1 INFO-VAX	Fri, 16 Dec 2005	Volume 2005 : Issue 698       Contents: A closing thought for 2005 Another Backup suggestion  Re: Another Backup suggestion  Re: Another Backup suggestion  Re: Another Backup suggestion  Re: Appletalk on VMS 7.3-2/ RE: Clustering: switches reliability/redundancy / Re: Clustering: switches reliability/redundancy / RE: Clustering: switches reliability/redundancy  Re: Database access from COBOL Re: Database access from COBOL RE: Database access from COBOL# Re: DECnet IV over Multinet Circuit  Re: DESTA Memory hog Re: DESTA Memory hog7 Re: HP announces new Integrity Blade Servers - OpenVMS? 7 Re: HP announces new Integrity Blade Servers - OpenVMS? 7 Re: HP announces new Integrity Blade Servers - OpenVMS? 7 Re: HP announces new Integrity Blade Servers - OpenVMS? 7 Re: HP announces new Integrity Blade Servers - OpenVMS? 7 Re: HP announces new Integrity Blade Servers - OpenVMS? 7 Re: HP announces new Integrity Blade Servers - OpenVMS? 7 Re: HP announces new Integrity Blade Servers - OpenVMS? 7 Re: HP announces new Integrity Blade Servers - OpenVMS? 7 Re: HP announces new Integrity Blade Servers - OpenVMS? 7 RE: HP announces new Integrity Blade Servers - OpenVMS? 7 RE: HP announces new Integrity Blade Servers - OpenVMS? 7 Re: HP announces new Integrity Blade Servers - OpenVMS? 7 RE: HP announces new Integrity Blade Servers - OpenVMS? 7 Re: HP announces new Integrity Blade Servers - OpenVMS?  Jon Andruszkiewicz+ Re: Location of initial page and swap files  Re: monitoring VMS log files Re: No more SETI on VMS ? B PHONE error - Invalid specification of node or person.  Try again.F Re: PHONE error - Invalid specification of node or person.  Try again.E Re: PHONE error - Invalid specification of node or person. Try again. E Re: PHONE error - Invalid specification of node or person. Try again. E Re: PHONE error - Invalid specification of node or person. Try again. & R400X: converting DSSI shelves to SCSI* Re: R400X: converting DSSI shelves to SCSI Re: SRM memtest  Re: SRM memtest  Re: SRM memtest  Re: SRM memtest  Re: SRM memtest  Re: SRM memtest  Re: SRM memtest  Re: SRM memtest  Re: SRM memtest  Re: Storageworks shelves Re: Storageworks shelves Re: Storageworks shelves Re: Storageworks shelves Re: Storageworks shelves Re: Storageworks shelves Re: Storageworks shelves Re: Storageworks shelves  Re: UK VMS HP contract customers  Re: UK VMS HP contract customers Updated VMS information & Re: Upgrade a 5300A controller via CD?& Re: Upgrade a 5300A controller via CD? Re: Upgrading TCPIP 9 Re: using SYS$ACM to authenticate arbitrary windows user? 9 Re: using SYS$ACM to authenticate arbitrary windows user?   F ----------------------------------------------------------------------    Date: 15 Dec 2005 11:47:13 -08009 From: "Alphaman" <alphaman-nix-spam@hsv.sungardtrust.com> # Subject: A closing thought for 2005 C Message-ID: <1134676033.077737.208240@z14g2000cwz.googlegroups.com>   B Have you ever been in a crowded restaurant where everyone all of aE sudden stops talking simultaneously?  Everyone pauses, not wanting to > be the first one to talk again, to break that golden moment...  A I'm convinced that one of these days, every Windoze PeeCee on the D planet will simultaneously crash.  All the 'Dozers will look at eachC other and say "that's weird".  All the VMS people will look up from 1 their work only long enough to say "told you so."   D Here's hoping the golden moment arrives in 2006!  Best wishes to all! for a happy cluster* of holidays!    Aaron   G * Everyone else is calling their stuff a cluster, why not the holidays?    ------------------------------  % Date: Thu, 15 Dec 2005 20:50:36 -0500 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> " Subject: Another Backup suggestion, Message-ID: <43A21D55.6E4354C4@teksavvy.com>  D While freezing my toes off in this very cold weather, I got to think/ about Guy Peleg's recent Backup improvements...    I would suggest the following:   BACKUP /LOG=INTERVAL=(DIR:3)D Writes a log line everytime it has completed the backing up of a 3rd* level directory, as well as higher levels. eg:   N 30-FEB-2006 07:43:25 dka200:[sys0.syshlp.examples]  9273 blocks  230202 blocksA 30-FEB-2006 08:20:21 dka200:[sys0.syshlp]           15000 blocks   3002383 blocksA 30-FEB-2006 09:10:34 dka200:[sys0]                  29392 blocks   3938282 blocksB (eg: give the number of blocks in that directory, and a cumulative+ number of blocks for the backup operation.)        BACKUP /LOG=INTERVAL=100  C This one is simpler, it issues a standard line (filename) every 100  files it has processed.       E The first one gives very informative messages about which directories G contain many blocks, and where the cutoff really happens between tapes.   E The second one just gives a rough idea of where the backup has gotten = without filling a log file with  agazillion lines, or without : monopolising a terminal with constant output of filenames.   ------------------------------  % Date: Thu, 15 Dec 2005 23:36:25 -0500 ( From: Wayne Sewell <wayne@tachysoft.com>& Subject: Re: Another Backup suggestion0 Message-ID: <00A4E558.D3D6C921.11@tachysoft.com>  & >Date: Thu, 15 Dec 2005 20:50:36 -0500. >From: JF Mezei <jfmezei.spamnot@teksavvy.com> >Organization: Indexed* >X-Mailer: Mozilla 8.2 (Macintosh; U; PPC)    E >While freezing my toes off in this very cold weather, I got to think 0 >about Guy Peleg's recent Backup improvements... >  >I would suggest the following:  >  >BACKUP /LOG=INTERVAL=(DIR:3) E >Writes a log line everytime it has completed the backing up of a 3rd + >level directory, as well as higher levels.  >eg: > O >30-FEB-2006 07:43:25 dka200:[sys0.syshlp.examples]  9273 blocks  230202 blocks B >30-FEB-2006 08:20:21 dka200:[sys0.syshlp]           15000 blocks  >3002383 blocks B >30-FEB-2006 09:10:34 dka200:[sys0]                  29392 blocks  >3938282 blocks C >(eg: give the number of blocks in that directory, and a cumulative , >number of blocks for the backup operation.) >  >  >  >BACKUP /LOG=INTERVAL=100  > D >This one is simpler, it issues a standard line (filename) every 100 >files it has processed. >  >  > F >The first one gives very informative messages about which directoriesH >contain many blocks, and where the cutoff really happens between tapes. > F >The second one just gives a rough idea of where the backup has gotten> >without filling a log file with  agazillion lines, or without; >monopolising a terminal with constant output of filenames.     M You can get the second mode with tapesys, using the PROGRESS parameter, which O is also the number of files between progress messages.  Since sysbak is a batch $ job, the messages are sent to opcom.    O %%%%%%%%%%%  OPCOM  15-DEC-2005 21:34:46.88  %%%%%%%%%%%    (from node HARDY at   15-DEC-2005 21:33:53.42) " Message from user TAPESYS on HARDYP BILLYDISK:[WEB_DOCS_FOR_ODS5.HTDOCS.tutos.php]user.pinc;1 processed on reel DLT0 62, drive _HARDY$MKB600:    L If you specify verify on the backup, you also get progress messages for that/ phase, substituting "verified" for "processed".    Wayne O =============================================================================== N Wayne Sewell, Tachyon Software Consulting  (281)812-0738   wayne@tachysoft.com; http://www.tachysoft.com/www/tachyon.html and wayne.html    O =============================================================================== P Jake Blues:"You traded the Caddy for a microphone? ...... Okay, I can buy that."   ------------------------------  % Date: Fri, 16 Dec 2005 00:49:43 -0500 ' From: Dave Froble <davef@tsoft-inc.com> & Subject: Re: Another Backup suggestion0 Message-ID: <11q4l8k8ko7m469@corp.supernews.com>   Wayne Sewell wrote: ' >>Date: Thu, 15 Dec 2005 20:50:36 -0500 / >>From: JF Mezei <jfmezei.spamnot@teksavvy.com>  >>Organization: Indexed + >>X-Mailer: Mozilla 8.2 (Macintosh; U; PPC)  >  >  > F >>While freezing my toes off in this very cold weather, I got to think1 >>about Guy Peleg's recent Backup improvements...  >>  >>I would suggest the following: >> >>BACKUP /LOG=INTERVAL=(DIR:3)F >>Writes a log line everytime it has completed the backing up of a 3rd, >>level directory, as well as higher levels. >>eg:  >>P >>30-FEB-2006 07:43:25 dka200:[sys0.syshlp.examples]  9273 blocks  230202 blocksC >>30-FEB-2006 08:20:21 dka200:[sys0.syshlp]           15000 blocks   >>3002383 blocksC >>30-FEB-2006 09:10:34 dka200:[sys0]                  29392 blocks   >>3938282 blocksD >>(eg: give the number of blocks in that directory, and a cumulative- >>number of blocks for the backup operation.)  >> >> >> >>BACKUP /LOG=INTERVAL=100 >>E >>This one is simpler, it issues a standard line (filename) every 100  >>files it has processed.  >> >> >>G >>The first one gives very informative messages about which directories I >>contain many blocks, and where the cutoff really happens between tapes.  >>G >>The second one just gives a rough idea of where the backup has gotten ? >>without filling a log file with  agazillion lines, or without < >>monopolising a terminal with constant output of filenames. >  >  > O > You can get the second mode with tapesys, using the PROGRESS parameter, which Q > is also the number of files between progress messages.  Since sysbak is a batch & > job, the messages are sent to opcom.  I A good opportunity to mention your product, but, while I don't know your  E prices, do you really consider JF a prospective customer?  (The same   would apply to me.)   ' Or perhaps you have a hobbyist program?    --  4 David Froble                       Tel: 724-529-04504 Dave Froble Enterprises, Inc.      Fax: 724-529-0596> DFE Ultralights, Inc.              E-Mail: davef@tsoft-inc.com 170 Grimplin Road  Vanderbilt, PA  15486    ------------------------------  % Date: Fri, 16 Dec 2005 01:42:38 -0500 ( From: Wayne Sewell <wayne@tachysoft.com>& Subject: Re: Another Backup suggestion/ Message-ID: <00A4E56A.75743398.1@tachysoft.com>   ( >From: Dave Froble <davef@tsoft-inc.com> >X-Newsgroups: comp.os.vms' >Subject: Re: Another Backup suggestion & >Date: Fri, 16 Dec 2005 00:49:43 -0500= >Organization: Posted via Supernews, http://www.supernews.com      >Wayne Sewell wrote:( >>>Date: Thu, 15 Dec 2005 20:50:36 -05000 >>>From: JF Mezei <jfmezei.spamnot@teksavvy.com> >>>Organization: Indexed, >>>X-Mailer: Mozilla 8.2 (Macintosh; U; PPC) >>   >>   >>  G >>>While freezing my toes off in this very cold weather, I got to think 2 >>>about Guy Peleg's recent Backup improvements... >>> ! >>>I would suggest the following:  >>>  >>>BACKUP /LOG=INTERVAL=(DIR:3) G >>>Writes a log line everytime it has completed the backing up of a 3rd - >>>level directory, as well as higher levels.  >>>eg: >>> Q >>>30-FEB-2006 07:43:25 dka200:[sys0.syshlp.examples]  9273 blocks  230202 blocks D >>>30-FEB-2006 08:20:21 dka200:[sys0.syshlp]           15000 blocks  >>>3002383 blocks D >>>30-FEB-2006 09:10:34 dka200:[sys0]                  29392 blocks  >>>3938282 blocks E >>>(eg: give the number of blocks in that directory, and a cumulative . >>>number of blocks for the backup operation.) >>>  >>>  >>>  >>>BACKUP /LOG=INTERVAL=100  >>> F >>>This one is simpler, it issues a standard line (filename) every 100 >>>files it has processed. >>>  >>>  >>> H >>>The first one gives very informative messages about which directoriesJ >>>contain many blocks, and where the cutoff really happens between tapes. >>> H >>>The second one just gives a rough idea of where the backup has gotten@ >>>without filling a log file with  agazillion lines, or without= >>>monopolising a terminal with constant output of filenames.  >>   >>   >>  P >> You can get the second mode with tapesys, using the PROGRESS parameter, whichR >> is also the number of files between progress messages.  Since sysbak is a batch' >> job, the messages are sent to opcom.  > J >A good opportunity to mention your product, but, while I don't know your F >prices, do you really consider JF a prospective customer?  (The same  >would apply to me.) >   M While my post is in response to JF, many other people are reading this thread K and not all of them are hobbyists.   Some of them may also be interested in I this capability, now that it has been pointed out to them.  True, this by J itself would probably not be a reason to buy the product, but it is a niceN feature that might tip someone over the edge, if they have read my other postsN about parallel backups and other tapesys capabilities far beyond what bare vmsL backup will do.  I really don't know how a large site can function without a tape management product.    ( >Or perhaps you have a hobbyist program? >   K I don't deal with sales or distribution, but I don't think so.  You can ask K them if they would be interested in starting one.  But like all vms-related . vendors, we really need more paying customers.   Wayne O =============================================================================== N Wayne Sewell, Tachyon Software Consulting  (281)812-0738   wayne@tachysoft.com; http://www.tachysoft.com/www/tachyon.html and wayne.html    O =============================================================================== P Jake Blues:"You traded the Caddy for a microphone? ...... Okay, I can buy that."   ------------------------------    Date: 15 Dec 2005 11:22:40 -0800$ From: "Ed Wilts" <ewilts@ewilts.org># Subject: Re: Appletalk on VMS 7.3-2 B Message-ID: <1134674560.191092.58960@g43g2000cwa.googlegroups.com>   healyzh@aracnet.com wrote:3 > Chris Sharman <chris.sharman@sorry.nospam> wrote:  > > Ed Wilts wrote: J > > > It worked fine for us before some of the 7.3-2 patches were applied. > > > C > > > Any issues in AppleTalk need to be fixed by the maintainer of A > > > AppleTalk.  Unfortunately, there is no active maintainer...  > 7 > > Did you get any further with this issue from July ? G > > Did you identify the problem patch, or find a workaround, or what ?  > J > > I'd like to move up to a supported version, but I really don't want to > > break appletalk. > M > What are you using Appletalk for?  I use it for accessing a couple of disks I > from my Mac, and for printing to a printer on an Ethertalk-to-Localtalk J > converter.  However, becuase of the news it breaks with one of the 7.3-2N > patches, and because if I upgrade to Mac OS X 10.4.x I can't talk to classicJ > Appletalk I recently installed a recent version of Samba, and unlike theK > antique version I was running, I'm able to use Samba Shares from the Mac.   E We have several hundred AppleTalk printers.  Many of these don't talk 8 nicely from DCPS in IP mode - some don't have IP at all.  M > I've also got a print server laying around that will support both Appletalk I > and other protocols for printing, I've had it for years, but have never M > bothered to install it since the existing solution works.  I'll have to get % > that installed one of these months.   E We need the instantaneous feedback that bidirectional AppleTalk gives F us.  Offloading the printing to a separate server makes that much more difficult, if not impossible.   M > I'd be interested in knowing what patches break the Appletalk support, as I 6 > really need to get the patches on my system updated.  F I don't know which one broke it and it's tough to find.  I crash aboutA once per month or two and do a *lot* of AppleTalk printing spread  across 2 nodes.   G You may find yourself in the situation where you can be current and not D crash at all.  On the other hand, maybe I'm the lucky one and you'll crash daily!  	    .../Ed    ------------------------------  + Date: Thu, 15 Dec 2005 23:01:25 +0000 (UTC) - From: klewis@OMEGA.MITRE.ORG (Keith A. Lewis) 8 Subject: RE: Clustering: switches reliability/redundancy. Message-ID: <dnssk5$6r5$1@newslocal.mitre.org>   "Main, Kerry" <Kerry.Main@hp.com> writes in article <FD827B33AB0D9C4E92EACEEFEE2BA2FB7735F5@tayexc19.americas.cpqcorp.net> dated Wed, 14 Dec 2005 17:35:34 -0500:  > : >> From: JF Mezei [mailto:jfmezei.spamnot@teksavvy.com]=20  H >> Are there considerations if a cluster nodes are all connected via theD >> same switch/hub ? It the later fails, the whole cluster hangs and >> becomes inaccessible.=20   K Yes.  It should un-hang when power is restored.  Of course, if the same hub C connects you to your network of clients, the cluster would still be 2 inaccessible regardless of whether it hung or not.  I >> Are hubs/switches considered "fault tolerant"  ? If not, what possible A >> steps would a good site planner take to ensure a cluster isn't ' >> jeoperdized by some $50 switch/hub ?  >>=20 < >> Are hubs considered more "fault tolerant" than switches ? >>=20 C >> Is it just a simple case of reserving spare ports on a backup=20  >> switch soG >> that cluster ethernet connectiosn can be moved one by one before the 7 >> main switch/hub is powered off for maintenance etc ?  >>=20   F >Simple solution is to establish VLAN between 2 trunked switch/routersG >and use separate NIC connections to each from each server. This causes G >the switch to appear as a logical unit or virtual cluster interconnect H >box. Entire switch/rtr fails and OpenVMS cluster keeps running i.e. not) >even any application failover issues.=20  > H >With the right IP failover config's in place, you would not even lose a >telnet connection.=20 > B >OpenVMS will load balance SCS across all configured and availableG >connections. By configured I mean not disabled with the SCACP utility.   L What we do is run 2 independent switches.  Can't v-lan them together becauseH the MAC addresses are the same across all cards due to DECNET phase IV. L With DECNET performance no longer important, Kerry's solution is better, I'd! change to that if I had the time.   J The configuration I describe is tolerant of any single fault, but you haveD to stay on top of them because a double fault can hang the cluster.   K Example:  Nodes A-C each have 2 NICs and are connected to switches 1 and 2. J If NIC A-1 fails, everything still works.  But if NIC B-2 fails before A-1K is fixed, nodes A and B can no longer communicate.  C can still see both of J them, so it wants to keep the cluster together, and each of {A,B} wants to7 kick the other out.  The result is a cluster-wide hang.   0 --Keith Lewis              klewis {at} mitre.org> The above may not (yet) represent the opinions of my employer.   ------------------------------  + Date: Fri, 16 Dec 2005 04:43:47 +0000 (UTC) 7 From: moroney@world.std.spaamtrap.com (Michael Moroney) 8 Subject: Re: Clustering: switches reliability/redundancy( Message-ID: <dntgm3$dno$1@pcls4.std.com>  / klewis@OMEGA.MITRE.ORG (Keith A. Lewis) writes:   L >Example:  Nodes A-C each have 2 NICs and are connected to switches 1 and 2.K >If NIC A-1 fails, everything still works.  But if NIC B-2 fails before A-1 L >is fixed, nodes A and B can no longer communicate.  C can still see both ofK >them, so it wants to keep the cluster together, and each of {A,B} wants to 8 >kick the other out.  The result is a cluster-wide hang.  I Cluster shouldn't hang.  One of A or B will be kicked out and the cluster H will continue.  I don't know the algorithm other than the one with fewerC votes will be kicked out if different.  C knows A and B cannot talk  to each other.   ------------------------------  % Date: Thu, 15 Dec 2005 23:45:35 -0500 ' From: "Main, Kerry" <Kerry.Main@hp.com> 8 Subject: RE: Clustering: switches reliability/redundancyR Message-ID: <FD827B33AB0D9C4E92EACEEFEE2BA2FB77368D@tayexc19.americas.cpqcorp.net>   > -----Original Message-----9 > From: Keith A. Lewis [mailto:klewis@OMEGA.MITRE.ORG]=20 ! > Sent: December 15, 2005 6:01 PM  > To: Info-VAX@Mvb.Saic.Com : > Subject: RE: Clustering: switches reliability/redundancy >=208 > "Main, Kerry" <Kerry.Main@hp.com> writes in article=20@ > <FD827B33AB0D9C4E92EACEEFEE2BA2FB7735F5@tayexc19.americas.cpqc1 > orp.net> dated Wed, 14 Dec 2005 17:35:34 -0500:  > > > > >> From: JF Mezei [mailto:jfmezei.spamnot@teksavvy.com]=3D20 >=20; > >> Are there considerations if a cluster nodes are all=20  > connected via the F > >> same switch/hub ? It the later fails, the whole cluster hangs and > >> becomes inaccessible.=3D20  >=20@ > Yes.  It should un-hang when power is restored.  Of course,=20 > if the same hub E > connects you to your network of clients, the cluster would still be 4 > inaccessible regardless of whether it hung or not. >=20@ > >> Are hubs/switches considered "fault tolerant"  ? If not,=20 > what possible C > >> steps would a good site planner take to ensure a cluster isn't ) > >> jeoperdized by some $50 switch/hub ? 	 > >>=3D20 > > >> Are hubs considered more "fault tolerant" than switches ?	 > >>=3D20 G > >> Is it just a simple case of reserving spare ports on a backup=3D20  > >> switch soA > >> that cluster ethernet connectiosn can be moved one by one=20  > before the9 > >> main switch/hub is powered off for maintenance etc ? 	 > >>=3D20  >=20H > >Simple solution is to establish VLAN between 2 trunked switch/routers@ > >and use separate NIC connections to each from each server.=20
 > This causes ? > >the switch to appear as a logical unit or virtual cluster=20  > interconnect< > >box. Entire switch/rtr fails and OpenVMS cluster keeps=20 > running i.e. not- > >even any application failover issues.=3D20  > > A > >With the right IP failover config's in place, you would not=20 
 > even lose a  > >telnet connection.=3D20 > > D > >OpenVMS will load balance SCS across all configured and available= > >connections. By configured I mean not disabled with the=20  > SCACP utility. >=20@ > What we do is run 2 independent switches.  Can't v-lan them=20 > together becauseB > the MAC addresses are the same across all cards due to DECNET=20 > phase IV.=20B > With DECNET performance no longer important, Kerry's solution=20 > is better, I'd# > change to that if I had the time.  >=20B > The configuration I describe is tolerant of any single fault,=20 > but you haveH > to stay on top of them because a double fault can hang the cluster.=20 >=20> > Example:  Nodes A-C each have 2 NICs and are connected to=20 > switches 1 and 2. > > If NIC A-1 fails, everything still works.  But if NIC B-2=20 > fails before A-1> > is fixed, nodes A and B can no longer communicate.  C can=20 > still see both of @ > them, so it wants to keep the cluster together, and each of=20 > {A,B} wants to9 > kick the other out.  The result is a cluster-wide hang.  >=202 > --Keith Lewis              klewis {at} mitre.org@ > The above may not (yet) represent the opinions of my employer. >=20   Keith,  C With one Cust with high availability 3 node ES45 cluster we did the 
 following:  ) - dual Cisco switch/rtrs trunked together G - 3 separate VLANS to isolate traffic protocols and increased security. G VLAN1 - normal TCPIP traffic (each connection was actually 2 NIC's, one  to each switch) F VLAN2 - SCS traffic (each connection was actually 2 NIC's, one to each switch) G VLAN3 - DECnet traffic (each connection was only one NIC to one switch)    Actually, we had 4th as well: H VLAN4 - mgmt monitoring, security and console manager (had network level; security controls on who could access devices on this VLAN)   G Hence with the above, if a switch failed, then DECnet to 1 or 2 servers < might be cut, but all normal cluster traffic would continue.  B And a misc note - while there are always some trade-offs, the CustH wanted to ensure that when the took 1 server down for planned reasons ifE one of the remaining servers crashed, then the last node should still I continue running. Hence, we simply made each system =3D 1 vote and made =  a  quorum disk =3D 2 votes.=20      Regards   
 Kerry Main Senior Consultant  HP Services Canada Voice: 613-592-4660  Fax: 613-591-4477  kerryDOTmainAThpDOTcom (remove the DOT's and AT)=20  4 OpenVMS - the secure, multi-site OS that just works.   ------------------------------    Date: 15 Dec 2005 14:27:26 -0800 From: dooleys@snowy.net.au' Subject: Re: Database access from COBOL C Message-ID: <1134685646.185094.127760@z14g2000cwz.googlegroups.com>   - Do you have these directories on your system? ! sys$sysroot:[syshlp.examples.rdb] $ sys$sysroot:[syshlp.examples.decrdb] Phil   ------------------------------  # Date: Fri, 16 Dec 2005 02:52:13 GMT # From: hoff@hp.nospam (Hoff Hoffman) ' Subject: Re: Database access from COBOL 0 Message-ID: <xZpof.755$NI5.481@news.cpqcorp.net>  a In article <43A221EB.A31E5019@oracle.com>, Norman Lastovica <norman.lastovica@oracle.com> writes: 1 :install Oracle Rdb and there ought to be several 1 :examples of database access with module language  :or embeded SQL.  =   With OpenVMS Alpha V7.3-2 and an Rdb V7.1-* variant loaded:   9 $ DIRECTORY SYS$COMMON:[SYSHLP.EXAMPLES.RDB71]*.COB,*.SCO   , Directory SYS$COMMON:[SYSHLP.EXAMPLES.RDB71]  N DBMS_ADD.COB;1      DBMS_DEL.COB;1      JOBINFOINT.COB;1    RDB_ADD_EAST.COB;1( RDB_DEL_WEST.COB;1  SQL_DIST_TRANS.COB;1N SQL_INSERT_DEGREES_PROGRAM.COB;1        TRANSFER_EAST.COB;1 RDB_ADD_WEST.SCO;1+ RDB_DEL_EAST.SCO;1  SQL_ALL_DATATYPES.SCO;1 = SQL_ALL_DATATYPES_DATE.SCO;1            SQL_CONNECTIONS.SCO;1 ? SQL_DIST_TRANS_ERROR.SCO;1              SQL_LOAD_COLLEGES.SCO;1 < SQL_LOAD_DEGREES.SCO;1                  SQL_LOAD_DEPTS.SCO;1> SQL_LOAD_EMPLOYEES.SCO;1                SQL_LOAD_JOBHIST.SCO;1L SQL_LOAD_JOBS.SCO;1 SQL_LOAD_SALHIST.SCO;1                  SQL_REPORT.SCO;1; SQL_RESUMES.SCO;1   SQL_TERMINATE.SCO;1 TRANSFER_WEST.SCO;1     C   Also see the HELP SQLPRE71 command output for information on the  C   Rdb SQL preprocessor (embedded SQL, the "exec sql" stuff that can B   be inserted into C, COBOL or a few other languages), or whatever?   equivalent documentation exists for the particular version of    Rdb you have installed.     N  ---------------------------- #include <rtfaq.h> -----------------------------K     For additional, please see the OpenVMS FAQ -- www.hp.com/go/openvms/faq N  --------------------------- pure personal opinion ---------------------------G        Hoff (Stephen) Hoffman   OpenVMS Engineering   hoff[\0100]hp.com    ------------------------------  % Date: Thu, 15 Dec 2005 22:45:23 -0500 ' From: "Main, Kerry" <Kerry.Main@hp.com> ' Subject: RE: Database access from COBOL R Message-ID: <FD827B33AB0D9C4E92EACEEFEE2BA2FB77368A@tayexc19.americas.cpqcorp.net>   > -----Original Message-----$ > From: bill@triangle.cs.uofs.edu=20A > [mailto:bill@triangle.cs.uofs.edu] On Behalf Of Bill Gunshannon " > Sent: December 15, 2005 10:52 AM > To: Info-VAX@Mvb.Saic.Com % > Subject: Database access from COBOL  >=20F > Well, it looks like this semester of student VAX use went well.  NowH > it is time to start building on this.  The students who just did data-H > structures will be moving on to their first database course.  The lastA > thing they did with COBOL on the VAX was ISAM.  Looks like a=20  > good lead-G > up to see why databases are the next step.  So,  can someone point me H > at some simple examples of database access from COBOL on VMS?  And letH > me know which manual I shuold expect to find more detailed informationA > about it?  (It would be nice if I had the time to pl,ay with=20  > it myself,@ > but I don't and I hate re-inventing the wheel anyway.)  Who=20 > knows, maybeH > we are headed towards a VMS Renaisance in Education!!  Maybe if we getE > through this next semester with the students working on VMS again I G > can convince the professor involved to consider a paper for the Small ? > Colleges Conference about using VMS to handle these concepts.  >=20 > bill >=20     Bill,   F Nice feedback .. If you'd like, perhaps you could add a few paragraphs' to the following Hobbyist and Edu site:   H http://h71000.www7.hp.com/openvms/license_software_programs/mainstorypag e.html  H Drop me a note offline if you are interested and I will get you in touch3 with the folks in charge of this web sites content.   E JF - you might be interested in this short note about using VAX's and  films:H http://h71000.www7.hp.com/openvms/license_software_programs/filmmaker.ht ml   :-)   
 Kerry Main Senior Consultant  HP Services Canada Voice: 613-592-4660  Fax: 613-591-4477  kerryDOTmainAThpDOTcom (remove the DOT's and AT)=20  4 OpenVMS - the secure, multi-site OS that just works.   ------------------------------  % Date: Thu, 15 Dec 2005 11:52:07 -0700 & From: Jim Mehlhop <mjim1025@qwest.net>, Subject: Re: DECnet IV over Multinet Circuit( Message-ID: <43A1BB57.6070803@qwest.net>  , Were you going through at NAT at either end?       healyzh@aracnet.com wrote:+ > Bill Gunshannon <bill@cs.uofs.edu> wrote:  > J >>Our networking people installed a (industry standard) firewall here someH >>time ago that has turned TCPIP in to a stateful protocol.  That is, itJ >>actually tracks sessions running through it and if a session is inactiveM >>for a (operator defined) period of time it stops passing data for it should J >>it become active again.  A real braindead idea, but I don;t have controlK >>over it and it has forced me to institute keep-alives on all our remotely  >  >  > Bill, N > Thanks!  That's exactly what the problem was, since I *do* have control overK > this firewall, I up'd the timeouts to 5 minutes 30 seconds, and it's been N > working for an hour or two.  Something tells me your networking people might6 > be using the same software I'm using for a firewall. >  > 		Thanks,  > 		    Zane >    ------------------------------  + Date: Thu, 15 Dec 2005 23:16:08 +0000 (UTC) - From: klewis@OMEGA.MITRE.ORG (Keith A. Lewis)  Subject: Re: DESTA Memory hog . Message-ID: <dnstfo$6r5$2@newslocal.mitre.org>   comp.os.vms@hotmail.com writes in article <1134655433.046524.155060@g43g2000cwa.googlegroups.com> dated 15 Dec 2005 07:01:22 -0800:  >>From a DS25 4GB machine. > & >$ pipe show sys |sear sys$input destaG >0000045D DESTA Director  HIB      6  1729374   0 00:11:02.78    189857 	 >34869 M   >  >This seems a tad excessive.  L I don't know what DESTA is or does, but a 4GB system can handle 34,000 pages4 without a problem.  I/O count looks a little high...  0 --Keith Lewis              klewis {at} mitre.org> The above may not (yet) represent the opinions of my employer.   ------------------------------  # Date: Fri, 16 Dec 2005 00:33:54 GMT # From: hoff@hp.nospam (Hoff Hoffman)  Subject: Re: DESTA Memory hog 0 Message-ID: <SXnof.748$xG5.739@news.cpqcorp.net>  c In article <1134655316.309623.163070@g47g2000cwa.googlegroups.com>, comp.os.vms@hotmail.com writes:  :>From a DS25 4GB machine. : & :$ pipe show sys |sear sys$input destaG :0000045D DESTA Director  HIB      6  1729374   0 00:11:02.78    189857 	 :34869 M   :  :This seems a tad excessive.  F   "DESTA Director" is part of the Distributed Enterprise Service ToolsC   Architecture; see the WEBES, System Event Analyzer (SEA, formerly 0   Compaq Analyze) and CCAT crash analysis tools.  .   This amount of memory does appear excessive.  I   Do check with the customer support center, and find out of a DESTA STOP G   command or similar would be appropriate for you, or if there is a fix 4   or an update or an ECO available for the software.   	--   I   DESTA is also a device name of an Ethernet transceiver widget, but that !   is not related to this process.   N  ---------------------------- #include <rtfaq.h> -----------------------------K     For additional, please see the OpenVMS FAQ -- www.hp.com/go/openvms/faq N  --------------------------- pure personal opinion ---------------------------G        Hoff (Stephen) Hoffman   OpenVMS Engineering   hoff[\0100]hp.com    ------------------------------  % Date: Thu, 15 Dec 2005 14:28:47 -0500 - From: William Webb <william.w.webb@gmail.com> @ Subject: Re: HP announces new Integrity Blade Servers - OpenVMS?I Message-ID: <8660a3a10512151128l6e00fbc9ib9645bb4a1f2ebf9@mail.gmail.com>   6 On 12/15/05, Jeff Cameron <roktsci@comcast.net> wrote:: > So HP has now announced the new Integrity Blade Servers: > L > http://h18004.www1.hp.com/products/servers/platforms/new11-01.html?jumpid= =3Dem  > _CTdec05_spot01  > G > They say it supports HPUX, Linux and Windows, but nothing of OpenVMS?  >  >   B It is my understanding that the entire blade concept was developed9 because Windows doesn't scale well when it moves into the  multiprocessor world...   2 I had the following exchange with our on-site FSE:  D Me:  "Do you know what the ultimate meaning of that rack jammed full of Blades is?"   FSE:  "No, what?"   = Me: "If that entire rack dies, you're in a WORLD of trouble."   
 (FSE laughs.)    WWWebb   --C NOTE: This email address is only used for noncommerical VMS-related  correspondence. C All unsolicited commercial email will be deemed to be a request for 8 services pursuant to the terms and conditions located at# http://bellsouthpwp.net/w/e/webbww/    ------------------------------    Date: 15 Dec 2005 11:59:04 -0800; From: "johnhreinhardt@yahoo.com" <johnhreinhardt@yahoo.com> @ Subject: Re: HP announces new Integrity Blade Servers - OpenVMS?C Message-ID: <1134676743.862722.114810@o13g2000cwo.googlegroups.com>    William Webb wrote: 8 > On 12/15/05, Jeff Cameron <roktsci@comcast.net> wrote:< > > So HP has now announced the new Integrity Blade Servers: > > P > > http://h18004.www1.hp.com/products/servers/platforms/new11-01.html?jumpid=em > > _CTdec05_spot01  > > I > > They say it supports HPUX, Linux and Windows, but nothing of OpenVMS?  > >  > >  > D > It is my understanding that the entire blade concept was developed; > because Windows doesn't scale well when it moves into the  > multiprocessor world...  >   4 One of the things I don't think Windows does well isD "compartmentization" (I know, I just now made that up), i.e. runningG multiple application  on the same machine.  I know it can happen on any B system, but Windows seems to be more likely to have problems.  ForE example, on a job I had a couple of years back a co-worker who was an B MSCE, etc and did a lot of playing with things at home had troubleF running Exchange (might have been RIS) on the same server that was hisE Domain Controller (Active Directory) (I think - the details are fuzzy F now).  Through research he discovered that Microsoft did not reccomendG such a configuration because, since they ran as Windows services, there C was no way to specifiy a startup order and therefore no way to make D sure that the Active Directory services started before Exchange did.D Maybe this was not the exact problem, but that's ths gist of it as IF understood it at the time.  If I remember right, he solved the problemE by using the Microsoft Virtual Server (beta releases) and put several C virtual machines on his one server and set them up each running one 
 application..    ------------------------------    Date: 15 Dec 2005 12:38:23 -0800! From: kenneth.randell@verizon.net @ Subject: Re: HP announces new Integrity Blade Servers - OpenVMS?C Message-ID: <1134679103.402150.279820@g14g2000cwa.googlegroups.com>    Jeff Cameron wrote: : > So HP has now announced the new Integrity Blade Servers: > N > http://h18004.www1.hp.com/products/servers/platforms/new11-01.html?jumpid=em > _CTdec05_spot01  > G > They say it supports HPUX, Linux and Windows, but nothing of OpenVMS?    Who's 'they'???   F When this was discussed at the HP Technology Forum, it was HP-UX only,G at least for the first release.  It's unlikely these blade server boxes 5 would ever run VMS native, but that's my own opinion.   A The blade servers supposedly will (does) support the HP Integrity F Virtual Machine, which I believe (for now) only runs HP-UX as well forG the guest OS.  Other VM OS support (Windows 2003 and Linux) comes later G (2006?), with VMS after that (it requires the Montecito chip).  The VMS 6 rolling road map points out some time 2007 time frame.   ------------------------------  % Date: Thu, 15 Dec 2005 16:02:14 -0500 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> @ Subject: Re: HP announces new Integrity Blade Servers - OpenVMS?, Message-ID: <43A1D9D1.88FAA360@teksavvy.com>   Re Blade servers.   H Don't blade servers require special drivers on Windows so it can use the2 special high speed interconnect between machines ?  G If that is the case, then VMS would also need to use special drivers to E allow clustering via that (new to VMS) interface. And that means some  VMS version in the future.  H Note that at the time HP announced this (a number of weeks ago), HP madeH a point that IA64 units could be mixed with 8086 based units in the sameH blade cabinet. So when people upgrade from IA64 to 8086, they can retain their blade infrastructure.   H And while, from a disaster recovery point of view, blades aren't useful,H they are useful for 7/24 operation since a cluster of VMS instances doesF allow you to make software upgrades on one node while other nodes keepH on servicing users. (and then reboot each node to use the new software).   ------------------------------  # Date: Fri, 16 Dec 2005 00:20:55 GMT # From: hoff@hp.nospam (Hoff Hoffman) @ Subject: Re: HP announces new Integrity Blade Servers - OpenVMS?/ Message-ID: <HLnof.747$xG5.61@news.cpqcorp.net>   [ In article <BFC6D88F.19046%roktsci@comcast.net>, Jeff Cameron <roktsci@comcast.net> writes: 9 :So HP has now announced the new Integrity Blade Servers:  ..F :They say it supports HPUX, Linux and Windows, but nothing of OpenVMS?  D   OpenVMS is not supported on these Itanium blades, and (as has beenE   stated in various customer meetings I've been in attendance at) the D   business management team is soliciting customer interest in adding+   OpenVMS support for Itanium-based blades.   C   Ask yourself not just if you want to see blade support in OpenVMS D   (a question that would receive an automatic "yes" answer from mostA   audiences), but particularly how many racks and how many blades >   are you looking to purchase for use with OpenVMS, of course.  N  ---------------------------- #include <rtfaq.h> -----------------------------K     For additional, please see the OpenVMS FAQ -- www.hp.com/go/openvms/faq N  --------------------------- pure personal opinion ---------------------------G        Hoff (Stephen) Hoffman   OpenVMS Engineering   hoff[\0100]hp.com    ------------------------------  % Date: Thu, 15 Dec 2005 19:57:45 -0500 2 From: "Stanley F. Quayle" <squayle@insight.rr.com>@ Subject: Re: HP announces new Integrity Blade Servers - OpenVMS?. Message-ID: <43A1CAB9.21232.57AD8AA@localhost>  + On 16 Dec 2005 at 0:20, Hoff Hoffman wrote: G > [...] particularly how many racks and how many blades are you looking / > to purchase for use with OpenVMS, of course.    @ A business case is essential, of course.  But I'd hate to see a B "crippled" Itanium box -- didn't anyone learn from the "crippled"  Alpha systems that DEC made?  . Gosh, a Multia is about the size of a blade...  
 --Stan Quayle  Quayle Consulting Inc.  
 ----------- Stanley F. Quayle, P.E. N8SQ  +1 614-868-1363 3 8572 North Spring Ct., Pickerington, OH  43147  USA 0 stan-at-stanq-dot-com       http://www.stanq.com) "OpenVMS, when downtime is not an option"    ------------------------------  % Date: Thu, 15 Dec 2005 20:02:38 -0500 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> @ Subject: Re: HP announces new Integrity Blade Servers - OpenVMS?, Message-ID: <43A21219.249DA587@teksavvy.com>   Hoff Hoffman wrote: E >   Ask yourself not just if you want to see blade support in OpenVMS F >   (a question that would receive an automatic "yes" answer from most >   audiences),   B Work should be focused on completing the port to the 8086 ASAP ;-)    B I assume you guys looked into supporting blades ? What is actually& involved at the VMS level to do this ?    H How versatile is the blade high speed interconnect ?  Could one use thisF for TCPIP traffic between a blade instance of VMS and a blade instanceA of Linux ? Or is it limited to SCS type of traffic ? (in whatecer   proprietary flavor for each OS).   ------------------------------    Date: 15 Dec 2005 20:20:57 -0600- From: Kilgallen@SpamCop.net (Larry Kilgallen) @ Subject: Re: HP announces new Integrity Blade Servers - OpenVMS?3 Message-ID: <bl1vSxTx6ts6@eisner.encompasserve.org>   c In article <43A1CAB9.21232.57AD8AA@localhost>, "Stanley F. Quayle" <squayle@insight.rr.com> writes: - > On 16 Dec 2005 at 0:20, Hoff Hoffman wrote: H >> [...] particularly how many racks and how many blades are you looking0 >> to purchase for use with OpenVMS, of course.  > B > A business case is essential, of course.  But I'd hate to see a D > "crippled" Itanium box -- didn't anyone learn from the "crippled"  > Alpha systems that DEC made?   Who said "crippled" ?   D "For use with (Open)VMS, of course." tells me (as if I did not know)D that HP is interested in how much additional hardware they will sell) by making VMS available on this hardware.   4 Is that such a bad question for them to care about ?   ------------------------------    Date: 15 Dec 2005 20:26:01 -0600- From: Kilgallen@SpamCop.net (Larry Kilgallen) @ Subject: Re: HP announces new Integrity Blade Servers - OpenVMS?3 Message-ID: <zxoK8otdTHcZ@eisner.encompasserve.org>   \ In article <43A21219.249DA587@teksavvy.com>, JF Mezei <jfmezei.spamnot@teksavvy.com> writes:  D > I assume you guys looked into supporting blades ? What is actually( > involved at the VMS level to do this ?  C I don't know this hardware (or any hardware much), but certainly it @ is considerably different from traditional systems that run VMS.F The hardware-specific code for I/O, etc. likely would require changes.  C Somebody has to set up an environment to test configurations of the 1 blades in the ways that customers would use them.   ? Itanium is supposed to be more strictly defined than Alpha from C what I read, but VMS was often better than Unix at finding hardware $ defects, which have to be addressed.  J > How versatile is the blade high speed interconnect ?  Could one use thisH > for TCPIP traffic between a blade instance of VMS and a blade instanceC > of Linux ? Or is it limited to SCS type of traffic ? (in whatecer " > proprietary flavor for each OS).   ------------------------------  % Date: Thu, 15 Dec 2005 19:56:36 -0800 ( From: Jeff Cameron <roktsci@comcast.net>@ Subject: Re: HP announces new Integrity Blade Servers - OpenVMS?0 Message-ID: <BFC77AF4.190BE%roktsci@comcast.net>  J On 12/15/05 1:02 PM, in article 43A1D9D1.88FAA360@teksavvy.com, "JF Mezei"% <jfmezei.spamnot@teksavvy.com> wrote:    > Re Blade servers.  > J > Don't blade servers require special drivers on Windows so it can use the4 > special high speed interconnect between machines ? > I > If that is the case, then VMS would also need to use special drivers to G > allow clustering via that (new to VMS) interface. And that means some  > VMS version in the future. > J > Note that at the time HP announced this (a number of weeks ago), HP madeJ > a point that IA64 units could be mixed with 8086 based units in the sameJ > blade cabinet. So when people upgrade from IA64 to 8086, they can retain > their blade infrastructure.  > J > And while, from a disaster recovery point of view, blades aren't useful,J > they are useful for 7/24 operation since a cluster of VMS instances doesH > allow you to make software upgrades on one node while other nodes keepJ > on servicing users. (and then reboot each node to use the new software).  C While I am not necessarily a Windows advocate, if you run a Windows K enterprise, blade servers can be quite valuable in DR when used with VMWare G (an EMC company www.VMWare.com). I have seen demonstrations where where J several windows servers failed over from one blade server rack to another,L even across large geographic distances with little interruption. VMWare also> supports Linux. The impressive part of VMSWare is not only theI virtualization of nodes, but the virtualization of network topology which 2 fails over with the failover of the virtual nodes.   ------------------------------  % Date: Thu, 15 Dec 2005 23:13:43 -0500 ' From: "Main, Kerry" <Kerry.Main@hp.com> @ Subject: RE: HP announces new Integrity Blade Servers - OpenVMS?R Message-ID: <FD827B33AB0D9C4E92EACEEFEE2BA2FB77368B@tayexc19.americas.cpqcorp.net>   > -----Original Message-----9 > From: JF Mezei [mailto:jfmezei.spamnot@teksavvy.com]=20 ! > Sent: December 15, 2005 4:02 PM  > To: Info-VAX@Mvb.Saic.Com B > Subject: Re: HP announces new Integrity Blade Servers - OpenVMS? >=20 > Re Blade servers.  >=20A > Don't blade servers require special drivers on Windows so it=20 
 > can use the 4 > special high speed interconnect between machines ? >=20A > If that is the case, then VMS would also need to use special=20  > drivers toG > allow clustering via that (new to VMS) interface. And that means some  > VMS version in the future. >=20? > Note that at the time HP announced this (a number of weeks=20  > ago), HP made A > a point that IA64 units could be mixed with 8086 based units=20 
 > in the same B > blade cabinet. So when people upgrade from IA64 to 8086, they=20 > can retain > their blade infrastructure.  >=20> > And while, from a disaster recovery point of view, blades=20 > aren't useful,> > they are useful for 7/24 operation since a cluster of VMS=20 > instances doesH > allow you to make software upgrades on one node while other nodes keepB > on servicing users. (and then reboot each node to use the new=20 > software). >=20  0 Re: blades .. From a purely speculative view :-)  G - Think of a rack full of blades (e.g. 100) and how many system disk it D takes.  [hint, at least 1 local disk for each blade with Windows andC Linux and most other UNIX's. Some OS's require local disks (esp for G clusters) as they do not yet support full booting over SAN] Think about C the effort to keep all of these systems in sync [hint tools need to F autocopy stuff all over the place. Each disk is a separate OS upgrade]  E With OpenVMS cluster this would be one system disk on the SAN for the E entire rack (well, two if you wanted higher availability with rolling A upgrades). Of course by one, I mean one HW partition on SAN. With @ OpenVMS, common files are updated once and change is immediately4 available to all systems because it's the same file.  C - Think about how a rack of blades would access common storage with & Windows/Linux. [hint - network access]  H With OpenVMS cluster on blades, each blade would do direct IO via shared file system and DLM.  D - think about how you would balance the batch jobs with rack full ofF blades. [hint - tough to do with Windows and Linux as there base batchA capabilities on one system is limited, never mind balancing batch " workloads across multiple systems]  H With OpenVMS cluster, submitting jobs to generic queue would then result= in job being executed on blade with most available cycles.=20   A Thinking about the above, perhaps HPTC, movies on demand, digital 1 theatres etc might be something to think about ..   0 Course, all of the above is pure speculation.=20   Regards   
 Kerry Main Senior Consultant  HP Services Canada Voice: 613-592-4660  Fax: 613-591-4477  kerryDOTmainAThpDOTcom (remove the DOT's and AT)=20  4 OpenVMS - the secure, multi-site OS that just works.   ------------------------------    Date: 15 Dec 2005 22:30:49 -0600- From: Kilgallen@SpamCop.net (Larry Kilgallen) @ Subject: RE: HP announces new Integrity Blade Servers - OpenVMS?3 Message-ID: <97UQh4TruJ9P@eisner.encompasserve.org>   | In article <FD827B33AB0D9C4E92EACEEFEE2BA2FB77368B@tayexc19.americas.cpqcorp.net>, "Main, Kerry" <Kerry.Main@hp.com> writes:  I > - Think of a rack full of blades (e.g. 100) and how many system disk it F > takes.  [hint, at least 1 local disk for each blade with Windows andE > Linux and most other UNIX's. Some OS's require local disks (esp for I > clusters) as they do not yet support full booting over SAN] Think about   C I hope nobody tells the HP Business folks about Kerry's notion that / VMS users of blades will spend less on disks...   E > the effort to keep all of these systems in sync [hint tools need to H > autocopy stuff all over the place. Each disk is a separate OS upgrade]    ...and fewer cycles on upgrades.  J > With OpenVMS cluster, submitting jobs to generic queue would then result? > in job being executed on blade with most available cycles.=20   I Is there a new queue manager coming ?  The existing one has traditionally G balanced job counts rather than CPU loading.  Attempting to balance CPU F loading without application input would seem dangerous, since load can vary over the life of a job.  C > Thinking about the above, perhaps HPTC, movies on demand, digital 3 > theatres etc might be something to think about ..   > I guess they get better movies up in Canada than down here :-)   ------------------------------  % Date: Fri, 16 Dec 2005 04:47:45 -0000 @ From: "Alex Daniels" <alexNOSPAMHERETHANKSdaniels@themail.co.uk>@ Subject: Re: HP announces new Integrity Blade Servers - OpenVMS?6 Message-ID: <43a246f0$0$23285$db0fefd9@news.zen.co.uk>  3 "Main, Kerry" <Kerry.Main@hp.com> wrote in message  L news:FD827B33AB0D9C4E92EACEEFEE2BA2FB77368B@tayexc19.americas.cpqcorp.net...H >- Think of a rack full of blades (e.g. 100) and how many system disk itE >takes.  [hint, at least 1 local disk for each blade with Windows and D >Linux and most other UNIX's. Some OS's require local disks (esp forH >clusters) as they do not yet support full booting over SAN] Think aboutD >the effort to keep all of these systems in sync [hint tools need toG >autocopy stuff all over the place. Each disk is a separate OS upgrade]  > F >With OpenVMS cluster this would be one system disk on the SAN for theF >entire rack (well, two if you wanted higher availability with rollingB >upgrades). Of course by one, I mean one HW partition on SAN. WithA >OpenVMS, common files are updated once and change is immediately 5 >available to all systems because it's the same file.   K I thought the (supported) limit on nodes in a cluster was 96 not 100, so I  M would need more than the 1 system disk you suggest, and in fact they couldnt  # be in the same (supported) cluster.    Alex     ------------------------------  % Date: Fri, 16 Dec 2005 00:00:02 -0500 ' From: "Main, Kerry" <Kerry.Main@hp.com> @ Subject: RE: HP announces new Integrity Blade Servers - OpenVMS?R Message-ID: <FD827B33AB0D9C4E92EACEEFEE2BA2FB77368E@tayexc19.americas.cpqcorp.net>   > -----Original Message-----9 > From: Larry Kilgallen [mailto:Kilgallen@SpamCop.net]=20 " > Sent: December 15, 2005 11:31 PM > To: Info-VAX@Mvb.Saic.Com B > Subject: RE: HP announces new Integrity Blade Servers - OpenVMS? >=20 > In article=20 @ > <FD827B33AB0D9C4E92EACEEFEE2BA2FB77368B@tayexc19.americas.cpqc5 > orp.net>, "Main, Kerry" <Kerry.Main@hp.com> writes:  >=20? > > - Think of a rack full of blades (e.g. 100) and how many=20  > system disk itH > > takes.  [hint, at least 1 local disk for each blade with Windows andG > > Linux and most other UNIX's. Some OS's require local disks (esp for B > > clusters) as they do not yet support full booting over SAN]=20
 > Think about  >=20E > I hope nobody tells the HP Business folks about Kerry's notion that 1 > VMS users of blades will spend less on disks...  >=20G > > the effort to keep all of these systems in sync [hint tools need to A > > autocopy stuff all over the place. Each disk is a separate=20 
 > OS upgrade]  >=20" > ...and fewer cycles on upgrades. >=20= > > With OpenVMS cluster, submitting jobs to generic queue=20  > would then result C > > in job being executed on blade with most available cycles.=3D20  >=20@ > Is there a new queue manager coming ?  The existing one has=20 > traditionally @ > balanced job counts rather than CPU loading.  Attempting to=20
 > balance CPU H > loading without application input would seem dangerous, since load can > vary over the life of a job. >=20  ' Like I said all pure speculation .. :-)   E > > Thinking about the above, perhaps HPTC, movies on demand, digital 5 > > theatres etc might be something to think about ..  >=20@ > I guess they get better movies up in Canada than down here :-) >=20  G Yeah, they discovered that electrons move quicker in the cold, hence it < makes it cheaper to download movies here. It was part of theE justification for laying high speed fibre across most of the country.    :-)    Regards,  
 Kerry Main Senior Consultant  HP Services Canada Voice: 613-592-4660  Fax: 613-591-4477  kerryDOTmainAThpDOTcom (remove the DOT's and AT)=20  4 OpenVMS - the secure, multi-site OS that just works.   ------------------------------  % Date: Thu, 15 Dec 2005 23:57:07 -0500 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> @ Subject: Re: HP announces new Integrity Blade Servers - OpenVMS?, Message-ID: <43A24900.69E109B8@teksavvy.com>   Larry Kilgallen wrote:F > "For use with (Open)VMS, of course." tells me (as if I did not know)F > that HP is interested in how much additional hardware they will sell+ > by making VMS available on this hardware.  > 6 > Is that such a bad question for them to care about ?   Yes.  = It is about vision and it is about leadership in the market.    G Did businesses knock on microsoft's doors to ask for flawed software to H run their enterprises ? No. Gates decided he wanted that market and wentH for it and shoved Wintel crap down enterprises throaths despite it beingE an illogical choice. And now he's making money from enterprise sales.   H HP has developped the hardware. It is there. HP has spent money on HP-UXG and Linux to have those run on those Blades. The announcement says "but G we dodn't think VMS is important enough". That is a very bad message to N send to a product that has been struggling for credibility in the marketplace.  F While Gartner may have been wrong all those years, they have succeededD in killing VMS in the marketplace and the various owners of VMS have% acted as if it had already happened.      7 Did HP see enough sales potential for HP-UX on Blades ?   C If HP doesn't see sufficient sales of VMS on Blades, it doesn't see ; sufficient sales of VMS. And that is very bad news for VMS.   H One shouldn't look at the number of existing VMS customers who would buy< Blades. One needs to look at the market potential for VMS inD small/medium size applications. It is a huge market which VMS is not participating in anymore.     B I can understand Tandem not participating in regular Blades due to/ Tandem meaning special fault tolerant hardware. F  But VMS was ported to that IA64 thing so it could benefit from all ofB the IA64 servers produced by its owners.  (And ironically, I thinkD tandem would benefit from Blades because its concept is to have many0 CPUs with applications running on multiple CPUs   G In the original announcement, HP mentioned that customers could mix and C match IA64 and 8086 blade servers in the same enclosure. On the one E hand, you can implement IA64 solutions by plugging in a single server E into an existing 8086 blade cabinet. On the other hand, buying a IA64 F blade system allows you to upgrade to 8086 servers one at a time until! you are rid of those IA64 things.   F That decision alone isn't enough to make conclusions about HP and VMS.H But when you look at the big pictures and all the small things that haveF happened since Sept 2001 when HP announced it was buying Compaq, thereC is a clear picture emerging: HP isn't killing VMS actively, it will G support customers, but it ain't about to spend a single penny to revive A VMS and fully leverage its full potential for growth, success and M profit. In other words, HP has decided to let VMS slowly evaporate over time.   @  It is bad enough that HP/Intel widthdrew IA64 from large marketG segments. But now, VMS can't even participate in all of what is left of G the IA64 market. That is a very bad signal to send to customers. And of F course, the concept of "potential customers" is becoming foreign sinceF HP has yet to begin to market VMS despite VMS now running on true blue HP hardware.    G There is one good aspect of having a small customer base: VMS engineers G can really have closer relationship with its customers and respond more G directly to their needs. But doing so does not necessarily allow VMS to D get back in the game because the remaining customer base is focuselyH narrowed to a few select market niches and their needs may not represent2 needs of potential customers outside those niches.   ------------------------------  % Date: Thu, 15 Dec 2005 19:42:43 -0500 " From: "Hal Kuff" <kuff@tessco.com> Subject: Jon Andruszkiewicz 0 Message-ID: <11q43c4oi8scma7@corp.supernews.com>  K Hi, if you are out there I would like to continue an earlier discussion on  % long distance shadows and journals...    kuff at tessco dot com     ------------------------------  % Date: Thu, 15 Dec 2005 15:57:07 -0500 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> 4 Subject: Re: Location of initial page and swap files, Message-ID: <43A1D89E.3EAFAEE8@teksavvy.com>  
 Many said:  I > PAge, swap and dump file are hard coded and accessed very early, before G > the file system is full up and running. If you boot without them, you 4 > can see the "Failed to map <name>" on the console.    A Thanks for confirmation. I saw the messages early in the boot and D realised that satellite_page.com wouldn't work since those page/swap files were already installed.   H However, at the time the messages are issued by sysboot, VMS has already? read the sysgen parameters, so I figured that perhaps there was @ something configured in sysgen that I couldn't see. But if it is- hardcoded, then I can't worry about it much !      Paul Repacholi wrote:   K > Very carfully... BTW, i use the file name AAA<SCSID>.PAGE|SWAP and put it G > in [000000].  This automajikally sticks it right in the centre of the 3 > disk next to the index file on an /image restore.     A Does putting a file in [000000] result in physical file placement ' difference versus any other directory ?    ------------------------------  % Date: Thu, 15 Dec 2005 10:35:52 -0800 ( From: Jeff Cameron <roktsci@comcast.net>% Subject: Re: monitoring VMS log files 0 Message-ID: <BFC6F788.19059%roktsci@comcast.net>   On 12/15/05 8:25 AM, in article D 1134663926.819306.74400@g49g2000cwa.googlegroups.com, "Ken Robinson" <kenrbnsn@gmail.com> wrote:    >  > tumblindice wrote: >> Hi, >>  E >>    We would like to monitor our VMS logfiles, operator.log and the H >> security files to a machine that could send out email to notify us ofI >> discrepancies. There are a few tools out there for UNIX but I have not ? >> seen anything for VMS. Any help or pointers would be greatly  >> appreciated. Thank you. > I > There are a number of products like that. Two that come to mind are the H > Unicenter Console Management for OpenVMS from CA and ConsoleWorks from. > TECsys Developement <http://www.tditx.com/>. >  > Ken  > I Both the OPCOM and The Security Auditor have a calling interface allowing ) your process to intercept these messages.    ------------------------------  % Date: Thu, 15 Dec 2005 19:27:23 -0500 % From: BRAD <bradhamilton@comcast.net> " Subject: Re: No more SETI on VMS ?* Message-ID: <43A209EB.6090603@comcast.net>   Robert Alan Byer wrote:  > $ > Yes, there is a BONIC SETI client. > G > I'm making progress, I'm hoping to have something working in the next  > few weeks. > I > Once I get my WWW server back up (waiting for new SCSI cable) I'll post  > what I have on my WWW site :)   " Thanks for taking on this project.   How is the client coming along?    ------------------------------    Date: 15 Dec 2005 19:20:26 -0800$ From: "Bob Armstrong" <bob@jfcl.com>K Subject: PHONE error - Invalid specification of node or person.  Try again. B Message-ID: <1134703226.736240.23520@f14g2000cwb.googlegroups.com>  G When you say, for example, "PHONE CODA::BOB", what does the PHONE error   F                   Invalid specification of node or person.  Try again.  G really mean?  CODA is a real DECnet node name, known to the local host, ; and the user BOB really exists on CODA.  If you look at the F NETSERVER.LOG in SYS$SPECIFIC:[PHONE$SERVER] on CODA, it just says the  same thing.  Not very helpful...  E   I suspect it's some kind of configuration or setup error, because I F get it when phoning between some systems but not others.  Both systems! are OVMS 7.3 and DECnet Phase IV.    Thanks,  Bob    ------------------------------    Date: 15 Dec 2005 21:52:23 -0600- From: Kilgallen@SpamCop.net (Larry Kilgallen) O Subject: Re: PHONE error - Invalid specification of node or person.  Try again. 3 Message-ID: <g+tqoZBDNFvl@eisner.encompasserve.org>   i In article <1134703226.736240.23520@f14g2000cwb.googlegroups.com>, "Bob Armstrong" <bob@jfcl.com> writes: I > When you say, for example, "PHONE CODA::BOB", what does the PHONE error  > H >                   Invalid specification of node or person.  Try again. >  > really mean?  B It can mean that the node your starting node thinks is CODA:: doesC not believe that its own name is CODA:: (via executive mode logical # name SYS$NODE in table LNM$SYSTEM).   ; > CODA is a real DECnet node name, known to the local host,   : But is it surely known to CODA:: that CODA:: is the name ?; That is the only failure possibility that I see, given your , other attestations of correct configuration.  = > and the user BOB really exists on CODA.  If you look at the H > NETSERVER.LOG in SYS$SPECIFIC:[PHONE$SERVER] on CODA, it just says the" > same thing.  Not very helpful... > G >   I suspect it's some kind of configuration or setup error, because I H > get it when phoning between some systems but not others.  Both systems# > are OVMS 7.3 and DECnet Phase IV.    ------------------------------    Date: 15 Dec 2005 20:10:22 -0800$ From: "Bob Armstrong" <bob@jfcl.com>N Subject: Re: PHONE error - Invalid specification of node or person. Try again.C Message-ID: <1134706222.495175.229830@g14g2000cwa.googlegroups.com>      Ah, that's it -    $sho log sys$node '    "SYS$NODE" = "::" (LNM$SYSTEM_TABLE)   ;   Next question - where is SYS$NODE supposed to be defined?    Bob    ------------------------------    Date: 15 Dec 2005 22:20:03 -0600- From: Kilgallen@SpamCop.net (Larry Kilgallen) N Subject: Re: PHONE error - Invalid specification of node or person. Try again.3 Message-ID: <RfInryjRLNW8@eisner.encompasserve.org>   j In article <1134706222.495175.229830@g14g2000cwa.googlegroups.com>, "Bob Armstrong" <bob@jfcl.com> writes: >  >   Ah, that's it -  >  > $sho log sys$node ) >    "SYS$NODE" = "::" (LNM$SYSTEM_TABLE)  > = >   Next question - where is SYS$NODE supposed to be defined?   C The DECnet node name should be in system parameter SCSNODE and also A in the DECnet database (exactly where depends on which DECnet you  are using).   M I do not know which one DECnet startup uses when it creates the logical name.    ------------------------------    Date: 15 Dec 2005 20:16:25 -0800$ From: "Bob Armstrong" <bob@jfcl.com>N Subject: Re: PHONE error - Invalid specification of node or person. Try again.B Message-ID: <1134706585.049759.39290@g49g2000cwa.googlegroups.com>  G Never mind - it looks like the executor node name wasn't defined in the  DECnet database.   $NCP DEFINE EXEC NAME CODA  D   takes care of it.  It probably got screwed up when I had to change" the DECnet address of the machine.   Thanks!  Bob    ------------------------------  % Date: Thu, 15 Dec 2005 16:15:29 -0500 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> / Subject: R400X: converting DSSI shelves to SCSI , Message-ID: <43A1DCEA.96A26C14@teksavvy.com>  C This week, I adopted a stray R400X with 5 drives (4*2gig, 1*1gig).    G Those drives are noisy and generate heat and eat electricity. (they are " the big 5.25" full height format).  H The specs for the R400X mention that the backplane is actually 3 layers:F power, SCSI and DSSI.  They also mention that it is possible to insertD SCSI drives next to DSSI drives, with the SCSI drives picking up theG signals frm the SCSI signals and the DSSI drives from the DSSI signals.  Neat design.  F Tough question: if I were to cannabalise the 1 gig DSSI drive, could IF reuse the mounting aparatus and get it to feed me SCSI signals ? (i.e:H does the connector into the backplane get all signals and then just feed the DSSI stuff to the drive ?   A I know that the mounting bracket provides power to the drive in a  standard power plug.    H Failing this, I could always string my own ribbon cable in the backplane+ and use that, but it wouldn't be as "neat".   E Also, are DSSI drives hot removable/replaceable  from the cabinet, or  must it be powered down ?     G My goal to keep my cluster up and web server running while I move to my B newsly acquired machines is proving more difficult than I though.   E Also, does anyone know if there are QBUS SCSI interfaces which allow  " clustering/dual access to drives ?  H Right now, the advantage of the DSSI drives is that I can get 2 vaxes toF access them directly. Just need to make sure that the DSSI controllersE have different BUS IDs. (and I think some SCS traffic travels through  the DSSI, right ?)  D Any chance I could do this with 1992 vintage SCSI controllers on the, QBUS ?  (2 vaxes accessing the same drives).   ------------------------------  + Date: Thu, 15 Dec 2005 21:48:30 +0000 (UTC) 7 From: moroney@world.std.spaamtrap.com (Michael Moroney) 3 Subject: Re: R400X: converting DSSI shelves to SCSI ( Message-ID: <dnsobe$rlv$1@pcls4.std.com>  / JF Mezei <jfmezei.spamnot@teksavvy.com> writes:   D >This week, I adopted a stray R400X with 5 drives (4*2gig, 1*1gig).  ... F >Also, does anyone know if there are QBUS SCSI interfaces which allow # >clustering/dual access to drives ?   E QBUS implies VAX, and multihost SCSI has never been supported on VAX, " and I don't think it works at all.  I >Right now, the advantage of the DSSI drives is that I can get 2 vaxes to G >access them directly. Just need to make sure that the DSSI controllers F >have different BUS IDs. (and I think some SCS traffic travels through >the DSSI, right ?)    Yes, SCS can and will use DSSI.   E >Any chance I could do this with 1992 vintage SCSI controllers on the - >QBUS ?  (2 vaxes accessing the same drives).    As stated, unlikely.   ------------------------------  % Date: Thu, 15 Dec 2005 15:59:11 -0800 % From: J.C. Roberts <unknown@abac.com>  Subject: Re: SRM memtest8 Message-ID: <eru3q1hlm4g970icg4tgugu5bd5hn93bsk@4ax.com>  F On Wed, 14 Dec 2005 21:05:27 GMT, hoff@hp.nospam (Hoff Hoffman) wrote:  a >In article <5s90q1pb9q75o5jj7hjhgdsu4nfd43ugl9@4ax.com>, J.C. Roberts <unknown@abac.com> writes: J >:So the system passed it's memory tests and all was well until I rebootedE >:the system. This put me into AlphaBIOS/ARC for some strange reason.  >.. E >:For some strange reason I ended up in AlphaBIOS/ARC again? This was H >:weird so I did the steps again, cold booted again, and sure enough, it( >:_still_ came up in AlphaBIOS/ARC mode? >  > I >  The usual trigger for this behaviour -- reverting to the AlphaBIOS/ARC G >  console selection -- is a failed or fully discharged battery on the  G >  BB_WATCH chip, or an indication that something stepped on the NVRAM, G >  as that is where the system information and the settings are stored.  > H >  In retrospect, it would have been better to have selected a code for H >  the AlphaBIOS/ARC console that differed not only from the SRM consoleH >  but also from the results of a discharged or a null-erased NVRAM, andF >  to have had the console perform some sort of simple verification ofF >  the data it finds stored in NVRAM.  But that opportunity for change >  is now long past. > H >  The battery on most (all?) VAX, Alpha and Integrity systems is a NiCdG >  pack or a lithium cell, and it does eventually fail.  AFAIK, either  H >  the battery pack can be replaced (when there is an external battery),G >  or the BB_WATCH NVRAM chip itself (which has an integrated battery)  $ >  is socketed, and can be replaced. > E >  If the system has been powered up for a couple of days and it does D >  not retain its settings over a subsequent power-cycle, then it isE >  likely the battery has failed.  If so, you'll probably also end up F >  required to enter the system time, too, since that's typically also' >  maintained by the same power source.  > O > ---------------------------- #include <rtfaq.h> ----------------------------- L >    For additional, please see the OpenVMS FAQ -- www.hp.com/go/openvms/faqO > --------------------------- pure personal opinion --------------------------- H >       Hoff (Stephen) Hoffman   OpenVMS Engineering   hoff[\0100]hp.com    G It seems I found a third way to trigger the reset to AlphaBIOS/ARC from $ the SRM Console on this PSW-433 box.     >>>runecu     H After further research it seems that memtest is actually broken in 7.2-1B on the PSW-433. Regardless of what I do, it steadfastly refuses to accept the -sa switch.  H memtest runs fine with the -rb switch since this switch tells memtest to ignore allocation errors.   9 But the -sa switch always results in an allocation error.      >>>memtest -sa 200000 D   unable to allocate memory of length 2000 at starting address 2f7e0  F Using the memtest -z switch only crashes the system, usually resultingG in a reboot. Regardless of the value assigned (hex) to the -sa starting F address switch, I get generally the same error message. The only thing; that sometimes differs in the error message is the address.      >>>dynamic -r C   zone       zone    used     used    free   free      utili-  high D   address    size    blocks   bytes   blocks bytes     zation  waterH   -------- --------- ------ --------- ------ --------- ------  ---------H   00038128    720832    352    287232     23    433623    39%     315072H   00030860 266551296      1        32      1 266551296     0%          0    G The overlaping of the two memory ranges is a bit odd, well, technically  more than a bit. ;-)   The system has 256MB of memory: #   256MB == 0x0FFFFFFF  == 268435456    Do a little math: ,   268435456 - 266551296 = 1884160 = 0x1CC000  F So it seems like a -sa 200000 should clearly be out of range. From oneD of the esoteric docs I've found avme5um.pdf/EK-VME54-UM it seems theG "real" starting address is what you supply minus 32 bytes for the alloc  headers.  : If anyone knows how to deal with this, please let me know.   Thanks,  JCR    ------------------------------  # Date: Fri, 16 Dec 2005 00:09:00 GMT # From: hoff@hp.nospam (Hoff Hoffman)  Subject: Re: SRM memtest0 Message-ID: <wAnof.746$xG5.283@news.cpqcorp.net>  Y In article <11q1kepgr2l051@corp.supernews.com>, Dave Froble <davef@tsoft-inc.com> writes:  :Robert Deininger wrote:  I :> No.  Source code licenses are available to developers, but they aren't F :> particularly cheap.  And you don't get the source code, you get theK :> compiler listings.  Some proprietary stuff is censored from the listings K :> kit.  In particular, machine-specific code for Alpha systems tends to be  :> censored.  C   Listings are available for purchase, and are arguably often as or G   more useful for their intended purpose.  Source code and particularly -   a license for OpenVMS itself is quote rare.   H :Is there any particular reason for censoring the machine specific code?  I   Yes, there is a reason.  The bulk of the code involved is not generally I   sensitive, but there is one machine-specific piece that does tend to be    considered to be sensitive.   5 :>>Is the source code for the SRM firmware available?  :>  I :> AFAIK, it was never available outside Digital/Compaq/HP, and was never ' :> widely available inside the company.  : I :Is there a reason this software could not be available to customers and   :hobbyists?   F   SRM has traditionally been provided by the Alpha hardware engineers,A   and is largely intended and treated as classic embedded-device  D   firmware.  This SRM firmware is in support of OpenVMS, of course, A   and effectively provides extensions to Alpha that are necessary ;   for OpenVMS operations within the current Alpha hardware.   D   In current times, the reasons to restrict SRM are rather less, butC   there are still areas of the firmware that would be considered to    be sensitive.   C   OpenVMS I64 has transitioned most of the tasks that are performed <   by SRM firmware on Alpha directly into OpenVMS, of course.  B   And yes, I am being somewhat deliberately vague here.  (Is that @   statement doubly ambiguous? :-)   The two questions posed here>   are actually rather more related than might initially appear   to be the case, too.    N  ---------------------------- #include <rtfaq.h> -----------------------------K     For additional, please see the OpenVMS FAQ -- www.hp.com/go/openvms/faq N  --------------------------- pure personal opinion ---------------------------G        Hoff (Stephen) Hoffman   OpenVMS Engineering   hoff[\0100]hp.com    ------------------------------  % Date: Thu, 15 Dec 2005 18:04:55 -0600 % From: Dan Foster <usenet@evilphb.org>  Subject: Re: SRM memtest5 Message-ID: <slrndq4157.40u.usenet@zappy.catbert.org>   U In article <wAnof.746$xG5.283@news.cpqcorp.net>, Hoff Hoffman <hoff@hp.nospam> wrote:  > I >:Is there any particular reason for censoring the machine specific code?  > K >   Yes, there is a reason.  The bulk of the code involved is not generally K >   sensitive, but there is one machine-specific piece that does tend to be  >   considered to be sensitive.   . To whomever (Mr. Froble?) asked that question:  C I can imagine a scenario where had that information been available, D people could buy expensive server hardware then hardcode a differentF model number -- that would then allow less ethical people to load much9 cheaper licenses, and HP would be out some serious money.   F E.g. buying a server, and hardcoding it to report as a workstation. Or< even as a lower priced licensing tier within a server class.  G Or another hypothetical situation, also with less ethical people, where G they defeat application-specific licensing by changing hardware details H on a second (or more) machine instead of purchasing additional licenses.  D These kind of issues might be less of a concern if HP electronicallyH burnt fuses to hardcode certain information in chips at the factory, but7 was probably judged as being impractical and expensive.    -Dan   ------------------------------  % Date: Fri, 16 Dec 2005 03:28:50 +0800  From: prep@prep.synonet.com  Subject: Re: SRM memtest- Message-ID: <87fyoup3bh.fsf@prep.synonet.com>   7 rdeininger@mindspringdot.com (Robert Deininger) writes:   B > The SRM console should accept any compliant I/O card. (Has thereF > ever been a compliant I/O card?)  It will only boot from a supported@ > card (or a close relative).  It shouldn't choke, except in theD > unfortunate cases where an unsupported card looks like one the SRM& > knows, but does behave the same way.  B I have a solid barf case. Plug a TGA2 twin card video into a 500auD and it stops solidly before SRM is even up... I wish I could remeber> the alphabet soup name of those things... It's the 24 bit one.   --  < Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda. @                                              West Australia 6076* comp.os.vms,- The Older, Grumpier Slashdot. Raw, Cooked or Well-done, it's all half baked.F EPIC, The Architecture of the future, always has been, always will be.   ------------------------------  % Date: Thu, 15 Dec 2005 16:48:40 -0800 # From: "Tom Linden" <tom@kednos.com>  Subject: Re: SRM memtest( Message-ID: <ops1upvemfzgicya@hyrrokkin>  J On Thu, 15 Dec 2005 18:04:55 -0600, Dan Foster <usenet@evilphb.org> wrote:  A > In article <wAnof.746$xG5.283@news.cpqcorp.net>, Hoff Hoffman    > <hoff@hp.nospam> wrote:  >>K >> :Is there any particular reason for censoring the machine specific code?  >>D >>   Yes, there is a reason.  The bulk of the code involved is not   >> generallyK >>   sensitive, but there is one machine-specific piece that does tend to    >> be   >>   considered to be sensitive. > 0 > To whomever (Mr. Froble?) asked that question: > E > I can imagine a scenario where had that information been available, F > people could buy expensive server hardware then hardcode a differentH > model number -- that would then allow less ethical people to load much; > cheaper licenses, and HP would be out some serious money.  > H > E.g. buying a server, and hardcoding it to report as a workstation. Or> > even as a lower priced licensing tier within a server class. > I > Or another hypothetical situation, also with less ethical people, where I > they defeat application-specific licensing by changing hardware details J > on a second (or more) machine instead of purchasing additional licenses. > F > These kind of issues might be less of a concern if HP electronicallyJ > burnt fuses to hardcode certain information in chips at the factory, but9 > was probably judged as being impractical and expensive.  >  > -Dan  G Reverse engineering of SRM, at least a portion of it, was done long ago J by Wolfgang Moeller when he did the images for estending the size of the   system disk on the 3100's   ------------------------------  % Date: Thu, 15 Dec 2005 19:20:54 -0600 % From: Dan Foster <usenet@evilphb.org>  Subject: Re: SRM memtest5 Message-ID: <slrndq45jm.40u.usenet@zappy.catbert.org>   D In article <ops1upvemfzgicya@hyrrokkin>, Tom Linden <tom@kednos.com> wrote: > E > Reverse engineering of SRM, at least a portion of it, was done long G > ago by Wolfgang Moeller when he did the images for estending the size # > of the  system disk on the 3100's   A Hmm. The 3100s had SRM? My impression is that SRM was Alpha-only?   E (Herr Moeller did do a fine job with that one, from my recollection.)   F But, yes, point taken. Things can indeed usually be reverse engineered= without source listings if one is sufficiently desperate. :-)   E And sometimes, things can be changed (e.g. model type) via SRM if one G knows what knobs to use -- someone posted a specific command to do that 5 in the case of the XP900 and DS10 just the other day.    -Dan   ------------------------------  % Date: Fri, 16 Dec 2005 01:42:38 -0000 @ From: "Alex Daniels" <alexNOSPAMHERETHANKSdaniels@themail.co.uk> Subject: Re: SRM memtest6 Message-ID: <43a21b8e$0$29560$da0feed9@news.zen.co.uk>  1 "Hoff Hoffman" <hoff@hp.nospam> wrote in message  * news:k5pof.753$YE5.678@news.cpqcorp.net...F >no VAX system I am aware of uses the Alpha SRM console as part of its >  system console.  M The VAX7000 (which also could be board swapped to an Alpha 8x00), had a very  6 SRM like console, at least in terms of it's interface.  F http://h18002.www1.hp.com/alphaserver/vax/download/ek-70cob-tm-201.pdf  M That link is to the manual for the VAX7000's console, and it even references  ( the Alpha Architecture Reference Manual.   Alex     ------------------------------  # Date: Fri, 16 Dec 2005 02:29:21 GMT # From: hoff@hp.nospam (Hoff Hoffman)  Subject: Re: SRM memtest0 Message-ID: <5Epof.754$0J5.573@news.cpqcorp.net>  y In article <43a21b8e$0$29560$da0feed9@news.zen.co.uk>, "Alex Daniels" <alexNOSPAMHERETHANKSdaniels@themail.co.uk> writes: N :The VAX7000 (which also could be board swapped to an Alpha 8x00), had a very 7 :SRM like console, at least in terms of it's interface.   D   Ayup, there are similarities between the VAX 7000 and the DEC 7000D   series and there was a path to move from the VAX processor(s) over>   to Alpha processor(s) within the enclosure, and there was an@   evolutionary progression from the DEC 7000 series up into the 9   AlphaServer 8200 and AlphaServer 8400 series platforms.   D   The VAX 7000 series development was operating in parallel with theD   Alpha effort, and the so-called Laser and Blazer platforms used onD   the VAX 7000 and DEC 7000 series were (obviously) common, as were D   many (but not all) of the peripheral I/O widgets for the platform.  D   The VAX 7000 console wasn't SRM, however.  (The command interface D   is just the tip of the iceberg, too, as there is a whole lot more E   to SRM than most folks realize -- both commands and utilities, and  A   the more obvious tie-in with the PALcode interfaces used by the '   host operating system, are involved.)    	--   B   VAX consoles didn't have any particular central design referenceC   or command specification, where the Alpha consoles tended to have B   various basic parts included in a central reference -- there wasB   far more variability between arbitrary VAX consoles than between@   arbitrary Alpha SRM console implementations.  (You can see theA   evolution of consoles as you look across VAX platforms, and you A   can see how these evolved into the SRM console.  There are some @   similarities between the SRM console and the EFI console, too,<   despite having completely and entirely different origins.)    N  ---------------------------- #include <rtfaq.h> -----------------------------K     For additional, please see the OpenVMS FAQ -- www.hp.com/go/openvms/faq N  --------------------------- pure personal opinion ---------------------------G        Hoff (Stephen) Hoffman   OpenVMS Engineering   hoff[\0100]hp.com    ------------------------------  % Date: Fri, 16 Dec 2005 00:31:57 -0500 ' From: Dave Froble <davef@tsoft-inc.com>  Subject: Re: SRM memtest0 Message-ID: <11q4k7bmak1gb78@corp.supernews.com>   Dan Foster wrote: W > In article <wAnof.746$xG5.283@news.cpqcorp.net>, Hoff Hoffman <hoff@hp.nospam> wrote:  > J >>:Is there any particular reason for censoring the machine specific code? >>K >>  Yes, there is a reason.  The bulk of the code involved is not generally K >>  sensitive, but there is one machine-specific piece that does tend to be  >>  considered to be sensitive.  >  > 0 > To whomever (Mr. Froble?) asked that question: > E > I can imagine a scenario where had that information been available, F > people could buy expensive server hardware then hardcode a differentH > model number -- that would then allow less ethical people to load much; > cheaper licenses, and HP would be out some serious money.  > H > E.g. buying a server, and hardcoding it to report as a workstation. Or> > even as a lower priced licensing tier within a server class. > I > Or another hypothetical situation, also with less ethical people, where I > they defeat application-specific licensing by changing hardware details J > on a second (or more) machine instead of purchasing additional licenses. > F > These kind of issues might be less of a concern if HP electronicallyJ > burnt fuses to hardcode certain information in chips at the factory, but9 > was probably judged as being impractical and expensive.  >  > -Dan   Good perspective.   C My question came from the perspective of being able to learn about  @ console systems and such.  Secondary would be the capability of , modifications to support additional devices.  G As for licenses, there are methods for circumventing such.  Those also  * would be what Steve considers 'sensitive'.   --  4 David Froble                       Tel: 724-529-04504 Dave Froble Enterprises, Inc.      Fax: 724-529-0596> DFE Ultralights, Inc.              E-Mail: davef@tsoft-inc.com 170 Grimplin Road  Vanderbilt, PA  15486    ------------------------------    Date: 15 Dec 2005 14:40:30 -0600- From: Kilgallen@SpamCop.net (Larry Kilgallen) ! Subject: Re: Storageworks shelves 3 Message-ID: <SBoziXkzvfIf@eisner.encompasserve.org>    In article <Bieof.21861$Y72.1770@bignews1.bellsouth.net>, "David Turner, Island Computers US Corp" <dbturner@icusc.com> writes: 2 > Perhaps someone can explain to me the following: > E > Why, when SCSI was implemented, did it have the 7 drive limitation?  > Why not 16 or 32?  > Of course UW> is 14 but why? > J > Did someone in the SCSI original design team think that noone would ever > need more than 7 drives?  B No, they thought (correctly) that multiple SCSI busses were viable on a single system.    ------------------------------   Date: 15 Dec 2005 21:34:36 GMT( From: bill@cs.uofs.edu (Bill Gunshannon)! Subject: Re: Storageworks shelves , Message-ID: <40e5rcF19guhdU1@individual.net>  3 In article <SBoziXkzvfIf@eisner.encompasserve.org>, 0 	Kilgallen@SpamCop.net (Larry Kilgallen) writes: > In article <Bieof.21861$Y72.1770@bignews1.bellsouth.net>, "David Turner, Island Computers US Corp" <dbturner@icusc.com> writes: 3 >> Perhaps someone can explain to me the following:  >>  F >> Why, when SCSI was implemented, did it have the 7 drive limitation? >> Why not 16 or 32? >> Of course UW> is 14 but why?  >>  K >> Did someone in the SCSI original design team think that noone would ever  >> need more than 7 drives?  > D > No, they thought (correctly) that multiple SCSI busses were viable > on a single system.   J But multiple SCSI busses doesn't have anything to do with the 7(15) deviceJ limit.   It has to do with the number of bits allocated for addressing andJ the real question is why did they only allow 3 originally and after seeing@ the way things were going, why only increase it by one more bit?   bill   --  J 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: Thu, 15 Dec 2005 15:51:23 -0600 (CST) * From: sms@antinode.org (Steven M. Schweda)! Subject: Re: Storageworks shelves 2 Message-ID: <05121515512367_2021CD8C@antinode.org>  ( From: bill@cs.uofs.edu (Bill Gunshannon)  L > But multiple SCSI busses doesn't have anything to do with the 7(15) deviceL > limit.   It has to do with the number of bits allocated for addressing andL > the real question is why did they only allow 3 originally and after seeingB > the way things were going, why only increase it by one more bit?  G    Perhaps someone thought that a bus with 64K disks on it might suffer 7 a little from congestion.  Not to mention cable length.   H ------------------------------------------------------------------------  4    Steven M. Schweda               (+1) 651-699-98183    382 South Warwick Street        sms@antinode-org     Saint Paul  MN  55105-2547    ------------------------------   Date: 16 Dec 2005 00:02:51 GMT' From: ruben@CCNMR.MIT.EDU (David Ruben) ! Subject: Re: Storageworks shelves ? Message-ID: <43a2042b$0$580$b45e6eb0@senator-bedfellow.mit.edu>    Bill Gunshannon wrote:  K >But multiple SCSI busses doesn't have anything to do with the 7(15) device K >limit.   It has to do with the number of bits allocated for addressing and K >the real question is why did they only allow 3 originally and after seeing A >the way things were going, why only increase it by one more bit?   M The SCSI protocol allows multiple devices to be bus masters. Therefore, there M must be a mechanism for arbitrating control of the bus. The mechanism defined K calls for each device requesting the bus to drive one of the data bus lines O at an appropriate time. The line driven by each device is the one corresponding L to its address. The higher order lines are monitored and the device which isM driving the highest order line wins. Since narrow SCSI has 8 data lines there L can be no more that 8 devices on that bus. Since wide SCSI has 16 data linesJ there can be no more than 16 devices on that bus. The controller counts asK one device and, I believe, one data line is lost for addressing on the wide H bus in order to allow narrow and wide devices to correctly interoperate.   D.R.     ------------------------------  % Date: Fri, 16 Dec 2005 03:44:04 +0800  From: prep@prep.synonet.com ! Subject: Re: Storageworks shelves - Message-ID: <87bqzip2m3.fsf@prep.synonet.com>   * bill@cs.uofs.edu (Bill Gunshannon) writes:  - > In article <dnq3pl1c2k@enews2.newsguy.com>,  > 	healyzh@aracnet.com writes:, >> Bill Gunshannon <bill@cs.uofs.edu> wrote:  D >>> No "Personality Module" in either of them. Just drives and powerD >>> supplies.  Behind the fans are the terminator and the connector.F >>> Bothe are configured for single bus and exert SHELF_OK (although I- >>> suspect the PDP doesn't care about that.)    C >> A BA350 doesn't use a personality module, those are for BA356's.   F > Aren't there "Personality Modules" that plug into the top slot whichC > is much narrower than the rest and has the two SCSI connectors in  > the back of it?   A The 350 has the personality module you have when you don't have a B personality. ;) It is just a pair of double Honda-50s so you don't> have to go bus diving to connect it up. No smarts at all, just mechanical.   C You don't have a bent and shorted out pin somewhere do you? That is . the cause of a seemingly dead shelt I've seen.   --  < Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda. @                                              West Australia 6076* comp.os.vms,- The Older, Grumpier Slashdot. Raw, Cooked or Well-done, it's all half baked.F EPIC, The Architecture of the future, always has been, always will be.   ------------------------------   Date: 16 Dec 2005 01:48:06 GMT( From: bill@cs.uofs.edu (Bill Gunshannon)! Subject: Re: Storageworks shelves , Message-ID: <40ekmmF17ipk2U1@individual.net>  - In article <87bqzip2m3.fsf@prep.synonet.com>,  	prep@prep.synonet.com writes:, > bill@cs.uofs.edu (Bill Gunshannon) writes: > . >> In article <dnq3pl1c2k@enews2.newsguy.com>, >> 	healyzh@aracnet.com writes: - >>> Bill Gunshannon <bill@cs.uofs.edu> wrote:  > E >>>> No "Personality Module" in either of them. Just drives and power E >>>> supplies.  Behind the fans are the terminator and the connector. G >>>> Bothe are configured for single bus and exert SHELF_OK (although I . >>>> suspect the PDP doesn't care about that.) >   D >>> A BA350 doesn't use a personality module, those are for BA356's. > G >> Aren't there "Personality Modules" that plug into the top slot which D >> is much narrower than the rest and has the two SCSI connectors in >> the back of it? > C > The 350 has the personality module you have when you don't have a D > personality. ;) It is just a pair of double Honda-50s so you don't@ > have to go bus diving to connect it up. No smarts at all, just
 > mechanical.  > E > You don't have a bent and shorted out pin somewhere do you? That is 0 > the cause of a seemingly dead shelt I've seen.  , I'll look tomorrow.  Thanks for the pointer.   bill      --  J 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: 16 Dec 2005 01:53:59 GMT( From: bill@cs.uofs.edu (Bill Gunshannon)! Subject: Re: Storageworks shelves , Message-ID: <40el1mF17ipk2U2@individual.net>  , In article <FOb6GpHe15Ys@malvm9.mala.bc.ca>,4 	nothome@spammers.are.scum (Malcolm Dunnett) writes:/ > In article <40e5rcF19guhdU1@individual.net>,  . >   bill@cs.uofs.edu (Bill Gunshannon) writes: >>  M >> But multiple SCSI busses doesn't have anything to do with the 7(15) device M >> limit.   It has to do with the number of bits allocated for addressing and M >> the real question is why did they only allow 3 originally and after seeing C >> the way things were going, why only increase it by one more bit?  >>  I >    ISTM that if your goal is to just have lots of data available on the K > system then larger SCSI drives makes more sense than more SCSI drives. If J > your goal is performance then the question becomes how much data the bus7 > can transfer rather than how many drives you can add.   F Actually, there is more to it than just bus bandwidth.  There are alsoH limits to the number of transactions a disk can handle in a given periodG of time.  Sometimes more spindles can also increase performance.  Isn't I that the reason why one of the modes of RAID is striping across spindles?    > M >   Wasn't the original SCSI bus only something like 1.5MB/sec in asynch mode K > and 5MB/sec in synch mode? That's not a lot of data transfer capacity and M > the designers probably felt that under most usage patterns the bus would be L > saturated before you could add more than 7 drives ( and also remember thatK > drives were very expensive in those days and SCSI was originally designed  > for use on "cheap" systems ).   D I have run a number of USENET News Servers and I can assure you moreB spindles is always better. Lots of small files and no way to know A where the next access is going to be from/to.  Can't think of any % time when I ran out of bus bandwidth.    > E >   At the time "wide SCSI" came out I believe the bus was only 10Mhz L > ( or 20MB/sec maximum ). Given that limitation at the time how many drivesK > would it be reasonable to support on the bus ( assuming they might all be P > accessed simultaneously ). Perhaps with Ultra320 available now that limitation7 > seems too restrictive, but that's hindsight isn't it?   ' Like I said, hindsight is always 20/20.    bill   --  J 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: Thu, 15 Dec 2005 21:40:39 -0600 / From: Chris Scheers <chris@applied-synergy.com> ! Subject: Re: Storageworks shelves . Message-ID: <43A23737.903@applied-synergy.com>  - David Turner, Island Computers US Corp wrote: 2 > Perhaps someone can explain to me the following: > E > Why, when SCSI was implemented, did it have the 7 drive limitation?  > Why not 16 or 32?  > Of course UW> is 14 but why? > J > Did someone in the SCSI original design team think that noone would ever > need more than 7 drives?  ! Time to use the way back machine.   F There are actually several reasons for the 7 drive limit.  (Actually, E you can have 8 drives on the bus, but they would be difficult to use    without a host adapter.  <grin>)  C Initially, SCSI was derived from the SASI (Shugart Associates) bus  F interface.  At the time SASI was developed, the disk drives used with B the interface were almost always full height 5 1/4" and 8" drives.  I Generally, drives were installed one or two drives to a case.  So if you  D figure cable lengths for multiple drives/cases remembering that the F drives were full height, you could quickly exceed legal cable lengths $ trying to connect just seven drives.  I Also, part of the protocol uses a bit oriented priority scheme.  With an  E eight bit bus (which is true for SASI and SCSI-1), there can only be   eight device bits.  H Also, this was a low end bus.  At that time, disks were so horrendously A expensive that, if you had the money, you would probably not use  % SASI/SCSI to connect a lot of drives.     D Finally, the 7 device limitation is not really a limitation of SCSI:  F With SCSI's definition of LUNs, you can easily have 56 devices on the G bus, but there hasn't been much interest in implementing this for disk   drives.   G Going further, if I use a HSZ controllers, I can have a huge number of   disks on a single SCSI bus.    --  G ----------------------------------------------------------------------- $ Chris Scheers, Applied Synergy, Inc.  B Voice: 817-237-3360            Internet: chris@applied-synergy.com    Fax: 817-237-3074   ------------------------------    Date: 15 Dec 2005 14:54:25 -0800) From: "Sue" <susan_skonetski@hotmail.com> ) Subject: Re: UK VMS HP contract customers C Message-ID: <1134687265.211243.186300@g47g2000cwa.googlegroups.com>    Dear Newsgroup,   D I had not heard anything about retirment of any Alpha's, but to make< sure I contacted a friend in Services, here is his response.   Sue,  @ To date we have not ended maintenance support on ANY AlphaServerF platforms; including those that were introduced in the early 1990s. HPC Policy dictates that support is continued at least five years after C product is off the new Product CPL. So clearly the AlphaServer DS10 C which was on the CPL in 2004 would not have maintenance support end 
 next year.   Dan        Simon Clubley wrote:H > Those of you who are HP contract customers in the UK and have upcomingJ > contract renewals may want to check that you have actually received your > renewal contracts. > G > It appears that my contracts never got sent out and may have been due  > to HP staff changeovers. > I > I would be interested in knowing if anyone else had problems, or if I'm  > just a one-off.  >  > Simon. >  > --= > Simon Clubley, clubley@remove_me.eisner.decus.org-Earth.UFP K > If Google's motto is "do no wrong", then how did we get Google Groups 2 ?    ------------------------------  % Date: Thu, 15 Dec 2005 23:42:46 -0500 * From: "FredK" <fred.nospam@nospam.dec.com>) Subject: Re: UK VMS HP contract customers , Message-ID: <43a245c7$1@usenet01.boi.hp.com>  G Perhaps you should send the name of this person to Sue.  It sounds like ; whoever this is has "at minimum" crossed an ethical line...     5 "Syltrem" <syltremzulu@videotron.ca> wrote in message * news:11q3a9ierr8jd94@corp.supernews.com... > H > "David Turner, Island Computers US Corp" <dbturner@icusc.com> wrote in= > message news:FvVnf.16454$MA2.6628@bignews6.bellsouth.net... K > >I heard in the USA that several platforms have had suport canned on them  > >  > > DS10 DS20 DS20e ES40 > >   > > Anyone able to confirm that? > > 	 > > David  > >  > K > Our HP representative assured us this is NOT the case, these machines CAN  be  > put on a maintenance contract. > J > But malicious companies try to get contracts based on false information.G > We got a call this morning, from a person that said HP will no longer ' > support these machines (that we own). I > He maintained that support will end by 2006, which is totally false. He  alsoL > tried to push some other hardware and suggested we go with something else, > software-wise. > K > Again that's bad for VMS business when such people call a bunch of places H > and tell them they use unsupported hardware, and also suggest that the, > software is not in a much better position. > B > You should probably call your clients and ask where they got the information K > from. Possibly not directly from HP ? Or else the situation in the USA is " > different than that in Canada...K > And HP should send a mailing (it's been a very long time since I received  a L > nice toy from DEC/Compaq/HP in the mail!) to tell their customers what theL > plans are. This info is currently only available on the web afaik, and notI > all people will check (believe it or not HP, some people will not spend  time: > hunting for such information, it must be given to them).J > So I'm waiting for a new OpenVMS t-shirt, and a copy of the HP statement onJ > AlphaServer support to go with it (and of course the necessary Integrity > servers flyer).  > 	 > Syltrem  >  >    ------------------------------    Date: 15 Dec 2005 14:57:27 -0800) From: "Sue" <susan_skonetski@hotmail.com>   Subject: Updated VMS informationC Message-ID: <1134687447.342172.122080@g14g2000cwa.googlegroups.com>    Dear Folks,   A Here is this weeks Updated VMS Information which I think you will  useful.   
 Warm Regards,  Sue    Index    >From VMS Engineering 	 Training:  My Favorite Web Sites this week  VMS Technical Journal  OpenVMS Hobbyist:  >From Encompass  >From our PartnersG Jobs - Please keep in mind I have nothing to do with hiring folks, I am % just passing on what gets sent to me.    _______________________  >From VMS Engineering E Earlier this year, after hearing from our existing BASEstar customers A as well as potentially new BASEstar customers, we decided to port B BASEstar Open to HP's new Integrity Servers.  Specifically we haveE ported BASEstar Open Server, OMNI, OSAP/AP, OSAP/H1, and the S7 DASes D to both HP-UX and OpenVMS I64. We have a select number of field testG sites who have shown interest in evaluating BASEstar Open and / or OMNI C on Integrity.  I'd like to extend the offer to have some additional F customer sites added for feedback to engineering on the product, so ifG you are interested and would like to participate please let me know. We > expect the formal release of the above BASEstar products to beE completed before end of this year.  They can be ordered now, the part D numbers and pricing are already in place.  The BASEstar products areC now licensed on a per-processor basis like most of the HP Integrity @ Server software products.  Trade-in allowances are available for@ customers who wish to move from existing platforms to Integrity,G details are available. Requirements for evaluation: Must be an existing C BASEstar Open or OMNI user. (The product and interfaces will be the G same as on existing platforms.) OpenVMS I64 An Integrity Server running B OpenVMS I64 V8.2 or V8.2-1 HP-UX An Integrity Server running HP-UX version is B.11.23 (or 11i v2). D 32 bit libraries are initially supported, both 32 and 64 bit support" will be provided in final release.  . ______________________________________________	 Training:   4 Host Based Volume Shadowing (HBVS) Basics and Beyond  B BRUDEN Corporation is proud to announce that it will be sponsoringE deliveries of a new course developed by John Andruszkiewicz (aka John E AtoZ).  John, a former member of OpenVMS Engineering, was responsible F for developing and maintaining the HBVS product.  The course is titledD Host Based Volume Shadowing (HBVS) Basics and Beyond. This course isG designed to assist the System Manager in implementing, maintaining, and @ troubleshooting Host Based Volume Shadowing (HBVS) in an OpenVMS< environment.  This lecture/lab course is 3.5 days in length. Scheduled course dates are: % February 21 through February 24, 2006  April 3 through April 6, 2006 B Please check the BRUDEN web site (www.BRUDEN.com) or contact Bruce Ellis % (Bruce.Ellis@BRUDEN.com) for details.   - _____________________________________________    VMS Technical Journal   C Reminder:  If you have not filled out the OpenVMS Technical Journal B Survey could you please do so before Monday at this point only 185A folks have responded.  Please visit www.openvms.org we value your 	 feedback.   E Based on survey feedback we have now added a total list of a table of . contents of all the OpenVMS Technical Journals2 http://h71000.www7.hp.com/openvms/journal/toc.html       ________________________ OpenVMS Hobbyist:   G Latest Count: 644025 Paks Generated to date, 40662 in the last 6 months   G The OpenVMS Hobbyist web site has been updated with new stories, thanks E to Jim Rainville and Warren Sander:  Please stop by and visit our web  site at:H http://h71000.www7.hp.com/openvms/license_software_programs/mainstorypag e.html  G http://www.bunkerofdoom.com this is a friend of mine not to mention run  on a VMS hobby license  6 Ann McQuaid Announces the winner of the hobby contest:9 http://www.openvms.org/stories.php?story=05/11/30/5749689    ________________________ >From Encompass   - Updates to the VMS Technical Resource page at C http://www.encompassus.org/resources/openvmsportal.html by adding a # section for VMS-related whitepapers    ___________________________  >From our Partners  
 Oracle Rdb  < We are pleased to announce the release of Oracle Rdb ReleaseA      7.1.4.3 (aka V7.1-43).  The kit is available on Metalink and       will be on OTN shortly.  8      Software Errors Fixed in Oracle Rdb Release 7.1.4.38      ---------------------------------------------------  7      Software Errors Fixed That Apply to All Interfaces 3        - Various Bugchecks Due to Memory Corruption 9        - Reserved Sequence Slots Not Properly Initialized 3        - User Attach Stalls Waiting for ALS Startup       SQL Errors Fixed F        - CREATE TABLE ... LIKE May Create Invalid NOT NULL ConstraintsF        - STDDEV and VARIANCE Generate Wrong Values in GROUP BY QueriesG        - SQL Module Language /PROTOTYPES Now Generates INT64 for BIGINT           Parameters <        - IMPORT DATABASE Looses SECURITY CHECKING Attributes       RMU Errors FixedB        - RMU/BACKUP of Database With Very Large Storage Area Count@        - RMU Backup and Restore Problems With Many Storage Areas#        - Corrections to RMU Extract       Row Cache Errors Fixed =        - RCS and User Memory Leak With Snapshots in Row Cache %      RMU Show Statistics Errors Fixed 2        - Excessive Erases Done for Fragmented Rows>        - Some RMU/SHOW STATISTICS Screens Not Cleared Properly       JOBS: F ______________________________________________________________________   Name: Vicki Tyler - Email: vicki.tyler@eurobase-international.com  Country/Region: GB* Subject: OpenVMS product business feedback  	 Comments: G I am working for an IT Recruitment agency and am vety eager to find VMS D System Administrators willing to work for a large investment bank upA London. If they could send their cv's through to me this would be  great.    urls: http://h71000.www7.hp.com/) Referring_URL: http://h71000.www7.hp.com/  company_name: Eurobase People   C ___________________________________________________________________ B >>> Paulette Rose Poloni <polonip@LAFAYETTE.EDU> 10/28 5:37 pm >>>E Since it is hard to find folks with VMS skills (especially those of a E system administrator) I am wondering if any of you use or have looked  at  D getting part or all of your VMS system administration support from a remote / outsourced service?  If so, who is the vendor? ) How has the arrangement been working out? 7 If not, how do you go about providing VMS sys admin?  I C s your current VMS sys admin also responsible for other op systems? ? Have you developed VMS sys admin skills by sending folks off to 	 training? . What was their original background when hired?   Thanks   Paulette Poloni    Paulette Rose Poloni. Director of Administrative Information Systems Lafayette College  634 Parson Street  Easton Pa 18042-1768D ____________________________________________________________________  * Date: Fri, 2 Dec 2005 15:47:09 -0500 (EST)* From: "John R. Covert" <covert@covert.org>* To: Sue Skonetski <susan.skonetski@hp.tom>F Subject: Jobs in CA and NM for sixty OpenVMS engineers with clearances  C There is an immediate need for some sixty VMS engineers, especially > those with internals and device driver experience, for a largeC government project in CA and NM.  Security clearances are required; G therefore I don't know the exact details of the work.  Please send your E resumes to Bonnie Huval at bonnie_huval@bigfoot.com or at the address  in her message below.    /john   @ From:	SMTP%"huvalbd@seneschalonline.com"  2-DEC-2005 09:58:15.18& To:	"Covert, John" <covert@covert.org> Subj:	OpenVMS Work   John:   G Sunday I met with a fellow who needs 40 VMS people in California and 20  atE Los Alamos NM. All of these openings require security clearances. But  Lou F has his finger on the pulse of a lot of VMS work, particularly for DOD and E NASA. Some that doesn't require clearance is likely to turn up, also.   D Would you or anyone you know want to do some of this work? Lou would likeG it very much if my company could offer some VMS people. He has known me  for G about nine years now and has high regard for the work I do and the type  of people I hire.  E Please pass this along to anyone you know of who might be interested.    Bonnie   ------------------------------  % Date: Thu, 15 Dec 2005 17:04:48 -0500 C From: "David Turner, Island Computers US Corp" <dbturner@icusc.com> / Subject: Re: Upgrade a 5300A controller via CD? 9 Message-ID: <ZElof.24967$Y72.6432@bignews1.bellsouth.net>   @ The DS5300a has the firmware (V3.40) included on the Firmware CD! (downloadable in ISO format here)   4 ftp://ftp.digital.com/pub/DEC/Alpha/firmware/v70.zip    J It auto-updates from the SRM when loading the new System Firmware (V7.0-3)           --     David B Turner Island Computers US Corp 2700 Gregory St, Suite 180 Savannah GA 31404  Tel: 912 447 6622 X201 Cell: 912 447 6622 X252  Fax: 912 201 0402  Email: dbturner@icusc.com  Web: http://www.islandco.com% ===================================== < All orders are subject to the following terms and conditions. of sale. These should be read before ordering.% http://www.islandco.com/warranty.html   I "Christoph Gartmann" <gartmann@non.immunbio.mpg.de.sens> wrote in message # news:dlvgld$nq4$1@news.BelWue.DE...  > Hello, > J > in my DS25 there is a raid controller 5300a. I downloaded a new firmware for I > this controller. Unfortunately the DS25 has no floppy. According to the  manualL > an upgrade via the CD/DVD should be possible as well. The question is: how do IK > do that? I burned an ISO CD on a PC and put the unzipped firmware file on  it. L > But it was not recognized. Thus, what else is necessary to create a proper CD?  > 
 > Regards, >    Christoph Gartmann  >  > --  G >  Max-Planck-Institut fuer      Phone   : +49-761-5108-464   Fax: -452  >  ImmunbiologieK >  Postfach 1169                 Internet: gartmann@immunbio dot mpg dot de  >  D-79011  Freiburg, Germany ; >                http://www.immunbio.mpg.de/home/menue.html    ------------------------------    Date: 15 Dec 2005 14:51:50 -0800; From: "johnhreinhardt@yahoo.com" <johnhreinhardt@yahoo.com> / Subject: Re: Upgrade a 5300A controller via CD? C Message-ID: <1134687110.407497.177650@g47g2000cwa.googlegroups.com>   - David Turner, Island Computers US Corp wrote: B > The DS5300a has the firmware (V3.40) included on the Firmware CD# > (downloadable in ISO format here)  > 6 > ftp://ftp.digital.com/pub/DEC/Alpha/firmware/v70.zip >  > L > It auto-updates from the SRM when loading the new System Firmware (V7.0-3) >   B I don't suppose this firmware could be loaded into a 5300 to get a 5300a?   ------------------------------  # Date: Thu, 15 Dec 2005 22:28:15 GMT 6 From: "Andy Bustamante" <a_c_bustamante@earthlink.net> Subject: Re: Upgrading TCPIP= Message-ID: <36mof.34133$q%.30866@newssvr12.news.prodigy.com>   E Cluster alias is replaced by failSAFE ip.  A good deal more flexible.    --       Andy Bustamante  Remove the ASCII 95s for e-mail     . "Tom Linden" <tom@kednos.com> wrote in message" news:ops1sozidbzgicya@hyrrokkin... > Currently running B >    Compaq TCP/IP Services for OpenVMS Alpha Version V5.1 - ECO 3@ >    on a COMPAQ Professional Workstation P running OpenVMS V7.3 > : > Anything to be on the lookout for when upgrading to 5.4?6 > Not planning VMS upgrade, need to have a 7.3 around. >  > Tom    ------------------------------    Date: 15 Dec 2005 16:17:56 -0600- From: Kilgallen@SpamCop.net (Larry Kilgallen) B Subject: Re: using SYS$ACM to authenticate arbitrary windows user?3 Message-ID: <5CSs07McGrJq@eisner.encompasserve.org>   ` In article <LyPYyRl3Xajj@malvm9.mala.bc.ca>, nothome@spammers.are.scum (Malcolm Dunnett) writes:  F > %ACME-E-NOEXTAUTH, principal name cannot be authenticated externally > / > Does anyone have an example of authenticating 3 > against a "raw" domain\user without a VMS mapping . > (or hints on how to make this work)? Is this > even possible?  B Reasoning from first principles, which UAF record would you expect@ to be the source of process quota information for such a login ?   ------------------------------    Date: 15 Dec 2005 16:46:17 -0600- From: Kilgallen@SpamCop.net (Larry Kilgallen) B Subject: Re: using SYS$ACM to authenticate arbitrary windows user?3 Message-ID: <syF0C390of5c@eisner.encompasserve.org>   ` In article <3NANGUftX2fx@malvm9.mala.bc.ca>, nothome@spammers.are.scum (Malcolm Dunnett) writes:e > In article <5CSs07McGrJq@eisner.encompasserve.org>, Kilgallen@SpamCop.net (Larry Kilgallen) writes: c >> In article <LyPYyRl3Xajj@malvm9.mala.bc.ca>, nothome@spammers.are.scum (Malcolm Dunnett) writes:  >>  H >>> %ACME-E-NOEXTAUTH, principal name cannot be authenticated externally >>> 1 >>> Does anyone have an example of authenticating 5 >>> against a "raw" domain\user without a VMS mapping 0 >>> (or hints on how to make this work)? Is this >>> even possible? >>  E >> Reasoning from first principles, which UAF record would you expect C >> to be the source of process quota information for such a login ?  > F >   Obviously I wouldn't - I am thinking of "authentication" in a moreH > general sense. In this particular case I'm thinking of the possibilityF > of tying it into the OSU webserver so that I could authorize certainB > web services for users who have a windows account but don't haveI > ( or need ) a VMS account. I would be maintaining my own "authorization @ > database" of which services each windows user was entitled to.  @ I think that would be a security mistake, since it would have to< bypass VMS auditing and other features like breakin evasion.  D > If that can't be done I can create a VMS account for each user andG > flag it for external authentication - but in many cases they'll never D > use the VMS account for anything other than authenticating against > these web services.   ? That would certainly be better for breakin evasion, since users @ would not be able to deny service to each other inappropriately.  F >   I suppose that's the broader question: is SYS$ACM intended to be aF > general "authentication" service or just a fancier way of processing
 > VMS logins?   ? The former, but for authorized VMS users.  One example would be C a _subsequent_ authentication for high value monetary transactions.    ------------------------------   End of INFO-VAX 2005.698 ************************                                                                                                                                                                                                                                                            