1 INFO-VAX	Sat, 29 Mar 2003	Volume 2003 : Issue 173       Contents: Re: 7.3-1 and EXBYTLIM <None>$ AS4100 / DEC Server 7300 differences Re: Conference On Clustering COV Sponsors Re: COV Sponsors Re: COV Sponsors Default Gateway  Re: Default Gateway  Re: Getting an Alphastation  Re: Getting an Alphastation  Re: Getting an Alphastation  Re: Getting an Alphastation  Help converting Alpha NT to VMS  Re: High RMS file activity Re: High RMS file activity Re: High RMS file activityL HP User Advocacy survey: Business Continuity and Disaster Recovery/Avoidance, Include an automated date within a printout.0 Re: Include an automated date within a printout.  Re: Macs seeing Pathworks shares Macs vs. Pathworks et. al. Re: Macs vs. Pathworks et. al. Re: Macs vs. Pathworks et. al.$ Re: Netscape gags on VMS FTP ServersG Re: OpenVMS v7.3-1 and Advanced Server v7.3A - Pathworks server crashes : Re: OpenVMS.org Survey:  Who's Been Knocking At Your Door?: Re: OpenVMS.org Survey:  Who's Been Knocking At Your Door? Overuse of LIB$SPAWN?  Re: Overuse of LIB$SPAWN?  Re: Overuse of LIB$SPAWN?  problems with accountn.dat file # Re: problems with accountn.dat file  Re: RamDisk futures  Re: RamDisk futures  Re: RamDisk futures  Re: RamDisk futures  Re: RamDisk futures  RamDisk question/ SIMH Emulation of VAX KA650-A, KA650-B, KA655-A # SIMH VAX build with MS Visual C++ 6  Re: Suggestion to Sue & Re: TCP/IP 5.1 and SMTP authentication/ Re: Three questions re: the VMS startup process / Re: Three questions re: the VMS startup process / Re: Three questions re: the VMS startup process / Re: Three questions re: the VMS startup process 5 Which PAK for DECNET Phase IV under Hobbyist Program?   F ----------------------------------------------------------------------  % Date: Sat, 29 Mar 2003 00:45:10 -0500   From: John Santos <JOHN@egh.com> Subject: Re: 7.3-1 and EXBYTLIM 4 Message-ID: <1030329002915.416F-100000@Ives.egh.com>  ( On Fri, 28 Mar 2003, Paul Webster wrote:  G > I have an application (but no source code) that works on VMS 7.1 (and & > I think 7.2) but is failing on 7.3-1 > = > The failure is %SYSTEM-F-EXBYTLM, exceeded byte count quota  > E > The account that it is running on is configured in same way on each G > system - but I tried pushing up the BYTLM anway - with no difference.  > E > I know that the code does a delete/create of a process logical name H > table and then stores some values in that table - and I also know thatA > the $CRELNT is being called with a Quota (size) of 0 (hopefully G > meaning unlimited) ... is there a chance that this no longer works in  > 7.3-1  > H > I can see that the Help text for $CRELNT has changed between 6 and 7 -B > could this mean that something that it is doing no longer works? > ? > Clearly the problem could be something totally different, but G > comparing some of the tracing output that the application makes leads C > me to think that it is somewhere around there that it is failing.  > --   > Rgds > Paul Webster  E Completely different thing that has worked fine for years, but is now ' dying on VMS 7.3-1 with the same error.   ? TCPWare IP-over-DECnet line to a remote site.  The local end is D an AlphaServer 1200 running VMS V7.3-1 (upgraded from V7.2-1 Tuesday@ night), DECnet-Plus, TCPware 5.6-2.  Remote end is VAX VMS V7.1,  DECnet-Plus V7.1, TCPware 5.3-3.  A The local end was the LISTENER.  Whenever TCPware tried to start, A it would get an %SYSTEM-F-EXBYTLM, exceeded byte count quota, and C TCPware would immediately shut itself down.  I removed the commands > to start the DECnet-0 line, and TCPware came up fine.  I tried@ executing them interactively and got the same error (but TCPware= doesn't shut itself down when you do it interactively, making @ testing much easier.)  I messed around a lot with SYSTEM's BYTLM: quota, eventually quadrupling it, without benefit.  Once I@ accidently left off the "/flags=listener", and it didn't die (or> maybe it got a different error, I don't remember).  Of course,> the line didn't start (no listener), but this was progress!  I< restarted the remote side as the listener, started the localA side as the active end, and the line came up and has been working  fine ever since.  B Since this time, I've installed the 7.3-1 DECnet ECO and rebooted,@ but I haven't tried swapping back to the original configuration.  ; Maybe there is a generic problem with BYTLM quota in 7.3-1?    --   John Santos  Evans Griffiths & Hart, Inc. 781-861-0670 ext 539   ------------------------------    Date: 28 Mar 2003 13:03:20 -0600; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler)  Subject: <None> 3 Message-ID: <b7uk1fX1wElg@eisner.encompasserve.org>   F    I know my Mac's can see MS servers at work.  If I set up the latestI    Pathworks or Advanced Server on my home cluster, will my Macs be able  G    to see its services?  (I don't want to try to track down and use the :    old version of Pathworks that supported Mac protocols).  )    (OBTW the Macs are running OS 8 or X).    ------------------------------  % Date: Fri, 28 Mar 2003 22:08:23 +0100 / From: Roland Barmettler <sendhere@spammail.com> - Subject: AS4100 / DEC Server 7300 differences ; Message-ID: <20030328220823.4abbf7dd.sendhere@spammail.com>    Hello   A I've been wondering whether it would be possible so load SRM onto C a DEC Server 7310 and run VMS on it. The systems seem very similar.  Thanks for any answers.  ;-)   Greetings, Roland    ------------------------------  % Date: Fri, 28 Mar 2003 15:48:36 -0500 * From: "Bill Todd" <billtodd@metrocast.net>% Subject: Re: Conference On Clustering 2 Message-ID: <PhmcnQNwady5KBmjXTWc3Q@metrocast.net>  5 "Nic Clews" <sendspamhere@127.0.0.1> wrote in message # news:3E84616F.AFDF0901@127.0.0.1...  > John Smith wrote:  > >  > > > > > "Hello and welcome to our technical session on clustering. > > H > > Without wanting to take up your valuable time on preamble, let's get > > started right away.  > > > > > Here is a rundown of what is available in the marketplace:5 > > a) IBM mainframe clusters work but are expensive. G > > b) All unix, linux, and Microsoft clusters don't work as advertised 1 > > and aren't worth the money, time, and effort. G > > c) Clusters that define the industry standard (tm) are VMS clusters G > > and we'll be spending the rest of this technical session discussing 
 > > them." > I > In reality I'm probably going to be slightly more tactful, in order not G > to be accused of being a XYZ product basher, I'll say what it can and A > can't do, and allow the listener to draw their own conclusions.   J Just be sure that you *know* what other clusters can and can't do.  One toJ be especially careful to understand is AIX clustering (and its distributedL GPFS file system).  Another is the Veritas clustering product suite.  And ofG course be sure to be up to date on the newest clustering mechanisms Sun  offers.    > , > Apart from that you're more or less right!  K No, he isn't:  there are several Unix-style clustering implementations that H offer approximate functional equivalence to VMS's facilities (even DLMs:I Tru64 acquired one recently, and AIX has had one for about a decade - and J donated it to Linux a couple of years ago), though usually with a bit moreL set-up folderol and with fail-over times that tend to be measured in tens ofK seconds rather than seconds.  Now that you've been notified, perhaps you've C got some studying to do if you don't want to run afoul of the trade  descriptions act.    - bill   ------------------------------    Date: 28 Mar 2003 20:30:45 -08001 From: susan_skonetski@hotmail.com (Sue Skonetski)  Subject: COV Sponsors = Message-ID: <857e9e41.0303282030.7bdb425a@posting.google.com>   D Does anyone else get a side bar with the COV sponser that says "help4 in migrating off VMS to Linux and a few other OS's"?   sue    ------------------------------    Date: 28 Mar 2003 23:13:01 -0600- From: Kilgallen@SpamCop.net (Larry Kilgallen)  Subject: Re: COV Sponsors 3 Message-ID: <BvVA3RVnXUcB@eisner.encompasserve.org>   q In article <857e9e41.0303282030.7bdb425a@posting.google.com>, susan_skonetski@hotmail.com (Sue Skonetski) writes: F > Does anyone else get a side bar with the COV sponser that says "help6 > in migrating off VMS to Linux and a few other OS's"?  6 Since I am using ANUNEWS on VMS to read the newsgroup, there is no "on the side".  M I don't suppose you are using a newsreader on some inferior OS, are you ? :-)    ------------------------------  % Date: Sat, 29 Mar 2003 00:28:16 -0500 $ From: Hein <hein_cov@eps.zk.dec.com> Subject: Re: COV Sponsors . Message-ID: <3E852EF0.6154355B@eps.zk.dec.com>  = I have send Sue a mail. We'll get her on an add-free internal  news-reader.F On the other hand... it is always good to know what 'the others' might
 be seeing!   Cheers,  Hein.    Sue Skonetski wrote:  F > Does anyone else get a side bar with the COV sponser that says "help6 > in migrating off VMS to Linux and a few other OS's"? >  > sue    ------------------------------  % Date: Fri, 28 Mar 2003 23:30:55 -0500 1 From: "routerwoman" <routerwoman@routerwoman.com>  Subject: Default Gateway0 Message-ID: <DOqcnQ3xPJwQvBijXTWcoQ@comcast.com>  L Is it possible to set up a default gateway on a VMS system.  I have a systemL that's running RIPv1 just to talk to the one and only router in the network.E Seems like a static route to the Gateway would stop all that chatter.  Thanks.    ------------------------------  + Date: Sat, 29 Mar 2003 04:58:49 +0000 (UTC) ) From: Dan Foster <dsf@globalcrossing.net>  Subject: Re: Default Gateway2 Message-ID: <slrnb8a9vt.qg.dsf@gaia.roc2.gblx.net>  a In article <DOqcnQ3xPJwQvBijXTWcoQ@comcast.com>, routerwoman <routerwoman@routerwoman.com> wrote: N > Is it possible to set up a default gateway on a VMS system.  I have a systemN > that's running RIPv1 just to talk to the one and only router in the network.G > Seems like a static route to the Gateway would stop all that chatter. 	 > Thanks.   I Sure can. The exact procedure depends on what VMS version and what TCP/IP I stack you use (Multinet, UCX -- now known as TCP/IP Services for OpenVMS, K Wollongong, TCPware, etc), as well as its version. You'd have to supply all J three pieces of information if you need a walk-through. Otherwise, you canK probably download or browse the administrator guide from either the OpenVMS 1 site or the TCP/IP stack manufacturer's web site.   J But sure, not a problem to set a default route. I just did it two days agoH for TCP/IP Services (formerly known as UCX) and then again for Multinet.0 Works great; also is simple and straightforward.   -Dan   ------------------------------  % Date: Fri, 28 Mar 2003 21:50:58 -0000 * From: "John Travell" <john@travell.uk.net>$ Subject: Re: Getting an Alphastation5 Message-ID: <b62g48$136om$1@ID-120847.news.dfncis.de>   D "Bradford J. Hamilton" <brad@.homeportal.2wire.net> wrote in message) news:xG_ga.263561$S_4.169561@rwcrnsc53... ? > In article <28602a55.0303280806.2a10b6b3@posting.google.com>, - mark.round7@ntlworld.com (Mark Round) writes:  > <snip>F > >I have found one supplier I have had good Unix dealings with in the past -K > >13w3.com, and they seem to offer the sort of thing I'm after in the form  of9 > >the AlphaStation 255/233Mhz. Specs are on this page :-  > >  > >http://www.13w3.com/dec/  > > H > >I know next to nothing about Alpha gear, my experiences with "exotic"J > >hardware being limited to Sparc-based systems. I'd plan on fitting moreH > >memory (256Mb in total) - but could anyone recommend this machine, or adviseI > >me to stear well clear ? I note that OpenVMS is supported on it. Would  thisJ > >be an OK beginners machine, or should I save my money for a better spec > >system ?  > L > "It depends" on what you plan to do with this machine.  Any use up to Java and L > Web browsing with Mozilla would be OK with this machine; any "serious" use ofF > Java and Mozilla is more tolerable with 512Mb and above, on a faster machine. > @ > The PWS500 pictured on the same page may fit the bill for more "heavy-duty"K > work, assuming it has the "correct" SCSI adapter installed.  You may also  wantJ > to check out the offerings at www.hpaq.net (formerly known as islandco). Mr. K > Turner is a "square dealer"; I use a PWS433au purchased from his company,  and I  > am a satisfied customer. > 0 Careful, those PWS boxes are 500-a's, NOT -au's.L They are not supported under VMS and according to the FAQ it may not work on them.   
 From the FAQ: D   "The Personal Workstation "-a" series is not supported by OpenVMS,<   while the Personal Workstation "-au" series is supported." and F   "OpenVMS can boot directly from a supported IDE CD-ROM only on thoseF   DIGITAL Personal Workstation systems with the Cypress PCI PeripheralH   controller (Miata/GL), and cannot boot from systems with the Intel SIOG   (Miata/MX5).  (Use the console command SHOW CONFIG to determine which I   of these you have.)  On systems with the Intel SIO, a SCSI CD-ROM drive    is required."   K However, I ran VMS on a Miata/MX5 with a SCSI CD for years. Unfortunately I - had to give it back when HP gave me the push.   I I have looked at these boxes from 13w3 myself, and raised a question with = them about which PCI peripheral controller chip thay contain.      -- John Travell  VMS crashdump expertise for hire john@travell.uk.net  http://www.travell.uk.net/       --- & Outgoing mail is certified Virus Free.: Checked by AVG anti-virus system (http://www.grisoft.com).A Version: 6.0.443 / Virus Database: 248 - Release Date: 10/01/2003    ------------------------------  # Date: Fri, 28 Mar 2003 22:10:48 GMT 7 From: brad@.homeportal.2wire.net (Bradford J. Hamilton) $ Subject: Re: Getting an Alphastation/ Message-ID: <IL3ha.267744$S_4.170081@rwcrnsc53>   b In article <b62g48$136om$1@ID-120847.news.dfncis.de>, "John Travell" <john@travell.uk.net> writes:E >"Bradford J. Hamilton" <brad@.homeportal.2wire.net> wrote in message * >news:xG_ga.263561$S_4.169561@rwcrnsc53...@ >> In article <28602a55.0303280806.2a10b6b3@posting.google.com>,. >mark.round7@ntlworld.com (Mark Round) writes: <snip>1 >Careful, those PWS boxes are 500-a's, NOT -au's. M >They are not supported under VMS and according to the FAQ it may not work on  >them. >  >From the FAQ:E >  "The Personal Workstation "-a" series is not supported by OpenVMS, = >  while the Personal Workstation "-au" series is supported."  >andG >  "OpenVMS can boot directly from a supported IDE CD-ROM only on those G >  DIGITAL Personal Workstation systems with the Cypress PCI Peripheral I >  controller (Miata/GL), and cannot boot from systems with the Intel SIO H >  (Miata/MX5).  (Use the console command SHOW CONFIG to determine whichJ >  of these you have.)  On systems with the Intel SIO, a SCSI CD-ROM drive >  is required."  G John is correct, which is why I qualified my endorsement of the PWS500:   N >> The PWS500 pictured on the same page may fit the bill for more "heavy-duty"> >> work, assuming it has the "correct" SCSI adapter installed.  O Just to put a fine point on it, I think that Mark specified that this was to be I a "beginner's" system, and that he was using it under the auspices of the M hobbyist license; if that is indeed the case, then the question of whether or O not it is an hp-supported configuration is moot.  Indeed, as Philip pointed out N in a subsequent reply to mine, the "a" boxes can, under certain circumstances,J run VMS quite nicely.  Again, as John points out, the FAQ contains enough M detail for Mark to determine if a system is "properly" configured to run VMS.   H Dave Turner at www.hpaq.net goes a step further in helping the potentialB hobbyist, by configuring his PWS machines with the proper hardwareJ up front, before he sells them; that's one of the reasons that I feel thatM islandco/hpaq may offer better value to the VMS "beginner" than a site which  / is oriented towards the DUnix/Tru64 "customer".    > L >However, I ran VMS on a Miata/MX5 with a SCSI CD for years. Unfortunately I. >had to give it back when HP gave me the push. > J >I have looked at these boxes from 13w3 myself, and raised a question with> >them about which PCI peripheral controller chip thay contain. >  >  >-- 
 >John Travell ! >VMS crashdump expertise for hire  >john@travell.uk.net >http://www.travell.uk.net/  >  >  >  >---' >Outgoing mail is certified Virus Free. ; >Checked by AVG anti-virus system (http://www.grisoft.com). B >Version: 6.0.443 / Virus Database: 248 - Release Date: 10/01/2003 >  >   A _________________________________________________________________ 0 Bradford J. Hamilton			"All opinions are my own"/ bMradAhamiPltSon@atMtAbi.cPoSm		"Lose the MAPS"    ------------------------------   Date: 29 Mar 2003 00:14:25 GMT2 From: "Zane H. Healy" <healyzh@shell1.aracnet.com>$ Subject: Re: Getting an Alphastation, Message-ID: <b62oh102jfs@enews3.newsguy.com>  , Mark Round <mark.round7@ntlworld.com> wrote:M > me to stear well clear ? I note that OpenVMS is supported on it. Would this I > be an OK beginners machine, or should I save my money for a better spec 
 > system ?  L It really depends on what you want to use the system for.  If you want to beH able to surf and run Java, then you really need a faster system.  If you0 want to use it for other things, then it's fine.  7 > Thanks again for any advice you may be able to offer!   H My *BIGGEST* piece of advice is to determine what type of RAM the system@ uses, and to make sure it takes easy to obtain RAM.  For exampleI AlphaStation 200's take 72-pin Parity RAM, it's fairly easy to obtain and K cheap, the DEC PWS's take DIMM's that are fairly easy to obtain.  OTOH, the J AlphaStation 500's take two different types of RAM (depending on the speed= of the system) and both types are hard to find and expensive.   L Something else to consider is what type of disks the system uses, and if youK want graphics, what type of graphics card it has.  Less cards are supported I under VMS than are supported under Tru64, and 24-bit cards can be hard to I get.  I found it cheaper to buy X-Windows software for my Mac a few years ' ago, than to try and get a 24-bit card!    			Zane    ------------------------------   Date: 29 Mar 2003 00:34:34 GMT2 From: "Zane H. Healy" <healyzh@shell1.aracnet.com>$ Subject: Re: Getting an Alphastation, Message-ID: <b62pmq02o0v@enews3.newsguy.com>  : Phillip Helbig <HELBPHI@sysdev.deutsche-boerse.com> wrote:7 > For disks, the BA350, BA353 and BA356 boxes are nice.   K I've been running my main system with a BA350 for several years now, and at G one point had two attached (found I needed the space more than the disk L space).  A few weeks ago I finally made use of the BA353 I've got, but thinkJ the next time I power that system up, I'll switch it to a BA350 as I thinkH the disk wasn't getting cooled properly (IIRC, you shouldn't use 7200RPM disks in a BA353).  I I've been wanting to get a BA356 or two, but am now seriously considering I putting a Sun 711 disk box on my system instead.  While a Sun 711 is more H expensive than a BA356, being able to simply slid in a SCA SCSI disk vs.I dealing with SBB canisters looks good.  What type of disks are in the -VW L canisters (68-pin or SCA)?  I've never opened one up, despite having a bunch that I can't currently use.   K I have to agree with your comments about volume shadowing, I really need to I get that setup, and it's one of the reason's I'm threatening to hook up a K Sun 711 (I might leave the BA350 attached also).  Anyone ever try running a J PWS 433 with two Qlogic UW SCSI controllers and a NCR810 Narrow Controller :^)    			Zane    ------------------------------    Date: 28 Mar 2003 16:41:15 -0800' From: nrt@nxrt.cncdsl.com (Norm Taylor) ( Subject: Help converting Alpha NT to VMS= Message-ID: <97e2760a.0303281641.73f4ee20@posting.google.com>   E I'm converting a PWS500au from it's original NT configuration to VMS. F I've make quite a bit of progress, including switching to the SRM bootC prompt, replacing the SCSI adaptor with one that is VMS compatible, B replacing the IDE CD drive with a SCSI CD, and replacing the video% card with one that is VMS compatible.   C My problem now is that at the boot prompt, when I do "show config", E the SCSI hard disk drives do not show up. The new SCSI CD show up, so D I know the adaptor is alive and generally working. The SCSI cable is6 terminated, and I've checked the SCSI address jumpers.   Help!    thanks.    ------------------------------  % Date: Fri, 28 Mar 2003 20:28:46 -0500 $ From: Hein <hein_cov@eps.zk.dec.com># Subject: Re: High RMS file activity . Message-ID: <3E84F6CE.6FEF25C9@eps.zk.dec.com>   Lee Y T Mah wrote:  L > Thanks for all the replies.  We do run the freebee daily.  I'll check into > it.    Good. J Now back to your subject. You mention RMS in the subject but nowhere else.K Why? Do you think you have an RMS performance problem? Any details on that?   ? Just today I submitted some stuff for the next VMS Freeware CD. N We also had it integrated (thanks Warren!) with the VMS V5.0 Freeware website:  ?     http://www.openvms.compaq.com/freeware/freeware50/rms_tools   J Perhaps you could/should check out the new RMS_TUNE_CHECK (.EXE and .TXT).   hth,
      Hein.   ------------------------------  % Date: Fri, 28 Mar 2003 21:25:17 -0700 $ From: Lee Y T Mah <lytmah@cha.ab.ca># Subject: Re: High RMS file activity ) Message-ID: <3E85202D.812ADCAB@cha.ab.ca>   K Our programs are in-house and our data is RMS indexed-sequentail.  We don't K suspect any RMS performance issues.  However, some of our shadow drives are R showing high i/o and long queue lengths.  We have to zero in on the volatile files) and move them over to less active drives.    Hein wrote:    > Lee Y T Mah wrote: > N > > Thanks for all the replies.  We do run the freebee daily.  I'll check into > > it.  >  > Good. L > Now back to your subject. You mention RMS in the subject but nowhere else.M > Why? Do you think you have an RMS performance problem? Any details on that?  > A > Just today I submitted some stuff for the next VMS Freeware CD. P > We also had it integrated (thanks Warren!) with the VMS V5.0 Freeware website: > A >     http://www.openvms.compaq.com/freeware/freeware50/rms_tools  > L > Perhaps you could/should check out the new RMS_TUNE_CHECK (.EXE and .TXT). >  > hth, >      Hein.   -- Lee   5 L Y T Mah                    Email:  lytmah@cha.ab.ca  Capital Health Authority Information Systems, RAH CSC Edmonton, Alberta, CANADA    ------------------------------  % Date: Fri, 28 Mar 2003 23:47:33 -0500 $ From: Hein <hein_cov@eps.zk.dec.com># Subject: Re: High RMS file activity . Message-ID: <3E852565.F8C558BD@eps.zk.dec.com>   Lee Y T Mah wrote:  D >> Our programs are in-house and our data is RMS indexed-sequentail./ >> We don't suspect any RMS performance issues.   	 Hi there!    More power to you!C But... I have recently run into a few customer situations where RMS > tuning gave a 10x improvement. Not 10%, no a magnitude faster!S Admittedly they had essentially neglected tuning, other then an occasional convert. - So they had silly, easily correctable errors.   T Still, for grins, run my rms_tune_check (script in the .txt file) on a critical disk or two and send me the output?R It has not been used that much and I am still collecting feedback as to how to set some knobs/rules.   Q > However, some of our shadow drives are showing high i/o and long queue lengths. T > We have to zero in on the volatile files and move them over to less active drives.  < Brute force the problem! SAME: Stripe And Mirror Everything.N Anything else is likely a waste of brainpower, and a recipe for future trouble  P Also... if you have high IO and long queue and your data is mostly stored in rms indexed N file, does that not really mean thatr you do appear to have an RMS performance problem?A Or can you justify every single IO and know it is not avoidable?! P Often that means you are near 100% write? : Writes = busines. Reads = time being wasted. P Too often I see RMS data entry applications running at 95% read, 5% write, Yuck.- I'd like to aim for read/write = 50/50 or so. @ As the writes generally can not be avoided, being at 50/50 meansC that you are at least half way to the absolute maximum performance. V Some truly large data files that just can not be cached will of course force a serious
 read rate.  V Finally, if you know RMS is tuned and you still have a hot disk or two, and there is a good reason why V you can not use SAME, then maybe you are in a  position to consider using VOLUME sets.  R They  allow you to spread a single file over volumes through CONVERT and FDL files! with AREAs and PLACEMENT control.    Regards, Hein.   M > Our programs are in-house and our data is RMS indexed-sequentail.  We don't M > suspect any RMS performance issues.  However, some of our shadow drives are T > showing high i/o and long queue lengths.  We have to zero in on the volatile files+ > and move them over to less active drives.  > 
 > Hein wrote:  >  > > Lee Y T Mah wrote: > > P > > > Thanks for all the replies.  We do run the freebee daily.  I'll check into	 > > > it.  > > 	 > > Good. N > > Now back to your subject. You mention RMS in the subject but nowhere else.O > > Why? Do you think you have an RMS performance problem? Any details on that?  > > C > > Just today I submitted some stuff for the next VMS Freeware CD. R > > We also had it integrated (thanks Warren!) with the VMS V5.0 Freeware website: > > C > >     http://www.openvms.compaq.com/freeware/freeware50/rms_tools  > > N > > Perhaps you could/should check out the new RMS_TUNE_CHECK (.EXE and .TXT). > >  > > hth, > >      Hein. >  > -- > Lee  > 7 > L Y T Mah                    Email:  lytmah@cha.ab.ca  > Capital Health Authority > Information Systems, RAH CSC > Edmonton, Alberta, CANADA    ------------------------------    Date: 28 Mar 2003 15:37:23 -08001 From: keithparris_NOSPAM@yahoo.com (Keith Parris) U Subject: HP User Advocacy survey: Business Continuity and Disaster Recovery/Avoidance = Message-ID: <cf15391e.0303281537.42daf2ca@posting.google.com>   F There's a new survey at the HP User Advocacy website, and better issueD submission/voting capabilities, according to this message I received today: --- & Date: Fri, 28 Mar 2003 10:27:11 -0600  From: admin@hpuseradvocacy.org  5 Subject: HP User Group Online Advocacy - Web Changes    3 We've made some changes at www.hpuseradvocacy.org!    E First, we've posted two new surveys regarding business continuity and F disaster recovery/avoidance to the HP User Group Online Advocacy's webA site. One survey was created exclusively for NonStop Server users D (http://www.hpuseradvocacy.org/surveys/2003MarchA/) and the other is) more general for all other platform users 4 (http://www.hpuseradvocacy.org/surveys/2003MarchB/).  F The surveys were created in order to collect data and develop a reportF that systems administrators can use to compare their state of disasterF recovery readiness. In addition, this report may also help to convinceD upper management in your company to increase its support of Business? Continuity(BC) and Disaster Recovery/Avoidance (DR) initiatives B because it is well known that the success of any Disaster RecoveryE Plan relies heavily on the extent of upper management's support of BC  and DR initiatives.   ; And while you are at the HP User Group Online Advocacy site ? (www.hpuseradvocacy.org), take a moment to look at the recently C improved issue submission and Me-Too voting features! We've made it D even easier for users to submit issues that are important to them asC well as to vote on issues submitted by their peers. And remember, 5 ? Me-Too votes on any issue guarantees that it will be sent to HP  management for review and 
 response.   D So take a minute and check out these exciting new features on the HP? User Group Online Advocacy's web site (www.hpuseradvocacy.org)!    ------------------------------    Date: 28 Mar 2003 17:42:42 -0800, From: JohnEllicottington@lycos.co.uk (Johno)5 Subject: Include an automated date within a printout. = Message-ID: <10822590.0303281742.3932d56c@posting.google.com>    Hi  E Does anyone know a way I can get a standard ASCII text file to output E to the printer on a VAX VMS/DCL system and contained with the output, 8 regardless when run, is the following Fridays date, e.g:   Day:	Fridays only  Date: 11th April 2003  xxxxxx xxxxxx  C The above date command being calculated automatically each time the  print is produced.   Many thanks    Johno    ------------------------------  % Date: Sat, 29 Mar 2003 01:19:13 -0400 0 From: JF Mezei <jfmezei.spamnot@vl.videotron.ca>9 Subject: Re: Include an automated date within a printout. / Message-ID: <3E852CD0.7CCB906D@vl.videotron.ca>    Johno wrote:G > Does anyone know a way I can get a standard ASCII text file to output G > to the printer on a VAX VMS/DCL system and contained with the output, : > regardless when run, is the following Fridays date, e.g:    M Use GOOGLE to find the recent threads on how to use DCL to generate a date on   the first monday of a month etc.   What you can do: $mynewdate = <whatever code> $OPEN/WRITE TEMP head.txt ' $write temp "Next Friday: ''mynewdate'"  $close temp    $PRINT HEAD.TXT,DATA.TXT  I This will print as a single job the header file you just created with the , existing DATA.TXT as if both files were one.   ------------------------------  % Date: Fri, 28 Mar 2003 19:49:47 -0700 " From: John Nebel <nebel@csdco.com>) Subject: Re: Macs seeing Pathworks shares ( Message-ID: <3E8509CB.5030403@csdco.com>   Bob Koehler wrote:H >    I know my Mac's can see MS servers at work.  If I set up the latestK >    Pathworks or Advanced Server on my home cluster, will my Macs be able  I >    to see its services?  (I don't want to try to track down and use the < >    old version of Pathworks that supported Mac protocols). > + >    (OBTW the Macs are running OS 8 or X).    Bob,  I OS X will see files on Pathworks as it has MS file sharing built in.  It   will also work with VMS NFS.  F OS 8 and 9 will work not in this manner.  Your MS servers at work are < probably serving via apple file sharing as well as natively.  B OS X is a pretty nice system as long as your Mac applications are  available under it.   A I've had the amusing experience of tracking down hacked PCs on a  G Microsoft network using a Mac laptop running nmap while the fellow who   set it up looked on.  
 John Nebel   ------------------------------    Date: 28 Mar 2003 13:10:15 -0600; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) # Subject: Macs vs. Pathworks et. al. 3 Message-ID: <uU25qbPaLHNs@eisner.encompasserve.org>   F    I know my Mac's can see MS servers at work.  If I set up the latestI    Pathworks or Advanced Server on my home cluster, will my Macs be able  G    to see its services?  (I don't want to try to track down and use the :    old version of Pathworks that supported Mac protocols).  )    (OBTW the Macs are running OS 8 or X).    ------------------------------   Date: 29 Mar 2003 00:45:02 GMT2 From: "Zane H. Healy" <healyzh@shell1.aracnet.com>' Subject: Re: Macs vs. Pathworks et. al. , Message-ID: <b62qae12o0v@enews3.newsguy.com>  < Bob Koehler <koehler@eisner.nospam.encompasserve.org> wrote:H >    I know my Mac's can see MS servers at work.  If I set up the latestK >    Pathworks or Advanced Server on my home cluster, will my Macs be able  I >    to see its services?  (I don't want to try to track down and use the < >    old version of Pathworks that supported Mac protocols).  + >    (OBTW the Macs are running OS 8 or X).   L I *think* the Mac's running OS X should be able to via Samba.  I think SambaH now comes as part of Mac OS X (I've yet to buy 10.2 as I'm still feeling. really negative about where the Mac is going).  E The Mac's running Mac OS 8 would need your home cluster to be running F Pathworks/Mac or whatever it's called.  I don't know about running theK latest Pathworks or Advanced Server, but I do know that when I tried using  H Mac OS 8.x and DAVE (software that lets the Mac talk to Windows systems)J that I was unable to access the Samba shares on my cluster.  Shortly afterG that attempt I loaded Pathworks/Mac on my Main VMS server and have been I using that without any problem.  I also use it so that the VMS server can E print to my HP5MP LaserJet, and I've even got Samba setup to act as a " printserver printing to the HP5MP.  K For Mac OS X you don't really need the added overhead of Pathworks/Advanced L Server, as you should simply be able to turn on NFS and mount the disks that way.  I BTW, the MS servers at work that you mention might very well be exportingoI the data via Appletalk, unless you're running Mac OS X and accessing theme
 via Samba.   			Zaneo   ------------------------------  # Date: Sat, 29 Mar 2003 04:03:08 GMTe% From: Bob Harris <harris@zk3.dec.com>l' Subject: Re: Macs vs. Pathworks et. al.e= Message-ID: <harris-B89B33.23032928032003@juggl7.zk3.dec.com>t  3 In article <uU25qbPaLHNs@eisner.encompasserve.org>,m=  koehler@eisner.nospam.encompasserve.org (Bob Koehler) wrote:h  H X    I know my Mac's can see MS servers at work.  If I set up the latestK X    Pathworks or Advanced Server on my home cluster, will my Macs be able eI X    to see its services?  (I don't want to try to track down and use thet< X    old version of Pathworks that supported Mac protocols). Xi+ X    (OBTW the Macs are running OS 8 or X).o X   I Disclaimer:  I've never tried connecting to the PC PATHWORKS or Advanced   Server on OpenVMS.  H Mac OS X can connect to SMB shares out of the box.  This is provided as $ part of Mac OS X.  No muss, no fuss.  @ For Mac OS 8.6 thu 9.2 you can get "Dave" from Thursby Software H <thursby.com>.  Actually you can get "Dave" for Mac OS X as well if you  prefer "Dave".  2                                         Bob Harris   ------------------------------  # Date: Fri, 28 Mar 2003 20:47:38 GMTu( From: Alder <PGDEHMKOKIMD@spammotel.com>- Subject: Re: Netscape gags on VMS FTP Servers ( Message-ID: <3E84B4E7.106@spammotel.com>   Joseph Huber wrote:cY > In article <3E7FDCA0.6020801@spammotel.com>, Alder <PGDEHMKOKIMD@spammotel.com> writes:- > S > I have given up to use the VMS TCPIP server because of the different assumptions m: > different clients have about (mis-)handling VMS servers.P > I have excellent experience with Hunter Goatleys HGFTP server: it can be setupO > to completely appear as a Unix server, and then works even with most Windows V) > and macintosh clients, Mozilla as well.  >  3  S I've now switched over to Hunter's HGFTP and configured it for full UNIX mode with:n 	HG_FTP_DO_UNIX_LS ALWAYSc 	HG_FTP_DO_UNIX_LS_INIT TRUE 	HG_FTP_HIDE_VMS_SYST TRUE  Q Netscape 7.01 for Windows can now navigate (no pun intended) the sub-directories eQ correctly, but will display a "0" where the name of the current directory should a8 be.  Is this "normal" behaviour under the circumstances?  N MSIE (version 6 anyway), is now the one which gags and refuses to display the K anonymous root directory at all.  Pathetic.  Even the MS-DOS FTP client in r Windows ME will work!t   Alderd   ------------------------------  # Date: Fri, 28 Mar 2003 23:09:56 GMT 6 From: peter@langstoeger.at (Peter 'EPLAN' LANGSTOEGER)P Subject: Re: OpenVMS v7.3-1 and Advanced Server v7.3A - Pathworks server crashes3 Message-ID: <8D4ha.15667$Xa3.185848@news.chello.at>   v In article <58ba0101.0303280653.6763e29e@posting.google.com>, andrew.rycroft@intrinsitech.com (Andrew Rycroft) writes:O >We have just upgraded to openVMS v7.3-1, and Pathworks Advnaced Server v7.3A. a  ! You know of the ECO 1 for V7.3A ?A  I >The Pathworks Server crashes without any apparent reason every two days.n: >No entry in the OPA0 nor in any Pathworks event log file.E >There are two dumpfiles in PWRK$LMROOT:[LANMAN.LOGS]PWRK$LMMCP.DMP;2    And what does they contain ?  ( >Has anybody had any similar behaviour ?  E No. Only hundreds of VMS crashes during pathworks startup caused by atG (maybe still existant) pool corruption bug in KNB (NetBIOS over TCP/IP)iM for over three years (and many pathworks versions/ECOs starting with V6.0)...h   -- l Peter "EPLAN" LANGSTOEGERa% Network and OpenVMS system specialist  E-mail  peter@langstoeger.atF A-1030 VIENNA  AUSTRIA              I'm not a pessimist, I'm a realist   ------------------------------    Date: 28 Mar 2003 14:34:08 -0800& From: jordan@ccs4vms.com (Rich Jordan)C Subject: Re: OpenVMS.org Survey:  Who's Been Knocking At Your Door?a= Message-ID: <cc5619f2.0303281434.1860a09b@posting.google.com>t  B Re: Question 3.  My mom always said "System Administrators are forC unix and pc and other such lesser systems... be a System Manager if.D you want to work with the best."  I agree, but sysadmin will have to do on the survey :(o   Rich   ------------------------------  # Date: Fri, 28 Mar 2003 22:40:50 GMTS7 From: brad@.homeportal.2wire.net (Bradford J. Hamilton)9C Subject: Re: OpenVMS.org Survey:  Who's Been Knocking At Your Door?0- Message-ID: <Sb4ha.243268$F1.45318@sccrnsc04>   f In article <cc5619f2.0303281434.1860a09b@posting.google.com>, jordan@ccs4vms.com (Rich Jordan) writes:C >Re: Question 3.  My mom always said "System Administrators are foroD >unix and pc and other such lesser systems... be a System Manager ifE >you want to work with the best."  I agree, but sysadmin will have tod >do on the survey :(  L Unfortunate, but true - if my job search were limited to "System Manager", IK would be assured of *never* finding a job.  On occasion, VMS System ManagermK positions will appear under the guise of "System Administrator", or "System0N Programmer".  This has both good and bad aspects - you could be mistaken for aF unix or mainframe support person, but sometimes the pay is better.	:-)    >  >Richv  A _________________________________________________________________r0 Bradford J. Hamilton			"All opinions are my own"/ bMradAhamiPltSon@atMtAbi.cPoSm		"Lose the MAPS"3   ------------------------------    Date: 28 Mar 2003 16:05:07 -08002 From: texascheryl@hotmail.com (Cheryl Hoefelmeyer) Subject: Overuse of LIB$SPAWN?= Message-ID: <7112dd15.0303281605.59583e8b@posting.google.com>3   Hi, all,D My colleagues at my present place of work are almost all new to VMS,D and I am trying to teach them various useful things that I know. OneD thing they always seem to want to do rather than learn to use systemF calls within programs is to spawn a subprocess and issue DCL commands.D For example, they would rather spawn a subprocess to call SORT/MERGEB than use the utility inside a program. I have never seen this doneE anywhere else, so I suspect it's not a good idea to spawn unnecessary D subprocesses right and left, but I don't know if this is so, or why.C So, does it matter whether a system routine is called from within am& program or if a subprocess is spawned?   Thanks,t Cheryl   ------------------------------  # Date: Sat, 29 Mar 2003 00:16:59 GMTlL From: winston@SSRL.SLAC.STANFORD.EDU ("Alan Winston - SSRL Admin Cmptg Mgr")" Subject: Re: Overuse of LIB$SPAWN?6 Message-ID: <00A1D8CD.EC90302B@SSRL.SLAC.STANFORD.EDU>  r In article <7112dd15.0303281605.59583e8b@posting.google.com>, texascheryl@hotmail.com (Cheryl Hoefelmeyer) writes:	 >Hi, all,   E >My colleagues at my present place of work are almost all new to VMS,sE >and I am trying to teach them various useful things that I know. OnerE >thing they always seem to want to do rather than learn to use systemiG >calls within programs is to spawn a subprocess and issue DCL commands.aE >For example, they would rather spawn a subprocess to call SORT/MERGEfC >than use the utility inside a program. I have never seen this done F >anywhere else, so I suspect it's not a good idea to spawn unnecessaryE >subprocesses right and left, but I don't know if this is so, or why. D >So, does it matter whether a system routine is called from within a' >program or if a subprocess is spawned?   F It mostly depends how often it happens and how many resources you haveL available.  It costs more to spawn a default subprocess (which inherits yourM symbols and logicals by getting told about them, slowly, via a mailbox).  ButgJ there's a lot of applications where that doesn't matter a lot.  If you're K sorting or merging large files, the sort/merge run time probably swamps the-H difference the between calling callable sort directly and spawning a DCLJ procedure to do that.  It's certainly cleaner and nicer to use the libraryL calls -- the pure in heart will always do it -- but for many applications itO doesn't make any noticeable difference.  _However_, if you want to do something H that's not necessarily supported by the CLI provided -- you see a lot ofN mail processing tools that want to use the  callable mail libraries to provideL functionality that VMS mail doesn't -- then you really have no choice but to  call the MAIL routines yourself.  O That said, it's possible to imagine cases where it makes more sense to  spawn a0M subprocess than to do some work in your own main line (although most of thosesO cases are more properly handled by kernel threads).  If you're writing a servereH process that can do, potentially, a lot of work for a bunch of differentL clients, you don't want your main line bogged down doing work in synchronousM system calls.  You could spawn a subprocess with some and set up some kind ofoN completion AST arrangement, or -- if you were using threading -- start anotherD thread and have your dispatcher thread go back and listen for calls.   -- Alan0  O ===============================================================================e0  Alan Winston --- WINSTON@SSRL.SLAC.STANFORD.EDUM  Disclaimer: I speak only for myself, not SLAC or SSRL   Phone:  650/926-3056hM  Paper mail to: SSRL -- SLAC BIN 99, 2575 Sand Hill Rd, Menlo Park CA   94025mO ===============================================================================-   ------------------------------  % Date: Fri, 28 Mar 2003 21:45:46 -0400 0 From: JF Mezei <jfmezei.spamnot@vl.videotron.ca>" Subject: Re: Overuse of LIB$SPAWN?/ Message-ID: <3E84FAB2.C4310B11@vl.videotron.ca>t   Cheryl Hoefelmeyer wrote:cF > subprocesses right and left, but I don't know if this is so, or why.E > So, does it matter whether a system routine is called from within a ( > program or if a subprocess is spawned?  K Sometime, it is easier to document a program with a single character bufferyF that has a DCL command in it used for a spawn, as opposed to using the, callable equivalent with complex item lists.  H But from an efficiency point of view, the callable methods are fare moreN efficient and sometimes give you greater facility to interface to your program and handle/detect errors.o  F However, if you do use spawned subprocesses, one way to make this moreK efficient is to create 2 mailboxes, the spawn a single subprocess (with thekM equivalent of /NOWAIT) whose input is the first mailbox and the output is the M second mailbox. This way, you can issue a command by doing a $QIO (write) theaM the first mailbox and read the output from the other mailbox. This allows you H toisseu many commands within the context of the same process and save on process creation overhead.   ------------------------------  % Date: Fri, 28 Mar 2003 17:46:36 -04003' From: "Jorge V.S." <jvalen@alefdata.cl>f( Subject: problems with accountn.dat file, Message-ID: <b62fqi$lr8$1@news.chilesat.net>   Hello:  C 	Last month we noticed that our accounting file doesn't recordered > anything  for tow days, why ?a  F 	Where to look for errors, corruption, audit information to ckech the  problem and solve it.t   				thanks in advancea   					Jorge V.S.'030328   ------------------------------  % Date: Fri, 28 Mar 2003 17:13:14 -0400p0 From: JF Mezei <jfmezei.spamnot@vl.videotron.ca>, Subject: Re: problems with accountn.dat file/ Message-ID: <3E84BAE2.A0D8893E@vl.videotron.ca>h   "Jorge V.S." wrote:o >  > Hello: > K >         Last month we noticed that our accounting file doesn't recorderede > anything  for tow days, why ?    SHOW ACCOUNTING   I then, dir/full sys$manager:accountng.dat to see just how big the file is.i  N You can rename the file, and see if a new one is created when the next process is created/terminated. a  ; HELP SET ACCOUNTING will give you more information as well.w   ------------------------------    Date: 28 Mar 2003 13:57:51 -0600+ From: young_r@encompasserve.org (Rob Young)1 Subject: Re: RamDisk futures3 Message-ID: <fGeaG2idoST4@eisner.encompasserve.org>y  a In article <QvzyfbVt+EEe@eisner.encompasserve.org>, young_r@encompasserve.org (Rob Young) writes:Ys > In article <cf15391e.0303261443.324c1390@posting.google.com>, keithparris_NOSPAM@yahoo.com (Keith Parris) writes:si >> young_r@encompasserve.org (Rob Young) wrote in message news:<TqkdHZQyBq8o@eisner.encompasserve.org>...u= >>> RamDisk has come a long way.  Back in the bad old days itt= >>> 	was limited to 512 MBytes.  With recent versions you can ? >>> 	do 2 Gbytes.  That said, in order to do large RamDisks youtB >>> 	have to resort to BINDing RamDisks together.  I hate BINDing. >> sA >> You might consider using host-based RAID software (Compaq RAID G >> Software for OpenVMS) as a way to combine DECram disks into a RAID 0r. >> array, instead of using a bound volume set. >    [snip]  ? > 	Having a max size of 2 Gig for a RamDisk in 2003 is crampin'm" > 	my style and is sooooooo 1990s.   	And so wrong.  E 	I think I am having an all time worst week for accuracy.  It doesn't.@ 	help that a co-worker confused me and "pointed" out we won't beC 	able to create a RamDisk very large (and then me stumbling through F 	documentation for a day as I "remembered" that also).  It also didn'tB 	help *me* (others of course probably didn't make the same readingG 	mistake) that the version 3.0 manual talks about using BIND to create   	larger RamDisks.  u  > 	Here is what I mean from the DECRAM user guide.  This section. 	is comparing and contrasting DECRAM with XFC:  7                                 Use as a Storage Device:  I               DECram_Features____________Disk_Cache_Features_____________0    B               Can hold as much as        Capable of holding entire3               4,294,967,296 blocks       databases.2#               of information on onet#               disk with Version 3.0f$               or higher. The maximum#               amount of informationt&               on one disk that Version%               2.3 can hold is 524,2806%               blocks for a VAX system.#               and 67,108,864 blocksm$               on an Alpha system. It#               is possible to create "               logically contiguous"               devices greater than'               either of these limits by &               creating multiple DECram&               devices and binding them               together.w    G 	[version 3.0 and higher is Alpha only and supports much larger sizes].   = 	A quick read of that highlights bind.  You most likely won'tD: 	need to bind anything on an Alpha as 4.294 billion blocksH 	is over 209 GBytes (not 2 GBytes as a co-worker pointed out).  Perhaps E 	a re-word for poorer readers (me) and/or those that get confused by t	 	numbers:h  G 	"You may find it necessary to bind together RAMDisks on VAX to create e> 	larger logically contiguous devices.  It will most likely be E 	unnecessary to do that on Alpha as you are able to create a RamDisk g5 	larger than 209 GBytes with version 3.0 or higher." m  @ 	Sheesh.  Sorry for confusing anyone.  Sorry for being confused!   				Rob    ------------------------------  + Date: Fri, 28 Mar 2003 21:24:54 +0100 (MET)t9 From: Phillip Helbig <HELBPHI@sysdev.deutsche-boerse.com>c Subject: Re: RamDisk futures; Message-ID: <01KU2M4IMDUU9H3B5Q@sysdev.deutsche-boerse.com>e  > > A quick read of that highlights bind.  You most likely won't; > need to bind anything on an Alpha as 4.294 billion blocksmI > is over 209 GBytes (not 2 GBytes as a co-worker pointed out).  Perhaps sF > a re-word for poorer readers (me) and/or those that get confused by 
 > numbers: > H > "You may find it necessary to bind together RAMDisks on VAX to create ? > larger logically contiguous devices.  It will most likely be dF > unnecessary to do that on Alpha as you are able to create a RamDisk 6 > larger than 209 GBytes with version 3.0 or higher."   G RAMDisk is the "solid-state disk in RAM" as opposed to the solid-state oI disk in the SBB, right?  If I can have a 209-GB RAMDisk, then presumably U$ that's only part of my total memory.  ? What's the current maximum memory?  What is the largest memory w, configuration which has actually been built?  G I was thinking 8 GB of RAM was a lot, but now I am feeling humble.  :-)    ------------------------------    Date: 28 Mar 2003 14:56:07 -0600+ From: young_r@encompasserve.org (Rob Young)  Subject: Re: RamDisk futures3 Message-ID: <rl7DhbavElOe@eisner.encompasserve.org>/  w In article <01KU2M4IMDUU9H3B5Q@sysdev.deutsche-boerse.com>, Phillip Helbig <HELBPHI@sysdev.deutsche-boerse.com> writes:_? >> A quick read of that highlights bind.  You most likely won't"< >> need to bind anything on an Alpha as 4.294 billion blocksJ >> is over 209 GBytes (not 2 GBytes as a co-worker pointed out).  Perhaps G >> a re-word for poorer readers (me) and/or those that get confused by t >> numbers:s >> nI >> "You may find it necessary to bind together RAMDisks on VAX to create t@ >> larger logically contiguous devices.  It will most likely be G >> unnecessary to do that on Alpha as you are able to create a RamDisk  7 >> larger than 209 GBytes with version 3.0 or higher." l > I > RAMDisk is the "solid-state disk in RAM" as opposed to the solid-state uK > disk in the SBB, right?  If I can have a 209-GB RAMDisk, then presumably t& > that's only part of my total memory. > & > What's the current maximum memory?      @ 	Without digging, I recall 256 GBytes with more density bringingA 	a 64 CPU EV7 to 1 TByte in the future (a year or so out , IIRC).M  I > I was thinking 8 GB of RAM was a lot, but now I am feeling humble.  :-)o    	8 GByte is now very affordable.   				Robt   ------------------------------  + Date: Fri, 28 Mar 2003 22:16:41 +0100 (MET) 9 From: Phillip Helbig <HELBPHI@sysdev.deutsche-boerse.com>t Subject: Re: RamDisk futures; Message-ID: <01KU2NVMGR369H3B5Q@sysdev.deutsche-boerse.com>s  A > Without digging, I recall 256 GBytes with more density bringingrB > a 64 CPU EV7 to 1 TByte in the future (a year or so out , IIRC).   Wow!  K > > I was thinking 8 GB of RAM was a lot, but now I am feeling humble.  :-)  > " > 	8 GByte is now very affordable.  D All things are relative.  Still, time to start saving to upgrade my  hobbyist system!  H As Ted Turner said when he donated $ 1 billion to the U.N.: It was only 3 a month's earnings, so I figured I could afford it!b   ------------------------------  % Date: Fri, 28 Mar 2003 20:18:34 -0500o$ From: Hein <hein_cov@eps.zk.dec.com> Subject: Re: RamDisk futures. Message-ID: <3E84F46A.523EF834@eps.zk.dec.com>   Phillip Helbig wrote:   @ > What's the current maximum memory?  What is the largest memory. > configuration which has actually been built? >lI > I was thinking 8 GB of RAM was a lot, but now I am feeling humble.  :-)s  0 8GB ... PER CPU is still a nice number today :-)  L We run the larger Unix TPC-C configs (aiming well over 300,000 tpm) on boxes with 256 GB, 32P. L Then we give all that memory away to  Oracle for an SGA of say 220 GB. Scary stuff.  I Try to checkpoint 100+GB of dirty buffers. (It takes a minute or more two6 :-)nL Now try to have that not influence your ongoing  production too much. (it'll take 10 minutes or so).sL (We now have configs clocked at 1 million 2KB sized IO/sec = 2GB/sec... just doing IO nothing else).c  D Real, production, systems are now frequently configured with 2GB/CPUL Simply because they can often use that memory, and furthermore that quantity  K often comes with the minimum interleaving to get the throughput and latencyn desired.  @ Terabyte sized main memory systems are just around the corner...     Cheers,o Hein.r   ------------------------------    Date: 28 Mar 2003 15:05:28 -0600+ From: young_r@encompasserve.org (Rob Young)  Subject: RamDisk questionl3 Message-ID: <lNxE0Znz2r3j@eisner.encompasserve.org>3  ( 	Beating the RamDisk horse a bit more...  ( 	Why am I seeing this in the User guide:  >               Enter the following command from the DCL prompt:  J               DECRAM CREATE DISK device-name /CAPACITY=blocks [/qualifier]                 where:   		[snip]                   /CAPACITY=blocksB               This required qualifier specifies the size in blocksF               (512 bytes=1 block) of the virtual disk to be created orF               modified. The capacity block limit in DECram Version 3.01               and higher is now 8,388,608 blocks.   D         I'm interpreting this to mean, the /CAPACITY switch is stuckC         in the past.  After all, we can create RamDisks very large:                    /SIZE=n G               The amount of memory (in blocks or 512 byte chunks) to bes'               allocated to this device.>  G                 _____________________ Restriction _____________________b  B                 With DECram for OpenVMS Version 3.0 or higher, theB                 limit on the DECram disk size has been extended toE                 4,294,967,296 blocks on OpenVMS Alpha systems runningd*                 Version 7.2-1H1 or higher.    3 	So why is there a limit of 8 million on /CAPACITY?w   				Rob    ------------------------------  % Date: Fri, 28 Mar 2003 22:30:04 -0500r& From: "Bran Furman" <bsfurman@att.net>8 Subject: SIMH Emulation of VAX KA650-A, KA650-B, KA655-A2 Message-ID: <4CGdnb7b6qkpjhijXTWc3w@speakeasy.net>  F In case anyone is interested, the VAX 3400, 3600 and 3900 use the sameG firmware except for identification codes in the boot ROM.  The ROM basecK address is 20040000.  The model is determined by the 6th byte (offset of 5)lF in the ROM (ka655.bin).  This is one field in a location that has been@ referred to as SIDEX.  You can examine it from the console with:    >>> EXAM/P/B 20040005  # It determines the model as follows:a       Code    Model   ID     ---------------------:     2       3400    KA640e     1       3600    KA650e     3       3900    KA655o  B However, you cannot run a KA640 under SIMH.  There is apparently aG difference in the hardware that is not emulated for the KA640.  Perhapsf DSSI?   L The 5th byte (offset of 4, address of 20040004) designates the model as A or0 B (server) and affects VMS licensing as follows:  7     Code    ID    Name        License Availability Code 7     ---------------------------------------------------m-     1       A     MicroVAX    A or E requiredo(     2       B     VAXserver   B required   End.   ------------------------------  % Date: Fri, 28 Mar 2003 22:45:47 -0500a& From: "Bran Furman" <bsfurman@att.net>, Subject: SIMH VAX build with MS Visual C++ 62 Message-ID: <vyKdnUP5PJ7-ihijXTWcoQ@speakeasy.net>  J Does anyone know the correct project settings to build SIMH with MS Visual C++ 6?  J For "Preprocessor definitions" I used the default WIN32,_CONSOLE,_MBCS and3 added USE_INT64 resulting in the following options:o  D /nologo /MLd /W3 /Gm /GX /ZI /Od /I ".\\" /I "..\\" /I "..\pdp11" /IG "..\net" /D "_DEBUG" /D "WIN32" /D "_CONSOLE" /D "_MBCS" /D "USE_INT64"r9 /Fp"Debug/VAX.pch" /YX /Fo"Debug/" /Fd"Debug/" /FD /GZ /c    This fails as follows:   KA655-B V5.3, VMB 2.7d Performing normal system tests.l( 40..39..38..37..36..35..34..33..32..31..   ?53 2 0A FF 00 0000l  ? P1=00000002  P2=00000028  P3=00000FF7  P4=00D40077  P5=00000001_? P6=FFFFFFFF  P7=00000000  P8=00000000  P9=00000000 P10=20051CE0n? r0=00000002  r1=20140110  r2=3CC66E04  r3=000301B1  r4=3CC66E0Eo? r5=2004E8F9  r6=00018678  r7=000186C8  r8=00019697 ERF=80000000e 30..29..28..27..26..25..@ 24..23..22..21..20..19..18..17..16..15..14..13..12..11..10..09.. 08..07..06..05..04..03.. Normal operation not possible.   >>>e  = Where does one find documentation on diagnostic codes anyway?.   Thanks.e   ------------------------------  % Date: Fri, 28 Mar 2003 16:15:14 -0400o0 From: JF Mezei <jfmezei.spamnot@vl.videotron.ca> Subject: Re: Suggestion to Sue/ Message-ID: <3E84AD4E.24F624F7@vl.videotron.ca>o   Fred Kleinsorge wrote:J > I'd rather Andrew just go away, or be replaced with someone who has more> > technical depth, in his thinly disguised Sun sales function.  N Yes, but in anty conference, don't you have to have at least one function that" is structured as "entertainment" ?  I Actually, havuing a Andrew vs Fred presentation might be very good in theaH context of such a conterence since it would provide pro-VMS folks with aI perspective on what Sun does  and how one can respond. In essence, such aa2 debate would teach us to do do battle against Sun.  H So, either the Fred-Andrew could be an entertaining catfight, or if bothM promise to be civil, it could be a most interesting factual debate that wouldeL be instructive. (But in that context, I am not sure Andrew wqould be willing to sacrifice himself).    N > another soapbox).  But aside from raised blood pressure on occasion, I don'tN > think I have a desire to use the persuader in the office next door (a nice 29 > inch diameter, rubber clad copper wire) on any of them.o  ' Is that persuader in Sue's office ? :-)o   ------------------------------  % Date: Sat, 29 Mar 2003 00:04:26 +0100o& From: Andreas Fassl <afassl@progis.de>/ Subject: Re: TCP/IP 5.1 and SMTP authentication / Message-ID: <b62kad$e46$03$1@news.t-online.com>e   Hi Mr. Ulmann,  $ how goes it? (Perfect germanism :-))G Just browsed through the 5.3 docs, couldn't find there a hint neither. w5 Have you got a fixed ip when connecting to your  ISP?r* Sounds like a question for the ... wizard.   Best regards  
 Andreas Fasslt; (I'm just setting up my mail system to use the new OpenVMS a$ TCPIP-Services 5.3 antispam options.     Bernd Ulmann wrote:    >Hi - F >I have the problem that my ISP is changing his SMTP server so that itE >will only accept mails sent with SMTP and authentication. They claim.G >this is because they run into quite severe spam problems and they want H >to be sure that the people using their smtp server are their customers.H > My problem is now, that I am doing everything at home on my (wonderfulG >:-) ) VAX-7820 running OpenVMS-VAX 7.2 and TCPIP 5.1. Up to now I sentPA >all my outgoing mails via smtp to the smtp server of my provideriB >(alternate gateway) with a simple domain substitution to have theI >"right" sender address appearing on my outgoing mails. Since my provider G >has changed his access rules, I am consequently not able to send mailssH >any longer, since I can not get TCPIP 5.1 into sending an smtp user and >passwort for authentication.@F > Is there any obvious solution for this which I totally missed? Or isI >there some clever trick, maybe getting the data out of the SMTP queue(s)eH >and sending it by a dedicated program (maybe in PERL?). To be honest, IF >am stuck at the moment, and somehow I have to get online (mail) againH >and I do not want to use another computer running what operating system >it might, just to send mails.2 > If someone has an idea, please drop me a note atF >ulmann(at)raven-infotech.com. Thank you very much for your help, best	 >regards,c >t >	Bernd Ulmann. :-)t >    >    ------------------------------    Date: 28 Mar 2003 13:51:33 -0800" From: horn@shsu.edu (James T Horn)8 Subject: Re: Three questions re: the VMS startup process= Message-ID: <843706dc.0303281351.4d29d18c@posting.google.com>e  C What I do to know what the systartup_vms.com file does, is have theeE output go to a log file, then I can look at it after the system is upr* and try to figure out what the problem is:  % At the begining of SYSTARTUP_VMS.COM:   ' $! Setup log file for SYSTARTUP_VMS.COMO8 $   If "''STOP_RECURSION'" .NES. "" THEN GOTO DO_STARTUP $   STOP_RECURSION == "YES"  $   SET VERIFYM $   @SYS$MANAGER:SYSTARTUP_VMS.COM/OUT=SYS$SPECIFIC:[SYSMGR]SYSTARTUP_VMS.LOGs -iG         "''P1'","''P2'","''P3'","''P4'","''P5'","''P6'","''P7'","''P8'"e= $   PURGE/KEEP=1/NOLOG SYS$SPECIFIC:[SYSMGR]SYSTARTUP_VMS.LOGu $   SET NOVERIFY $   EXIT $DO_STARTUP:
 $ SET NOON  R brandon@dalsemi.com wrote in message news:<03032808104499@dscis6-0.dalsemi.com>...N > > 1. How do I configure LAT to run at startup?  Right now, I have to executeQ > > SYS$STARTUP:LAT$STARTUP.COM manually after every reboot, and simply adding itt > > toK > > the end of SYS$SYSROOT:[MANAGER]SYSTARTUP_VMS.COM doesn't seem to work.e > Q > This should work, make sure that there SYSTARTUP_VMS is completing - i.e., thataM > there is not a syntax or logic error that forces SYSTARTUP_VMS to bomb.  ToMP > find out, define the console output at the begining of your SYSTARTUP_VMS to a% > log file for your viewing pleasure:l > ) > $ open /write console systartup_vms.logh >  > N > > 2. I have a similar question for configuring X terminal support.  Should IA > > just add "DEFINE/SYSTEM/EXEC DECW$INSTALL_XTERMINAL TRUE" and N > > "@SYS$STARTUP:DECW$STARTUP XTERMINAL" to SYSTARTUP_VMS.COM?  The VXT-2000+ > > installation manualaB > > (http://h71000.www7.hp.com/freeware/freeware40/vxt/post_docs/v! > > xt_sysman_install_openvms.ps) R > > implies that if I define the DECW$INSTALL_XTERMINAL logical, DECwindows shouldN > > automatically setup X terminal support, but if I just put the logical into< > > SYSTARTUP_VMS.COM, I still have to run DECW$STARTUP.COM. > ( > Place the logical in SYLOGICALS.COM;   > N > > 3. I'm trying to get a VXT-2000+ to MOP-boot off my Alpha.  Unfortunately, > > evenL > > though I'm doing "MCR NCP DEFINE NODE <nodename> ADDRESS <decnet-addr>  . > > HARDWARE ADDRESS <ethernet-addr> LOAD FILER > > SYS$SYSDEVICE:[VXT$LIBRARY]<vxt-image-filename>" and updating NCP's persistentR > > database, the changes are erased when I restart.  What am I doing wrong?  Do IM > > have to write out the configuration changes somehow after doing a "DEFINEc
 > > NODE"? > : > I use to do this, but that was years ago.  Someone else? >  >  > John Brandon > VMS Systems Administratorr > Dallas Semiconductor > john.brandon@dalsemi.com > 972.371.4172 wkn > 972.371.4003 fxi   ------------------------------  % Date: Fri, 28 Mar 2003 16:04:39 -0600a From: brandon@dalsemi.como8 Subject: Re: Three questions re: the VMS startup process1 Message-ID: <03032816043940@dscis6-0.dalsemi.com>o  ) > $! Setup log file for SYSTARTUP_VMS.COMh: > $   If "''STOP_RECURSION'" .NES. "" THEN GOTO DO_STARTUP > $   STOP_RECURSION == "YES"t > $   SET VERIFYO > $   @SYS$MANAGER:SYSTARTUP_VMS.COM/OUT=SYS$SPECIFIC:[SYSMGR]SYSTARTUP_VMS.LOGn > -lI >         "''P1'","''P2'","''P3'","''P4'","''P5'","''P6'","''P7'","''P8'"v? > $   PURGE/KEEP=1/NOLOG SYS$SPECIFIC:[SYSMGR]SYSTARTUP_VMS.LOG  > $   SET NOVERIFY
 > $   EXIT > $DO_STARTUP: > $ SET NOON  9 A little bit more complicated than it needs to be, but...o  G Do you have a readable output of the startup of LAT?  What does it say?62 Are you performing a LAT$STARTUP or LAT$SYSTARTUP?& Are you executing DECnet prior to LAT?         John Brandon VMS Systems Administratorm Dallas Semiconductor john.brandon@dalsemi.com 972.371.4172 wk  972.371.4003 fxa   ------------------------------  % Date: Fri, 28 Mar 2003 20:23:10 -0600b1 From: "David J. Dachtera" <djesys.nospam@fsi.net> 8 Subject: Re: Three questions re: the VMS startup process' Message-ID: <3E85038E.ECB2A997@fsi.net>t   James T Horn wrote:8 > E > What I do to know what the systartup_vms.com file does, is have theTG > output go to a log file, then I can look at it after the system is ups, > and try to figure out what the problem is: > ' > At the begining of SYSTARTUP_VMS.COM:e > ) > $! Setup log file for SYSTARTUP_VMS.COMk: > $   If "''STOP_RECURSION'" .NES. "" THEN GOTO DO_STARTUP > $   STOP_RECURSION == "YES"d > $   SET VERIFYO > $   @SYS$MANAGER:SYSTARTUP_VMS.COM/OUT=SYS$SPECIFIC:[SYSMGR]SYSTARTUP_VMS.LOG  > -aI >         "''P1'","''P2'","''P3'","''P4'","''P5'","''P6'","''P7'","''P8'"o? > $   PURGE/KEEP=1/NOLOG SYS$SPECIFIC:[SYSMGR]SYSTARTUP_VMS.LOGk > $   SET NOVERIFY
 > $   EXIT > $DO_STARTUP: > $ SET NOON  > There's an easier way to do that. Just add these at the top of SYSTARTUP_VMS:  
 $ SET NOON $ SET VERIFY1 $ DEFINE SYS$OUTPUT SYS$MANAGER:SYSTARTUP_VMS.LOGu  2 Everything after that will appear in the log file.   -- a David J. Dachtera  dba DJE Systemst http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/x   ------------------------------  % Date: Sat, 29 Mar 2003 00:06:21 -0500u  From: John Santos <JOHN@egh.com>8 Subject: Re: Three questions re: the VMS startup process4 Message-ID: <1030328235823.416E-100000@Ives.egh.com>  B It looks like no one has caught on yet to the fact that you (i.e.,@ the original poster, not James Horn) are probably changing files in the wrong place!i  . SYS$MANAGER: is *not* SYS$SYSROOT:[MANAGER]...   It is SYS$SYSROOT:[SYSMGR]...r  A This (unless it was merely a typo in your original post) explains # why your changes are being ignored.e  ? (DECNET config changes are a separate issue -- The commands you ; are using are for DECnet Phase IV.  If Phase 5, you need toc> do the "@net$configure" thing, or manually edit the appropriat .NCL files in sys$manager:.)  ? Lots of other good advice here about logging the startup, whereR to define logical names, etc.    Hope this helps!  # On 28 Mar 2003, James T Horn wrote:m  E > What I do to know what the systartup_vms.com file does, is have theeG > output go to a log file, then I can look at it after the system is upc, > and try to figure out what the problem is: > ' > At the begining of SYSTARTUP_VMS.COM:  > ) > $! Setup log file for SYSTARTUP_VMS.COMe: > $   If "''STOP_RECURSION'" .NES. "" THEN GOTO DO_STARTUP > $   STOP_RECURSION == "YES"o > $   SET VERIFYO > $   @SYS$MANAGER:SYSTARTUP_VMS.COM/OUT=SYS$SPECIFIC:[SYSMGR]SYSTARTUP_VMS.LOGm > -II >         "''P1'","''P2'","''P3'","''P4'","''P5'","''P6'","''P7'","''P8'"n? > $   PURGE/KEEP=1/NOLOG SYS$SPECIFIC:[SYSMGR]SYSTARTUP_VMS.LOGi > $   SET NOVERIFY
 > $   EXIT > $DO_STARTUP: > $ SET NOON > T > brandon@dalsemi.com wrote in message news:<03032808104499@dscis6-0.dalsemi.com>...P > > > 1. How do I configure LAT to run at startup?  Right now, I have to executeS > > > SYS$STARTUP:LAT$STARTUP.COM manually after every reboot, and simply adding it  > > > toM > > > the end of SYS$SYSROOT:[MANAGER]SYSTARTUP_VMS.COM doesn't seem to work.o  & Wrong directory!-------------^^^^^^^^^   > > S > > This should work, make sure that there SYSTARTUP_VMS is completing - i.e., thatdO > > there is not a syntax or logic error that forces SYSTARTUP_VMS to bomb.  ToPR > > find out, define the console output at the begining of your SYSTARTUP_VMS to a' > > log file for your viewing pleasure:e > > + > > $ open /write console systartup_vms.log  > >  > > P > > > 2. I have a similar question for configuring X terminal support.  Should IC > > > just add "DEFINE/SYSTEM/EXEC DECW$INSTALL_XTERMINAL TRUE" andeP > > > "@SYS$STARTUP:DECW$STARTUP XTERMINAL" to SYSTARTUP_VMS.COM?  The VXT-2000+ > > > installation manualUD > > > (http://h71000.www7.hp.com/freeware/freeware40/vxt/post_docs/v# > > > xt_sysman_install_openvms.ps))T > > > implies that if I define the DECW$INSTALL_XTERMINAL logical, DECwindows shouldP > > > automatically setup X terminal support, but if I just put the logical into> > > > SYSTARTUP_VMS.COM, I still have to run DECW$STARTUP.COM. > > * > > Place the logical in SYLOGICALS.COM;   > > P > > > 3. I'm trying to get a VXT-2000+ to MOP-boot off my Alpha.  Unfortunately,
 > > > evenN > > > though I'm doing "MCR NCP DEFINE NODE <nodename> ADDRESS <decnet-addr>  0 > > > HARDWARE ADDRESS <ethernet-addr> LOAD FILET > > > SYS$SYSDEVICE:[VXT$LIBRARY]<vxt-image-filename>" and updating NCP's persistentT > > > database, the changes are erased when I restart.  What am I doing wrong?  Do IO > > > have to write out the configuration changes somehow after doing a "DEFINEr > > > NODE"? > > < > > I use to do this, but that was years ago.  Someone else?   -- o John Santos  Evans Griffiths & Hart, Inc. 781-861-0670 ext 539   ------------------------------  + Date: Sat, 29 Mar 2003 04:51:38 +0000 (UTC).) From: Dan Foster <dsf@globalcrossing.net>L> Subject: Which PAK for DECNET Phase IV under Hobbyist Program?2 Message-ID: <slrnb8a9if.qg.dsf@gaia.roc2.gblx.net>  ? I can't quite seem to determine which of the various PAKs wouldy? apply for DECNET Phase IV; it comes on the Hobbyist CDROM, so IPA imagine a corresponding license is also provided as a part of the 6 program... but isn't intuitively obvious which one :-)  @ I don't just dump the contents of the email sent to register all@ these licenses because it's a lot easier for me to keep track of@ the ones I really need by installing them on an as-needed basis.' (I rarely like brute-forcing anything.)r  ' Any pointers would be much appreciated.'   -Dan   ------------------------------   End of INFO-VAX 2003.173 ************************