1 INFO-VAX	Mon, 06 Sep 2004	Volume 2004 : Issue 495       Contents:: Re: Changing "DECnet Phase IV Prefix" under DECnet Phase V: Re: Changing "DECnet Phase IV Prefix" under DECnet Phase V- cluster member crashes during boot (again...) 1 Re: cluster member crashes during boot (again...) 1 Re: cluster member crashes during boot (again...) 1 Re: cluster member crashes during boot (again...) 1 Re: cluster member crashes during boot (again...) 1 Re: cluster member crashes during boot (again...) 1 Re: cluster member crashes during boot (again...)  Re: cluster problems (again) Re: cluster problems (again) Re: cluster problems (again) Re: cluster problems (again) Re: cluster problems (again) Re: Definition for SOR$GK_* ?  Re: EV8 advertised by accident Re: Future of Decnet Re: Future of Decnet Re: Future of Decnet Re: Future of Decnet Re: Future of DecnetB Re: How to monitor RAM usage and automatically send warning email?! Re: NCL command for SHOW OBJECT ?  VMS Cluster and IP Subnetting ! RE: VMS Cluster and IP Subnetting ! Re: VMS Cluster and IP Subnetting   F ----------------------------------------------------------------------  % Date: Mon, 06 Sep 2004 08:15:42 +0200  From: Dirk Munk <munk@home.nl>C Subject: Re: Changing "DECnet Phase IV Prefix" under DECnet Phase V 2 Message-ID: <chgvai$2ls$1@news2.tilbu1.nb.home.nl>   Paul Sture wrote:   J > A real pain to work through the NET$CONFIGURE menu for even one client, A > let alone multiple ones, so direct editing was the way to go...  > N This system already was configured, and then a complete configuration is very Q simple. After changing the address information, the procedure will discover that  P   there are scripts for all other subjects, and it will ask if you want to keep R them. Just answer "yes", and the whole configuration will take less then a minute.   ------------------------------  % Date: Mon, 06 Sep 2004 12:46:39 +0200 * From: Paul Sture <nospam@sture.homeip.net>C Subject: Re: Changing "DECnet Phase IV Prefix" under DECnet Phase V * Message-ID: <2q2tgjFqusk4U1@uni-berlin.de>   John Santos wrote: > Paul Sture wrote:  >  >> Dirk Munk wrote:  >>4 >>> This will work, but only untill the next reboot. >>> B >>> You will have to do a complete reconfiguration of your DECnet. >>>  >>>  >>H >> Totally unsupported, and I don't think I'd dream of trying to change I >> the prefix this way, but this may help understand what goes on during  ! >> the startup of DECnet Phase V.  >>F >> In brief, NET$CONFIGURE.COM generates scripts which are run during G >> startup and does checksums on them. If a checksum doesn't match on a 5 >> reboot, a freshly modified script file is ignored.  >> > J > My experience has been that DECNET DOES execute them on reboot, but theyH > get ignored the next time you run NET$CONFIGURE it tells you the filesF > have invalid checksums, and they get reset to default configuration. >   . I believe you are right, so I stand corrected.  E > To preserve changes across NET$CONFIGURE.COM, you need to make them  > with NET$CONFIGURE.  > C > However, I haven't tried the "Fix the checksums" trick below.  It E > ought to work, but the danger is if there is a typo, or if you make F > a change that NET$CONFIGURE can't parse (I don't know if it uses the@ > same parsing tables as NCL, so it is possible that there existB > syntactically valid commands that NET$CONFIGURE can't cope with.F > If that happens, NET$CONFIGURE might ignore your changes or die when > it tries to parse them.    That is indeed the danger.   > . >> Here's a post I did in 2001 on the subject: >> >> http://tinyurl.com/6jamb  >>G >> --------------------------------------------------------------------  >>G >> This is almost certainly totally unsupported, but in SYS$SYSTEM you  F >> will find some NET$CHECKSUM*.DAT files. It is possible to edit the I >> various startup .NCL files and then preserve them by running CHECKSUM  C >> against the new version, and putting the result in the relevant   >> NET$CHECKSUM*.DAT file. >>J >> I just went through this last week when I was setting up a development I >> cluster with multiple versions of VMS. I discovered that the MOP info  H >> had been set up to load the system image from DISK$LABEL: instead of  >> SYS$SYSDEVICE:  >>C >> A real pain to work through the NET$CONFIGURE menu for even one  J >> client, let alone multiple ones, so direct editing was the way to go... >>G >> --------------------------------------------------------------------  >  > E > I often hand-edit them, then save a copy of the new version.  After F > running NET$CONFIGURE, I compare my saved versions with the new ones@ > produced by NET$CONFIGURE and restore (or edit and restore if # > appropriate) my changed versions.  > D > This of course is completely unsupported, but is often quicker andH > less error-prone for making one-line changes or additions than walkingC > through a slew of questions in obscure submenus in NET$CONFIGURE.   H And it carries the health warning that if someone other than you should > try to run NET$CONFIGURE they could innocently mess things up.  E > IMHO, NET$CONFIGURE is not so much your friend as a cranky neighbor ? > who will sometimes give you a ride to pick up your car at the H > repair shop, or do other favors if you ask nice (but sometimes won't.) >    True.    ------------------------------  % Date: Mon, 06 Sep 2004 16:34:36 +0100 + From: Rodrigo Ventura <yoda@isr.ist.utl.pt> 6 Subject: cluster member crashes during boot (again...)) Message-ID: <m3k6v7if77.fsf@pixie.isrnet>   ? This time I added with SYSGEN two swapfiles and one pagefile in E [SYSn.SYSEXE], and booted with -fl 0,20000, but no avail: it crashes! < It is so verbose that I can't figure out exactly when/how it- crashes. The anal/crash is not very explicit:   F ---------------------------------------------------------------------- -System crash information  ------------------------- Time of system crash:  6-SEP-2004 14:42:53.76 H Version of system: OpenVMS (TM) Alpha Operating System, Version V7.3-1  % System Version Major ID/Minor ID: 3/0 - VMScluster node: xxxxxx, a DEC 3000 Model 400 # Crash CPU ID/Primary CPU ID:  00/00 4 Bitmask of CPUs active/available:  00000001/00000001 CPU bugcheck codes: 1         CPU 00 -- PROCGONE, Process not in system   " CPU 00 Processor crash information" ----------------------------------; CPU 00 reason for Bugcheck: PROCGONE, Process not in system 0 Process currently executing on this CPU: SYSINIT Current image file: SYSINIT.EXE  Current IPL: 0  (decimal)  CPU database address:  8142E000 E ---------------------------------------------------------------------   E Moreover there is also a SYS$ERRLOG.DMP file -- how can I examine it?   F And my main question: how can I redirect all output of the log process? (including the increased verbosity with -fl 0,20) to a logfile?   E And by the way: regarding another node, in despair, I created lots of  swap/pagefiles:   F ----------------------------------------------------------------------L Swap File Usage (8KB pages):                   Index        Free        SizeP   DISK$ALPHASYS:[SYS10.SYSEXE]SWAPFILE.SYS                                      L                                                    1           0          32P   DISK$ALPHASYS:[SYS10.SYSEXE]SWAPFILE1.SYS;1                                   L                                                    2           0          24P   DISK$xxxxxx_1026:[SYSEXE]SWAPFILE.SYS;1                                       L                                                    3         360         496  L   Total size of all swap files:                                          552  L Paging File Usage (8KB pages):                 Index        Free        SizeP   DISK$xxxxxx_1026:[SYSEXE]PAGEFILE.SYS;1                                       L                                                  253        8243        8704P   DISK$ALPHASYS:[SYS10.SYSEXE]PAGEFILE.SYS                                      L                                                  254          69         536  L   Total size of all paging files:                                       9240L   Total committed paging file usage:                                    4979F ----------------------------------------------------------------------  A (this is a sattelite node, with one local disk; DISK$ALPHASYS: is 2 remote and DISK$xxxxxx_1026: is local to the node)  C Well, the first thing you'll certainly find odd is the sizes of the F page/swapfiles -- I don't know about units here, I supposed it was allC measured in blocks in VMS, but here it seems it is not. So the size A one puts in SYSGEN to create the files is not in blocks. What are ? reasonable sizes for temporary swap and page files in a cluster - sattelite? The node's memory is about 32MB...   > My second problem is that my intention was to create temporaryD page/swapfiles to help the node boot without memory problems (beforeD mounting the localdisk page/swap files). After booting, is there anyF way to move everything to the local disk page/swapfiles and disable to remote ones?   Cheers,    Rodrigo    --    : *** Rodrigo Martins de Matos Ventura <yoda@isr.ist.utl.pt>. ***  Web page: http://www.isr.ist.utl.pt/~yoda0 ***   Teaching Assistant and PhD Student at ISR:7 ***    Instituto de Sistemas e Robotica, Polo de Lisboa 4 ***     Instituto Superior Tecnico, Lisboa, PORTUGALH *** PGP fingerprint = 0119 AD13 9EEE 264A 3F10  31D3 89B3 C6C4 60C6 4585   ------------------------------   Date: 6 Sep 2004 18:14:45 +0200 . From: huber@NOBODY-mppmu.mpg.de (Joseph Huber): Subject: Re: cluster member crashes during boot (again...)+ Message-ID: <hIeOdV1E4Cpa@vms.mppmu.mpg.de>   W In article <m3k6v7if77.fsf@pixie.isrnet>, Rodrigo Ventura <yoda@isr.ist.utl.pt> writes:  > A > This time I added with SYSGEN two swapfiles and one pagefile in G > [SYSn.SYSEXE], and booted with -fl 0,20000, but no avail: it crashes! G > And by the way: regarding another node, in despair, I created lots of  > swap/pagefiles:  > H > ----------------------------------------------------------------------N > Swap File Usage (8KB pages):                   Index        Free        SizeR >   DISK$ALPHASYS:[SYS10.SYSEXE]SWAPFILE.SYS                                      N >                                                    1           0          32R >   DISK$ALPHASYS:[SYS10.SYSEXE]SWAPFILE1.SYS;1                                   N >                                                    2           0          24R >   DISK$xxxxxx_1026:[SYSEXE]SWAPFILE.SYS;1                                       N >                                                    3         360         496 > N >   Total size of all swap files:                                          552 > N > Paging File Usage (8KB pages):                 Index        Free        SizeR >   DISK$xxxxxx_1026:[SYSEXE]PAGEFILE.SYS;1                                       N >                                                  253        8243        8704R >   DISK$ALPHASYS:[SYS10.SYSEXE]PAGEFILE.SYS                                      N >                                                  254          69         536 > N >   Total size of all paging files:                                       9240N >   Total committed paging file usage:                                    4979H > ---------------------------------------------------------------------- > C > (this is a sattelite node, with one local disk; DISK$ALPHASYS: is 4 > remote and DISK$xxxxxx_1026: is local to the node) > @ > My second problem is that my intention was to create temporaryF > page/swapfiles to help the node boot without memory problems (beforeF > mounting the localdisk page/swap files). After booting, is there anyH > way to move everything to the local disk page/swapfiles and disable to > remote ones?  E First, remove all page/swap files on the system disk (disk$alphasys), D they are anyway too small to be usefull: the size is in Alpha pages,1 so multiply it by 8192 to get the size in KBytes. 8 If the satellite is not running : just delete the files,J if running , rename the files to something, then after a satellite reboot, delete the files.   I Since You showed a "SHOW memory" of the satellite, it was running already 8 earlier, no ? After which action did You get the crash ?  D In general I think, a 32MB memory Alpha is not able to run VMS 7.3x, at least not supported.     --  6    Joseph Huber, Muenchen  http://www.huber-joseph.de/   ------------------------------  % Date: Mon, 06 Sep 2004 17:43:45 +0100 + From: Rodrigo Ventura <yoda@isr.ist.utl.pt> : Subject: Re: cluster member crashes during boot (again...)) Message-ID: <m3brgjibzy.fsf@pixie.isrnet>   B >>>>> "Joseph" == Joseph Huber <huber@NOBODY-mppmu.mpg.de> writes:  @     Joseph> First, remove all page/swap files on the system diskE     Joseph> (disk$alphasys), they are anyway too small to be usefull: E     Joseph> the size is in Alpha pages, so multiply it by 8192 to get B     Joseph> the size in KBytes.  If the satellite is not running :C     Joseph> just delete the files, if running , rename the files to @     Joseph> something, then after a satellite reboot, delete the     Joseph> files.  E The idea was to provide the booting node with some page/swap space in F order to prevent it from running out of memory. But after boot, it can solely use the local disk.  E In case of not being possible to remove one file from swap/paging, is D it possible to assign priorities? For instance, local disk page/swap. files having higher priority than remote ones.  E (curiously in linux it is fairly easy to do either operation: you can D remove/add swap partitions/files at will with the swapon and swapoff? commands, and you can assign priorities with the same commands)     E     Joseph> Since You showed a "SHOW memory" of the satellite, it was D     Joseph> running already earlier, no ? After which action did You     Joseph> get the crash ?   E Sorry, I was talking about two different satellite nodes. My setup is @ one master node (DEC 3000/500 w/ 64MB) with the system disk, oneA satellite (DEC 3000/500 w/ 64MB) that crashed last friday, and is B running with all those swap/page files, and another satellite (DECD 3000/400 w/ 32MB) which crashes during boot. My last post refered toA different problems in each node, sorry for not making that clear.     F     Joseph> In general I think, a 32MB memory Alpha is not able to run-     Joseph> VMS 7.3x, at least not supported.    B Hum, I read somewhere that the minimum memory to run VMS was aboutC 4MB, which I found quite amazing -- in fact I have here a couple of E old VAXstations (3100 and 2000) which I was considering to add to the D cluster, at the end. But since I can't get all those alphas running,& I'm not worried with that right now...   Cheers,    Rodrigo    --    : *** Rodrigo Martins de Matos Ventura <yoda@isr.ist.utl.pt>. ***  Web page: http://www.isr.ist.utl.pt/~yoda0 ***   Teaching Assistant and PhD Student at ISR:7 ***    Instituto de Sistemas e Robotica, Polo de Lisboa 4 ***     Instituto Superior Tecnico, Lisboa, PORTUGALH *** PGP fingerprint = 0119 AD13 9EEE 264A 3F10  31D3 89B3 C6C4 60C6 4585   ------------------------------  % Date: Mon, 06 Sep 2004 17:49:21 +0100 + From: Rodrigo Ventura <yoda@isr.ist.utl.pt> : Subject: Re: cluster member crashes during boot (again...)) Message-ID: <m37jr7ibqm.fsf@pixie.isrnet>   @ >>>>> "Rodrigo" == Rodrigo Ventura <yoda@isr.ist.utl.pt> writes:  E     Rodrigo> Hum, I read somewhere that the minimum memory to run VMS D     Rodrigo> was about 4MB, which I found quite amazing -- in fact I  	 Found it!   F ---------------------------------------------------------------------- 1.1.6 Restrictions  1 Restrictions for OpenVMS VAX Version 7.2 include:   E     * The minimum memory requirement for OpenVMS Vax Version 7.2 is 4 A megabytes. Additional components, such as DECnet for OpenVMS VAX, E DIGITAL TCP/IP Services, or VAXcluster, require more memory to ensure  satisfactory performance. F ----------------------------------------------------------------------9 (http://h71000.www7.hp.com/doc/72final/6521/6521pro.html)   ? Ok, my mistake, this is for VAX and for 7.2. I've been using an : hobbyist licence of 7.3-1 for alpha. Is it much different?   Cheers,    Rodrigo    --    : *** Rodrigo Martins de Matos Ventura <yoda@isr.ist.utl.pt>. ***  Web page: http://www.isr.ist.utl.pt/~yoda0 ***   Teaching Assistant and PhD Student at ISR:7 ***    Instituto de Sistemas e Robotica, Polo de Lisboa 4 ***     Instituto Superior Tecnico, Lisboa, PORTUGALH *** PGP fingerprint = 0119 AD13 9EEE 264A 3F10  31D3 89B3 C6C4 60C6 4585   ------------------------------   Date: 6 Sep 2004 19:01:23 +0200 . From: huber@NOBODY-mppmu.mpg.de (Joseph Huber): Subject: Re: cluster member crashes during boot (again...)+ Message-ID: <f9S00EuGQawi@vms.mppmu.mpg.de>   W In article <m3brgjibzy.fsf@pixie.isrnet>, Rodrigo Ventura <yoda@isr.ist.utl.pt> writes: G > The idea was to provide the booting node with some page/swap space in H > order to prevent it from running out of memory. But after boot, it can > solely use the local disk. > G > In case of not being possible to remove one file from swap/paging, is F > it possible to assign priorities? For instance, local disk page/swap0 > files having higher priority than remote ones. > G > (curiously in linux it is fairly easy to do either operation: you can F > remove/add swap partitions/files at will with the swapon and swapoffA > commands, and you can assign priorities with the same commands)   A I don't know if ever in the early boot-phase a pagefile is needed  (somebody proves me wrong ?). G Anyhow, once a file is in use, You can't remove it in a running system.   D > Hum, I read somewhere that the minimum memory to run VMS was aboutE > 4MB, which I found quite amazing -- in fact I have here a couple of G > old VAXstations (3100 and 2000) which I was considering to add to the F > cluster, at the end. But since I can't get all those alphas running,( > I'm not worried with that right now...  + This true for Vaxen (and VMS prior to 6.0), 0  but certainly not for Alphas, and it never was.? I don't have the SPD for VMS 7.3-2 ready, but I'm sure it needs  something beyond 32MB .     --  6    Joseph Huber, Muenchen  http://www.huber-joseph.de/   ------------------------------   Date: 6 Sep 2004 19:29:31 +0200 . From: huber@NOBODY-mppmu.mpg.de (Joseph Huber): Subject: Re: cluster member crashes during boot (again...)+ Message-ID: <HxsVOFUkYUHg@vms.mppmu.mpg.de>   W In article <m37jr7ibqm.fsf@pixie.isrnet>, Rodrigo Ventura <yoda@isr.ist.utl.pt> writes: A >>>>>> "Rodrigo" == Rodrigo Ventura <yoda@isr.ist.utl.pt> writes:  > G >     Rodrigo> Hum, I read somewhere that the minimum memory to run VMS F >     Rodrigo> was about 4MB, which I found quite amazing -- in fact I >  > Found it!  > H > ---------------------------------------------------------------------- > 1.1.6 Restrictions > 3 > Restrictions for OpenVMS VAX Version 7.2 include:  > G >     * The minimum memory requirement for OpenVMS Vax Version 7.2 is 4 C > megabytes. Additional components, such as DECnet for OpenVMS VAX, G > DIGITAL TCP/IP Services, or VAXcluster, require more memory to ensure  > satisfactory performance. H > ----------------------------------------------------------------------; > (http://h71000.www7.hp.com/doc/72final/6521/6521pro.html)  > A > Ok, my mistake, this is for VAX and for 7.2. I've been using an < > hobbyist licence of 7.3-1 for alpha. Is it much different? >   8 The SPD for Vax 7.3, Alpha 7.3-1/-2 (SPD 25.01.68) says:    Minimum memory for Alphas: 64 MB                    Vaxen:   14 MB    --  6    Joseph Huber, Muenchen  http://www.huber-joseph.de/   ------------------------------  % Date: Mon, 06 Sep 2004 18:31:51 +0100 + From: Rodrigo Ventura <yoda@isr.ist.utl.pt> : Subject: Re: cluster member crashes during boot (again...)) Message-ID: <m33c1vi9rs.fsf@pixie.isrnet>   B >>>>> "Joseph" == Joseph Huber <huber@NOBODY-mppmu.mpg.de> writes:  F     Joseph> I don't know if ever in the early boot-phase a pagefile isD     Joseph> needed (somebody proves me wrong ?). Anyhow, once a file?     Joseph> is in use, You can't remove it in a running system.   @ It was an hypothesis from me that the problem was lack of memoryC before mounting the local disk swap/page files. Since it crashed as F before, it may have nothing to do. The hypothesis followed a series of9 lack of memory warnings that seemed to lead to the crash.     E     Joseph> This true for Vaxen (and VMS prior to 6.0), but certainly C     Joseph> not for Alphas, and it never was.  I don't have the SPD @     Joseph> for VMS 7.3-2 ready, but I'm sure it needs something     Joseph> beyond 32MB .   C I see. In the master node, which has 64MB, the SHOW MEMORY commands B tells that 3441 pages are permanently allocated to OpenVMS. If theA page size is 8K, this means 26MB. In a 32MB machine, 5MB seems an D awfully small space for user memory... Although, this master node is@ running DECwindows, while I don't want DECwindows running on the satellites.   D Is it possible to tune the system parameters such that VMS uses lessF memory? One problem I had with DECwindows was lack of free GBLSECTION;D so I had to modify MODPARAMS.DAT to add a few more. Since this files= seems common to all nodes, I could reduce this (and/or other) 
 parameter(s).    Cheers,    Rodrigo    --    : *** Rodrigo Martins de Matos Ventura <yoda@isr.ist.utl.pt>. ***  Web page: http://www.isr.ist.utl.pt/~yoda0 ***   Teaching Assistant and PhD Student at ISR:7 ***    Instituto de Sistemas e Robotica, Polo de Lisboa 4 ***     Instituto Superior Tecnico, Lisboa, PORTUGALH *** PGP fingerprint = 0119 AD13 9EEE 264A 3F10  31D3 89B3 C6C4 60C6 4585   ------------------------------  % Date: Mon, 06 Sep 2004 11:28:40 +0100 + From: Rodrigo Ventura <yoda@isr.ist.utl.pt> % Subject: Re: cluster problems (again) ) Message-ID: <m38ybnk7xj.fsf@pixie.isrnet>   > >>>>> "Keith" == Keith A Lewis <lewis@PROBE.MITRE.ORG> writes:  @     Keith> For satellite nodes, CLUSTER_CONFIG sets up a file in;     Keith> [SYSn.SYSEXE] called SATELLITE_PAGE.COM which is D     Keith> responsible for mounting the local disk(s) and installing     Keith> the files.   F I noticed that file, but I can't figure out whether/how it is run. HowD calls it? I haven't found any references to it in either startup.com nor systartup_vms.com.  F Hum, I've read somewhere that there is a file specifying which scripts
 are called...    Cheers,    Rodrigo    --    : *** Rodrigo Martins de Matos Ventura <yoda@isr.ist.utl.pt>. ***  Web page: http://www.isr.ist.utl.pt/~yoda0 ***   Teaching Assistant and PhD Student at ISR:7 ***    Instituto de Sistemas e Robotica, Polo de Lisboa 4 ***     Instituto Superior Tecnico, Lisboa, PORTUGALH *** PGP fingerprint = 0119 AD13 9EEE 264A 3F10  31D3 89B3 C6C4 60C6 4585   ------------------------------  % Date: Mon, 06 Sep 2004 12:50:55 +0200 * From: Paul Sture <nospam@sture.homeip.net>% Subject: Re: cluster problems (again) * Message-ID: <2q2tofFqusk4U2@uni-berlin.de>   Rodrigo Ventura wrote:> >>>>>>"Keith" == Keith A Lewis <lewis@PROBE.MITRE.ORG> writes: >  > B >     Keith> For satellite nodes, CLUSTER_CONFIG sets up a file in= >     Keith> [SYSn.SYSEXE] called SATELLITE_PAGE.COM which is F >     Keith> responsible for mounting the local disk(s) and installing >     Keith> the files.  > H > I noticed that file, but I can't figure out whether/how it is run. HowF > calls it? I haven't found any references to it in either startup.com > nor systartup_vms.com. > H > Hum, I've read somewhere that there is a file specifying which scripts > are called...  >   H On Alpha V7.3-2 it's called from VMS$DEVICE_STARTUP.COM, but only if it  exists.    ------------------------------  * Date: Mon, 6 Sep 2004 11:27:51 +0000 (UTC)% From: Dan Foster <usenet@evilphb.org> % Subject: Re: cluster problems (again) 6 Message-ID: <slrncjoig0.pr0.usenet@gaia.roc2.gblx.net>  V In article <m38ybnk7xj.fsf@pixie.isrnet>, Rodrigo Ventura <yoda@isr.ist.utl.pt> wrote:? >>>>>> "Keith" == Keith A Lewis <lewis@PROBE.MITRE.ORG> writes:  > A >    Keith> For satellite nodes, CLUSTER_CONFIG sets up a file in < >    Keith> [SYSn.SYSEXE] called SATELLITE_PAGE.COM which isE >    Keith> responsible for mounting the local disk(s) and installing  >    Keith> the files. > H > I noticed that file, but I can't figure out whether/how it is run. HowF > calls it? I haven't found any references to it in either startup.com > nor systartup_vms.com. > H > Hum, I've read somewhere that there is a file specifying which scripts > are called...   6 $ search sys$sysdevice:[000000...]*.com satellite_page   ...shows it to be:  > SYS$SYSDEVICE:[VMS$COMMON.SYS$STARTUP]VMS$DEVICE_STARTUP.COM;1  G $if f$search("sys$specific:[sysexe]satellite_page.com") .nes. "" then - D @sys$specific:[sysexe]satellite_page "''p1'" "''p2'" "''p3'" "''p4'" "''p5'" "''   H To see some of what scripts and utilities your VMS box runs during boot:   $ mcr sysman, SYSMAN> STARTUP SET DATA STARTUP$STARTUP_VMS SYSMAN> STARTUP SHOW FILE  SYSMAN> exit  5 You'll see VMS$DEVICE_STARTUP.COM being in that list.   H VMS$DEVICE_STARTUP.COM executes SYCONFIG, SYLOGICALS, SYPAGSWPFILES, and other important things.   H STARTUP.COM reads from a database of things to start up, which is how it4 learns about VMS$DEVICE_STARTUP.COM and executes it.   -Dan   ------------------------------  * Date: Mon, 6 Sep 2004 11:29:32 +0000 (UTC)% From: Dan Foster <usenet@evilphb.org> % Subject: Re: cluster problems (again) 6 Message-ID: <slrncjoij5.pr0.usenet@gaia.roc2.gblx.net>  V In article <2q2tofFqusk4U2@uni-berlin.de>, Paul Sture <nospam@sture.homeip.net> wrote: > J > On Alpha V7.3-2 it's called from VMS$DEVICE_STARTUP.COM, but only if it 	 > exists.   @ Ah, I didn't realize that behavior was specific to V7.3-2/Alpha.   Duly noted, thanks. :)  B Just out of curiosity, how was it handled in previous V7 versions?   -Dan   ------------------------------  % Date: Mon, 06 Sep 2004 14:00:00 +0200 * From: Paul Sture <nospam@sture.homeip.net>% Subject: Re: cluster problems (again) * Message-ID: <2q31qfFqhaerU1@uni-berlin.de>   Dan Foster wrote: X > In article <2q2tofFqusk4U2@uni-berlin.de>, Paul Sture <nospam@sture.homeip.net> wrote: > J >>On Alpha V7.3-2 it's called from VMS$DEVICE_STARTUP.COM, but only if it 	 >>exists.  >  > B > Ah, I didn't realize that behavior was specific to V7.3-2/Alpha. >  > Duly noted, thanks. :) > D > Just out of curiosity, how was it handled in previous V7 versions? >   A Sorry, I'm not saying it was handled any differently in previous  F versions; simply that I only looked on a V7.3-2 system. But I've just G checked on a VAX and Alpha, both running V7.2, and it's the same there.   H Things in the startup can change with versions, but you are more likely G to see differences between Alpha and VAX, which is why I mentioned the   architecture and version.    ------------------------------  % Date: Mon, 06 Sep 2004 15:45:20 +0100 - From: John Laird <nospam@laird-towers.org.uk> & Subject: Re: Definition for SOR$GK_* ?8 Message-ID: <3vtoj0p6g5jkcvffsnlu22l3rs3lf2u69j@4ax.com>  6 On Fri, 3 Sep 2004 17:58:57 +0000 (UTC), Dale Dellutri! <ddelQQQlutr@panQQQix.com> wrote:   F >A web search seems to indicate that SOR$GK_RECORD = 1, but I'd reall 5 >like to get the standard definition from the system.    INTEGER   SORTYP EXTERNAL  SOR$GK_RECORD    SORTYP = %LOC(SOR$GK_RECORD)  F should always work.  It is, as you have discovered, equal to 1.  StillJ missing from the FORSYSDEF library at 7.3-2 with Fortran 77 V7.5 installedG (and the source files I checked go back to about 1993 if not before !).    --   I dont nead no speling cheker!     Mail john rather than nospam...    ------------------------------  # Date: Mon, 06 Sep 2004 14:58:36 GMT 6 From: "Kenneth Farmer" <kfarmer@NOSPAM.spyderbyte.com>' Subject: Re: EV8 advertised by accident = Message-ID: <wW__c.8585$683.1017088@twister.southeast.rr.com>   5 "Rich Jordan" <duodec@speakeasy.net> wrote in message , news:Rr-dnTcFAIkdTqbcRVn-ug@speakeasy.net... > Kenneth Farmer wrote: 7 > > "Rich Jordan" <jordan@ccs4vms.com> wrote in message ; > > news:cc5619f2.0409031131.7bf3ffb4@posting.google.com...  > > E > >>On the last content page, HPTC Scalable Products, the last column ; > >>lists the HP AlphaServer SC45.  Processor listed is....  > >>* > >>64-bit Alpha EV8 processor at 1.25 GHz > >> > > $ > >>but since its Tru64 only I guess$ > >>its not worth the effort anyway. > >  > >  > > Take this as humor...  > > I > > Ever run Tru64 in a production environment?  If you had, you wouldn't  makeF > > that statement.  If you did, and you had problems with Tru64, then whomeverJ > > handled the servers wasn't good at it.  Second best in many regards toL > > OpenVMS? Yes.  It could be easily claimed as the most reliable UNIX, andG > > that comes from someone who has used them all, Solaris, AIX, HP-UX,  Linux,K > > etc. Tru64 doesn't deserve that kinda of disrespectful comment.  I mean L > > after all, we are talking Alpha chips here anyway, right?  Tru64 is part of > > the family.  > > G > > That's like looking at your brother on his death bed and saying, oh  well, K > > that cure they were going to try might have worked, but he's not really  > > worth it anyway. > >  > > :) > >  > > Ken  > I > Aww, Ken, come on.  My point was only that the SC is not supported with G > VMS.  Tru64 may very well be the best unix that will ever be, but its J > not VMS, and therefore is not of primary interest here.  Just like a boxH > that will only run Win-NT Alpha, and not VMS (or Tru-64 Unix) is of no > real interest here either.  F I know.  I tried to make it light with the intro statement and smiley.F After it was posted and I read it, it did sound a little too harsh.  I/ pressed the send button too quickly, sorry.  :)    Ken   # OpenVMS.org, http://www.openvms.org  Tru64.org, http://www.tru64.org  ________________________" Kenneth R. Farmer <>< 336-736-7376$ SpyderByte: http://www.SpydeByte.com   ------------------------------  % Date: Mon, 06 Sep 2004 09:54:54 +0100 - From: Roy Omond <Roy.Omond@BlueBubble.UK.Com>  Subject: Re: Future of Decnet * Message-ID: <2q2n3hFq0mskU1@uni-berlin.de>   Alan Greig wrote:    >> [... snip ...]  >  > But first there was ANF-10  " Now there's a blast-from-the-past.   What are you up to Alan ?    ------------------------------  % Date: Mon, 06 Sep 2004 11:13:19 +0100 + From: Rodrigo Ventura <yoda@isr.ist.utl.pt>  Subject: Re: Future of Decnet ) Message-ID: <m3d60zk8n4.fsf@pixie.isrnet>   P >>>>> "David" == David J Dachtera <djesys.nospam@NeOaSrPtAhMlNiOnWk.net> writes:  4     Roy> Yep, I know.  Adds credence to what I said.  B     David> Um, well, not really in the kernel. It's an add-on, andA     David> probably uses a kernel-loadable driver or some similar E     David> module, but I rather doubt that it is otherwise integrated B     David> into the Linux kernel very intimately (I could be wrong     David> about that...).  E Although the linux-decnet project was born separately from the kernel ? (probably via a loadable-module), current linux kernel versions F integrate that code into the kernel. I've never used it, but I've seenD it, for instance, in the kernel configuration options. Moreover, theD decnet code resides on the linux/net/decnet directory of the kernel;D it can be either compiled as a kernel module or be integrated in theE kernel image; in the former case, when it is loaded (modprobe decnet) " it issues a message like this one:  I NET4: DECnet for Linux: V.2.5.68s (C) 1995-2003 Linux DECnet Project Team 9 DECnet: Routing cache hash table of 1024 buckets, 8Kbytes " NET: Registered protocol family 12   Cheers,    Rodrigo    --    : *** Rodrigo Martins de Matos Ventura <yoda@isr.ist.utl.pt>. ***  Web page: http://www.isr.ist.utl.pt/~yoda0 ***   Teaching Assistant and PhD Student at ISR:7 ***    Instituto de Sistemas e Robotica, Polo de Lisboa 4 ***     Instituto Superior Tecnico, Lisboa, PORTUGALH *** PGP fingerprint = 0119 AD13 9EEE 264A 3F10  31D3 89B3 C6C4 60C6 4585   ------------------------------   Date: 6 Sep 2004 07:05:26 -0700 . From: fabiopenvms@yahoo.com.br (Fabio Cardoso) Subject: Re: Future of Decnet = Message-ID: <f30679fb.0409060605.165635ba@posting.google.com>   \ Rodrigo Ventura <yoda@isr.ist.utl.pt> wrote in message news:<m3d60zk8n4.fsf@pixie.isrnet>...R > >>>>> "David" == David J Dachtera <djesys.nospam@NeOaSrPtAhMlNiOnWk.net> writes: > 6 >     Roy> Yep, I know.  Adds credence to what I said. > D >     David> Um, well, not really in the kernel. It's an add-on, andC >     David> probably uses a kernel-loadable driver or some similarbG >     David> module, but I rather doubt that it is otherwise integratedtD >     David> into the Linux kernel very intimately (I could be wrong >     David> about that...). > G > Although the linux-decnet project was born separately from the kernelsA > (probably via a loadable-module), current linux kernel versionshH > integrate that code into the kernel. I've never used it, but I've seenF > it, for instance, in the kernel configuration options. Moreover, theF > decnet code resides on the linux/net/decnet directory of the kernel;F > it can be either compiled as a kernel module or be integrated in theG > kernel image; in the former case, when it is loaded (modprobe decnet)m$ > it issues a message like this one: > K > NET4: DECnet for Linux: V.2.5.68s (C) 1995-2003 Linux DECnet Project Teamn; > DECnet: Routing cache hash table of 1024 buckets, 8KbytesP$ > NET: Registered protocol family 12 > 	 > Cheers,- > 	 > Rodrigo-    > May be its a good solution in case of lack of Decnet Routers.% Does it support DDCMP for instance ? a   Regardse   FC   ------------------------------  % Date: Mon, 06 Sep 2004 15:26:49 +0100n+ From: Rodrigo Ventura <yoda@isr.ist.utl.pt>p Subject: Re: Future of DecnetF) Message-ID: <m3vferiic6.fsf@pixie.isrnet>n  A >>>>> "Fabio" == Fabio Cardoso <fabiopenvms@yahoo.com.br> writes:s  @     Fabio> May be its a good solution in case of lack of Decnet9     Fabio> Routers.  Does it support DDCMP for instance ?c  C I have no idea of what is that, but looking for that acronym in the 1 project FAQ (July 2004) one can find this answer:s  F ---------------------------------------------------------------------- 1.8 Does it do DDCMP?-  F We don't currently support DDCMP but some work is going on (slowly) toB add it to the feature list. If you need DDCMP then don't hold your4 breath it could be a very long time before it works.  C If you want to help with the development then please contact us viaw> the mailing list and I'll tell you what thoughts I have on the subject.F ----------------------------------------------------------------------   Cheers,>   Rodrigos   -- d  : *** Rodrigo Martins de Matos Ventura <yoda@isr.ist.utl.pt>. ***  Web page: http://www.isr.ist.utl.pt/~yoda0 ***   Teaching Assistant and PhD Student at ISR:7 ***    Instituto de Sistemas e Robotica, Polo de Lisboaw4 ***     Instituto Superior Tecnico, Lisboa, PORTUGALH *** PGP fingerprint = 0119 AD13 9EEE 264A 3F10  31D3 89B3 C6C4 60C6 4585   ------------------------------  % Date: Mon, 06 Sep 2004 19:36:18 +0200x From: Dirk Munk <munk@home.nl> Subject: Re: Future of Decnet 2 Message-ID: <chi76s$4gr$1@news2.tilbu1.nb.home.nl>   Fabio Cardoso wrote:   >  >  > @ > May be its a good solution in case of lack of Decnet Routers.' > Does it support DDCMP for instance ? s > 	 > RegardsC >  > FCN You can use any Alpha VMS system as DECnet-Plus router. There are intelligent I PCI WAN cards available too. However, DDCMP is no longer supported AFAIK.h   ------------------------------   Date: 6 Sep 2004 04:15:51 -0700a% From: dterzic@hemo.net (Dejan Terzic)eK Subject: Re: How to monitor RAM usage and automatically send warning email?a= Message-ID: <3fbb7bf7.0409060315.298e60dc@posting.google.com>p  
 Thank you!  X Andreas Gruhl <gruhl@isidata.de> wrote in message news:<41382D42.4F3FB183@isidata.de>... > Dejan Terzic wrote:' >  > > Hi!d > >0@ > > I need to create command procedure which will monitor systemG > > parameters (RAM memory usage) for OpenVMS cluster consisting of twoTH > > nodes and if it drops below 70% send a warning e-mail to a specifiedI > > address. It can be a batch job which will start every 10 minutes. ButoH > > I don't know how to extract RAM usage to a variable and how to start< > > mail utility from command procedure. Can anyone help me? > 2 > Here is a small example (working on VMS 7.3-2) : >  > $ OPEN/WRITE F MEMAIL.TMPg9 > $ WRITE  F "Memory Size    = ", F$GETSYI("TOTAL_PAGES")S8 > $ WRITE  F "Free Pages     = ", F$GETSYI("FREE_PAGES")< > $ WRITE  F "Modified Pages = ", F$GETSYI("MODIFIED_PAGES") > $ CLOSE  F2 > $ MAIL /SUBJECT="Memory Size"  MEMAIL.TMP  GRUHL >  > Good luck  > Andreas Gruhlo   ------------------------------  % Date: Mon, 06 Sep 2004 09:44:59 +0100-- From: Roy Omond <Roy.Omond@BlueBubble.UK.Com>-* Subject: Re: NCL command for SHOW OBJECT ?* Message-ID: <2q2mguFqjgo2U1@uni-berlin.de>   Larry Kilgallen wrote:A > After such good luck getting an answer to my previous question,$5 > let me ask how with NCL one does the equivalent of:p >  > 	NML> SHOW OBJECT FAL   7 Larry, I know Dave Sneddon already answered your query.n  7 This is just to remind folks that there exists a usefuls> utility in Sys$Update:DECNET_MIGRATE.Exe to aid in translating8 NCP commands into the equivalent NCL one.  Not being all5 that au fait with NCL, I find it invaluable.  In your 3 case you'd have got the equivalent answer as below:s   $ run sys$update:decnet_migratet1 DECNET_MIGRATE> convert command "show object fal"i  2 ! *** Converting the following NCP command to NCL: !     SHOW OBJECT FALo< SHOW NODE 0 SESSION CONTROL APPLICATION {{{FAL}}} ALL STATUS $   : (I've never really understood why you get these gratuitous# curly brackets round the "FAL" ...)   	 Roy Omond. Blue Bubble Ltd.   ------------------------------   Date: 6 Sep 2004 06:52:40 -070076 From: andrew.rycroft@intrinsitech.com (Andrew Rycroft)& Subject: VMS Cluster and IP Subnetting= Message-ID: <58ba0101.0409060552.7d76a357@posting.google.com>    Hi,   7 We have 2 sites that each run on different IP sub nets.s  E We are now going to implement an OpenVMS cluster between the 2 sites.-D The first site will be in one subnet, and the second site in anotherD subnet. Will this cause problems for the VMS cluster, and additional@ routing overhead ? We do have an IP alias. Are there any generalB rules/ best business practises for configuring TCP/IP in a cluster
 environment ?C   With regards Andrew   ------------------------------  $ Date: Mon, 6 Sep 2004 11:25:59 -0400' From: "Main, Kerry" <kerry.main@hp.com> * Subject: RE: VMS Cluster and IP SubnettingR Message-ID: <FD827B33AB0D9C4E92EACEEFEE2BA2FB3E02FC@tayexc19.americas.cpqcorp.net>   > -----Original Message-----B > From: Andrew Rycroft [mailto:andrew.rycroft@intrinsitech.com]=20! > Sent: September 6, 2004 9:53 AM  > To: Info-VAX@Mvb.Saic.ComP( > Subject: VMS Cluster and IP Subnetting >=20 > Hi,- >=209 > We have 2 sites that each run on different IP sub nets.i >=20G > We are now going to implement an OpenVMS cluster between the 2 sites. F > The first site will be in one subnet, and the second site in anotherF > subnet. Will this cause problems for the VMS cluster, and additionalB > routing overhead ? We do have an IP alias. Are there any generalD > rules/ best business practises for configuring TCP/IP in a cluster > environment ?8 >=20 > With regards > Andrew >=20   Andrew -=20   A The primary requirement is for there to be a dual connection (HA) E bridged network between the two sites for cluster SCS communications. H What some sites are doing these days is to implement multiple VLANS thatF stretch both sites. They typically separate the network traffic into aD separate VLAN for TCPIP traffic, cluster SCS traffic and a Mgmt VLAN3 (security and monitoring of servers e.g. consoles).   E Some additional resources to research the planning and design of this  multi-site OpenVMS cluster:l  - TCPIP Services High Availability planning:=20 ? http://h71000.www7.hp.com/openvms/journal/v2/articles/tcpip.pdfe  - A bit dated (VMS V7.3-1 base), but check out:.: http://h71000.www7.hp.com/openvms/fibre/fc_hbvs_dtc_wp.pdfB http://h71000.www7.hp.com/openvms/fibre/index.html (general stuff)   V8.2 Information (now in beta)H http://www.hp.com/products1/evolution/alpha_retaintrust/openvms/v82featu res.html=20 F http://h71000.www7.hp.com/doc/os82ft_index.html (doc's - see shadowingE guide for info on HBMM - this will be very important for anyone doing  multi-site clusters)  A Also, Keith Parris has some excellent presentations on multi-site- strategy at:  http://www2.openvms.org/kparris/  @ Keith also wrote an article for one of the VMS Tech Journals at:< http://h71000.www7.hp.com/openvms/journal/v1/disastertol.pdf  H And in addition, there is some information on multi-site clusters in the formal cluster doc's at:. http://h71000.www7.hp.com/doc/os732_index.htmlH http://h71000.www7.hp.com/doc/732FINAL/DOCUMENTATION/PDF/aa-q28lg-tk.PDF (V7.3-2 cluster guide)    C Btw, biased plug, but HP also has services available to assist witha! multi-site engagements like this.    :-)    Regardsn  
 Kerry Main Senior Consultants HP Services Canada Voice: 613-592-4660e Fax: 613-591-4477t kerryDOTmainAThpDOTcom (remove the DOT's and AT)=20  $ "OpenVMS has always had integrity .. Now, Integrity has OpenVMS .."   ------------------------------  % Date: Mon, 06 Sep 2004 19:29:53 +0200o From: Dirk Munk <munk@home.nl>* Subject: Re: VMS Cluster and IP Subnetting2 Message-ID: <chi6qq$f6r$1@news3.tilbu1.nb.home.nl>   Andrew Rycroft wrote:d > Hi,u > 9 > We have 2 sites that each run on different IP sub nets.L > G > We are now going to implement an OpenVMS cluster between the 2 sites.rF > The first site will be in one subnet, and the second site in anotherF > subnet. Will this cause problems for the VMS cluster, and additionalB > routing overhead ? We do have an IP alias. Are there any generalD > rules/ best business practises for configuring TCP/IP in a cluster > environment ?6 >  > With regards > Andrew  N Unless this problem has been resolved in the latest eco's of TCP/IP services, K you may encounter a problem with the routing configuration. For some silly 0O reason TCPIP$ROUTE.DAT is placed in SYS$SYSDEVICE:[VMS$COMMON.SYSEXE], and not rQ in a SYS$SPECIFIC directory branch. Creating a default route for cluster members -O in different subnets is impossible this way. Apparently the IP engineers never  " thought of multi-site clusters....   ------------------------------   End of INFO-VAX 2004.495 ************************