1 INFO-VAX	Tue, 15 Jun 2004	Volume 2004 : Issue 330       Contents:, Re: Accessing NFS from OS X. Was: NFS how to, Re: Accessing NFS from OS X. Was: NFS how to, Re: Accessing NFS from OS X. Was: NFS how to, Re: Accessing NFS from OS X. Was: NFS how to, Re: Accessing NFS from OS X. Was: NFS how to Re: DEC PWS RAM  Re: DecServer password reset FTP Problem 0 Re: Quadrics adds lower-end supercomputer switch0 Re: Quadrics adds lower-end supercomputer switch7 Re: Tru64 --> OpenVMS rather than another unix platform = Re: Understanding Peak virtual size VS UAF pgflquota+WSEXTENT 5 Re: URL for latest Library and Tape Tools version 3.5 " VMS Gets Short Shrift at HP World.* [OpenVMS Alpha V7.3-2] VMS732_UPDATE-V0200. Re: [OpenVMS Alpha V7.3-2] VMS732_UPDATE-V0200. Re: [OpenVMS Alpha V7.3-2] VMS732_UPDATE-V0200. Re: [OpenVMS Alpha V7.3-2] VMS732_UPDATE-V0200. Re: [OpenVMS Alpha V7.3-2] VMS732_UPDATE-V0200N Re: [TCPIP] How is the routing table updated when ROUTED or GATED aren't used?  F ----------------------------------------------------------------------  + Date: Mon, 14 Jun 2004 20:20:53 +0000 (UTC) % From: News Reader <nomail@nomail.vms> 5 Subject: Re: Accessing NFS from OS X. Was: NFS how to $ Message-ID: <cal1b5$rsj$1@online.de>  G On  Mon, 14 Jun 2004 16:56:34 GMT, Paul Anderson <paul.anderson@hp.com>  wrote: > ? >If someone knows of a way to get a Finder window of an OpenVMS ) >directory via NFS, I'd love to know how!  >  >Paul   F Can't speak for OS X but use Finder windows of OpenVMS directories via  NFS in Classic Mac OS routinely.   ------------------------------  + Date: Mon, 14 Jun 2004 20:31:24 +0000 (UTC) % From: News Reader <nomail@nomail.vms> 5 Subject: Re: Accessing NFS from OS X. Was: NFS how to $ Message-ID: <cal1ur$rsj$2@online.de>  I On Sun, 13 Jun 2004 09:33:07, Paul Sture <nospam@sture.homeip.net> wrote:  > F >For the record, in case anyone else wants to do this, here is where IE >ended up in my quest to access my VMS system from my Mac running OSX  > 2 >On the VMS side:, map and export the system disk:  
 [..snip..]  < >In a file called by SYSTARTUP_VMS.COM I have the following: > / >$!      Serve the system disk as an NFS device  >$! I >$ TCPIP mount dnfs1: "/alpha/" /host='node' /system/path="/alpha/000000"   A That's VMS as NFS client.  I thought this thread was about VMS as  NFS server.  >  > . >filenames containing a dollar fool the shell: >  >cd vms$common.dir! >tcsh: common: Undefined variable  >   < Shell metacharacters in filenames - no great surprises here.   ------------------------------  + Date: Mon, 14 Jun 2004 21:00:35 +0000 (UTC) % From: News Reader <nomail@nomail.vms> 5 Subject: Re: Accessing NFS from OS X. Was: NFS how to $ Message-ID: <cal3lj$8gp$1@online.de>   > < >                   Accessing NFS from OS X. Was: NFS how to( >                                       . >   From: Paul Sture <nospam@sture.homeip.net> >   Reply to: Paul Sture) >   Date: Sun, 13 Jun 2004 09:33:07 +0200  >   Newsgroups:  >          comp.os.vms >   Followup to: newsgroup(s)  >   References: 0 >          <00A32D30.681CB91C@SendSpamHere.ORG> * >          <2iaui6FkijpvU1@uni-berlin.de> / >          <00A32D89.B3816670@SendSpamHere.ORG>  > F >For the record, in case anyone else wants to do this, here is where IE >ended up in my quest to access my VMS system from my Mac running OSX 
 >(10.3.4). >  > 2 >On the VMS side:, map and export the system disk: >  >TCPIP> show config map * >             Filesystem Map Configuration< >Pathname                                Logical File System > 7 >/alpha                                  SYS$SYSDEVICE:  >TCPIP> show export  > 2 >File System                             Host name > * >/alpha/000000                           * >TCPIP> show proxy > < >VMS User_name     Type      User_ID    Group_ID   Host_name > A >SYSTEM            OND             1           4   alpha1, alpha2 6 >SYSTEM            OND           501          20   mac4 >TCPIP$NFS         OND             0           1   *4 >TCPIP$NOBODY      OND            -2          -2   * > A >The first SYSTEM entry enables STARTUP to mount the DNFS device. E >The second SYSTEM entry is the User_ID/Group_ID of my Mac account as  >shown in Netinfo -> users.  > F >IMPORTANT: TCP/IP proxy modifications only take effect after all DNFS% >disks are dismounted then remounted.  > < >In a file called by SYSTARTUP_VMS.COM I have the following: > / >$!      Serve the system disk as an NFS device  >$! I >$ TCPIP mount dnfs1: "/alpha/" /host='node' /system/path="/alpha/000000"  > G >On the Mac side, to do an automount at startup, I followed the Netinfo  >instructions at > 4 >http://astcomm.net/mac/tech/nfs_howto/client/#final >  >Some notes on that: > C >1. I added a value of -P (use a privileged port) in the opts value A >    field. Without that I could only get things working with the ! >    /private/mumble mount point.  > = >2. The note on /private/mnt is a bit misleading. It could be ? >    /private/anything as well, and it's possible to create the C >    dir entry as /Users/sture/alpha, /Volumes/alpha and just plain  >    /alpha  > E >3. The example given for nidump doesn't work. In fact there's no way H >    I can get nidump to report anything. Any syntax I throw at it comes@ >    back with help text as if I'd entered no parameters at all. > E >Lastly, there's still no way I can access the VMS system via Finder; H >only via the CLI. Finder simply tells me I don't have the privilege, no# >matter where I put my mount point.  > H >Initially I suspected that Finder was using a different UID/GID pair toI >access the VMS system, but TCPTRACE tells me it's using the same one the 
 >CLI uses. > B >To end on a light note, navigating around an NFS mounted VMS disk8 >(ODS/2) using OS X is something of a new experience :-) >  >To move down a directory: >  >     cd syslost.dir >  >list all subdirectories:  > ; >     ls *.dir   # lists the contents of all subdirectories  > . >filenames containing a dollar fool the shell: >  >cd vms$common.dir! >tcsh: common: Undefined variable  > ' >so you need to escape the dollar sign:  >  >cd vms\$common  > H >Files with only one version show up without the version number. If moreG >than one version exists then the version number is included too - e.g.  >  >x.com >x.x.2 >x.x.3 >  >Feedback welcome.   ------------------------------  + Date: Mon, 14 Jun 2004 21:04:13 +0000 (UTC) % From: News Reader <nomail@nomail.vms> 5 Subject: Re: Accessing NFS from OS X. Was: NFS how to $ Message-ID: <cal3sc$8gp$2@online.de>   REPOSTI On Sun, 13 Jun 2004 09:33:07, Paul Sture <nospam@sture.homeip.net> wrote:  >TCPIP> show config map * >             Filesystem Map Configuration< >Pathname                                Logical File System7 >/alpha                                  SYS$SYSDEVICE:  >  ...  >Feedback welcome.  F I'd never NFS export something quite so high up on any NFS server; too much of a security risk.  E Also, I don't know what file formats single/double etc OS X uses, but C you may well find that Finder creates its own files to cache folder C information, window state etc in each directory you open in Finder.   D You might not want your entire directory tree peppered with Finder'sD private files as the price for filesystem exploration MAC GUI style.   ------------------------------   Date: 15 Jun 2004 01:37:30 GMT From: healyzh@aracnet.com 5 Subject: Re: Accessing NFS from OS X. Was: NFS how to , Message-ID: <caljsq022ei@enews3.newsguy.com>  & News Reader <nomail@nomail.vms> wrote: > >TCPIP> show config map , > >             Filesystem Map Configuration> > >Pathname                                Logical File System9 > >/alpha                                  SYS$SYSDEVICE:   H > I'd never NFS export something quite so high up on any NFS server; too > much of a security risk.  J Agreed, but will TCPIP V5.4 allow you to export subdirectories?  From whatG little I've played with it, I get the impression you now have to either / export the entire disk, or use container files.   G > Also, I don't know what file formats single/double etc OS X uses, but E > you may well find that Finder creates its own files to cache folder E > information, window state etc in each directory you open in Finder.   F > You might not want your entire directory tree peppered with Finder'sF > private files as the price for filesystem exploration MAC GUI style.  L I know this is a problem when using Appletalk to access OpenVMS disks.  I'veF "solved/ignored" this problem by putting the following in my LOGIN.COM  7 $ dir :== dir/exclude=(*msaf*.*;*,*.msaf*;*)/versions=1 1 $ dd :== DIR *.DIR/exclude=(*msaf*.*;*,*.msaf*;*)   H If Mac OS X does something like this to NFS disks (I don't think that it3 does), it might be possible to use a simular trick.   K Personally for the time being I think I'll continue to use Appletalk.  It's L also worth noting that Mac OS X supports Samba.  It might now be possible toK access samba shares on a VMS server.  I do know that in the Mac OS 8/9 days H trying to do this using DAVE didn't work as the Mac tried to create some/ really funky files to keep track of attributes.   = All in all, it would really help if Apple wasn't so anti-NFS.    		Zane   ------------------------------   Date: 14 Jun 2004 22:03:38 GMT From: healyzh@aracnet.com  Subject: Re: DEC PWS RAM, Message-ID: <cal7bq311q4@enews2.newsguy.com>  ) Rich Jordan <duodec@speakeasy.net> wrote: G > Don't honestly know.  When I was searching for memory for a PC66 PWS  J > 333i+, there were a lot of warnings (2002 timeframe) about watching out H > for the 'newer' PC100 memory as it would not work, and a lot of older H > messages saying to buy PC100 because it worked fine, cost the same or G > less, and could be used in a newer system when you upgraded (1998-99  K > timeframe).  I got some to try and it didn't work.  Later found the same  J > brand/model/CL# etc but made a couple years earlier (new old stock on a  > shelf) and it worked fine.  B > Same problems with PC133 memory in a friend's PC100 system; the G > "original" PC133 worked fine, but then current production would not.  I > The newsgroup articles and forums we reviewed indicated it was because  C > of manufacturing updates/changes to the actual memory chips that  D > tightened up the tolerances of the speed range they would work at.  K It might be because some of the newer RAM is set to be one of the following H two configurations depeding on if it's in a system that expects PC100 orC PC133.  I've no idea how this type of memory would work in a system  expecting PC66.    	PC100 CL2 or PC133 CL3  	PC133 CL2 or PC100 CL2    		Zane   ------------------------------    Date: 14 Jun 2004 18:59:41 -0700+ From: glen.thompson@cox.net (Glen Thompson) % Subject: Re: DecServer password reset = Message-ID: <2a92c50f.0406141759.318a771c@posting.google.com>   D It pretty much goes through the normal testing sequence, memory bankC 1, memory bank 2, internal test 1, 2, 3, ... etc.  Finally finishes D with the swirling display as normal.  I've tried it connected to the  network as well as disconnected.  E Thanks for the link to the manual.  I had found it and that's where I  got the reset button procedure.    glen% ===================================== v williamwebb@openvms-rocks.com (William Webb) wrote in message news:<bf98c417.0406110852.1259b6b@posting.google.com>... > M > What LED sequences do you get when you boot it up connected to the network? + > Does it get all the way to the "crazy 8"?  > M > You are doing that with the DECserver connected to the network, aren't you?  > T > Replace the ones in the following line with slashes to get the URL for the manual. > Q > http:11www.dnpg.com1dr1decserver1decserver-platforms1documentation1dsrvw-mg.pdf  >  > HTH  >  > WWWebb >    ------------------------------   Date: 15 Jun 2004 03:11:32 GMT From: healyzh@aracnet.com  Subject: FTP Problem+ Message-ID: <calpd40b8p@enews1.newsguy.com>   J I've discovered that there seems to be some sort of wierd problem with FTPM on my VMS server.  I've tried both active and passive with the same results.  J It might be something to do with my firewall, but FTP is working from UnixC systems.  I've pasted the errors below, does anyone have any ideas?    Thanks, 	      Zane        FTP> show status 211-dux418 FTP server status:  HP ASL ftpd, version(323)   Connected to xxxxxxx.xxxxxxx.xxx Logged in anonymously 3 TYPE: Image; STRUcture: File; transfer MODE: Stream  No data connection 211 End of status   Connected to: dux418.den.hp.com  VMS Plus mode disabledA Mode = stream , Type = binary, Form = non_print, Structure = file  Error level is SUCCESS& Passive is AUTO (IPv4: OFF, IPv6: ON).   FTP> dir 200 PORT command successful.3 150 Opening ASCII mode data connection for /bin/ls. H %TCPIP-E-FTP_DATACONF, cannot establish data connection with remote host4 -SYSTEM-F-REJECT, connect to network object rejected 226 Transfer complete. FTP>     ------------------------------  % Date: Mon, 14 Jun 2004 16:29:49 -0400 2 From: "Jonathan Boswell" <jsb.NOSP@M.cdrh.fda.gov>9 Subject: Re: Quadrics adds lower-end supercomputer switch 4 Message-ID: <3Vnzc.391$Ny6.1281@mencken.net.nih.gov>  > "Keith Parris" <keithparris_NOSPAM@yahoo.com> wrote in message7 news:cf15391e.0406101556.55bc3f19@posting.google.com... A > While it would theoretically be possible to run OpenVMS Cluster B > traffic over such an interconnect, there are no plans to supportE > another proprietary cluster interconnect, although investigation is A > underway about the possibility of using a new industry-standard  > interconnect in the future.   C How about LVD SCSI?  If DSSI was handling this traffic 10 years ago F (including SCS for crying out loud), why has no industry standard SCSIB interface ever been patched up and qualified in a similar fashion?   ------------------------------    Date: 14 Jun 2004 17:40:10 -0600- From: Kilgallen@SpamCop.net (Larry Kilgallen) 9 Subject: Re: Quadrics adds lower-end supercomputer switch 3 Message-ID: <KSjUdCceYyHd@eisner.encompasserve.org>   i In article <3Vnzc.391$Ny6.1281@mencken.net.nih.gov>, "Jonathan Boswell" <jsb.NOSP@M.cdrh.fda.gov> writes: @ > "Keith Parris" <keithparris_NOSPAM@yahoo.com> wrote in message9 > news:cf15391e.0406101556.55bc3f19@posting.google.com... B >> While it would theoretically be possible to run OpenVMS ClusterC >> traffic over such an interconnect, there are no plans to support F >> another proprietary cluster interconnect, although investigation isB >> underway about the possibility of using a new industry-standard >> interconnect in the future. > E > How about LVD SCSI?  If DSSI was handling this traffic 10 years ago H > (including SCS for crying out loud), why has no industry standard SCSID > interface ever been patched up and qualified in a similar fashion?  > Probably because Ethernet is so cheap, and these days so fast.   ------------------------------    Date: 14 Jun 2004 13:30:19 -07001 From: keithparris_NOSPAM@yahoo.com (Keith Parris) @ Subject: Re: Tru64 --> OpenVMS rather than another unix platform= Message-ID: <cf15391e.0406141230.31c20700@posting.google.com>   W "John Smith" <a@nonymous.com> wrote in message news:<pI6dnbGRj8DH31fdRVn_iw@igs.net>... J > Since people are being forced to migrate from Tru64, why doesn't HP even$ > suggest OpenVMS as an alternative?  A HP is aware of a number of Tru64 customers who decided to move to E OpenVMS when it was announced that OpenVMS would be ported to Itanium  and that Tru64 would not.    ------------------------------  % Date: Mon, 14 Jun 2004 18:25:05 -0400   From: John Santos <JOHN@egh.com>F Subject: Re: Understanding Peak virtual size VS UAF pgflquota+WSEXTENT- Message-ID: <1040614181405.404B@Ives.egh.com>   . On 14 Jun 2004 briggs@encompasserve.org wrote:  o > In article <b096a4ee.0406131911.48a461b@posting.google.com>, spamsink2001@yahoo.com (Alan E. Feldman) writes: Q > > mckinneyj@cpva.saic.com wrote in message news:<igPu+FzOLZp+@cpva.saic.com>... K > >> I believe that John is correct in suspecting backing stores other than G > >> pagefiles here. When running an image part of your virtual address J > >> space may consist of global sections or installed image sections (the > >> RTLs).  > > E > > Yes, of course -- virtual memory consists of physical memory plus E > > pagefile, swapping file, image files, and section files. But what H > > about data files? Shouldn't they also be included in this? The Perf.% > > manual only lists the first four.  > 9 > Data files mapped as global sections count as number 4. E > Data file virtual block number 4,257,312 has a fixed memory address " > associated with it in this case. > ? > Data files accessed through RMS do not occupy virtual memory. @ > Data file virtual block number 4,257,312 does not have a fixed1 > memory address associated with it in this case.   C The data files don't occupy virtual memory, but the RMS buffers do. > They count as 1, 2, or 3 (eirther in physical memory or in the page or swap file.)   C Just to confuse things a bit, if the files have RMS global buffers, A the global buffers are in temporary global sections, count agains A GBLPAGFIL, and are backed by the pagefiles.  They are part of the C processes virtual memory, but don't count against its pgflquo (page # file quota), since they are shared.    --   John Santos  Evans Griffiths & Hart, Inc. 781-861-0670 ext 539   ------------------------------  % Date: Mon, 14 Jun 2004 21:36:35 -0500 2 From: David J Dachtera <djesys.nospam@comcast.net>> Subject: Re: URL for latest Library and Tape Tools version 3.5+ Message-ID: <40CE60B3.ACA458FD@comcast.net>    Peter Weaver wrote:  >  > Peter Weaver wrote:  > > ... = > > Snipped rant about the HP Library and Tape Tools for VMS.  > > ...  > C > One person who posts here and works for HP sent me a private note J > telling me that I should make sure that I pass my rant on to my Gold TAMJ > since HP does not always listen to their own engineers about things likeH > this. I told him that my post was sent to my Gold TAM moments before I: > posted here. Friday morning I received this answer back; > 7 > | It appears L&TT 3.5 is only supported on VMS 7.3-2. ; > | As well, the terminfo database which is part of the GNV 8 > | unix emulation package, not present by default under	 > | OVMS.  > G > I just checked the documentation again, the web pages report that you I > need "OpenVMS 7 (Alpha Platform)." It is interesting that you mentioned F > that you need GNV since the instructions give two different commandsC > that you need depending on whether you are using GNV nor not. The G > machine I installed this on has GNV V1.14.8(0) installed but I do not . > see any file named "TERMINFO" on the system.  3 If the VMS port was complete, it'd use SMG instead.    D.J.D.   ------------------------------  % Date: Mon, 14 Jun 2004 21:58:16 -0500 2 From: David J Dachtera <djesys.nospam@comcast.net>+ Subject: VMS Gets Short Shrift at HP World. + Message-ID: <40CE65C8.8381A794@comcast.net>   C Well, looks like hp is continuing its VMS-cide efforts even with HP D World (the current incarnation of "DECUS"). I looked at the completeE session catalog in its current form (to see if any of my ten sessions F were scheduled - none were found) and was surprised at how few OpenVMS sessions there are.   G I just took a fast count. I show 319 sessions, of which only 20 mention F VMS (using SEARCH on OpenVMS on text copied from the web page). I make that as 6.2% of the total.   Pretty sorry state of affairs.  1 It seems VMS is getting the short end once again.   F ...unless VMS and VMS-related services make up only 6.2% of hp's total profits (doubtful)...    D.J.D.   P.S.: F Perhaps VMSUS can rise and make up for this somehow. Let me ask aroundG and see what the chances are of getting a VMS-focused symposium of some G kind together before the next bootcamp. No promises - just the last act  of a desperate clan.   ------------------------------  + Date: Mon, 14 Jun 2004 20:32:23 +0000 (UTC) 6 From: peter@langstoeger.at (Peter 'EPLAN' LANGSTOEGER)3 Subject: [OpenVMS Alpha V7.3-2] VMS732_UPDATE-V0200 0 Message-ID: <newscache$hefbzh$rdn$1@news.sil.at>   <RANT>@ Today I downloaded the latest ECO VMS732_UPDATE-V0200 from ITRC.  8 1) I tried to install it on my system at work. It failed  P -PCSI-E-REQVOLSPC, 218722 blocks required; 138276 blocks free; -80446 blocks netW -PCSI-I-DELRDSETS, you can reclaim disk space by using the DELETE RECOVERY_DATA command E Terminating is strongly recommended.  Do you want to terminate? [YES]     Ok, not enough disk space. Alas.& But some lines later during 'rollback'   Portion done: 0%/ %PCSI-E-S_OPCAN, operation cancelled by request   K %PCSI-I-CREDIR, created directory DSA0:[SYS0.SYSCOMMON.SYSMGR.PCSI$WRKA657] T %PCSI-I-CREDIR, created directory DSA0:[SYS0.SYSCOMMON.SYSMGR.PCSI$WRKA657.][SYSUPD]Z %PCSI-E-READERR, error reading DSA0:[INSTALL_KITS]DEC-AXPVMS-VMS732_UPDATE-V0200--4.PCSI;2Z -PCSI-E-INVDOCMTL, internal error - product document has invalid product material ordering  F Hhmmm. Where did I see the INVDOCMTL the last time ? I'll know soon...    K 2) After some ZIPping to get free diskspace, I tried again. It failed after  10% with  Z %PCSI-E-READERR, error reading DSA0:[INSTALL_KITS]DEC-AXPVMS-VMS732_UPDATE-V0200--4.PCSI;2Z -PCSI-E-INVDOCMTL, internal error - product document has invalid product material ordering" %PCSI-E-OPFAILED, operation failedE Terminating is strongly recommended.  Do you want to terminate? [YES] T %PCSI-E-CANCEL_WIP, termination resulted in an incomplete modification to the system/ %PCSI-E-S_OPCAN, operation cancelled by request   K %PCSI-I-CREDIR, created directory DSA0:[SYS0.SYSCOMMON.SYSMGR.PCSI$WRKA665] T %PCSI-I-CREDIR, created directory DSA0:[SYS0.SYSCOMMON.SYSMGR.PCSI$WRKA665.][SYSUPD]   Recovery pass starting ...  Z %PCSI-E-READERR, error reading DSA0:[INSTALL_KITS]DEC-AXPVMS-VMS732_UPDATE-V0200--4.PCSI;2Z -PCSI-E-INVDOCMTL, internal error - product document has invalid product material ordering  > Hhhmm. The INVDOCMTL again. Where did I see it the last time ?@ Now I know. It was a corrupt PCSI file. Corrupted by download orE by unpacking ? How will I know this ? A new download/unpack fixed it.     E 3) I downloaded the ECO at home and surprise installation failed with    Portion done: 60%   P %PCSI-E-OPENOUT, error opening DISK$VMSSYS:[SYS0.SYSCOMMON.][SYSLIB]DECW$SVEXT_G LX_SW.EXE; as output; -RMS-F-FUL, device full (insufficient space for allocation) " %PCSI-E-OPFAILED, operation failedE Terminating is strongly recommended.  Do you want to terminate? [YES] P %PCSI-E-CANCEL_WIP, termination resulted in an incomplete modification to the sy stemG %PCSI-I-DELFIL, deleted DSA0:[SYS0.SYSMGR.PCSI$WRK202002B9]SYSUPD.DIR;1   B So, the PCSI kit is not corrupted this time. At least it seems so.K (Then where comes the corruption from ? Was the kit on the FTP server fixed L in the meantime [2h] ? Was it the download ? Why without any error message ?F Or was it the DCX unpacking ? Why also without a warning ? Why can VMSK engineering not also use ZIPSFX instead of the error prone DCX ? Please !!)   E But it also seems, that the freespace check is insufficient. (@%$&^*)     $ What the heck is the problem today ?6 Can an ECO not successfully install at the first try ? </RANT>    Good luck, folks   --   Peter "EPLAN" LANGSTOEGER % Network and OpenVMS system specialist  E-mail  peter@langstoeger.atF A-1030 VIENNA  AUSTRIA              I'm not a pessimist, I'm a realist   ------------------------------  % Date: Mon, 14 Jun 2004 18:41:16 -0400   From: John Santos <JOHN@egh.com>7 Subject: Re: [OpenVMS Alpha V7.3-2] VMS732_UPDATE-V0200 4 Message-ID: <1040614183510.404C-100000@Ives.egh.com>  = Similar problem with the new DECnet ECO (V7.3, ECO 4) for VAX < on my uVAX 3600 at home.  Downloaded some time ago, tried toA install on Saturday.  About 60% of the way through, it complained B about a bad component in the PCSI kit; it may have been INVDOCMTL, but I don't remember for sure.  @ Anyway, I downloaded it again, re-expanded, and the 2nd time, it? worked fine.  Was it fixed on the FTP server between 29-May and 4 12-Jun, or is FTP (from ftp.itrc.hp.com) unreliable?  5 On Mon, 14 Jun 2004, Peter 'EPLAN' LANGSTOEGER wrote:    > <RANT>B > Today I downloaded the latest ECO VMS732_UPDATE-V0200 from ITRC. > : > 1) I tried to install it on my system at work. It failed > R > -PCSI-E-REQVOLSPC, 218722 blocks required; 138276 blocks free; -80446 blocks netY > -PCSI-I-DELRDSETS, you can reclaim disk space by using the DELETE RECOVERY_DATA command G > Terminating is strongly recommended.  Do you want to terminate? [YES]  > " > Ok, not enough disk space. Alas.( > But some lines later during 'rollback' >  > Portion done: 0%1 > %PCSI-E-S_OPCAN, operation cancelled by request  > M > %PCSI-I-CREDIR, created directory DSA0:[SYS0.SYSCOMMON.SYSMGR.PCSI$WRKA657] V > %PCSI-I-CREDIR, created directory DSA0:[SYS0.SYSCOMMON.SYSMGR.PCSI$WRKA657.][SYSUPD]\ > %PCSI-E-READERR, error reading DSA0:[INSTALL_KITS]DEC-AXPVMS-VMS732_UPDATE-V0200--4.PCSI;2\ > -PCSI-E-INVDOCMTL, internal error - product document has invalid product material ordering > H > Hhmmm. Where did I see the INVDOCMTL the last time ? I'll know soon... >  > M > 2) After some ZIPping to get free diskspace, I tried again. It failed after 
 > 10% with > \ > %PCSI-E-READERR, error reading DSA0:[INSTALL_KITS]DEC-AXPVMS-VMS732_UPDATE-V0200--4.PCSI;2\ > -PCSI-E-INVDOCMTL, internal error - product document has invalid product material ordering$ > %PCSI-E-OPFAILED, operation failedG > Terminating is strongly recommended.  Do you want to terminate? [YES] V > %PCSI-E-CANCEL_WIP, termination resulted in an incomplete modification to the system1 > %PCSI-E-S_OPCAN, operation cancelled by request  > M > %PCSI-I-CREDIR, created directory DSA0:[SYS0.SYSCOMMON.SYSMGR.PCSI$WRKA665] V > %PCSI-I-CREDIR, created directory DSA0:[SYS0.SYSCOMMON.SYSMGR.PCSI$WRKA665.][SYSUPD] >  > Recovery pass starting ... > \ > %PCSI-E-READERR, error reading DSA0:[INSTALL_KITS]DEC-AXPVMS-VMS732_UPDATE-V0200--4.PCSI;2\ > -PCSI-E-INVDOCMTL, internal error - product document has invalid product material ordering > @ > Hhhmm. The INVDOCMTL again. Where did I see it the last time ?B > Now I know. It was a corrupt PCSI file. Corrupted by download orG > by unpacking ? How will I know this ? A new download/unpack fixed it.  >  > G > 3) I downloaded the ECO at home and surprise installation failed with  >  > Portion done: 60%  > R > %PCSI-E-OPENOUT, error opening DISK$VMSSYS:[SYS0.SYSCOMMON.][SYSLIB]DECW$SVEXT_G > LX_SW.EXE; as output= > -RMS-F-FUL, device full (insufficient space for allocation) $ > %PCSI-E-OPFAILED, operation failedG > Terminating is strongly recommended.  Do you want to terminate? [YES] R > %PCSI-E-CANCEL_WIP, termination resulted in an incomplete modification to the sy > stemI > %PCSI-I-DELFIL, deleted DSA0:[SYS0.SYSMGR.PCSI$WRK202002B9]SYSUPD.DIR;1  > D > So, the PCSI kit is not corrupted this time. At least it seems so.M > (Then where comes the corruption from ? Was the kit on the FTP server fixed N > in the meantime [2h] ? Was it the download ? Why without any error message ?H > Or was it the DCX unpacking ? Why also without a warning ? Why can VMSM > engineering not also use ZIPSFX instead of the error prone DCX ? Please !!)  > G > But it also seems, that the freespace check is insufficient. (@%$&^*)  >  > & > What the heck is the problem today ?8 > Can an ECO not successfully install at the first try ?	 > </RANT>  >  > Good luck, folks >  > --   > Peter "EPLAN" LANGSTOEGER ' > Network and OpenVMS system specialist  > E-mail  peter@langstoeger.atH > A-1030 VIENNA  AUSTRIA              I'm not a pessimist, I'm a realist >  >    --   John Santos  Evans Griffiths & Hart, Inc. 781-861-0670 ext 539   ------------------------------  % Date: Mon, 14 Jun 2004 21:41:05 -0500 2 From: David J Dachtera <djesys.nospam@comcast.net>7 Subject: Re: [OpenVMS Alpha V7.3-2] VMS732_UPDATE-V0200 + Message-ID: <40CE61C0.9F930A4B@comcast.net>    John Santos wrote: > ? > Similar problem with the new DECnet ECO (V7.3, ECO 4) for VAX > > on my uVAX 3600 at home.  Downloaded some time ago, tried toC > install on Saturday.  About 60% of the way through, it complained D > about a bad component in the PCSI kit; it may have been INVDOCMTL,  > but I don't remember for sure. > B > Anyway, I downloaded it again, re-expanded, and the 2nd time, itA > worked fine.  Was it fixed on the FTP server between 29-May and 6 > 12-Jun, or is FTP (from ftp.itrc.hp.com) unreliable?   FWIW...   C See the HELP for PRODUCT. You can unpack the kit into a "reference" G (opposite of "sequential") kit without actually installing it. That may + tell you whether the kit is in fact intact.   A Don't have a bad one here to test with. Sorry. Just a suggestion.    D.J.D.   ------------------------------  % Date: Tue, 15 Jun 2004 07:13:27 +0200 2 From: Karl Rohwedder <emil.mustermann@t-online.de>7 Subject: Re: [OpenVMS Alpha V7.3-2] VMS732_UPDATE-V0200 ) Message-ID: <cam0he$kdp2@doiweb4.b2x.vwg>     Peter 'EPLAN' LANGSTOEGER wrote: > <RANT>B > Today I downloaded the latest ECO VMS732_UPDATE-V0200 from ITRC. > : > 1) I tried to install it on my system at work. It failed >    ... some cuts ...   L I have installed this patch on 4 systems until now without any difficulties.   ------------------------------  + Date: Tue, 15 Jun 2004 05:39:51 +0000 (UTC) 6 From: peter@langstoeger.at (Peter 'EPLAN' LANGSTOEGER)7 Subject: Re: [OpenVMS Alpha V7.3-2] VMS732_UPDATE-V0200 0 Message-ID: <newscache$uq4czh$lpu$1@news.sil.at>  ` In article <40CE61C0.9F930A4B@comcast.net>, David J Dachtera <djesys.nospam@comcast.net> writes:D >See the HELP for PRODUCT. You can unpack the kit into a "reference"H >(opposite of "sequential") kit without actually installing it. That may, >tell you whether the kit is in fact intact.  M Of course you can. And as a burnt kid I do it quite often. But not this time. K And I also didn't check the freespace size and compared it with the docs...   B >Don't have a bad one here to test with. Sorry. Just a suggestion.   Now I have another one. ? Funny. I did delete the last one yesterday for space reasons...    --   Peter "EPLAN" LANGSTOEGER % Network and OpenVMS system specialist  E-mail  peter@langstoeger.atF A-1030 VIENNA  AUSTRIA              I'm not a pessimist, I'm a realist   ------------------------------    Date: 14 Jun 2004 22:45:56 -0700$ From: n8wxs@arrl.net (Jeff Campbell)W Subject: Re: [TCPIP] How is the routing table updated when ROUTED or GATED aren't used? = Message-ID: <1a40a0b2.0406142145.556aae24@posting.google.com>   t "Matt Muggeridge" <Matt.Muggeridge@hp.com> wrote in message news:<Gx7yc.1720$sj4.1322@news-server.bigpond.net.au>...N > This may be due to route redirects and is usually indicative of a network or > router setup problem.  > I > A network diagram along with subnet addressing and host addresses would L > help.  Output from "netstat -nr" on both the PC and server would shed moreH > light on the situation, along with route information for the local and > remote routers.  > L > If the Flags field in the output from "netstat -nr" contains "D", then theM > route was dynamically created as the result of an ICMP redirect.  You don't K > usually want to disable redirects, as they provide a very useful service.  > H > Without more info, it sounds like your VMS node attempts to send an IPL > datagram to a remote PC and first sends to the default (.4) router (as youG > would expect it to).  When the default router determines the next hop N > address, it forwards the datagram out the same interface that it received itL > on, which is the signal to send an ICMP redirect back to VMS telling it toJ > use 192.168.10.1 in the future.  VMS then attempts to send to the remoteN > router directly, which of course fails, since it is not on the same network. > M > I would be inclined to look very closely at the subnetting and router table  > on the two routers.  >  > Matt.  >  > --  ? > -------------------------------------------------------------  > OpenVMS TCP/IP Engineering > Enterprise Computing Group > Hewlett-Packard Company  > Gold Coast, AUSTRALIA ? > -------------------------------------------------------------  >   = Thanks Matt, and to the others who responded to my questions.   # Here is basically the interconnect:      local net 192.168.12/24 0  T--+------+--------------------------------+--T-     |      |                                |sJ   DS20   CISCO <---T1---> CISCO            WAN <---> (picture of cloud 8-)-    .9     .4            remote net         .1c$                        192.168.10/24  L I have discovered that the remote end of the T1 line suffered a power outageL prior to the routing issue becoming apparent. What seems to have happened isB that the local router lost contact with the remote end and startedI redirecting traffic to the 192.168.10 net to it's default route, which isxO the WAN router. Once contact between the two ends of the T1 was re-established,RJ subsequent traffic to the 192.168.10 net was correctly routed down the T1.  N The problem is that it shouldn't have done that. 8-) The router's route to theO remote end of the T1 should be immutable. If packets get dropped due to loss ofAN connectivity, so be it. The local end router should not be redirecting traffic. intended for the remote end of the T1 circuit.  P Manually issuing $TCPIP SET NOROUTE 192.168.10.xxx for the 5 bad routes returned the Alpha to normal operation.  , Jeff n8wxs@arrl.net  	 [snipped]    ------------------------------   End of INFO-VAX 2004.330 ************************