 (1 system disk and 7 data).K This is running VMS6.2 with Oracle 7.0 of some ilk. VMS is fully patched as K of about three months ago. The backups are done to a DLT III 10/20GB device K which is quite capable of taking the entire system on a tape with plenty to I spare. The system is nominally 24 x 7 but it is fairly quiet at night the % backup runs at 3am Monday - Saturday.   E The current backup is rather erratic and I've changed it to do /IMAGE J backups of all the DSA devices but I now want to improve it. This is where I'm less sure.  L 1). At it's easiest breaking the shadow sets and mounting them privately andK doing backups of those before introducing them back into the shadow. As the K system is quiet at night nobody would notice the loss of performance as the 9 copies are done. Compaq don't seem to like this approach.   F 2). Compaq seem to prefer dismounting the shadow volume then remount aI single disk as a shadow and backup from the other one then introduce that G one back into the shadow. I doubt this would work as Oracle will almost   certainly have some files opens.  J 3). Shutting Oracle down briefly to do 1 or 2 is unlikely as the system isL needed 24 x 7 but I might be able to persuade the management otherwise if it was for about ten minutes only.   < Is option 1 really an option despite Compaq's protestations?  J In all instances do the disks have be forced to dismount as files could be open on the disks?   ------------------------------  % Date: Fri, 15 Feb 2002 21:43:40 +0100  From: Dirk Munk <munk@home.nl> Subject: Re: Backup query.& Message-ID: <3C6D72FC.3010809@home.nl>  C Your problem is not the backup, but the restore (if that should be  G necessary). After a restore you would like to have a working database.  F Closing everything down, and dismounting the disks will make sure you G have valid files etc. that can be restored. So "the Compaq way" is the  F most certain way to do this. Just breaking a shadow set will give you H the opportunity to make backups, but you should ask yourself if you are E sure you can restore a running system with those backups. I doubt it  F very much, because you can never can be sure that all data is written 5 back to disk at that moment. I would go for option 3.    Leigh G. Bowden wrote:  L >I've been asked to improve the backups on a AS2100 with 2 CPU's and 1 GB ofK >RAM it has 16 x 2.1 GB disks but shadowed as 8 (1 system disk and 7 data). L >This is running VMS6.2 with Oracle 7.0 of some ilk. VMS is fully patched asL >of about three months ago. The backups are done to a DLT III 10/20GB deviceL >which is quite capable of taking the entire system on a tape with plenty toJ >spare. The system is nominally 24 x 7 but it is fairly quiet at night the& >backup runs at 3am Monday - Saturday. > F >The current backup is rather erratic and I've changed it to do /IMAGEK >backups of all the DSA devices but I now want to improve it. This is where  >I'm less sure.  > M >1). At it's easiest breaking the shadow sets and mounting them privately and L >doing backups of those before introducing them back into the shadow. As theL >system is quiet at night nobody would notice the loss of performance as the: >copies are done. Compaq don't seem to like this approach. > G >2). Compaq seem to prefer dismounting the shadow volume then remount a J >single disk as a shadow and backup from the other one then introduce thatH >one back into the shadow. I doubt this would work as Oracle will almost! >certainly have some files opens.  > K >3). Shutting Oracle down briefly to do 1 or 2 is unlikely as the system is M >needed 24 x 7 but I might be able to persuade the management otherwise if it   >was for about ten minutes only. > = >Is option 1 really an option despite Compaq's protestations?  > K >In all instances do the disks have be forced to dismount as files could be  >open on the disks?  >  >    ------------------------------  % Date: Fri, 15 Feb 2002 15:52:33 -0500 - From: "John Eisenschmidt" <jeisensc@aaas.org>  Subject: Re: Backup query.+ Message-ID: <sc6d2ec5.080@AAASMTA.aaas.org>   & My personal preference with databases:  H The datafiles as they exist on disk are worthless to me in a different =L environment (i.e. after a restore). Instead of trying to take cold backups =I (which I've had bad luck with) that are little good after a restore, or = J trying to take hot backups while the system is online, I prefer snapshot =F backups. I have Oracle run an export at some time and then take that =I (along with everything else) off to take. If the system dies, I restore = L everything, recreate the database from its create scripts, and restore the =I export. I've found it to be a much faster and more reliable method that = E twitching data files and dictionaries to work with what they have.=20   E I use a similar strategy with MS SQL Server. While I've never had a = K catostrophic failure of Oracle on VMS, I have had several with SQL/NT - I = + can recover from them in less than an hour.   H To that end, you have a snapshot with all your structures that is 100% =H consistant at the point in time it was taken. Getting that back up and = running is trivial.=20  
 Good Luck, John  6 >>> Dirk Munk <munk@home.nl> 02/15/2002 3:43:40 PM >>>E Your problem is not the backup, but the restore (if that should be=20 I necessary). After a restore you would like to have a working database.=20 H Closing everything down, and dismounting the disks will make sure you=20I have valid files etc. that can be restored. So "the Compaq way" is the=20 H most certain way to do this. Just breaking a shadow set will give you=20J the opportunity to make backups, but you should ask yourself if you are=20G sure you can restore a running system with those backups. I doubt it=20 H very much, because you can never can be sure that all data is written=205 back to disk at that moment. I would go for option 3.    Leigh G. Bowden wrote:  K >I've been asked to improve the backups on a AS2100 with 2 CPU's and 1 GB =  ofF >RAM it has 16 x 2.1 GB disks but shadowed as 8 (1 system disk and 7 = data).K >This is running VMS6.2 with Oracle 7.0 of some ilk. VMS is fully patched =  asG >of about three months ago. The backups are done to a DLT III 10/20GB =  deviceK >which is quite capable of taking the entire system on a tape with plenty =  toJ >spare. The system is nominally 24 x 7 but it is fairly quiet at night the& >backup runs at 3am Monday - Saturday. > F >The current backup is rather erratic and I've changed it to do /IMAGEG >backups of all the DSA devices but I now want to improve it. This is =  where  >I'm less sure.  > K >1). At it's easiest breaking the shadow sets and mounting them privately =  and J >doing backups of those before introducing them back into the shadow. As = the J >system is quiet at night nobody would notice the loss of performance as = the : >copies are done. Compaq don't seem to like this approach. > G >2). Compaq seem to prefer dismounting the shadow volume then remount a J >single disk as a shadow and backup from the other one then introduce thatH >one back into the shadow. I doubt this would work as Oracle will almost! >certainly have some files opens.  > J >3). Shutting Oracle down briefly to do 1 or 2 is unlikely as the system = isL >needed 24 x 7 but I might be able to persuade the management otherwise if = it  >was for about ten minutes only. > = >Is option 1 really an option despite Compaq's protestations?  > J >In all instances do the disks have be forced to dismount as files could = be >open on the disks?  >  >    ------------------------------  % Date: Fri, 15 Feb 2002 15:57:33 -0500 > From: "Koska, John C. (LNG-MBC)" <John.C.Koska@lexisnexis.com> Subject: RE: Backup query.M Message-ID: <3D35AD137AAAD411A6BA0008C7B1B12D016027B0@MBCALBEXC03.BENDER.COM>    > -----Original Message-----B > From: Leigh G. Bowden [mailto:LGBowden@bowdenfamily.fsnet.co.uk]) > Sent: Friday, February 15, 2002 3:21 PM  > To: Info-VAX@Mvb.Saic.Com  > Subject: Backup query. > < > I've been asked to improve the backups on a AS2100 with 2  > CPU's and 1 GB of @ > RAM it has 16 x 2.1 GB disks but shadowed as 8 (1 system disk  > and 7 data).= > This is running VMS6.2 with Oracle 7.0 of some ilk. VMS is   > fully patched as? > of about three months ago. The backups are done to a DLT III   > 10/20GB device? > which is quite capable of taking the entire system on a tape   > with plenty to? > spare. The system is nominally 24 x 7 but it is fairly quiet   > at night the' > backup runs at 3am Monday - Saturday.  > G > The current backup is rather erratic and I've changed it to do /IMAGE ? > backups of all the DSA devices but I now want to improve it.   > This is where  > I'm less sure. > < > 1). At it's easiest breaking the shadow sets and mounting  > them privately and? > doing backups of those before introducing them back into the   > shadow. As the; > system is quiet at night nobody would notice the loss of   > performance as the; > copies are done. Compaq don't seem to like this approach.  > H > 2). Compaq seem to prefer dismounting the shadow volume then remount a= > single disk as a shadow and backup from the other one then   > introduce that> > one back into the shadow. I doubt this would work as Oracle 
 > will almost " > certainly have some files opens. > ? > 3). Shutting Oracle down briefly to do 1 or 2 is unlikely as a > the system is ? > needed 24 x 7 but I might be able to persuade the management p > otherwise if itp! > was for about ten minutes only.a > > > Is option 1 really an option despite Compaq's protestations? > > > In all instances do the disks have be forced to dismount as  > files could be > open on the disks?  A I could be wrong, but a backup of an open file is a "no-no".  Youw> might get away with it some of the time, but not all the time.  @ And as to a forced dismount, would that not leave the file in an@ uncertain state as compared to making sure the files are closed  first.    A As to Oracle, I would think you need to have the DBA do an exporteD or whatever it/they is/are called, and have them be responsible for < the export and journal files to recover the Oracle database.  B For my shop, which uses RMS, we make sure we close our applicationD fully (no users, no open files, deinstall images from any data disk,C make sure print spool files are on a non-data disk, no other batch  > jobs, nothing running that opens a file, etc, etc.) before we @ dismount our shadowsets with write bitmaps.  Once we have them, C dismounted cleanly with write bitmaps, then we run the application nF and whatever else run on one side of shadow, with backup on the other.E With write bit maps, the reshadowing times are significantly reduced  E with OpenVMS 7.3. It should only take a few (as in 5 or less I shouldsF think with the amount of storage you are talking) minutes to dismount H and remount.  We take the opportunity during the exclusive period where 0 nothing is open to analyze and repair our disks.  I Whatever you do, make sure you test recovery of your backups (preferably aE on a different system, in case the backups are not good), before you iB bless your backups as good.  I would suggest periodic test also.    . You should consult volume shadowing manual at:  - http://www.openvms.compaq.com:8000/index.htmln   :) jck
 John Koska Matthew Bender & Co., Inc. -"   A Member of the LexisNexis Group
 1275 Broadwaye Albany, NY  12204o USAt 518-487-3255 John.C.Koska@lexisnexis.comC  ) I post personal opinion only, and all thel* disclaimers one could imagine apply.  That( includes, I speak for myself only and my) views in no way represent my employer(s).s+ One should also take note of the Electronicr) Communications Privacy Act of 1986, whichi+ imposes civil and criminal liability on any ( person who intentionally intercepts "any( wire, oral or electronic communication."   ------------------------------  % Date: Fri, 15 Feb 2002 22:22:36 +0100l, From: "Bart Zorn" <B.Zorn@TrueBit.nospam.nl> Subject: Re: Backup query.* Message-ID: <a4ju6t$c6t$1@news1.xs4all.nl>  F "Leigh G. Bowden" <LGBowden@bowdenfamily.fsnet.co.uk> wrote in message) news:a4jqbt$6l8$2@newsg4.svr.pol.co.uk...tJ > I've been asked to improve the backups on a AS2100 with 2 CPU's and 1 GB ofL > RAM it has 16 x 2.1 GB disks but shadowed as 8 (1 system disk and 7 data).J > This is running VMS6.2 with Oracle 7.0 of some ilk. VMS is fully patched asF > of about three months ago. The backups are done to a DLT III 10/20GB deviceJ > which is quite capable of taking the entire system on a tape with plenty toK > spare. The system is nominally 24 x 7 but it is fairly quiet at night the6' > backup runs at 3am Monday - Saturday.e > G > The current backup is rather erratic and I've changed it to do /IMAGEaL > backups of all the DSA devices but I now want to improve it. This is where > I'm less sure. > J > 1). At it's easiest breaking the shadow sets and mounting them privately andcI > doing backups of those before introducing them back into the shadow. Asc thehI > system is quiet at night nobody would notice the loss of performance as  thef; > copies are done. Compaq don't seem to like this approach.a  5 I am not sure why Compaq wouldn't like this approach.   H > 2). Compaq seem to prefer dismounting the shadow volume then remount aK > single disk as a shadow and backup from the other one then introduce thateI > one back into the shadow. I doubt this would work as Oracle will almosti" > certainly have some files opens.  J This seems rather silly to me. As long as Oracle has files open, you can't dismount the shadow sets.e  L > 3). Shutting Oracle down briefly to do 1 or 2 is unlikely as the system isK > needed 24 x 7 but I might be able to persuade the management otherwise ifl it! > was for about ten minutes only.o >t> > Is option 1 really an option despite Compaq's protestations? >cL > In all instances do the disks have be forced to dismount as files could be > open on the disks?   I believe the best option is:e   - shut down Oracle1 - dismount one member of the (each) shadow set(s)i - restart Oracle= - mount the dismounted member private /NOWRITE and back it upn. - dismount it and remount it to the shadow set  J The down time for Oracle will be minimal and you will have a clean backup.K There may be possibilities within Oracle to flush al committed transactions-K to disk, at which point you can dismount the member disk. I know Oracle Rdbc1 has this, but I don't know about Oracle "Oracle".s  L Note that if you use OpenVMS V7.3 (or V7.2 with the Shadowing enhancements),G you can make use of the minicopy feature, wich will greatly improve thelK speed of the shadow copy after you remount the members back into the shadowi sets.o   HTHi  	 Bart Zorne   ------------------------------  % Date: Fri, 15 Feb 2002 15:25:42 -0600h+ From: Don Rogstad <Don.Rogstad@dalsemi.com>P Subject: RE: Backup query.H Message-ID: <1809DA15308DD51180EE00508BCF2194C96C39@misnts1.dalsemi.com>  J This message is in MIME format. Since your mail reader does not understand< this format, some or all of this message may not be legible.  ' ------_=_NextPart_001_01C1B667.528020E0- Content-Type: text/plain;1 	charset="iso-8859-1"   J As others have stated, the problem is what disk activity is occurring whenJ you break the shadow sets.   As Compaq has stated, the only "official 100%L certain" way to ensure that no disk activity is occurring is to dismount the: disk, break the shadow sets, and then remount the disks.    I    However, in a 24x7 shop, this is not always feasible or available.   I H have had success with the following when working with an Oracle database backups from shadow set disks:  J 1) Put Oracle into the "Hot Backup" mode.  (This stops Oracle from writingK directly to the data base disks and redirects the writes to the "redo logs"d( (or "archive logs", I currently forget).  > 2) Break the shadow sets containing the Oracle databases files  ' 3) Clear Oracle from "Hot Backup" mode.:  K 4) Mount "broken" shadow disks read-only, back up the data, then re-add thej  disks back into the shadow sets.  K By putting Oracle into the "hot Backup" mode, disk activity to the database J files is stopped and no I/O's the data disks occur.   NOTE: This backup isL from the time you put Oracle into the "Hot backup" mode.  By also backing upI the archive logs from the time the shadow sets were broken, allows you toa* get a full backup of the database cleanly.  I NOTE:  This method will NOT be endorsed by either Compaq or Oracle, but I.I have been able to recover from any database problems (user's deleting thepI wrong information, database table drops, full database restores) with thed above procedure.  G   I have had problems just "breaking the shadow sets" from Oracle whileTK Oracle was running normally.    As mentioned above, you can not ensure whataJ disk writes are occurring, so the internal Oracle times on the tables getsG corrupted.   By using the "hot backup" method, this keeps the tables inoE sync.   Read up on the Hot backup method under Oracle and think aboutmF dismounting shadow sets as the "backup" being performed for the OracleI database.  Just because it takes two seconds to backup (remove the shadoweF set for backup later) rather than two hours, the "backup" is the same.   Don Rogstad  Dallas Semiconductor   -----Original Message-----@ From: Leigh G. Bowden [mailto:LGBowden@bowdenfamily.fsnet.co.uk]' Sent: Friday, February 15, 2002 2:21 PM  To: Info-VAX@Mvb.Saic.Como Subject: Backup query.    K I've been asked to improve the backups on a AS2100 with 2 CPU's and 1 GB ofaJ RAM it has 16 x 2.1 GB disks but shadowed as 8 (1 system disk and 7 data).K This is running VMS6.2 with Oracle 7.0 of some ilk. VMS is fully patched as1K of about three months ago. The backups are done to a DLT III 10/20GB deviceuK which is quite capable of taking the entire system on a tape with plenty toaI spare. The system is nominally 24 x 7 but it is fairly quiet at night the % backup runs at 3am Monday - Saturday.   E The current backup is rather erratic and I've changed it to do /IMAGE J backups of all the DSA devices but I now want to improve it. This is where I'm less sure.  L 1). At it's easiest breaking the shadow sets and mounting them privately andK doing backups of those before introducing them back into the shadow. As thegK system is quiet at night nobody would notice the loss of performance as then9 copies are done. Compaq don't seem to like this approach.m  F 2). Compaq seem to prefer dismounting the shadow volume then remount aI single disk as a shadow and backup from the other one then introduce that G one back into the shadow. I doubt this would work as Oracle will almosts  certainly have some files opens.  J 3). Shutting Oracle down briefly to do 1 or 2 is unlikely as the system isL needed 24 x 7 but I might be able to persuade the management otherwise if it was for about ten minutes only.o  < Is option 1 really an option despite Compaq's protestations?  J In all instances do the disks have be forced to dismount as files could be open on the disks?    ' ------_=_NextPart_001_01C1B667.528020E0l Content-Type: text/html; 	charset="iso-8859-1"h+ Content-Transfer-Encoding: quoted-printablet  1 <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2//EN">  <HTML> <HEAD>9 <META HTTP-EQUIV=3D"Content-Type" CONTENT=3D"text/html; =h charset=3Diso-8859-1">@ <META NAME=3D"Generator" CONTENT=3D"MS Exchange Server version =
 5.5.2653.12">r  <TITLE>RE: Backup query.</TITLE> </HEAD>o <BODY>  C <P><FONT SIZE=3D2>As others have stated, the problem is what disk = F activity is occurring when you break the shadow sets.&nbsp;&nbsp; As =F Compaq has stated, the only &quot;official 100% certain&quot; way to =D ensure that no disk activity is occurring is to dismount the disk, =D break the shadow sets, and then remount the disks.&nbsp; </FONT></P>  E <P><FONT SIZE=3D2>&nbsp;&nbsp; However, in a 24x7 shop, this is not =fG always feasible or available.&nbsp;&nbsp; I have had success with the =pH following when working with an Oracle database backups from shadow set = disks:</FONT></P>c  A <P><FONT SIZE=3D2>1) Put Oracle into the &quot;Hot Backup&quot; =-G mode.&nbsp; (This stops Oracle from writing directly to the data base =jA disks and redirects the writes to the &quot;redo logs&quot; (or =c9 &quot;archive logs&quot;, I currently forget).</FONT></P>a  B <P><FONT SIZE=3D2>2) Break the shadow sets containing the Oracle = databases files</FONT> </P>  ? <P><FONT SIZE=3D2>3) Clear Oracle from &quot;Hot Backup&quot; =h mode.</FONT> </P>  G <P><FONT SIZE=3D2>4) Mount &quot;broken&quot; shadow disks read-only, =s> back up the data, then re-add the disks back into the shadow = sets.</FONT> </P>  E <P><FONT SIZE=3D2>By putting Oracle into the &quot;hot Backup&quot; = G mode, disk activity to the database files is stopped and no I/O's the = F data disks occur.&nbsp;&nbsp; NOTE: This backup is from the time you =H put Oracle into the &quot;Hot backup&quot; mode.&nbsp; By also backing =G up the archive logs from the time the shadow sets were broken, allows =w< you to get a full backup of the database cleanly.</FONT></P>  C <P><FONT SIZE=3D2>NOTE:&nbsp; This method will NOT be endorsed by =eC either Compaq or Oracle, but I have been able to recover from any =tD database problems (user's deleting the wrong information, database =5 table drops, full database restores) with the above =s procedure.</FONT></P>c  F <P><FONT SIZE=3D2>&nbsp; I have had problems just &quot;breaking the =8 shadow sets&quot; from Oracle while Oracle was running =I normally.&nbsp;&nbsp;&nbsp; As mentioned above, you can not ensure what =sG disk writes are occurring, so the internal Oracle times on the tables = I gets corrupted.&nbsp;&nbsp; By using the &quot;hot backup&quot; method, =tF this keeps the tables in sync.&nbsp;&nbsp; Read up on the Hot backup =D method under Oracle and think about dismounting shadow sets as the =H &quot;backup&quot; being performed for the Oracle database.&nbsp; Just =C because it takes two seconds to backup (remove the shadow set for = D backup later) rather than two hours, the &quot;backup&quot; is the = same.</FONT></P>  $ <P><FONT SIZE=3D2>Don Rogstad</FONT>. <BR><FONT SIZE=3D2>Dallas Semiconductor</FONT> </P>  3 <P><FONT SIZE=3D2>-----Original Message-----</FONT>s. <BR><FONT SIZE=3D2>From: Leigh G. Bowden [<A =I HREF=3D"mailto:LGBowden@bowdenfamily.fsnet.co.uk">mailto:LGBowden@bowden=e family.fsnet.co.uk</A>]</FONT>A <BR><FONT SIZE=3D2>Sent: Friday, February 15, 2002 2:21 PM</FONT>n3 <BR><FONT SIZE=3D2>To: Info-VAX@Mvb.Saic.Com</FONT>a0 <BR><FONT SIZE=3D2>Subject: Backup query.</FONT> </P> <BR>  F <P><FONT SIZE=3D2>I've been asked to improve the backups on a AS2100 = with 2 CPU's and 1 GB of</FONT>dF <BR><FONT SIZE=3D2>RAM it has 16 x 2.1 GB disks but shadowed as 8 (1 = system disk and 7 data).</FONT>lH <BR><FONT SIZE=3D2>This is running VMS6.2 with Oracle 7.0 of some ilk. = VMS is fully patched as</FONT>I <BR><FONT SIZE=3D2>of about three months ago. The backups are done to a =  DLT III 10/20GB device</FONT>yG <BR><FONT SIZE=3D2>which is quite capable of taking the entire system =q on a tape with plenty to</FONT>.D <BR><FONT SIZE=3D2>spare. The system is nominally 24 x 7 but it is =  fairly quiet at night the</FONT>? <BR><FONT SIZE=3D2>backup runs at 3am Monday - Saturday.</FONT>o </P>  I <P><FONT SIZE=3D2>The current backup is rather erratic and I've changed =l it to do /IMAGE</FONT>E <BR><FONT SIZE=3D2>backups of all the DSA devices but I now want to =   improve it. This is where</FONT>( <BR><FONT SIZE=3D2>I'm less sure.</FONT> </P>  D <P><FONT SIZE=3D2>1). At it's easiest breaking the shadow sets and =" mounting them privately and</FONT>H <BR><FONT SIZE=3D2>doing backups of those before introducing them back = into the shadow. As the</FONT>E <BR><FONT SIZE=3D2>system is quiet at night nobody would notice the =0! loss of performance as the</FONT>CD <BR><FONT SIZE=3D2>copies are done. Compaq don't seem to like this = approach.</FONT> </P>  D <P><FONT SIZE=3D2>2). Compaq seem to prefer dismounting the shadow = volume then remount a</FONT>F <BR><FONT SIZE=3D2>single disk as a shadow and backup from the other = one then introduce that</FONT>I <BR><FONT SIZE=3D2>one back into the shadow. I doubt this would work as =t Oracle will almost</FONT> : <BR><FONT SIZE=3D2>certainly have some files opens.</FONT> </P>  D <P><FONT SIZE=3D2>3). Shutting Oracle down briefly to do 1 or 2 is =  unlikely as the system is</FONT>F <BR><FONT SIZE=3D2>needed 24 x 7 but I might be able to persuade the =! management otherwise if it</FONT>h9 <BR><FONT SIZE=3D2>was for about ten minutes only.</FONT>  </P>  A <P><FONT SIZE=3D2>Is option 1 really an option despite Compaq's =y protestations?</FONT>e </P>  C <P><FONT SIZE=3D2>In all instances do the disks have be forced to =r! dismount as files could be</FONT>/, <BR><FONT SIZE=3D2>open on the disks?</FONT> </P>   </BODY>o </HTML> ) ------_=_NextPart_001_01C1B667.528020E0--)   ------------------------------  % Date: Fri, 15 Feb 2002 15:39:26 -0600a From: rogstad@dalsemi.come Subject: Re: Backup query.1 Message-ID: <02021515392673@dscis6-0.dalsemi.com>t  E As others have stated, the problem is what disk activity is occurringa@ when you break the shadow sets.   As Compaq has stated, the only> "official 100% certain" way to ensure that no disk activity isB occurring is to dismount the disk, break the shadow sets, and then remount the disks.    F However, in a 24x7 shop, this is not always feasible or available.   I? have had success with the following when working with an Oracleb' database backups from shadow set disks:h  B 1) Put Oracle into the "Hot Backup" mode.  (This stops Oracle fromC writing directly to the data base disks and redirects the writes to 8 the "redo logs" (or "archive logs", I currently forget).  ? 2) Break the shadow sets containing the Oracle databases files t  ( 3) Clear Oracle from "Hot Backup" mode.   @ 4) Mount "broken" shadow disks read-only, back up the data, then, re-add the disks back into the shadow sets.   B By putting Oracle into the "hot Backup" mode, disk activity to theD database files is stopped and no I/O's the data disks occur.   NOTE:A This backup is from the time you put Oracle into the "Hot backup" C mode.  By also backing up the archive logs from the time the shadowvA sets were broken, allows you to get a full backup of the database  cleanly.  C NOTE:  This method will NOT be endorsed by either Compaq or Oracle,pB but I have been able to recover from any database problems (user'sC deleting the wrong information, database table drops, full databaseh# restores) with the above procedure.t  E I have had problems just "breaking the shadow sets" from Oracle whilebF Oracle was running normally.    As mentioned above, you can not ensureC what disk writes are occurring, so the internal Oracle times on theoE tables gets corrupted.   By using the "hot backup" method, this keepssC the tables in sync.   Read up on the Hot backup method under Oracle = and think about dismounting shadow sets as the "backup" beingeE performed for the Oracle database.  Just because it takes two secondstB to backup (remove the shadow set for backup later) rather than two hours, the "backup" is the samer   ------------------------------    Date: 16 Feb 2002 06:17:29 +0800, From: Paul Repacholi <prep@prep.synonet.com> Subject: Re: Backup query.0 Message-ID: <87eljm5r0m.fsf@k9.prep.synonet.com>  = "Leigh G. Bowden" <LGBowden@bowdenfamily.fsnet.co.uk> writes:   E > I've been asked to improve the backups on a AS2100 with 2 CPU's andtB > 1 GB of RAM it has 16 x 2.1 GB disks but shadowed as 8 (1 systemC > disk and 7 data).  This is running VMS6.2 with Oracle 7.0 of somee > ilk.  B If that is RDB, then you are wasting your time using backup on theD data-base files. You must use RMU. So check what the DB system needs< done. I'm not sure what the magik mix for 'orible Oricle is.  D > VMS is fully patched as of about three months ago. The backups areC > done to a DLT III 10/20GB device which is quite capable of takingrA > the entire system on a tape with plenty to spare. The system is-E > nominally 24 x 7 but it is fairly quiet at night the backup runs atr > 3am Monday - Saturday.  m@ > The current backup is rather erratic and I've changed it to doA > /IMAGE backups of all the DSA devices but I now want to improve " > it. This is where I'm less sure.  ? The system disk should be backed up as a 'base' system, and theo< logs etc you need to keep archived. Re-do the base backup ifA you do any changes. Test you can get it up and running on anothern= machine if needed. Take two copys so you can burn one down :)s And hope you don't need them.T  @ > 1). At it's easiest breaking the shadow sets and mounting themC > privately and doing backups of those before introducing them back.F > into the shadow. As the system is quiet at night nobody would noticeF > the loss of performance as the copies are done. Compaq don't seem to > like this approach.n   F > 2). Compaq seem to prefer dismounting the shadow volume then remount> > a single disk as a shadow and backup from the other one thenE > introduce that one back into the shadow. I doubt this would work as 5 > Oracle will almost certainly have some files opens.s  oB > 3). Shutting Oracle down briefly to do 1 or 2 is unlikely as the= > system is needed 24 x 7 but I might be able to persuade them< > management otherwise if it was for about ten minutes only.  C This is the key to *all* the rest of the backup. BTW, if you do notdE have the files flushed out and consistant, you are wasting your time. B Don't bother to back it up at all if it is not done right! See the@ Oracle docs for 'the right thing'. Also, plan what you are going8 to do about the db logs etc. Again, see the orible docs.  h> > Is option 1 really an option despite Compaq's protestations?   No, not in this case.y   C > In all instances do the disks have be forced to dismount as filese > could be open on the disks?t   If you want to be sure, yes.   --  < Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda. @                                              West Australia 6076. Raw, Cooked or Well-done, it's all half baked.F EPIC, The Architecture of the future, always has been, always will be.   ------------------------------  % Date: Fri, 15 Feb 2002 15:45:06 -0500i From: William_Bochnik@acml.com Subject: Re: Backup query.> Message-ID: <OF98A0EED4.A338315D-ON85256B61.0071D87B@acml.com>  = read up on how to do online backups in oracle with redo logs,s? etc.  you should convince management that this is important, askA you might as well use nl: as the output device for the backup, aslA your backups probably wont be restorable, no matter how quiet then
 system is.  = there are ways to backup oracle with no downtime, and minimal  impact.h      c                                                                                                    rc                       "Leigh G. Bowden"                                                            nc                       <LGBowden@bowdenfamily.f                To:  Info-VAX@Mvb.Saic.Com           3c                       snet.co.uk>                             cc:                                   c                                                        Subject: Backup query.                      Ec                       02/15/2002 03:20 PM                                                          dc                       Please respond to "Leigh                                                      c                       G. Bowden"                                                                   hc                       <LGBowden@bowdenfamily.f                                                     rc                       snet.co.uk>                                                                  rc                                                                                                    ic                                                                                                    r      ? I've been asked to improve the backups on a AS2100 with 2 CPU'ss and 1 GB of)A RAM it has 16 x 2.1 GB disks but shadowed as 8 (1 system disk andd 7 data).@ This is running VMS6.2 with Oracle 7.0 of some ilk. VMS is fully
 patched as< of about three months ago. The backups are done to a DLT III 10/20GB deviceA which is quite capable of taking the entire system on a tape withc	 plenty tom? spare. The system is nominally 24 x 7 but it is fairly quiet atk	 night the % backup runs at 3am Monday - Saturday.e  > The current backup is rather erratic and I've changed it to do /IMAGEA backups of all the DSA devices but I now want to improve it. Thisc is where I'm less sure.  > 1). At it's easiest breaking the shadow sets and mounting them
 privately andd< doing backups of those before introducing them back into the shadow. As the8 system is quiet at night nobody would notice the loss of performance as the9 copies are done. Compaq don't seem to like this approach.n  < 2). Compaq seem to prefer dismounting the shadow volume then	 remount at: single disk as a shadow and backup from the other one then introduce that@ one back into the shadow. I doubt this would work as Oracle will almost  certainly have some files opens.  @ 3). Shutting Oracle down briefly to do 1 or 2 is unlikely as the	 system ist< needed 24 x 7 but I might be able to persuade the management otherwise if it  was for about ten minutes only.   < Is option 1 really an option despite Compaq's protestations?  A In all instances do the disks have be forced to dismount as files  could be open on the disks?            F ______________________________________________________________________;  The information contained in this transmission may contain @ privileged and confidential information and is intended only forA the use of the person(s) name above.  If you are not the intendeds= recipient, or an employee or agent responsible for deliveringd3 this message to the intended recipient, any review,b@ dissemination, distribution or duplication of this communication? is strictly prohibited.  If you are not the intended recipient, A please contact the sender immediately by reply e-mail and destroya# all copies of the original message.o   ------------------------------  % Date: Fri, 15 Feb 2002 22:13:10 -0500 1 From: Michael Austin <maustin@firstdbasource.com>a Subject: Re: Backup query.2 Message-ID: <3C6DCE46.259624A0@firstdbasource.com>  H sounds like you need an Oracle DBA Consultant to come in and help defineD and implement the database backups. If you don't do it correct, your2 backups might as well be bits in the bit-bucket.     -- e Regards,  7 Michael Austin            Registered Linux User #261163 7 First DBA Source, Inc.    http://www.firstdbasource.comt= Sr. Consultant  -- Specializing in all things Oracle and VMS.l 704-947-1089 (Office)l 704-236-4377 (Mobile).     "Leigh G. Bowden" wrote: > M > I've been asked to improve the backups on a AS2100 with 2 CPU's and 1 GB of L > RAM it has 16 x 2.1 GB disks but shadowed as 8 (1 system disk and 7 data).M > This is running VMS6.2 with Oracle 7.0 of some ilk. VMS is fully patched ascM > of about three months ago. The backups are done to a DLT III 10/20GB device?M > which is quite capable of taking the entire system on a tape with plenty tosK > spare. The system is nominally 24 x 7 but it is fairly quiet at night theT' > backup runs at 3am Monday - Saturday.H > G > The current backup is rather erratic and I've changed it to do /IMAGErL > backups of all the DSA devices but I now want to improve it. This is where > I'm less sure. > N > 1). At it's easiest breaking the shadow sets and mounting them privately andM > doing backups of those before introducing them back into the shadow. As theeM > system is quiet at night nobody would notice the loss of performance as thel; > copies are done. Compaq don't seem to like this approach.  > H > 2). Compaq seem to prefer dismounting the shadow volume then remount aK > single disk as a shadow and backup from the other one then introduce that2I > one back into the shadow. I doubt this would work as Oracle will almosth" > certainly have some files opens. > L > 3). Shutting Oracle down briefly to do 1 or 2 is unlikely as the system isN > needed 24 x 7 but I might be able to persuade the management otherwise if it! > was for about ten minutes only.i > > > Is option 1 really an option despite Compaq's protestations? > L > In all instances do the disks have be forced to dismount as files could be > open on the disks?   ------------------------------  % Date: Fri, 15 Feb 2002 13:47:28 -0500>( From: David Froble <davef@tsoft-inc.com>A Subject: Re: Batch job termination - OpenVMS v7.3 v. OpenVMS v7.1Z, Message-ID: <3C6D57C0.4060000@tsoft-inc.com>   > ----- Original Message -----: > From: "Andrew Rycroft" <andrew.rycroft@intrinsitech.com> > Newsgroups: comp.os.vmsy) > Sent: Friday, February 15, 2002 1:26 PMu? > Subject: Batch job termination - OpenVMS v7.3 v. OpenVMS v7.1n >  >  >  >>Hi,a >>H >>I have just upgrade to OpenVMS v7.3 from openVMS v7.1. We have a batchH >>job that runs continuously. On occasion we stop it, and restart it. To >>stop is we use >> >>$ Delete/entry=n  1 So, Ok, The BIG HAMMER approach works many times.   I There are shall I say, cleaner methods to stop a job that is not running lI interactively.  Just one example would be a system or group logical, for c example:   PROCESS_XYZ_STOP_NOW = "NO"'  F The logical value is normally 'NO' or 'N' or whatever.  Possibly just E it's existance, or lack thereof.  The application would periodically >> chack the logical, and when a specified condition is met, the . application will shut down in a normal manner.  I Other possibilities, definitely not a complete list, would be mailbox or iG socket communications, the contents of a communications disk file, etc.e   The logical is rather simple.g   Dave   -- b4 David Froble                       Tel: 724-529-04504 Dave Froble Enterprises, Inc.      Fax: 724-529-0596> DFE Ultralights, Inc.              E-Mail: davef@tsoft-inc.com6 T-Soft, Inc.  170 Grimplin Road  Vanderbilt, PA  15486   ------------------------------  % Date: Fri, 15 Feb 2002 21:48:46 -0000o- From: wspencer@ap.nospam.org (Warren Spencer)&A Subject: Re: Batch job termination - OpenVMS v7.3 v. OpenVMS v7.1o7 Message-ID: <91B6AE454warrenspencer1977@209.249.90.100><  D davef@tsoft-inc.com (David Froble) wrote in <3C6D57C0.4060000@tsoft-	 inc.com>:E   >> ----- Original Message -----R; >> From: "Andrew Rycroft" <andrew.rycroft@intrinsitech.com>e >> Newsgroups: comp.os.vms* >> Sent: Friday, February 15, 2002 1:26 PM@ >> Subject: Batch job termination - OpenVMS v7.3 v. OpenVMS v7.1 >> @ >> a >> m >>>Hi, >>>>I >>>I have just upgrade to OpenVMS v7.3 from openVMS v7.1. We have a batchDI >>>job that runs continuously. On occasion we stop it, and restart it. ToG >>>stop is we useF >>>I >>>$ Delete/entry=nx >12 >So, Ok, The BIG HAMMER approach works many times. >aJ >There are shall I say, cleaner methods to stop a job that is not running J >interactively.  Just one example would be a system or group logical, for 	 >example:  >  >PROCESS_XYZ_STOP_NOW = "NO" > G >The logical value is normally 'NO' or 'N' or whatever.  Possibly just nF >it's existance, or lack thereof.  The application would periodically ? >chack the logical, and when a specified condition is met, the i/ >application will shut down in a normal manner.  >sJ >Other possibilities, definitely not a complete list, would be mailbox or H >socket communications, the contents of a communications disk file, etc. >  >The logical is rather simple. >i >Dave  >o  E There are many possiblities, but the logical name method can lead to sK problems... Application comes up, logical name is already defined from the aJ previous shutdown, application see it and goes down immediately, operator > scratches his head.  Repeat until operator wakes someone up...   ws     -- >   Warren Spencer' Senior Software Engineer (not a writer)n The Associated Press  < ** Time flies like an arrow.  Fruit flies like a bananna. **   ------------------------------  # Date: Fri, 15 Feb 2002 22:10:21 GMT=* From: Jose Arroyo <jose.arroyo@pandora.be># Subject: Re: Carly interview on CNNN* Message-ID: <3C6D8724.FDA884AC@pandora.be>  N I guess that if Carly sells all non-PC/non-HP stuff she would safe a lot more!P The chances that DELL would hit even harder are not slim. Personally, OpenVMS isO what I'm most interested in of Compaq. Regarding PC's I have another choice andSD that one starts with a D* . But ofcourse, that is a personal choice.O I thank that if the merger goes on and the top-management decide to concentrateeH more and more on everything but OpenVMS servers they will loose a lot ofP customers who probably will go elsewhere to port their applications on some Unix	 language.s       JF Mezei wrote:s   > Additional tidbit: >PO > Carly says that by end of 2003, $2 billion in costs will have been saved, andi) > an additional $500 million during 2004.e   ------------------------------  # Date: Sat, 16 Feb 2002 00:00:27 GMT_- From: "John E. Malmberg" <wb8tyw@qsl.network>-L Subject: Re: compaq logo on DECwindows login screen...how to get the old one* Message-ID: <3C6DA3CE.5060106@qsl.network>   Roland Barmettler wrote:  I > Thanks for all the answers and the colourful discussion which I enjoyednA > reading. However, I've had trouble with my news server for overyG > a week, so that part (how to change the logo without CDE) is missing.n0 > :-( Could someone re-post that article to me ?
 > Thanks !  > This newsgroup/mailing list and many others are searchable at 4 http://www.google.com for past articles and threads.   -Johnk wb8tyw@qsl.network Personal Opinion Only_   ------------------------------  # Date: Sat, 16 Feb 2002 03:31:59 GMT 1 From: "David J. Dachtera" <djesys.nospam@fsi.net>iL Subject: Re: compaq logo on DECwindows login screen...how to get the old one' Message-ID: <3C6DD432.40E26402@fsi.net>i   Lyndon Bartels wrote:e > G > I'd be happy with "OpenVMS" for a logo, at least that way it wouldn'te; > have to change no matter what happened with the "merger."m > B > Of course, a phoenix wouldn't be too bad either... You know, the> > beautiful bird born from the reborn from it's own ashes..... > & > VMS .... reborn.... Not bad..... ;-)  3 Then, I guess that makes me a born-again VMS bigot.e   Say, "Hallelujah, Brothers!"   -- u David J. Dachterad dba DJE Systemsu http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/a   ------------------------------  , Date: Fri, 15 Feb 2002 21:43:57 -0200 (BRST) From: valdemir-@uol.com.br Subject: Creating a mailbox...5 Message-ID: <200202152343.VAA27413@orwell.uol.com.br>h  7 I=B4m having problems to create a mailbox in Pascal.=20 ; I need create a mailbox in Pascal to get all Opcom messagesl( send to operator=B4s console. My doubts:= - How create a mailbox in Pascal and enable this mailbox with I   Operator privilege to catch all messages send to Operator=B4s console ? * - How read data buffered in this mailbox ?6 - How clean up old messages buffered in this mailbox ?E Sorry for the insistence, but in my new job I have to read "zillions"sH  of messages sent to my terminal. I "need" create a program to catch allE  messages and search for string "Error - Batch job JOB_NAME finish=20 C  abnormally !"  - If I don=B4t catch batch error messages in my =20 ,  terminal I=B4m seriously  advertised !  =20A  A complete program=B4s example it will be gracefull apreciate...a  Thanks in advance...h   ------------------------------  % Date: Sat, 16 Feb 2002 01:28:04 +0000i0 From: Mark W J Redding <grawlfang@ukhackers.net>" Subject: Re: Creating a mailbox...- Message-ID: <3C6DB5A4.81BF0631@ukhackers.net>n   valdemir-@uol.com.br wrote:d  4 > Im having problems to create a mailbox in Pascal.= > I need create a mailbox in Pascal to get all Opcom messagesl( > send to operators console. My doubts:? > - How create a mailbox in Pascal and enable this mailbox with I >   Operator privilege to catch all messages send to Operators console ? , > - How read data buffered in this mailbox ?8 > - How clean up old messages buffered in this mailbox ?G > Sorry for the insistence, but in my new job I have to read "zillions"nJ >  of messages sent to my terminal. I "need" create a program to catch allD >  messages and search for string "Error - Batch job JOB_NAME finish? >  abnormally !"  - If I dont catch batch error messages in myC' >  terminal Im seriously  advertised !aA >  A complete programs example it will be gracefull apreciate...p >  Thanks in advance...-  F I used to be able to do this in Pascal (back in about 1988).. I've got- examples of this in Fortran if that helps ???:     -- Mark W J Redding  F Chaos reigns within.  Reflect, repent and reboot.  Order shall return.   ------------------------------  # Date: Fri, 15 Feb 2002 23:14:59 GMTo' From: Don Sykes <anonymous@pacbell.net>o Subject: Re: DCL procedure's+ Message-ID: <3C6D9770.C2D910BB@pacbell.net>r  A You can download a bunch at http://alphase.com/vms/VMSTools.html.     
 tim wrote: >  > Hello, > / > i am looking for some usefull DCL-procedure'sn% > maybe you wanna share some programs  >  > many thanx >  > tim    --     Have VMS. Will Travel. Wire Paladin @alphase.com 
 San Franciscot   ------------------------------  # Date: Sat, 16 Feb 2002 04:28:48 GMTt1 From: "David J. Dachtera" <djesys.nospam@fsi.net>b Subject: Re: DCL procedure's' Message-ID: <3C6DE159.2052B04A@fsi.net>"  
 tim wrote: >  > Hello, > / > i am looking for some usefull DCL-procedure'st% > maybe you wanna share some programsl >  > many thanx  2 Poke around at http://www.djesys.com/freeware/vms/  E The web page at http://www.djesys.com/vms/freeware.html explains mostt (but not all) of those files.o   -- p David J. Dachtera  dba DJE Systemsk http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/g   ------------------------------   Date: 16 Feb 2002 00:37:31 GMT& From: peter@abbnm.com (Peter da Silva)6 Subject: Re: Did Intel Bite Off More Than It Can Chew?% Message-ID: <a4k9kb$7sn@web.nmti.com>   % In article <3C690A60.6634450@gmx.de>, * Bernd Paysan  <bernd.paysan@gmx.de> wrote:I > > > > This means she's going to do to HP what Cortez did to the Aztecs?r   > > > No, to Compaq.   > > Both seem likely.v  @ > Ah, like the Conquistadores were formerly Reconquista soldiersE > ("freeing" Spain from the quite liberal and economically prospering0J > arabs and installing a fundamentalist katholic regime, which adhered theH > poverty ideal by making most of the inhabitants poor), the destructionI > of HP is something she already had succeeded in. What does she think HPoD > is? A *printer company*! She even stopped the calculator business.  I I also understand that the sudden influx of New World gold really screwedb* up the Spanish economy in the longer term.  L Speaking of HP, I just found in my mailslot the first marketing material forF the HP 3000 line I can ever recall seeing. The thrust of the flyer wasG reminiscent of Monty Python's Black Knight yelling "I'm not dead yet! I G can still fight! Chicken! Chicken!". If they'd started doing this a few  years ago, who knows...    --  +  `-_-'   In hoc signo hack, Peter da Silva.oE   'U`    "A well-rounded geek should be able to geek about anything."pL                                                        -- nicolai@esperi.org          Disclaimer: WWFD?   ------------------------------  % Date: Fri, 15 Feb 2002 14:11:19 -0500y0 From: "Syltrem" <syltrem@videotron.spammenot.ca> Subject: Re: Gordon Bell siteo3 Message-ID: <Z2db8.5242$EI.28188@tor-nn1.netcom.ca>i  J The Redmond Marketing Company trying to boost their sales some more? M$ do better marketing than software.b TGIF.  --   SyltremeI http://pages.infinit.net/syltrem (OpenVMS related web site - en franais)a> To reply to myself directly, remove .spammenot from my address  J "JF Mezei" <jfmezei.spamnot@videotron.ca> a crit dans le message de news:! 3C6D59BA.197A983D@videotron.ca...  > Syltrem wrote: > >t4 > > now how did this end up on a microsoft web site? >eF > Microsoft is in charge of rewriting history for its Encarta product.   ------------------------------  % Date: Fri, 15 Feb 2002 11:47:27 -0800C+ From: Mark Crispin <MRC@CAC.Washington.EDU>. Subject: Re: Gordon Bell site N Message-ID: <Pine.WNT.4.50.0202151146550.1016-100000@Shimo-Tomobiki.Panda.COM>  # On Fri, 15 Feb 2002, Syltrem wrote: 2 > now how did this end up on a microsoft web site?  A Maybe it has something to do with Gordon Bell being at Microsoft.o  
 -- Mark --   http://staff.washington.edu/mrcsF Science does not emerge from voting, party politics, or public debate.   ------------------------------    Date: 15 Feb 2002 12:12:47 -08003 From: Eric Smith <eric-no-spam-for-me@brouhaha.com>l Subject: Re: Gordon Bell sitep0 Message-ID: <qh4rkiwlkw.fsf@ruckus.brouhaha.com>  3 bill@triangle.cs.uofs.edu (Bill Gunshannon) writes: F > Which, of course, isn't true.  I still get the periodic catalog withG > $5000 386's.  Quality is still available to those who need it and area > willing to pay for it.  E Are you *sure* that those $5000 386s are really of any better qualitym% than your typical $2000 Dell machine?o  F In my experience, there are certainly more expensive things available,) but quality isn't usually the difference.a   ------------------------------   Date: 15 Feb 2002 20:43:11 GMT1 From: bill@triangle.cs.uofs.edu (Bill Gunshannon)y Subject: Re: Gordon Bell siteo, Message-ID: <a4jrsv$1p54$1@info.cs.uofs.edu>  3 In article <sLcb8.5238$EI.28292@tor-nn1.netcom.ca>,s3  "Syltrem" <syltrem@videotron.spammenot.ca> writes: 3 |> now how did this end up on a microsoft web site?c  < Apparently, he works for the evil empire now.  go figure....     bill   -- wJ Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>   n   ------------------------------   Date: 15 Feb 2002 20:50:52 GMT1 From: bill@triangle.cs.uofs.edu (Bill Gunshannon)- Subject: Re: Gordon Bell sitee, Message-ID: <a4jsbc$1p54$2@info.cs.uofs.edu>  0 In article <qh4rkiwlkw.fsf@ruckus.brouhaha.com>,6  Eric Smith <eric-no-spam-for-me@brouhaha.com> writes:6 |> bill@triangle.cs.uofs.edu (Bill Gunshannon) writes:I |> > Which, of course, isn't true.  I still get the periodic catalog withhJ |> > $5000 386's.  Quality is still available to those who need it and are |> > willing to pay for it.l |> .H |> Are you *sure* that those $5000 386s are really of any better quality( |> than your typical $2000 Dell machine? |>  E Well, I've never bought one :-) but I would guess based on the targetsC audience that they probably are.  These are industrial PC's, mostlypD rack mount and designed for dirty, and hostile environments.  SealedH boxes with filtered cooling fans.  Membrane encased keyboards.  Usually,E no mouse. :-) Probably (or, at least I hope) much more filtered powero supplies, usually redundant.  E When you read about a Power Plant or a Navy Ship using NT for controluG I certainly hope people don't think this means they bought from a cow!!a  pI |> In my experience, there are certainly more expensive things available,o, |> but quality isn't usually the difference.  H There are places where the survival of a home style PC would be measuredI in seconds.  You can get quality, but you will pay a real premium for it.i   bill   -- oJ Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>       ------------------------------  % Date: Fri, 15 Feb 2002 16:51:28 -0500T9 From: "Douglas H. Quebbeman" <dougq@ixnayamspayiglou.com>f Subject: Re: Gordon Bell sitei' Message-ID: <3c6d82e4_1@news.iglou.com>s  > "Bill Gunshannon" <bill@triangle.cs.uofs.edu> wrote in message& news:a4jrsv$1p54$1@info.cs.uofs.edu...5 > In article <sLcb8.5238$EI.28292@tor-nn1.netcom.ca>,t5 >  "Syltrem" <syltrem@videotron.spammenot.ca> writes:05 > |> now how did this end up on a microsoft web site?n >e> > Apparently, he works for the evil empire now.  go figure....  ; Yeah, but salary aside, he has a perk of being able to host 8 as much stuff as he wants- I mean, Dr. Bell can host not1 merely terabytes, but petabytes of data for free!u   -dql   --K -Douglas Hurst Quebbeman (dougq@ixsnayamspayiglou.com)     [Call me "Doug"] J   Surgically excise the pig-latin from my e-mail address in order to replyH   "The large print giveth, and the small print taketh away."  -Tom Waits   ------------------------------    Date: 15 Feb 2002 22:53:30 +0100- From: Neil Franklin <neil@franklin.ch.remove>e Subject: Re: Gordon Bell site / Message-ID: <6ug042pg2t.fsf@chonsp.franklin.ch>t  2 "Syltrem" <syltrem@videotron.spammenot.ca> writes:   [corrected wrongly quoted post]e  D > "Alan Greig" <a.greig@virgin.net> a crit dans le message de news:/ > koqp6uclolvv3jevp4ok52sb4j96dofiav@4ax.com...n > >eJ > > Just saw a reference to this site in comp.arch and hadn't known it wasA > > there. Check out http://research.microsoft.com/~gbell and, inp >o2 > now how did this end up on a microsoft web site?  : Because Gorden Bell works at Microsofts research division?     --? Neil Franklin, neil@franklin.ch.remove http://neil.franklin.ch/y? Hacker, Unix Guru, El Eng HTL/BSc, Sysadmin, Archer, Roleplayer / - Intellectual Property is Intellectual Robberyr   ------------------------------  % Date: Fri, 15 Feb 2002 15:03:10 -0700r+ From: Ben Franchuk <bfranchuk@jetnet.ab.ca>t Subject: Re: Gordon Bell sitet, Message-ID: <3C6D859E.A2A6FF05@jetnet.ab.ca>   Neil Franklin wrote:  < > Because Gorden Bell works at Microsofts research division?  E Mabie they need to move him to the Software division. He could be the # only one who knows how to fix bugs!e -- a% Ben Franchuk - Dawn * 12/24 bit cpu *?+ www.jetnet.ab.ca/users/bfranchuk/index.htmla   ------------------------------  % Date: Fri, 15 Feb 2002 17:31:14 -0500t* From: WILLIAM WEBB <WWEBB1@email.usps.gov> Subject: RE: Gordon Bell site - Message-ID: <0033000053332045000002L052*@MHS>9   =0AThey can't do that.  H If they fix all the bugs then nobody will buy the bugfixes, er upgrades= .i   WWWebb   -----Original Message-----/ From: Info-VAX-Request@Mvb.Saic.Com at INTERNETu' Sent: Friday, February 15, 2002 5:06 PM-B To: Webb, William W Raleigh, NC; Info-VAX@Mvb.Saic.Com at INTERNET Subject: RE: Gordon Bell sitec     Neil Franklin wrote:  < > Because Gorden Bell works at Microsofts research division?  E Mabie they need to move him to the Software division. He could be thet# only one who knows how to fix bugs!a --% Ben Franchuk - Dawn * 12/24 bit cpu * , www.jetnet.ab.ca/users/bfranchuk/index.html=   ------------------------------   Date: 15 Feb 2002 22:40:48 GMT( From: peter@taronga.com (Peter da Silva) Subject: Re: Gordon Bell sitei1 Message-ID: <a4k2pg$38b$1@citadel.in.taronga.com>   3 In article <sLcb8.5238$EI.28292@tor-nn1.netcom.ca>, / Syltrem <syltrem@videotron.spammenot.ca> wrote: 1 >now how did this end up on a microsoft web site?   G Microsoft's research division actually does a lot of really interesting I stuff, and has a lot of cool people. I suspect they push anyone who comeskB up with innovative ideas over there before they can do any damage.   --  @ Rev. Peter da Silva, ULC.	                                 WWFD?  F "Be conservative in what you generate, and liberal in what you accept" 	-- Matthew 10:16 (l.trans)    ------------------------------  # Date: Fri, 15 Feb 2002 22:53:16 GMThL From: winston@SSRL.SLAC.STANFORD.EDU ("Alan Winston - SSRL Admin Cmptg Mgr") Subject: Re: Gordon Bell site 8 Message-ID: <00A099B8.F6B50362@SSRL04.SLAC.STANFORD.EDU>  \ In article <a4k2pg$38b$1@citadel.in.taronga.com>, peter@taronga.com (Peter da Silva) writes:4 >In article <sLcb8.5238$EI.28292@tor-nn1.netcom.ca>,0 >Syltrem <syltrem@videotron.spammenot.ca> wrote:2 >>now how did this end up on a microsoft web site? > H >Microsoft's research division actually does a lot of really interestingJ >stuff, and has a lot of cool people. I suspect they push anyone who comesC >up with innovative ideas over there before they can do any damage.   M Well, they hire the same kind of people who would have been Apple Fellows (oraM have spent entire careers at Xerox PARC) directly into the research division,sL presumably on the promise that they will never have to soil their hands with  any currently-shipping products.  I This enables them to recruit respected computer people, get patents, and yI have some claim to innovation, even if little of that innovation shows up  in actual products.r   -- Alani  O ===============================================================================b0  Alan Winston --- WINSTON@SSRL.SLAC.STANFORD.EDUM  Disclaimer: I speak only for myself, not SLAC or SSRL   Phone:  650/926-3056rM  Physical mail to: SSRL -- SLAC BIN 69, PO BOX 4349, STANFORD, CA  94309-0210dO ===============================================================================    ------------------------------  % Date: Fri, 15 Feb 2002 18:11:03 -0400o+ From: Tim Shoppa <shoppa@trailing-edge.com>l Subject: Re: Gordon Bell sites1 Message-ID: <3C6D4F37.724F97F5@trailing-edge.com>n   John Smith wrote:k > 4 > "Alan Greig" <a.greig@virgin.net> wrote in message4 > news:koqp6uclolvv3jevp4ok52sb4j96dofiav@4ax.com...J > > Just saw a reference to this site in comp.arch and hadn't known it was
 > > there.F > http://research.microsoft.com/~gbell/Digital/timeline/dechistory.htm > >  > H > Interesting to reflect on the comparative values of the equipment eachH > company sold, and the relative amount of research that each undertook. > C > Reminds me a lot of the book written by Harvard's Michael Porter,- > 'Competitive Strategy'.- > L > Comapq pursued the 'low-value high volume' model, whereas Digital followed$ > the 'low-volume high value model'.  ? In the early 60's, DEC *was* the "low-value high volume" model.6= They sold pre-stuffed PCB's that the user could plug togethere> into his own system, and then started selling "Programmed Data? Processors" rather than "Computers" because they were targeting ; a market that didn't have the purchasing authority to buy ao "Computer".u  ; Things, of course, are relative, and both the world and DEC = had greatly changed by the late 80's when the DEC sales forcesA was ignoring PO's from anyone not high enough on the Fortune 500.u   Tim.   ------------------------------   Date: 15 Feb 2002 22:48 CSTi' From: carl@gerg.tamu.edu (Carl Perkins)w Subject: Re: Gordon Bell site-- Message-ID: <15FEB200222483462@gerg.tamu.edu>   , peter@taronga.com (Peter da Silva) writes...4 }In article <sLcb8.5238$EI.28292@tor-nn1.netcom.ca>,0 }Syltrem <syltrem@videotron.spammenot.ca> wrote:2 }>now how did this end up on a microsoft web site? } H }Microsoft's research division actually does a lot of really interestingJ }stuff, and has a lot of cool people. I suspect they push anyone who comesC }up with innovative ideas over there before they can do any damage.a }  }-- A }Rev. Peter da Silva, ULC.	                                 WWFD?p  F It seems to me that their research is often quite good. Unfortunately,A the people who decide which of the results of the research shouldmB be implemented are aparently nuts and the people who do the actual implementing are sloppy.   Or, in a nutshell: Smart researchers  Crazy management Sloppy programmers   --- Carl   ------------------------------    Date: 16 Feb 2002 01:03:10 +0800, From: Paul Repacholi <prep@prep.synonet.com>H Subject: Re: Hello Compaq!  What happens to us VMS users if Itanic dies?0 Message-ID: <87it8y7k4x.fsf@k9.prep.synonet.com>  - young_r@encompasserve.org (Rob Young) writes:   = > 	They don't have to.  IA32s in the millions and IA64 in thed? > 	hundred thousands.  The initial startup costs for IA64 won'tl; > 	be recouped for a very long time.  > Today, they do muchl? > 	smaller volumes of Xeon/Foster than they do P4.  The marginso? > 	on IA64 will be much higher than IA32 but I am sure they arep> > 	making money today, recouping ALL the costs is sometime far6 > 	out there.  They don't have to make a lot on it for= > 	sometime.... they are still quite profitable with IA32 andoA > 	they will be profitable with IA64 and the cost of Itanium witho@ > 	two go-rounds of shrinks (Itanium->McKinley at .18 -> Madison% > 	at .13) will be considerably less.   J The itanic can't even cover the cost of making the masks. So where is thisH wonderous revinue? IA-64 is a massive loss, even without the development costs.  A > 	The way I see it working is what we see today.  That Dell IA64nB > 	server we keep coming back to is quite a bit cheaper than most.G > 	3-4 years it is still quite a bit cheaper and on par for performancek= > 	and on-chip switches ensure it scales quite readily to 64+t > 	processors.  D The Dell is so cheap you can't buy it, it has gone. Its dead Fred...C (OK, possible not the best use of the expression. Not you Fred ;) )p    ? > 	The main reason Itanium is so stinking expensive is the sameo? > 	reason Pentium Pro and initial Xeon were so high... off-chip @ > 	L2.  McKinley should be cheaper as L2 moves on-chip.  MadisonA > 	, next year moderately less expensive as the die shrinks going 
 > 	to .13.  C Have you seen the papers on what they are planning? And the size ofi McMonster and its MadSon?   C However, in intels favour, they should be getting out of the shadowhE of the SGV demise and will soon be able to move away from phase-shiftD- masks and that will be a big boost for them.     -- l< Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda.o@                                              West Australia 6076. Raw, Cooked or Well-done, it's all half baked.F EPIC, The Architecture of the future, always has been, always will be.   ------------------------------  % Date: Fri, 15 Feb 2002 14:48:53 -0500e- From: JF Mezei <jfmezei.spamnot@videotron.ca>-H Subject: Re: Hello Compaq!  What happens to us VMS users if Itanic dies?, Message-ID: <3C6D6614.9BBDC42F@videotron.ca>   Paul Repacholi wrote:tL > The itanic can't even cover the cost of making the masks. So where is thisJ > wonderous revinue? IA-64 is a massive loss, even without the development > costs.  : Engineers make it compute, accountants make it profitable.  N Intel only need to combine the 8086 and the IA64 divisions/cost centres into aN single unit so that the unit will be profitable (IA64 subsidized by the 8086).J If Compaq can do it for VMS/wintel, Shirley Intel can do it for IA64/8086.  M The big question is whether Intel wants the pride of succeeding in a new chipr9 or whether Intel will make decidions to increase profits.W   ------------------------------  # Date: Fri, 15 Feb 2002 19:45:29 GMTa* From: "Bill Todd" <billtodd@metrocast.net>L Subject: Re: Re: Hello Compaq!  What happens to us VMS users if Itanic dies?A Message-ID: <tzdb8.43993$Re2.3821424@bin6.nnrp.aus1.giganews.com>h  8 "Rob Young" <young_r@encompasserve.org> wrote in message- news:CiMg6GiqVcnJ@eisner.encompasserve.org...iD > In article <00A099A8.B4EA922C.8@maxwell.ph.kcl.ac.uk>, Nigel Arnot% <sysmgr@maxwell.ph.kcl.ac.uk> writes:    ...s  I > > And even if you can break out Intel's x86 processor division from thei rest,kI > > there's also the underlying cost structures to compare. My prejudiceso alwaysL > > favour a young, lean and mean company that's growing its market share atF > > the expense of an older and (usually) fatter competitor. Time will tell.... > >e >0@ > I haven't.  We do know they are in memories and other segmentsC > (chipsets).  To clarify.... I would guess IA32 *related* revenuestA > are 90%, leaving $3 billion to other segments, maybe it is 70%.  > Whatever it is , it is large.t  J If you want to compare 'IA32 *related* revenues' to something, then you'llL have to include VIA, Micron, and many other competitors in your comparisons,I since a great deal of such revenue comes from ancillary products that AMDoJ either doesn't make at all (but others do, both for its processors and forK Intel's) or doesn't make as the sole source.  But that's getting fairly far J afield from anything remotely related to supporting your contentions aboutH the relative efficiency of Intel's processor manufacturing vs. AMD's, so
 don't bother.i  C It does, however, highlight the fact that Intel's dominance in IA32-F processors undoubtedly leverages a great deal of additional revenue inE related products, which in turn will make Intel all the more ready tomL sacrifice Itanic on the altar of Yamhill if Yamhill turns out to be vital to' its continued success in the IA32 area.w  J Over the years you established a clear record of uncritical enthusiasm forI Alpha's eventual success despite increasing danger signs to the contrary.7I While it's not clear why you suddenly performed a complete about-face andGL switched your allegiance to Itanic, it is clear that your enthusiasm remainsJ at least equally uncritical (and hence equally suspect).  'Blind faith' isG not at all a bad characterization of such an attitude:  one may meet atlK least the legal definition of blindness while still being able to perceive, % dimly, a hand in front of one's face.    - bill   ------------------------------  % Date: Fri, 15 Feb 2002 13:58:50 -0500h( From: David Froble <davef@tsoft-inc.com>H Subject: Re: Hello Compaq!  What happens to us VMS users if Itanic dies?, Message-ID: <3C6D5A6A.6060601@tsoft-inc.com>   Rob Young wrote:  W > In article <3C6C2BC1.60402@tsoft-inc.com>, David Froble <davef@tsoft-inc.com> writes:1 >  > O >>>So far, isn't HP the only one to have bet the farm on IA64 ?  Have any otheryL >>>companies prematurely terminated products such as Alpha, Tru64 and MPE inM >>>anticipation for a potential performance of a bloated chip sometime in the  >>>future ?- >>>i >>>nI >>I have to say that I respect many of Rob's views.  However, he appears .F >>to be applying 'blind faith' to IA-64.  I really don't see anybody, G >>other than Compaq, bowing to the inevitable conquest of all by Intel.i >> >> > H > 	Actually, not blind faith.  There are some key technologies in Intel & > 	process and Alpha that will be key. >  > H >>Intel has decreed, "resistance is futile".  IBM, AMD, and others have  >>replied, "Yeah, Right!"f >> > @ > 	AMD isn't even on the radar screen.  As Paul Jacobi points to< > 	this article, here is a summary of where the billions go: > 6 > http://www.anandtech.com/cpu/showdoc.html?i=1584&p=2  G Ok, Intel has R&D.  I'd hope so, but with respect to IA-64, it appears oH that either the R&D isn't too effective, or the R&D is ignored.  Having I R&D, good or bad, and also having the ineffective disasted that is IA-64 $ today sure doesn't impress me.  E You think IBM doesn't have good R&D?  DEC had good R&D, until Compaq eI castrated it.  I'm sure that Compaq felt that they couldn't compete with d= Intel's R&D, once they killed off all (or much of) their own.i  B If the IA-64 story is the result of Intel's R&D, then Intel isn't # getting much of ROI from their R&D.p  F No, I still do not believe that ONLY Intel can advance the science of 
 computing.   Dave   -- N4 David Froble                       Tel: 724-529-04504 Dave Froble Enterprises, Inc.      Fax: 724-529-0596> DFE Ultralights, Inc.              E-Mail: davef@tsoft-inc.com6 T-Soft, Inc.  170 Grimplin Road  Vanderbilt, PA  15486   ------------------------------  % Date: Fri, 15 Feb 2002 15:10:43 -0500r- From: JF Mezei <jfmezei.spamnot@videotron.ca>iH Subject: Re: Hello Compaq!  What happens to us VMS users if Itanic dies?, Message-ID: <3C6D6B31.4C8547D3@videotron.ca>   Bill Todd wrote:K > While it's not clear why you suddenly performed a complete about-face and % > switched your allegiance to Itanic,#  L That is something for the National Equirer or World Weekly news to speculateG on. If Mr Young does not wish to divulge what event made him switch hisa/ allegiance to the dark side that is his choice.s  ( There are many possibilities, including:E 	-having an affair with Andy Grove's daughter and/or Capellas's wife.$  Y 	-having source of revenus from Compaq or Intel or HP which means that he must prostitutee1 	 himself and openly support all their decisions.n  > 	-having been invaded by an alien mind out to undermine earth.  R 	-married into the Gates family and wants his part of the fortune when Billy dies.  L 	-blackmailed by Compaq who found a photo of him dating in the 1970s with MyI Young dressed in polyester orange disco bell pants with the disco hairdo.D  I 	-saw Star Wars Episode 1 and decided it was best to join the "dark side" M early since there are better chances of promotion when you join early into ane new organisation.,  K 	-the real Rob Young was kidnapped, and it is Mike Winkler who impersonates-? him on the net to make it look like he is still alive and well.r   ------------------------------  % Date: Fri, 15 Feb 2002 15:20:25 -0500.- From: JF Mezei <jfmezei.spamnot@videotron.ca>vH Subject: Re: Hello Compaq!  What happens to us VMS users if Itanic dies?, Message-ID: <3C6D6D76.97304354@videotron.ca>   David Froble wrote:oH > Ok, Intel has R&D.  I'd hope so, but with respect to IA-64, it appearsB > that either the R&D isn't too effective, or the R&D is ignored.   I Intel engineers were given a mandate to build one chip that embodied manyeM chips. I am not sure that you can blame the engineers. They were given a veryr  difficult if not impossible job.  L Remember that at the time, the biggest problem was that constructing new FABK plants was becoming way too expensive for each chip. That is one reason whyaK DEC decided to get out of fabbing sicne it couldn't afford to build the new1% fab needed to keep Alpha competitive.@  K I think that where HP/Intel erred is in trying to share the CHIP instead ofjK each having their own chip but sharing the FAB plant. If each had their ownoJ chip, then Intel would have been free to build a 64 bit 8086 that wouldn't% have had as many compromises as IA64.s   ------------------------------    Date: 15 Feb 2002 14:16:57 -0600+ From: young_r@encompasserve.org (Rob Young)-L Subject: Re: Re: Hello Compaq!  What happens to us VMS users if Itanic dies?3 Message-ID: <hcY6pNaxKcV0@eisner.encompasserve.org>r  n In article <tzdb8.43993$Re2.3821424@bin6.nnrp.aus1.giganews.com>, "Bill Todd" <billtodd@metrocast.net> writes:   > L > Over the years you established a clear record of uncritical enthusiasm forK > Alpha's eventual success despite increasing danger signs to the contrary.eK > While it's not clear why you suddenly performed a complete about-face andCN > switched your allegiance to Itanic, it is clear that your enthusiasm remainsL > at least equally uncritical (and hence equally suspect).  'Blind faith' isI > not at all a bad characterization of such an attitude:  one may meet atwM > least the legal definition of blindness while still being able to perceive,e' > dimly, a hand in front of one's face.- >   9 	EV7 in Marvel will be at the top of the performance heapaA 	for quite sometime, but after that no more.  Allegiance?  If you-> 	look at my postings and ruminations over the last 15 years to= 	cov, you'll notice that newer technologies have been pretty s> 	fascinating to me.  IA64 is new and it will be interesting to6 	see what technologies get incorporated as it matures.   				Roba   ------------------------------  % Date: Fri, 15 Feb 2002 15:07:56 -0500a( From: David Froble <davef@tsoft-inc.com>H Subject: Re: Hello Compaq!  What happens to us VMS users if Itanic dies?, Message-ID: <3C6D6A9C.5050300@tsoft-inc.com>   Jose Arroyo wrote:  o > bob@instantwhip.com (Bob Ceculski) wrote in message news:<d7791aa1.0202132026.4857ccd9@posting.google.com>...  > J >>I have been trying to get Compaq to answer this question, along w/HP ...J >>Capellas, Marcellas have been quiet, as has Carly ... how irresponsible!G >>They don't even respond to a serious question posed about where theirbK >>best os customer base goes if itanic dies ... do they restart alpha? doesrN >>intel just develop alpha and forget itanic (this would make the most sense)?G >>We need answers to plan for the future Compaq!  Were is Sue or Terry? + >>No one is providing info or answers here!a >> > G > I think a company should give some explanation toward issues that arepF > living among their clients. In not doing so, what can happen is that< > some clients might loose confident in their manufacturers. > D > How would you feel if, for example, the political party or anotherH > kind of group you belong to, does not keep you informed? Or, how would@ > you feel if you sensed that your questions and concerns remain
 > unanswered?.= > Would you be so willing to give them 'your vote' next time?  > G > I am aware that some things kept 'secret', however too much closenessi2 > may have a more negative effect in the long run. > B > I do not like to bet, yet if I had to, I would bet at least on 21 > sides, for if I would bet all and loose ... :-)D > @ > Personally I would continue the Alpha processor alongside withC > Itanic-processors and made a more final decision once the clientsS) > decide which platform they prefer most.= > F > In 'imposing' a specific platform at best you may gain more clients,C > but then again you may also loose a lot that might step over to a E > totally different operating system. For Intel this may not be a big  > issue, yet for Compaq ...5 > F > The question I pose to Compaq is in fact "why stop the production ofG > the Alpha processor", the only one that remains after VAX was stoppedo > some years ago now?f. > Are there problems with it? Not performing? H > I have seen reports where this processor (Alpha) is ranked at the top.? > Its useful to perform changes when something does not performpC > sufficient, but I am afraid it seems as it is the other way rounde	 > here...r) > Sorry, but this I do not understand...   > D > Could Compaq shine some light on this? Your clients are asking it. >   H The above questions appear to come from confusion, which in itself is a  problem.  H Compaq is NOT stopping production of Alpha CPUs at this time.  The EV7, H which is not even released to customers yet, is expected to have a long H and useful life.  Compaq has stated this many times.  (I won't get into F the issue of Compaq and commitments.)  With process shrinks and other H improvements, EV7 could still be a viable product as fat ahead as 2010.    (Speculation.)  F What has happened is that work on EV8, the follow on to EV7, has been C discontinued at this time.  For all but a few situations, EV7 will m8 supply the needs of VMS customers for a number of years.  H Compaq has decided to move (consolidate) all servers to one architecure G in the future.  They have chosen Intel, and the IA-64.  Some feel that 1I before committing to such a path that IA-64 should prove itself capable. eI   This is some of the debate, the chosing of IA-64 before it is a proven e@ product.  That IA-64 seems to be a lemon makes the issue larger.  B Compaq intends to derive significant cost savings by dropping EV8 ; research at this time.  Ok, a short term business decision.p  G I've maintained, and Compaq has proven me pretty accurate, that Compaq eG doesn't want to be in the chip business, and that they finally found a  D way to exit that business.  I don't look for Compaq to reverse that 	 decision.f  ) What Intel will do should be interesting.o   Dave   -- f4 David Froble                       Tel: 724-529-04504 Dave Froble Enterprises, Inc.      Fax: 724-529-0596> DFE Ultralights, Inc.              E-Mail: davef@tsoft-inc.com6 T-Soft, Inc.  170 Grimplin Road  Vanderbilt, PA  15486   ------------------------------  # Date: Fri, 15 Feb 2002 21:07:12 GMTa* From: "Bill Todd" <billtodd@metrocast.net>H Subject: Re: Hello Compaq!  What happens to us VMS users if Itanic dies?A Message-ID: <4Meb8.51076$eS3.3820470@bin3.nnrp.aus1.giganews.com>.  8 "Rob Young" <young_r@encompasserve.org> wrote in message- news:8vhYiY5TyD1O@eisner.encompasserve.org...sI > In article <RX4b8.45877$eS3.3317925@bin3.nnrp.aus1.giganews.com>, "Bill & Todd" <billtodd@metrocast.net> writes: > >r >l > >   The margins on IA64 willA > >> be much higher than IA32 but I am sure they are making moneya > >> today,r > > I > > Then you're abjectly ignorant:  at the volumes they're selling today,  theyE > > can't even pay the janitorial staff that's keeping the fab clean.e PerhapslH > > if/when they start selling hundreds of thousands of units (though at currentoK > > prices the lack of performance will make this rather difficult), *then*)J > > they'll start breaking even at the margin (not become profitable, mind you,5 > > nor begin paying back the development $billions).  > >v >o; > The cost of manufacturing for Itanium is estimated in theiH > $300 ballpark (without numbers, I believe it is in that neighborhood). >c5 > They are making money on everyone they are selling.   L You might as well offer up the observation that the cost of the sand used inK making each Itanic is under a penny:  it's about as realistic a comparison,e& since it also ignores fixed overheads.  L *If* Itanic starts selling in quantity *then* they'll start making money (ifJ you continue to ignore both already-sunk and on-going development expenses: but do include fixed manufacturing overheads) on each one.   >L > > ...l > >oC > >> The way I see it working is what we see today.  That Dell IA64lD > >> server we keep coming back to is quite a bit cheaper than most. > > K > > That Dell IA64 server you keep coming back to isn't available any more: L > > Dell decided that Itanic did not generate, and had no immediate prospect of7 > > generating, sufficient volume to be worth carrying.h > >  >k. > You are confusing workstations with servers:  K Mea culpa:  that's exactly what I did.  Though that still doesn't alter the-F fact that the configuration you quote is price/performance-competitiveJ neither with an Alpha (unless you jigger the comparison by including a tonK of artificially-costly Compaq memory, which has nothing to do with Itanic'sbI ability to compete on price/performance) nor with Itanic's IA32 brethren.0   - bill   ------------------------------  % Date: Fri, 15 Feb 2002 16:18:32 -0500h- From: JF Mezei <jfmezei.spamnot@videotron.ca>-H Subject: Re: Hello Compaq!  What happens to us VMS users if Itanic dies?, Message-ID: <3C6D7B10.DC54B6A5@videotron.ca>   David Froble wrote: I > Compaq is NOT stopping production of Alpha CPUs at this time.  The EV7, I > which is not even released to customers yet, is expected to have a longe > and useful life.  N We have to wait a few months to find out what HP really intends to do. So far,M it is a state of limbo until the new owners of Compaq make a clear statement.M  G > the issue of Compaq and commitments.)  With process shrinks and othertI > improvements, EV7 could still be a viable product as fat ahead as 2010.l  M Depends on how much IBM and Sun make their chips go. EV7 will definitely have L a longer life against the slow IA64, but not against the true competitors in the performance market.a  I > Compaq has decided to move (consolidate) all servers to one architecure2  M No. Compaq has decided to drop Alpha. It will continue to make wintel servers F on the 8086 platform. Compaq's product line will be far more complex : 	-windows 8086 	-windows ia64 	-Tandem - Mips, 	-Tandem - IA64s 	-Tru64 - Alphal 	-VMS - VAXs
 	-VMS - Alpha  	-VMS - IA64 	-HP UX - PArisc 	-HP UX - IA64 	-linux 8086
 	-linux alphat 	-linux ia64 	-WinCE - Strongarm   " Compared to the current situation: 	-windows 8086 	-tandem - mips  	-tru64 -Alpha 	-vms -Alpha
 	-vms -VAX 	-WinCe - Strongarmh    J The reason I list all those is that because IA64 does not offer sufficientM performance to replace Alpha, HP will have to maintain Alpha for much longer,s0 so its product lineup will be far more complex.   L The solution to this is to prematurely kill the old platforms as part of theN promised product rationalisation. So until HP makes clear statements about itsM product roadmaps once it is "allowed" to do so, it is pointless to speculate.pV We just have to be patient, wait and see what HP will do with what is left of Digital.  N Consider this: if you have more VAX-VMS customers paying support than you haveN IA64-VMS customers, and you need to rationalize the product suite, what do you do ?  C > Compaq intends to derive significant cost savings by dropping EV8s= > research at this time.  Ok, a short term business decision.i  G Since the details of the deal with Intel have not been made public, anyaP speculation on the financial aspects of the Alpha murder are impossible to make.   ------------------------------    Date: 15 Feb 2002 15:32:43 -0600+ From: young_r@encompasserve.org (Rob Young) H Subject: Re: Hello Compaq!  What happens to us VMS users if Itanic dies?3 Message-ID: <W$JbbykSi9VJ@eisner.encompasserve.org>d  \ In article <3C6D6B31.4C8547D3@videotron.ca>, JF Mezei <jfmezei.spamnot@videotron.ca> writes: > Bill Todd wrote:L >> While it's not clear why you suddenly performed a complete about-face and& >> switched your allegiance to Itanic, > N > That is something for the National Equirer or World Weekly news to speculateI > on. If Mr Young does not wish to divulge what event made him switch his 1 > allegiance to the dark side that is his choice.i >   E 	Read what I wrote.  I'm not happy Alpha was deep-sixed.  IA64 is theo  	decision.  Make the most of it.  * > There are many possibilities, including:G > 	-having an affair with Andy Grove's daughter and/or Capellas's wife.c  A 	I'm wondering if the lack of sunlight at certain latitudes wipese. 	out the ability to even remotely be humorous.   				Robt   ------------------------------  % Date: Fri, 15 Feb 2002 15:31:44 -0700t+ From: "Barry Treahy, Jr." <Treahy@mmaz.com>HH Subject: Re: Hello Compaq!  What happens to us VMS users if Itanic dies?' Message-ID: <3C6D8C50.6080903@mmaz.com>s  & --------------0507010506040904050207099 Content-Type: text/plain; charset=us-ascii; format=flowedh Content-Transfer-Encoding: 7bit    Bill Todd wrote:   >>>m; >>The cost of manufacturing for Itanium is estimated in the'H >>$300 ballpark (without numbers, I believe it is in that neighborhood). >>5 >>They are making money on everyone they are selling.> >> > M >You might as well offer up the observation that the cost of the sand used ineL >making each Itanic is under a penny:  it's about as realistic a comparison,' >since it also ignores fixed overheads.l >,M >*If* Itanic starts selling in quantity *then* they'll start making money (ifiK >you continue to ignore both already-sunk and on-going development expenses ; >but do include fixed manufacturing overheads) on each one.  >eB It seems that this has gone around in circles because the general D consensus is that IPF is a nitch processor which means that it will E never hit high-volumes and as my original posting pointed out, there -? appears to be a lot of noise in the back-room regarding an IPF eH replacement that is not part of the same lineage which still leaves the / unanswered question for Compaq, what about VMS?n   Barry,   -- h  @ Barry Treahy, Jr  *  Midwest Microwave  *  Vice President & CIO   A E-mail: Treahy@mmaz.com * Phone: 480/314-1320 * FAX: 480/661-7028m      & --------------050701050604090405020709) Content-Type: text/html; charset=us-asciio Content-Transfer-Encoding: 7bitr   <html> <head> </head>T <body> Bill Todd wrote:W <blockquote type="cite" cite="mid:4Meb8.51076$eS3.3820470@bin3.nnrp.aus1.giganews.com">d   <blockquote type="cite">     <blockquote type="cite">       <pre wrap=""><br></pre>-       </blockquote>        <pre wrap="">The cost of manufacturing for Itanium is estimated in the<br>$300 ballpark (without numbers, I believe it is in that neighborhood).<br><br>They are making money on everyone they are selling.<br></pre>r       </blockquote>t      <pre wrap=""><!----><br>You might as well offer up the observation that the cost of the sand used in<br>making each Itanic is under a penny:  it's about as realistic a comparison,<br>since it also ignores fixed overheads.<br><br>*If* Itanic starts selling in quantity *then* they'll start making money (if<br>you continue to ignore both already-sunk and on-going development expenses<br>but do include fixed manufacturing overheads) on each one.</pre>p       </blockquote>cK It seems that this has gone around in circles because the general consensussP is that IPF is a nitch processor which means that it will never hit high-volumesJ and as my original posting pointed out, there appears to be a lot of noiseJ in the back-room regarding an IPF replacement that is not part of the sameI lineage which still leaves the unanswered question for Compaq, what aboutr VMS?<br>
       <br>	 Barry<br> 
       <br>8       <pre class="moz-signature" cols="$mailwrapcol">--   D Barry Treahy, Jr  *  Midwest Microwave  *  Vice President &amp; CIO    E-mail: <a class="moz-txt-link-abbreviated" href="mailto:Treahy@mmaz.com">Treahy@mmaz.com</a> * Phone: 480/314-1320 * FAX: 480/661-7028</pre> 
       <br>
       </body>-
       </html>   ( --------------050701050604090405020709--   ------------------------------  # Date: Fri, 15 Feb 2002 22:35:09 GMTh* From: "Bill Todd" <billtodd@metrocast.net>H Subject: Re: Hello Compaq!  What happens to us VMS users if Itanic dies?A Message-ID: <x2gb8.47668$LY3.4056703@bin4.nnrp.aus1.giganews.com>s  8 "Rob Young" <young_r@encompasserve.org> wrote in message- news:W$JbbykSi9VJ@eisner.encompasserve.org...e7 > In article <3C6D6B31.4C8547D3@videotron.ca>, JF Mezeiu& <jfmezei.spamnot@videotron.ca> writes:   ...g  , > > There are many possibilities, including:H > > -having an affair with Andy Grove's daughter and/or Capellas's wife. >BB > I'm wondering if the lack of sunlight at certain latitudes wipes/ > out the ability to even remotely be humorous.n  F Well, *I* found it at least a bit humorous.  Though since I don't knowL either of the ladies in question, it's possible that if you do know them you  have good reason to be offended.   - bill   ------------------------------    Date: 16 Feb 2002 05:41:44 +0800, From: Paul Repacholi <prep@prep.synonet.com>H Subject: Re: Hello Compaq!  What happens to us VMS users if Itanic dies?0 Message-ID: <87it8y5so7.fsf@k9.prep.synonet.com>  - young_r@encompasserve.org (Rob Young) writes::  = Do you think you could add the DS10 and DS20 prices so we cans( all enjoy how much cheaper life will be?  @ > http://rcommerce.us.dell.com/rcomm/cat_mini.asp?brand_id=PEDGE >  > Processor   Price  d+ >   Intel Itanium Processor, 733-800 MHz   a> >  with Windows Advanced Server, Limited Edition  $14,996.70   > + >   Intel Itanium Processor, 733-800 MHz   g" >  with RedHat Linux  $16,737.70   >  > Additional processors ~$3800 > W > http://rcommerce.us.dell.com/rcomm/config.asp?order_code=H1054W&conum=70&ConfigType=3  >  > PowerEdge 7150 >  >  Specifications  >    d > TECH SPECS >  e8 >  Up to 4 Intel Itanium Processors at 733MHz and 800MHz >  s >  1GB to 64GB ECC SDRAM memoryw >  e  > 292GB maximum internal storage >  h	 > 				Robe >    --  < Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda. @                                              West Australia 6076. Raw, Cooked or Well-done, it's all half baked.F EPIC, The Architecture of the future, always has been, always will be.   ------------------------------    Date: 15 Feb 2002 19:53:09 -0600+ From: young_r@encompasserve.org (Rob Young)sH Subject: Re: Hello Compaq!  What happens to us VMS users if Itanic dies?3 Message-ID: <YWovXVcFV+77@eisner.encompasserve.org>@  _ In article <87it8y5so7.fsf@k9.prep.synonet.com>, Paul Repacholi <prep@prep.synonet.com> writes:D/ > young_r@encompasserve.org (Rob Young) writes:  > ? > Do you think you could add the DS10 and DS20 prices so we can<* > all enjoy how much cheaper life will be? >   < 	Sure... with this 4 processor quote I'm working up for you,0 	which would you prefer, 32 or 64 GBytes of RAM?   			Rob    A >> http://rcommerce.us.dell.com/rcomm/cat_mini.asp?brand_id=PEDGEr >> q >> Processor   Price  , >>   Intel Itanium Processor, 733-800 MHz   ? >>  with Windows Advanced Server, Limited Edition  $14,996.70  p >> d, >>   Intel Itanium Processor, 733-800 MHz   # >>  with RedHat Linux  $16,737.70    >> e >> Additional processors ~$3800  >> oX >> http://rcommerce.us.dell.com/rcomm/config.asp?order_code=H1054W&conum=70&ConfigType=3 >> n >> PowerEdge 7150  >>   >>  Specifications c >>    
 >> TECH SPECS, >>  9 >>  Up to 4 Intel Itanium Processors at 733MHz and 800MHz  >>    >>  1GB to 64GB ECC SDRAM memory >>  ! >> 292GB maximum internal storaget >>  
 >> 				Rob >>   >  > -- e> > Paul Repacholi                               1 Crescent Rd.,9 > +61 (08) 9257-1001                           Kalamunda.nB >                                              West Australia 60760 > Raw, Cooked or Well-done, it's all half baked.H > EPIC, The Architecture of the future, always has been, always will be.   ------------------------------  # Date: Sat, 16 Feb 2002 03:30:29 GMTe1 From: "David J. Dachtera" <djesys.nospam@fsi.net> H Subject: Re: Hello Compaq!  What happens to us VMS users if Itanic dies?' Message-ID: <3C6DD3A4.F70175C4@fsi.net>    Rob Young wrote: > ] > In article <3C6C8D53.4C3CBCB8@fsi.net>, "David J. Dachtera" <djesys.nospam@fsi.net> writes:t > > Rob Young wrote: > >> [snip]  > >>         Numbers please! > >  > > He doesn't report to you.  > >a& > > (See how stupid that sounds, Rob?) > >s > @ >         But in the context of being responsible for a businessI >         (i.e. the folks that know the numbers... the ones Gartner can'toH >         translate from one piece of paper to another) that was *those*E >         other numbers you are referring to.  The ones we won't see,t2 >         nor should we, competive in nature, etc.  G I'll have to apologize to the group for being so extremely dense that IcE simply canNOT understand how it is possible that trumpeting one's ownrD successes could be detrimental to one's self and/or helpful to one's5 competition, injurious to national security, etc. ...   ( Could someone please explain that to me?  E ...and remember, I'm a low-grade moron. Be sure to use small words soo3 you can be reasonably certain that I'll understand.t   > [snip]# >         So does it sound stupid? a   Did then, still does.f  ( > [snip] Little sore or what?  Sheesh...  
 Oh, I get it.h  F It's alright when *YOU* ask someone to back up a statement; but *I* do it, it's not alright.l  
 I got it now.d  ' Thank you so much for clearing that up.)   -- t David J. Dachteraa dba DJE Systemse http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/u   ------------------------------  % Date: Fri, 15 Feb 2002 23:29:46 -0500d( From: David Froble <davef@tsoft-inc.com>H Subject: Re: Hello Compaq!  What happens to us VMS users if Itanic dies?, Message-ID: <3C6DE03A.3040606@tsoft-inc.com>   Paul Repacholi wrote:i  / > young_r@encompasserve.org (Rob Young) writes:  > ? > Do you think you could add the DS10 and DS20 prices so we cant* > all enjoy how much cheaper life will be?     I'll attempt to fix that.  :-)    @ >>http://rcommerce.us.dell.com/rcomm/cat_mini.asp?brand_id=PEDGE >> >>Processor   Price  m+ >>  Intel Itanium Processor, 733-800 MHz   c> >> with Windows Advanced Server, Limited Edition  $14,996.70      H Lets see, last I heard, a low end DS10 with VMS was in the $8,000 price G range.  No tape drive, but everything else.  Let's say I'm optimistic, s; and place the cost of the DS10 in round numbers at $10,000.l  E So, in my simple math, I'm saving what, -$5,000 with the Dell system?     And I get windoz instead of VMS?    " Gee, thanks for such a great deal.     -- n4 David Froble                       Tel: 724-529-04504 Dave Froble Enterprises, Inc.      Fax: 724-529-0596> DFE Ultralights, Inc.              E-Mail: davef@tsoft-inc.com6 T-Soft, Inc.  170 Grimplin Road  Vanderbilt, PA  15486   ------------------------------  % Date: Fri, 15 Feb 2002 13:39:59 -0600 ! From: Dan Moore <dmoore@sosu.edu>n4 Subject: Re: How to become a Hobbyist (the hard way)( Message-ID: <3C6D640F.68AFED79@sosu.edu>   That was great! Thanks for sharing. I picked up a stack of 4 vaxes for $10 at the Sale in Dallas, TX a while back, I had some similar experiences getting VMS going.   Dank   Mark Redding wrote:r  J > I've recently installed Hobbyist VMS 7.2 onto my VAXstation 4000 VLC ... >   > http://www.aragorn.org.uk/vax/ >rE > ...Is my diary of events in the hope that it may be of use to othera > struggling Hobbyists.o >n > Mark.t   ------------------------------  % Date: Fri, 15 Feb 2002 18:46:10 -0500a1 From: "Island (hpaq.net)" <dbturner@islandco.com>g7 Subject: Re: How to get prices on CPQ OpenVMS products? / Message-ID: <u6r7dmpjkkb325@news.supernews.com>)  @ List prices etc. Call or email us and we'll be happy to tell you  + (And we'll send a quote with discount too!)i   :0)i   -- Island Computers US Corp.h 2700 Gregory Streetn Savannah GA 31404s Tel: 877 636 4332h International: 001 912 447 6622n  Facsimile:      001 912 201 0096 dbturner@hpaq.netc www.hpaq.net  G Koska, John C. (LNG-MBC) <John.C.Koska@lexisnexis.com> wrote in messagevG news:3D35AD137AAAD411A6BA0008C7B1B12D016027A8@MBCALBEXC03.BENDER.COM...a > > -----Original Message-----@ > > From: wspencer@ap.nospam.org [mailto:wspencer@ap.nospam.org]- > > Sent: Thursday, February 14, 2002 1:09 PM  > > To: Info-VAX@Mvb.Saic.Com>; > > Subject: RE: How to get prices on CPQ OpenVMS products?6 > >  > >rC > > John.C.Koska@lexisnexis.com (Koska, John C. (LNG-MBC)) wrote in F > > <3D35AD137AAAD411A6BA0008C7B1B12D016027A7@MBCALBEXC03.BENDER.COM>: > > ? > > >Try http://www.openvms.compaq.com/swcat/us/cocplsovms.html , > > >for pricing on C++ license for OpenVMS. > >b% > > Thanks John - just what I needed.  > >s< > > Also, I'd like to ask a more fundamental question - what > > phone number do us; > > "corporate" guys phone to talk to CPQ salespeople aboutt > > OpenVMS products?cE > > There seemed nothing appropriate a www.compaq.com / "Contact Us".i >u8 > I have Compaq Inside Enterprise person, who I reach atC > 1-800-277-8988.  But if you do not have an assigned customer carec? > person, then it is 1-800-282-6672.  They will query you as to-D > your location, name, bussiness, etc, and get you to proper supportD > person.  You should inquire as to who your customer care person isC > and to their phone extension, so you can reach them at the 1-800-5 > 277-8988 number. > D > http://www.compaq.com/corporate/overview/world_offices.html  shows$ > most of the contact phone numbers. >  > For U.S.:u >m? > Pre-sales Product Information: 1-800-AT-Compaq (800-282-6672)v >oE > Product Service & Technical Support: 1-800-OK-Compaq (800-652-6672)t > Main Operator:  281-370-0670 > Main Fax:       281-514-1740 >a > :) jck > John Koska > Matthew Bender & Co., Inc. -$ >   A Member of the LexisNexis Group > 1275 Broadway  > Albany, NY  12204d > USAf > 518-487-3255 > John.C.Koska@lexisnexis.comn >u+ > I post personal opinion only, and all thev, > disclaimers one could imagine apply.  That* > includes, I speak for myself only and my+ > views in no way represent my employer(s).o- > One should also take note of the Electronic2+ > Communications Privacy Act of 1986, whichw- > imposes civil and criminal liability on any8* > person who intentionally intercepts "any* > wire, oral or electronic communication."   ------------------------------  % Date: Fri, 15 Feb 2002 18:50:48 -0500-1 From: "Island (hpaq.net)" <dbturner@islandco.com>U7 Subject: Re: In stock: ES40 Model 2 USD7895 VMS Ready !i/ Message-ID: <u6r7mjhcj984d5@news.supernews.com>t  H For what it's worth, I am getting 5 brand new (well.. unused) systems in  within the next couiple of weeks  K They won't be as cheap as the ultimate workstations, but THEY WILL RUN VMS!rG Of course, you'll need a license AND and SMP license for the second CPU     K As for price - I would rather send a quote after having discussed disks etce   Davidk    5 Peter LANGSTOEGER <eplan@kapsch.net> wrote in messageo/ news:IKY98.590820$5G5.4247123@news.chello.at...oH > In article <3c6854a7.35523447@news.charter.net>, ben_myers@charter.net (Ben Myers) writes: A > >"just my 0.02"  What currency was that?  Euros?  ... Ben Myerso >i, > Yup. Are you green of envy cause of that ? >  > --> > Peter "EPLAN" LANGSTOEGER           Tel.    +43 1 81111 2651= > Network and OpenVMS system manager  Fax.    +43 1 81111 888p> > KAPSCH AG      Wagenseilgasse 1     E-mail  eplan@kapsch.netJ > A-1121 VIENNA  AUSTRIA              "I'm not a pessimist, I'm a realist"   ------------------------------  # Date: Fri, 15 Feb 2002 22:26:17 GMTo* From: Jose Arroyo <jose.arroyo@pandora.be> Subject: Re: Itanium troubles-* Message-ID: <3C6D8AE1.60B59088@pandora.be>  & --------------8AB876BECB82CE8D0F0CF412* Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bitu  - Bob, as you said it, they don't get it (yet).i? Sometimes one has to hit hard a wall before changing direction.o- This is exactly what most likely will happen.wJ And when one hits hard a wall one will become aware of precious time lost.W Maybe it is time that some people burn their fingers (figurative speaking ofcourse) :-)h     Bob Ceculski wrote:r  c > JF Mezei <jfmezei.spamnot@videotron.ca> wrote in message news:<3C643902.20BD2144@videotron.ca>...iR > > Interesting that Alpha was seen as DEC's saviour.  The chip that would propell > > DEC back to the front. > >-Q > > But for IA64, there is no such predictions. Intel is already at the front andi& > > may in fact hurt itself with IA64. > I > I think Intel should just produce Alphas and pick up on EV8 and go fromkF > there ... VMS and tru64 should remain w/hp unix customers eventuallyI > ported to tru64 ... this would save everyone time and money and the top,N > 2 64 bit platforms vms and tru64 would continue to dominate the high end ...M > of course common sense never seems to win in braindead windoze pc companiesDL > like compaq and hp, so don't count on it ... I'm sure hp unix could easilyK > be ported to alpha ... but anyone who thinks pc's are going to be a moneydJ > maker in the future unfortunately doesn't understand what's going on ...M > as pc sales continue to drop as home users move to smart cable boxes, theretL > will remian only business to sell to, and they are also trying to get awayN > from end user pc costs w/thin clients ... high end vms (I hope) and unix andI > linux boxes will drive internet/intranet apps and give home users theircN > freedom from buying pc's and messing around w/software and hardware ... homeK > users want what's coming, a wireless keyboard/remote control unit to surfnL > the web on their 50 digital flat wall tv's and send/recv mail from it, andL > download movies on demand, and have their disk space provided to them likeO > aol or any other service provider does now ... businesses want to consolidateiO > their 80,000 windoze boxes into several vms galaxies or  unix (gag!) or linux J > mainframes w/sans and use thin clients to access apps ... pc's are dead!K > Unfortunately, Capellas doesn't understand this, and I suggest anyone who Q > owns compaq/hp stock sell after merger or no merger, because they will fail ...   & --------------8AB876BECB82CE8D0F0CF412) Content-Type: text/html; charset=us-ascii: Content-Transfer-Encoding: 7bitm  > <!doctype html public "-//w3c//dtd html 4.0 transitional//en"> <html>- Bob, as you said it, they don't get it (yet).iJ <br><i>Sometimes one has to hit hard a wall before changing direction.</i>1 <br>This is exactly what most likely will happen.uF <br><i>And when one hits hard a wall one will become aware of precious time lost.</i>G <br><i>Maybe it is time that some people burn their fingers (figurativef speaking ofcourse) :-)</i>
 <br>&nbsp; <p>Bob Ceculski wrote:F <blockquote TYPE=CITE>JF Mezei &lt;jfmezei.spamnot@videotron.ca> wrote6 in message news:&lt;3C643902.20BD2144@videotron.ca>...F <br>> Interesting that Alpha was seen as DEC's saviour.&nbsp; The chip that would propell <br>> DEC back to the front. <br>>bI <br>> But for IA64, there is no such predictions. Intel is already at then	 front andi( <br>> may in fact hurt itself with IA64.J <p>I think Intel should just produce Alphas and pick up on EV8 and go fromH <br>there ... VMS and tru64 should remain w/hp unix customers eventuallyG <br>ported to tru64 ... this would save everyone time and money and the  top/H <br>2 64 bit platforms vms and tru64 would continue to dominate the high end ...sO <br>of course common sense never seems to win in braindead windoze pc companiesvG <br>like compaq and hp, so don't count on it ... I'm sure hp unix could  easilyG <br>be ported to alpha ... but anyone who thinks pc's are going to be aa moneyeH <br>maker in the future unfortunately doesn't understand what's going on ...>I <br>as pc sales continue to drop as home users move to smart cable boxes,= thererI <br>will remian only business to sell to, and they are also trying to getp awayG <br>from end user pc costs w/thin clients ... high end vms (I hope) andc unix andK <br>linux boxes will drive internet/intranet apps and give home users their G <br>freedom from buying pc's and messing around w/software and hardware  ... homeH <br>users want what's coming, a wireless keyboard/remote control unit to surfF <br>the web on their 50 digital flat wall tv's and send/recv mail from it, andeI <br>download movies on demand, and have their disk space provided to thema likeQ <br>aol or any other service provider does now ... businesses want to consolidate@F <br>their 80,000 windoze boxes into several vms galaxies or&nbsp; unix (gag!) or linuxjF <br>mainframes w/sans and use thin clients to access apps ... pc's are dead!sI <br>Unfortunately, Capellas doesn't understand this, and I suggest anyonel whowJ <br>owns compaq/hp stock sell after merger or no merger, because they will fail ...</blockquote>e </html>e  ( --------------8AB876BECB82CE8D0F0CF412--   ------------------------------    Date: 16 Feb 2002 01:06:39 +0100* From: eplan@kapsch.net (Peter LANGSTOEGER)? Subject: Re: JDK 1.3.1 past due (feature not found in the beta)-* Message-ID: <3c6da28f$1@news.kapsch.co.at>  W In article <3C67AE16.95735256@theblakes.com>, Colin Blake <colin@theblakes.com> writes:h >Peter LANGSTOEGER wrote:if >> In article <Vpj9GTHgb$qR@eisner.encompasserve.org>, koehler@encompasserve.org (Bob Koehler) writes: >> >I >> >   "late 2001" having come and gone, anyone have info on when we will / >> >   get a JDK for VMS that supports mozilla?r >>L >> Sorry, no. Personally, I did expect the Plug-In to come with the official; >> V1.3.1-1 but alas, it was only a guess, and I was wrong.w >d >Incorrect.,  J So I stand corrected (happens much too often the last months, I'm afraid).  g >V1.3.1-1 DOES ship the OJI support that allows Mozilla to run Java applets. You need the final kit (itc >wasn't in any of the betas).H  ; So, one has to tell it to the webmaster of the Q JAVA pagesi  / 	http://www.compaq.com/java/download/index.htmlI  E where the Plug-In (which I assume is OJI) is only mentioned for Tru64RF and I also have to find out, why my MOZILLA 0.9.8 doesn't do JAVA with> the final JAVA V1.3.1-1 I've installed a couple of days ago...   --  < Peter "EPLAN" LANGSTOEGER           Tel.    +43 1 81111 2651; Network and OpenVMS system manager  Fax.    +43 1 81111 888 < KAPSCH AG      Wagenseilgasse 1     E-mail  eplan@kapsch.netH A-1121 VIENNA  AUSTRIA              "I'm not a pessimist, I'm a realist"   ------------------------------    Date: 15 Feb 2002 13:48:16 -08001 From: keithparris_NOSPAM@yahoo.com (Keith Parris)m8 Subject: Re: Load Balancing DNA across 2*2MBit WAN links= Message-ID: <cf15391e.0202151348.74685dfd@posting.google.com>7  r adrian_ogden@hotmail.com (Scumbag Adie) wrote in message news:<91445c94.0202141520.1cd064c1@posting.google.com>...B > If both ends of the link have three nodes and I'm only using the= > link to shadow one member of a shadow set would the cluster B > performance at one end depend on the performance of the link (orD > be affected adversely) assumming no shadow copies are in progress?  F You spoke of a 200km inter-site distance.  Assuming the actual circuitA path length wasn't any longer than that (although I've seen casesiF where it was twice the highway driving distance between sites), you'reF looking at perhaps a 2-3 millisecond round trip time between sites dueF to the speed of light over that distance.  (If a circuit is already inE place, an IP Ping operation (or a DECnet LOOP NODE done say 100 times>A with some DCL code wrapped around it to time it) to a node at the ; opposite site will give you the actual round-trip latency.)a  E Any writes to a cross-site shadowset will incur a penalty of at leastmB 2 round trips, plus maybe 1 millisecond going through the VMS MSCPE Server on the other end.  (To put this into perspective, a non-cached1; disk write of 1 block with a modern disk might take say 5-7 B milliseconds; a 1-block write to an HSx controller with write-back  cache might take 1 millisecond.)  ? For regular locking activity, based on what you've said, you'll D probably not incur any penalty, since if a given application is onlyD running at one site at a time, the lock master nodes for applicationD lock trees will tend to be chosen from among the local nodes, and so* all lock requests will stay within a site.  D But note that one-half of your lock directory lookups will incur theE inter-site penalty of 1 round trip, because the mapping for directory E lookups includes all nodes based on their LOCKDIRWT values and so, onaF the average, half of the time you'll be hitting a node at the oppositeE site for this.  (Note that directory lookups are only needed when you C acquire the first lock for a given lock tree on a given node; so if ? you're just converting a lock between modes, you still know thetD lock-master node and don't have to do a directory lookup.  But sinceE locking behavior is typically buried deep in the code of applicationshC and RMS, you tend to have little control over how locking is done.)d  F For things like the queue manager, you may want to specify that it run8 on a node local to where the most activity is occurring.  A > One end would be a standby for the other and not doing any workaA > until the other end had its applications shutdown and restartedo> > at the standby. I'm trying here to understand what bandwidth@ > would be used by the actual shadowing v's that needed by other > cluster processes.  ? To try to estimate the amount of inter-site traffic from normalsE activity, we'd need to know the size of machines involved and an idea C of their I/O activity (particularly the volume of I/O activity, andg, the read/write ratios) and locking activity.  B The criteria I normally advise folks to use in choosing inter-site= links is to provide enough bandwidth to do a full shadow-copy-F operation in a reasonable amount of time, like 8 hours or so, to allowB a recovery of data disaster tolerance in a timely fashion.  If youF meet that requirement, then normally a full-copy operation is probably> the only time you're likely to see full utilization of all theC available inter-site bandwidth.  But then I've been working with E30A class links (actually DS-3/T3 at 45Mb) instead of E1 class links.   F You could do a test by taking one of the 3 nodes at a site, linking it; to the other 2 with only a LAN link, and then measuring thenF utilization on that LAN during normal activity.  If you created shadowC sets on that one node also, you could measure shadow-copy times and @ LAN utilization for that.  The only factor you couldn't simulateF without some trouble is the inter-site latency (although in your case,A 200km of fiber might not actually be cost-prohibitive for testings
 purposes).? ---------------------------------------------------------------?? Keith Parris | parris at encompasserve dot org | Consulting on: > Clusters, Disaster Tolerance, Performance, I/O, Storage & SANs   ------------------------------  % Date: Fri, 15 Feb 2002 12:02:51 -0700v$ From: Lee Y T Mah <lytmah@cha.ab.ca>A Subject: Re: Longest VMS _cluster_ uptime, was: Re: 1 Year uptime-) Message-ID: <3C6D5B5A.5EA21AA8@cha.ab.ca>   7 Thanks for your post.  FYI, here is our cluster uptime.o  C     This cluster has been operational since 31-JAN-1993 04:12:54.36 F     and this system has been operational since 20-SEP-2001 17:02:28.00     "Mark D. Jilson" wrote:X  G > Yep, if you don't have a cluster then CLUSTER_FTIME = 0 = 17-NOV-1858  >k' > This might be better for folks to usel > . > $ IF F$GETSYI("CLUSTER_MEMBER") .EQS. "TRUE" > $ THEN WRITE SYS$OUTPUT -gF > "This cluster has been operational since ",F$GETSYI("CLUSTER_FTIME") > $      WRITE SYS$OUTPUT -bD > "and this system has been operational since ",F$GETSYI("BOOTTIME") > $ ELSE WRITE SYS$OUTPUT - @ > "This system has been operational since ",F$GETSYI("BOOTTIME")	 > $ ENDIFr >s > Oswald Knoppers wrote: > >s > > "Mark D. Jilson" wrote:s > > >s2 > > > $ WRITE SYS$OUTPUT F$GETSYI("CLUSTER_FTIME") > > >p > >  > > Ok then, i hold the record:t > >h0 > > $ WRITE SYS$OUTPUT F$GETSYI("CLUSTER_FTIME") > > 17-NOV-1858 00:00:00.00n > > $n > > % > > But then this is not a cluster...a > >1
 > > Oswald >  > --H > Jilly   - Working from Home in the Chemung River Valley - Lockwood, NYL >         - jilly@clarityconnect.com                      - Brett Bodine fanL >         - Mark.Jilson@Compaq.com                        - since 1975 or so5 >         - http://www.jilly.baka.com               -w   -- Leen  ; Lee Y T Mah                        Capital Health Authority ? Email: lytmah@cha.ab.ca            Information Systems, RAH CSC 4 Phone:  (780) 477-4725, 477-4233   10240 Kingsway NW? Fax:      (780) 491-5119, 491-5619    Edmonton, AB, CAN  T5H3V97   ------------------------------    Date: 15 Feb 2002 12:02:25 -0800< From: alphaman-nixspam@hsv.sungardtrust.com (Aaron Sakovich)  Subject: OpenVMS Times -- RIP???= Message-ID: <8af17fe1.0202151202.47d82316@posting.google.com>   C Just read the latest couple of SKC issues.  According to Terry, theuD OpenVMS Times is going the way of so many other things OpenVMS.  TheA newsletter will get a name change and be de-focused from OpenVMS.w  7  "Consistent with Compaq's decision to merge all thingsE3  Alpha, Tru64 UNIX, OpenVMS, and HPTC under VP RichB4  Marcello's High Performance Systems Group, the firm6  elected to re-invent the 'OpenVMS Times' as the 'High  Performance Systems Times'."s  C We can now go back to scouring publications for sidebars on page 56rF that mention OpenVMS in a footnote...  *sigh*  Remember how excited weD used to get when we found OpenVMS mentioned in such a rag!?!  What a
 thrill... ;^Po  6 Dilute and obfuscate seems to be Compaq's OpenVMS way.  ? It sure was nice while it lasted.  I'll miss having a dedicatedc OpenVMS newsletter.A  
 "This sucks."- Aaron    ------------------------------  + Date: Fri, 15 Feb 2002 21:08:46 +0100 (MET)n9 From: Phillip Helbig <HELBPHI@sysdev.deutsche-boerse.com>a$ Subject: Re: OpenVMS Times -- RIP???; Message-ID: <01KEBFAB0ZEG8ZKA1M@sysdev.deutsche-boerse.com>2  8 > "Consistent with Compaq's decision to merge all things4 > Alpha, Tru64 UNIX, OpenVMS, and HPTC under VP Rich5 > Marcello's High Performance Systems Group, the firmg7 > elected to re-invent the 'OpenVMS Times' as the 'High  > Performance Systems Times'."  E Which, of course, we will appropriately abbreviate to "High Times".  nI After all, how often, after hearing the latest announcement from Compaq, y? has your reaction been "what's THAT dude been smoking', man?!".l   ------------------------------  % Date: Fri, 15 Feb 2002 15:23:02 -0500r- From: JF Mezei <jfmezei.spamnot@videotron.ca>h$ Subject: Re: OpenVMS Times -- RIP???, Message-ID: <3C6D6E13.7D8E76AE@videotron.ca>   Aaron Sakovich wrote:r9 >  "Consistent with Compaq's decision to merge all thingsa5 >  Alpha, Tru64 UNIX, OpenVMS, and HPTC under VP Richn6 >  Marcello's High Performance Systems Group, the firm8 >  elected to re-invent the 'OpenVMS Times' as the 'High >  Performance Systems Times'."d  K And before too long, that publication will be merged into "Inform" and willeM feature wintel stuff. So, what's new ? Nobody should be surprised about this.cM This is HP taking control of Compaq and many more such decisions will be maden in the next 12 months.  D On TV, Carly only mentioned keeping Compaq's enterprise PC business.   ------------------------------  % Date: Fri, 15 Feb 2002 22:34:32 -0500A2 From: rdeininger@mindspring.com (Robert Deininger)$ Subject: Re: OpenVMS Times -- RIP???K Message-ID: <rdeininger-1502022234320001@1cust197.tnt3.nashua.nh.da.uu.net>1  = In article <8af17fe1.0202151202.47d82316@posting.google.com>,h= alphaman-nixspam@hsv.sungardtrust.com (Aaron Sakovich) wrote:t  D >Just read the latest couple of SKC issues.  According to Terry, theE >OpenVMS Times is going the way of so many other things OpenVMS.  The:B >newsletter will get a name change and be de-focused from OpenVMS. >-8 > "Consistent with Compaq's decision to merge all things4 > Alpha, Tru64 UNIX, OpenVMS, and HPTC under VP Rich5 > Marcello's High Performance Systems Group, the firmo7 > elected to re-invent the 'OpenVMS Times' as the 'Highi > Performance Systems Times'."  E I believe Terry's information is not accurate in this case.  It is my > understanding that the OpenVMS Times newsletter will continue.   ------------------------------  % Date: Fri, 15 Feb 2002 20:00:28 -0000 ; From: "Leigh G. Bowden" <LGBowden@bowdenfamily.fsnet.co.uk>n Subject: Oracle problem./ Message-ID: <a4jqbs$6l8$1@newsg4.svr.pol.co.uk>e  K I posted a problem here a few weeks ago about Oracle 7 on OpenVMS 6.2 on an J AS2100 with 2 CPU's and 1GB of RAM which periodically locked up. I finally got to the bottom of this thus   SDA> show pool/page/full  K showed that the LNM entry was incresing at about 5% per day and PAGEDYN was  doing the same - obviously.i   >show log/struct/out=t.t >search t.t tns_/noout/stath  B Revealed about 95% of the logical spaces (is it called that?) wereL TNS_<number>. The original supplier of the machine applied the fix today andF the system hasn't lost anymore memory and has actually reclaimed some.   ------------------------------  # Date: Sat, 16 Feb 2002 04:13:51 GMTe1 From: "David J. Dachtera" <djesys.nospam@fsi.net> $ Subject: Re: OT: "Crashless Windows"& Message-ID: <3C6DDDCF.F10B378@fsi.net>   Bill Gunshannon wrote: > ) > In article <3C6C897C.7FA7A450@fsi.net>,s6 >  "David J. Dachtera" <djesys.nospam@fsi.net> writes: > |>1 > |> Better not try that in the healthcare field!d > |> > E > I hate to burst your bubble, but how many hospitals still run VMS??e  F AFAIK, all three o fthe major healthcare organizations here in Chicago (NMH, Advocate, U of I).   > How many run Windows??  D Hhmmm. Good point. i better find out and make 100% certain that I amA *NEVER* treated at any facility that trusts it patients' lives toa< Micro$hit crapware without a VMS back-end guarding the data!  9 > Looks to me like they not only tried it, they liked it.oE > (I only visit any hospital once every couple of years.  My last twohD > visits were a number of years ago to have my knees scoped one year@ > apart for each.  When I went in for the first, they were still@ > running VMS.  When I went in a year later for the second, theyE > were running NT 4.0.  Both trips went completely without surprises.oJ > As much as people here dislike MS junk, it is time for a reality check.)  H That reality check *WILL* come - probably beginning in the local circuitE court. It's only a matter of time until someone claims an injury or anD loved one's death was due to a Wintel computer that crashed during aC critical phase of someone's care. Heaven help the poor S.O.B. who's> gotta defend that case!    -- w David J. Dachterai dba DJE Systemss http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/p   ------------------------------    Date: 16 Feb 2002 03:37:49 +0100* From: eplan@kapsch.net (Peter LANGSTOEGER)/ Subject: Re: OT: Proliant DL and DS-10 cabinets>( Message-ID: <3c6dc5fd@news.kapsch.co.at>  W In article <14FEB200222521817@gerg.tamu.edu>, carl@gerg.tamu.edu (Carl Perkins) writes:pt >In article <20020214183741.15798.qmail@web20209.mail.yahoo.com>, Fabio Cardoso <fabiopenvms@yahoo.com.br> writes...4 >}Just an personal opinion to the engineering people- >}at Compaq which develops these 1U servers..p >} e5 >}We are instaling a Beowulf cluster with 32 ProliantM. >}DL servers and what mess of cables....power, >}network and storage (FC).  >} h. >}Why dont Compaq use a "backplane" as used by' >}the DNPG products - ex-DEC products ?4 >} d7 >}These DL and DS server should have the possibility too1 >}install a backplane. Each 5, 10 for example....p >o? >Perhaps you should have gone with the new BL series Proliants.r: >They are the new "Blade" type system which fits 20 sinlge: >processor (700MHz at the moment) systems into a single 3U >rack mount enclosure.  . Are they Alphas ? Can we run OpenVMS on them ?" If not, then I'm not interested...   -- v< Peter "EPLAN" LANGSTOEGER           Tel.    +43 1 81111 2651; Network and OpenVMS system manager  Fax.    +43 1 81111 888m< KAPSCH AG      Wagenseilgasse 1     E-mail  eplan@kapsch.netH A-1121 VIENNA  AUSTRIA              "I'm not a pessimist, I'm a realist"   ------------------------------   Date: 15 Feb 2002 23:01 CST ' From: carl@gerg.tamu.edu (Carl Perkins)s/ Subject: Re: OT: Proliant DL and DS-10 cabinetse- Message-ID: <15FEB200223014198@gerg.tamu.edu>    eplan@kapsch.net writes...X }In article <14FEB200222521817@gerg.tamu.edu>, carl@gerg.tamu.edu (Carl Perkins) writes:u }>In article <20020214183741.15798.qmail@web20209.mail.yahoo.com>, Fabio Cardoso <fabiopenvms@yahoo.com.br> writes...05 }>}Just an personal opinion to the engineering people . }>}at Compaq which develops these 1U servers.. }>} 6 }>}We are instaling a Beowulf cluster with 32 Proliant/ }>}DL servers and what mess of cables....power,g }>}network and storage (FC). }>} / }>}Why dont Compaq use a "backplane" as used byt( }>}the DNPG products - ex-DEC products ? }>} 8 }>}These DL and DS server should have the possibility to2 }>}install a backplane. Each 5, 10 for example.... }>@ }>Perhaps you should have gone with the new BL series Proliants.; }>They are the new "Blade" type system which fits 20 sinlge$; }>processor (700MHz at the moment) systems into a single 3Ue }>rack mount enclosure.n } / }Are they Alphas ? Can we run OpenVMS on them ? # }If not, then I'm not interested...c }  }-- = }Peter "EPLAN" LANGSTOEGER           Tel.    +43 1 81111 2651>   Of course not.  > Neither are the systems (Proliant DL series) he said they were using.  ? On the other hand, the "next next" generation of Alpha systems  = (i.e. the one after Marvel) will probably include something al? lot like them - or, at least, a lot like the way "them" will beo when that time comes.p   --- Carl   ------------------------------    Date: 15 Feb 2002 20:07:41 -0600- From: Kilgallen@SpamCop.net (Larry Kilgallen)Y@ Subject: re: Reading OPCOM messages (was: Creating a mailbox...)3 Message-ID: <lXNBKeG7Zzbc@eisner.encompasserve.org>e  R In article <200202152343.VAA27413@orwell.uol.com.br>, valdemir-@uol.com.br writes:9 > I=B4m having problems to create a mailbox in Pascal.=20 = > I need create a mailbox in Pascal to get all Opcom messagesa* > send to operator=B4s console. My doubts:? > - How create a mailbox in Pascal and enable this mailbox with4K >   Operator privilege to catch all messages send to Operator=B4s console ?   B I think you really need to create a pseudo-terminal by calling the PTD$ routines.  , > - How read data buffered in this mailbox ?  D Having Created the pseudo-terminal, open it on a channel and provide9 the REPLY/ENABLE command.  You will have to log in first.:  8 > - How clean up old messages buffered in this mailbox ?  @ Read from the channel on which the terminal is opened.  The onlyE data you receive on the pseudo-terminal should be broadcast messages.1   ------------------------------  + Date: Fri, 15 Feb 2002 16:19:26 -0600 (CST)' From: sms@antinode.org> Subject: Re: Security for VMS systems visable to the  Internet) Message-ID: <02021516192628@antinode.org>e  1 From: bill@triangle.cs.uofs.edu (Bill Gunshannon)n. > In article <3C6C400C.9090808@tsoft-inc.com>,- >  David Froble <davef@tsoft-inc.com> writes:u > [...]tL > |> Yep, there appears to be those with nothing better to do than probe IP  > |> addresses.  w > J > Get used to it.  They have a purpose and they are very persistent.  TheyJ > are warez vendors looking for the next sucker^H^H^H^H^H^H open anonymousK > ftp site so they can upload all their illegal junk and then advertise it.   H    We have a WU-FTP server (on Solaris) at work with a publicly writableC directory for use by our customers, and it was a regular target foraH these FTP scripts.  (We block downloads from the upload directory, so weF were not a useful site for the clowns, but they did frequently depositH test files, and then learn that they were not accessible.)  The easy fixG there was to change the name of the upload directory from "incoming" tocF anything else.  The typical script is pretty narrow-minded.  I haven't seen any junk come in since.  A    At home, with VMS and TCPIP, my FTP server supplies VMS-formatnF directory listings, and the scripts (like MSIE) can't cope.  It's alsoB tough to create a directory named ".020214225837p" on an ODS2 fileF system.  I get a lot of log entries like the following, but it doesn't
 hurt much:  G 13-FEB-2002 17:01:21.97 User:anonymous logged in ident:Kgpuser@home.com   from Host:209.27.156.194p= 13-FEB-2002 17:01:25.32 User:anonymous ident:Kgpuser@home.comk2  status:00010001 CWD dir:SYS$SYSDEVICE:[ANONYMOUS]H 13-FEB-2002 17:01:26.57 User:anonymous ident:Kgpuser@home.com logged out  H ------------------------------------------------------------------------  C    Steven M. Schweda               (+1) 651-699-9818  (voice, home) C    382 South Warwick Street        (+1) 763-781-0308  (voice, work)hG    Saint Paul  MN  55105-2547      (+1) 763-781-0309  (facsimile, work) 9    sms@antinode.org                sms@provis.com  (work)p   ------------------------------    Date: 15 Feb 2002 17:16:38 -0600- From: Kilgallen@SpamCop.net (Larry Kilgallen),> Subject: Re: Security for VMS systems visable to the  Internet3 Message-ID: <P4DQU0KJmlyA@eisner.encompasserve.org>a  B In article <02021516192628@antinode.org>, sms@antinode.org writes:3 > From: bill@triangle.cs.uofs.edu (Bill Gunshannon)e/ >> In article <3C6C400C.9090808@tsoft-inc.com>,w. >>  David Froble <davef@tsoft-inc.com> writes: >> [...]M >> |> Yep, there appears to be those with nothing better to do than probe IP t >> |> addresses.   >> cK >> Get used to it.  They have a purpose and they are very persistent.  TheyrK >> are warez vendors looking for the next sucker^H^H^H^H^H^H open anonymousrL >> ftp site so they can upload all their illegal junk and then advertise it. > J >    We have a WU-FTP server (on Solaris) at work with a publicly writableE > directory for use by our customers, and it was a regular target foraJ > these FTP scripts.  (We block downloads from the upload directory, so weH > were not a useful site for the clowns, but they did frequently deposit< > test files, and then learn that they were not accessible.)  @ Another possibility would be to make the default file protection? unreadable and then have a background process to make new files 5 readable only after considerable binary modification.    ------------------------------  # Date: Fri, 15 Feb 2002 23:38:07 GMTn6 From: "Andy Bustamante" <a_c_bustamante@earthlink.net>> Subject: Re: Security for VMS systems visable to the  InternetF Message-ID: <zZgb8.12361$P21.1159740@newsread1.prod.itd.earthlink.net>  H That's life on the Wild Wild Web.  As noted you can restrict services toH known hosts or networks.  I've had a customers run Alphas as web serversL without a firewall.  I locked down Telnet and FTP to known sites for supportH and ran without incident.  Eventually they installed firewalls for other reasons than Alpha security.  D Our FTP server (Alpha 800) sits behind a fiewall which filters otherH traffic, but still gets Anonymous (no such user) attempts and "guess theJ ID/password" connections.  I use a AMD 300 with Red Hat to firewall my DSLF line.  Lots of probes there, especailly since the IIS worms started to spawn.   --   Andy Bustamante ( remove the ascii-95's to reply by e-mail      5 "David Froble" <davef@tsoft-inc.com> wrote in messagea& news:3C6C400C.9090808@tsoft-inc.com...I > Well, I've read about security, and I've talked about security, but I'mE= > pretty sure that it's what I don't know that could hurt me.s > A > I've recently made one of my VMS systems visable to my internetWI > connection, via settings in my router.  It's a rather plain venella VMSCJ > V7.2-1 system with TCPIP V5.0A, DECnet IV, etc.  Since doing this, I see$ > the following several times a day. > : > %%%%%%%%%%%  OPCOM  14-FEB-2002 17:28:39.08  %%%%%%%%%%%$ > Message from user INTERnet on JODII > INTERnet ACP FTP Accept Request from Host: 209.127.72.194   Port: 49637  >sH > Yep, there appears to be those with nothing better to do than probe IPG > addresses.  I haven't yet been able to identify the domain for any ofRJ > the IP addresses I've seen.  Most probes are FTP connect requests, but I > did see one TELNET attempt.a > I > I've got only a few user accounts, some with privs, and others without.oJ >   It's my feeling that reasonable username and password data, along withF > break-in detection and countermeasures, should be adequate security.G > There's nothing important on the system, but it could be a gateway to G > other systems (once someone's in) that are a bit more valuable to me.  >iB > Keeping up with the hackers and their techniques isn't one of myD > activities.I do development work.  Security has never been a dailyG > activity for me.  It's the things that I'm not aware of that will get- > me, if anything does.2 >:F > I don't think I really need to lock the system down in ways that areG > available to VMS.  If I thought that, I wouldn't have allowed outside1H > access in the first place.  Any suggestions on things that I should be > considering? >u > Dave >i > --6 > David Froble                       Tel: 724-529-04506 > Dave Froble Enterprises, Inc.      Fax: 724-529-0596@ > DFE Ultralights, Inc.              E-Mail: davef@tsoft-inc.com8 > T-Soft, Inc.  170 Grimplin Road  Vanderbilt, PA  15486 >r >l   ------------------------------  % Date: Sat, 16 Feb 2002 02:10:06 +0000e0 From: Mark W J Redding <grawlfang@ukhackers.net> Subject: Re: Software Question- Message-ID: <3C6DBF7E.99ADDDE4@ukhackers.net>    Andy,i  I In the dim and distant past (VMS 5.5 / 6.?) I has a VMSINSTAL script thatnA installed OBJect code, and linked it before placing in it's finaloN destination...Why, because I could not be certain of the version of the targetM systems VMS... When you're miles from the office the very last thing you needuL is to be stuck with an older version of VMS and only .EXE files to play withN (believe me...I've been half way around the world and encountered this (and atL that distance you can't just courier up a new version))....I also maintainedJ an old [very] DECstation 3100 (the one with the greyscale screen) with theF oldest version of VMS that any customer of mine had, just to test that. *really* old VMS versions would still work....   ;<)o   Andy Proctor wrote:   ; > "Mark Redding" <grawlfang@ukhackers.net> wrote in messageo) > news:3C6CFC82.4649DC0D@ukhackers.net...t	 > > Andy,e > >nI > > In my experience it's always been safest to link the .OBJ code on they > targetJ > > system, especially when that target is running an older version of VMS
 > than the > > source system. > >0 >1M > I understand, thanks. The reason for the question was that up until now alleM > the compiling/linking was done under VMS6.2-1h3 and it looks like this unitoH > (Alpha 200) is about to die, we only have PWS433 and above running VMSL > 7.1-1h2+ apart from this, but it looks like we will use a DS10 and VMS 7.3K > in furture. Problem is we still have to support, and provide software for - > 6.2-1h3 versions. :( Your info helps though  >n > Cheers >l > Andy   -- Mark W J Redding  F Chaos reigns within.  Reflect, repent and reboot.  Order shall return.   ------------------------------    Date: 16 Feb 2002 01:50:55 +0100* From: eplan@kapsch.net (Peter LANGSTOEGER)- Subject: Re: Ultimate Workstation vs. AS 1200e( Message-ID: <3c6dacef@news.kapsch.co.at>  Y In article <%TZ98.37489$Pz4.225063@rwcrnsc53>, "Mark E. Levy" <mlevy70@attbi.com> writes:i >Not any more... > K >> So I stand corrected (and amused, because there is currently one at ebayn >;-)  L Yup, you've been very fast, but look again, there is the next one ($1200 ;-)   -- 7< Peter "EPLAN" LANGSTOEGER           Tel.    +43 1 81111 2651; Network and OpenVMS system manager  Fax.    +43 1 81111 888e< KAPSCH AG      Wagenseilgasse 1     E-mail  eplan@kapsch.netH A-1121 VIENNA  AUSTRIA              "I'm not a pessimist, I'm a realist"   ------------------------------  % Date: Sat, 16 Feb 2002 01:24:52 +0000'0 From: Mark W J Redding <grawlfang@ukhackers.net> Subject: Very Very Slow Telnet.h- Message-ID: <3C6DB4E4.D5BC43E9@ukhackers.net>t  H I've a Linux desktop machine and a VAXstation (4000 VLC) running VMS 7.2 and TCP/IP .D When I telnet from the Linux box to the VMS box the session responseG time is dreadful. (it takes 15-20 seconds to prompt for a username, anduE the same for a password, for example) ..If I boot into Windows (which G goes against the grain) and telnet to the VAX the response time is veryj good.aF I've a small four port Trust hub, and when the Linux telnet session isE active the lights on the ports connected to the Linux and VAX flicker D like crazy (and they don't when connecting the box when booted under	 windows).o@ In each case I'm using a non-routable 192.168.?.? address space.@ I'm running a Linux 2.4.6 kernel under Slackware (and Win98 SE).  : Has anyone encountered a similar problem and resolved it ?  D This wouldn't be such as issue but the direct serial port connectionD between the VAX and the PC is not responding, so UCX is the only wayD into the machine (okay, I've connected a very old psion XP [so thatsE where M$ stole the name] to the RS-232 port [TTA2] but a 2 line by 16  char display is not ideal).5     -- Mark W J Redding  F Chaos reigns within.  Reflect, repent and reboot.  Order shall return.   ------------------------------  % Date: Sat, 16 Feb 2002 01:48:20 +0000t0 From: Mark W J Redding <grawlfang@ukhackers.net># Subject: Re: Very Very Slow Telnet.e- Message-ID: <3C6DBA64.92B8861E@ukhackers.net>h    The next piece of this jigsaw...  
 on the VAX...,   # ucxm TCPIP> ping 192.168.2.1M- PING 192.168.2.1 (192.168.2.1): 56 data bytese8 64 bytes from 192.168.2.1: icmp_seq=0 ttl=255 time=40 ms; 64 bytes from 192.168.2.1: icmp_seq=1 ttl=255 time=13470 msd; 64 bytes from 192.168.2.1: icmp_seq=2 ttl=255 time=26950 msf *CANCEL*   On the Linux PC...   root@osiris:~# ping 192.168.2.3o. PING 192.168.2.3 (192.168.2.3): 56 octets data9 64 octets from 192.168.2.3: icmp_seq=0 ttl=64 time=1.8 msl9 64 octets from 192.168.2.3: icmp_seq=1 ttl=64 time=3.6 msw9 64 octets from 192.168.2.3: icmp_seq=2 ttl=64 time=3.2 mso9 64 octets from 192.168.2.3: icmp_seq=3 ttl=64 time=3.0 mss9 64 octets from 192.168.2.3: icmp_seq=4 ttl=64 time=2.4 msa9 64 octets from 192.168.2.3: icmp_seq=5 ttl=64 time=2.5 mss  # --- 192.168.2.3 ping statistics --- 9 6 packets transmitted, 6 packets received, 0% packet lossa' round-trip min/avg/max = 1.8/2.7/3.6 ms    -- Mark W J Redding  F Chaos reigns within.  Reflect, repent and reboot.  Order shall return.   ------------------------------  % Date: Fri, 15 Feb 2002 22:21:58 -0500r1 From: Michael Austin <maustin@firstdbasource.com>e# Subject: Re: Very Very Slow Telnet.t2 Message-ID: <3C6DD056.7AEA5836@firstdbasource.com>  D on the PC do a tracert (traceroute) to the VAX.  What is the default gateway?  What version TCPIP?o  F I run 7.2-1 on my Alpha 2100's, Caldera Linux 2.3 with a hub connectedC to a router/hub to ADSL modem.. I actually have 6 systems connected:C (although spaceheater #2, I mean the second Alpha, is powered down)oC (2 Linux, 2 Windows98/ME and when the airconditioning is set reallyeG cold, 2 Alphas.)  Pretty slick... I use reflections and Pathworks32 for2E emulators on the PC, and command-line telnet on the Linux boxes.  The % Linux boxes are Pent-133 and 166Mhz. e   Mark W J Redding wrote:h > J > I've a Linux desktop machine and a VAXstation (4000 VLC) running VMS 7.2 > and TCP/IP .F > When I telnet from the Linux box to the VMS box the session responseI > time is dreadful. (it takes 15-20 seconds to prompt for a username, anduG > the same for a password, for example) ..If I boot into Windows (whichpI > goes against the grain) and telnet to the VAX the response time is veryr > good.sH > I've a small four port Trust hub, and when the Linux telnet session isG > active the lights on the ports connected to the Linux and VAX flickeryF > like crazy (and they don't when connecting the box when booted under > windows).SB > In each case I'm using a non-routable 192.168.?.? address space.B > I'm running a Linux 2.4.6 kernel under Slackware (and Win98 SE). > < > Has anyone encountered a similar problem and resolved it ? > F > This wouldn't be such as issue but the direct serial port connectionF > between the VAX and the PC is not responding, so UCX is the only wayF > into the machine (okay, I've connected a very old psion XP [so thatsG > where M$ stole the name] to the RS-232 port [TTA2] but a 2 line by 162 > char display is not ideal).a >  > -- > Mark W J Redding > H > Chaos reigns within.  Reflect, repent and reboot.  Order shall return.   -- > Regards,  7 Michael Austin            Registered Linux User #261163o7 First DBA Source, Inc.    http://www.firstdbasource.comn Sr. Consultant 704-947-1089 (Office)o 704-236-4377 (Mobile)h   ------------------------------  # Date: Sat, 16 Feb 2002 03:59:41 GMT  From: danco@pebble.org# Subject: Re: Very Very Slow Telnet.w- Message-ID: <slrna6rm9c.qb9.danco@pebble.org>y  = >> Has anyone encountered a similar problem and resolved it ?-  D This sounds suspiciously like a known bug in UCX's TELNET.  However,A my memory is not good enough to recall any actual specifics other A than that it might have had something to do with a problem in the>" terminal type setting negotiation.   - Dant   ------------------------------  # Date: Fri, 15 Feb 2002 21:04:16 GMTeL From: winston@SSRL.SLAC.STANFORD.EDU ("Alan Winston - SSRL Admin Cmptg Mgr")6 Subject: Re: What is the best NFS package for OpenVMS?8 Message-ID: <00A099A9.BC699687@SSRL04.SLAC.STANFORD.EDU>  h In article <d7791aa1.0202150622.4f650790@posting.google.com>, bob@instantwhip.com (Bob Ceculski) writes:  C >I have been on vms for over 16 years now "without" an os crash ...eE >We have been on purveyor for two years now w/o a crash or lockup ...a? >We have been on TCPware for over 2 years now w/o a problem ... F >These are called testimonials ... do you have any, including your ownF >that suggests that wasd is reliable ... thats all I am asking for ...G >My point is, you have a job, and wasd is not your primary job, it is abF >side hobby, and it is great that you are trying to help others w/freeH >shareware, but software that you pay for is generally better (i.e. vms)I >than shareware (i.e. linux) because you are paying for it (the exceptionlE >to this is microsoft) ... wasd may be an exception, and testimonialsr0 >would answer any questions ... do you have any?    I (1) Actually, if you go to the WASD site you can find a nice presentation N from a company that's built its web strategy on WASD, using Python (interfaced2 to Rdb) for CGIs; that's a lot like a testimonial.  I (2) "Software that you pay for is generally better" but there are a _lot_ L of counter-examples.  Perl is free and it's a very, very effective scriptingL language that's used in production many places, and has good (free) support.N Python is free, and is also a competent and capable piece of software.  ApacheL is used in production lots and lots of places, and I think is unquestionablyN better than IIS.  (And CSWS seems to be a _lot_ better than Netscape FastTrack server.)   -- Alant  O ===============================================================================u0  Alan Winston --- WINSTON@SSRL.SLAC.STANFORD.EDUM  Disclaimer: I speak only for myself, not SLAC or SSRL   Phone:  650/926-3056.M  Physical mail to: SSRL -- SLAC BIN 69, PO BOX 4349, STANFORD, CA  94309-0210 O ===============================================================================y   ------------------------------  # Date: Sat, 16 Feb 2002 03:16:54 GMT^$ From: "MarkHill" <mhill@stny.rr.com> Subject: Re: XP 1000 won't bootu: Message-ID: <Gakb8.87103$QG.18946143@typhoon.nyroc.rr.com>  2 Just a guess here cause there is no info posted...I VMS Boot drivers are sensitive... They dont handle directory sizes biggertK than 127 blocks very well and they dont like directories that have multiplyiE allocated blocks that defraggers may have moved among other things...n  K try booting with R5 set to 10000 or 20000.... I forgot which bit is exactlyaL which but these will give you some better idea where your crapping out... On( an XP it would look something like this:  I (Assuming a whole bunch of things like you're booting from sys0, you dont F want to stop at sysboot, and bootdef_dev is set to something sensible)   >>>b -fl 20000  ; "Tony Barker" <tony@dartfactordata.ltd.uk> wrote in messageS0 news:B891C685.8CFB%tony@dartfactordata.ltd.uk...2 > I have a XP 1000 workstation running ovms V7.2-1K > It has been up and running for several months - including several reboots2I > but this morning following a shutdown (@sys$system:shutdown) the systemo nowsC > fails to restart.  Just after the boot starts - it announces ovms6	 starting,oL > then performs a crash dump, displays the message "inconsistent boot driver- > state" and then reboots again, and again...eJ > Can anyone please tell me ehat's wrong and how I can go about fixing it? >r > thanks >  > Tony >6   ------------------------------   End of INFO-VAX 2002.091 ************************le, work) 9    sms@antinode.org                sms@provis.com  (work)p   ------------------------------    Date: 15 Feb 2002 17:16:38 -0600- From: Kilgallen@SpamCop.net (Larry Kilgallen),> 