1 INFO-VAX	Fri, 29 Nov 2002	Volume 2002 : Issue 660       Contents:2 ??== PAKS, Campus License, Samba, Adavnced Server.6 Re: ??== PAKS, Campus License, Samba, Adavnced Server.6 Re: ??== PAKS, Campus License, Samba, Adavnced Server.6 Re: ??== PAKS, Campus License, Samba, Adavnced Server.( Alternatives to "DEC LMF PAK GEN V1.3" ?, Re: Alternatives to "DEC LMF PAK GEN V1.3" ?, Re: Alternatives to "DEC LMF PAK GEN V1.3" ?, Re: Alternatives to "DEC LMF PAK GEN V1.3" ?3 Re: Cracking OpenVMS passwords with John the Ripper 3 Re: Cracking OpenVMS passwords with John the Ripper 3 Re: Cracking OpenVMS passwords with John the Ripper 3 Re: Cracking OpenVMS passwords with John the Ripper 3 Re: Cracking OpenVMS passwords with John the Ripper   DS10 VMS 7.3-1 DECW$SERVER crash RE: Galaxy and ES47 vs ES80.... 8 Re: Gigabit Ethernet DEGPA-TA as second network problems HELP: DCL and quotation marks!" Re: HELP: DCL and quotation marks!" RE: HELP: DCL and quotation marks!> INCONSTATE at SYS$FGEDRIVER+04558 with VMS731_FIBRE_SCSI-V0100) RE: Independent Consultants + OpenVMS.org / Information Needed -Cross Compiler Availability 3 Re: Information Needed -Cross Compiler Availability 3 RE: Information Needed -Cross Compiler Availability % Re: poor Gigabit Ethernet performance " Reason to use MMS instead of MMK ?& Re: Reason to use MMS instead of MMK ? SRM for multiboot  RE: SRM for multiboot ' Re: System uptime - The Uptimes Project ' Re: System uptime - The Uptimes Project ' Re: System uptime - The Uptimes Project " Re: Wanted: Second hand Alpha DS10  F ----------------------------------------------------------------------  % Date: Fri, 29 Nov 2002 12:04:53 +0100 . From: "Hans M. Aus" <aus@vim.uni-wuerzburg.de>; Subject: ??== PAKS, Campus License, Samba, Adavnced Server. B Message-ID: <aus-B48B4A.12045329112002@wrzx08.rz.uni-wuerzburg.de>  : Which PAKS do I need to use the Advanced Server and Samba?  4 Our University computer center has a campus license.  I The license administrator tells me that there is no PAK for the Advanced  4 Server and that Pathworks 6.1 requires an extra fee.   --  B Cheers, Hans M. Aus, Wuerzburg, Germany,  aus@vim.uni-wuerzburg.de   ------------------------------  % Date: Fri, 29 Nov 2002 12:16:16 +0100 9 From: Jan-Erik =?iso-8859-1?Q?S=F6derholm?= <aaa@aaa.com> ? Subject: Re: ??== PAKS, Campus License, Samba, Adavnced Server. ' Message-ID: <3DE74C80.7A407BB0@aaa.com>   + SAMBA don't have any PAK, as far as I know.  Jan-Erik Sderholm.    "Hans M. Aus" wrote: > < > Which PAKS do I need to use the Advanced Server and Samba? > 6 > Our University computer center has a campus license. > J > The license administrator tells me that there is no PAK for the Advanced6 > Server and that Pathworks 6.1 requires an extra fee. >  > --D > Cheers, Hans M. Aus, Wuerzburg, Germany,  aus@vim.uni-wuerzburg.de   ------------------------------  # Date: Fri, 29 Nov 2002 11:20:39 GMT . From: peter@langstoeger.at (Peter LANGSTOEGER)? Subject: Re: ??== PAKS, Campus License, Samba, Adavnced Server. 4 Message-ID: <b4IF9.159765$up.1689567@news.chello.at>  s In article <aus-B48B4A.12045329112002@wrzx08.rz.uni-wuerzburg.de>, "Hans M. Aus" <aus@vim.uni-wuerzburg.de> writes: ; >Which PAKS do I need to use the Advanced Server and Samba?   " Nothing for SAMBA. It is freeware.  " 	PWLMXXXCA06.00 for PATHWORKS V6.0> 	PWLMXXXCA07.02 for Advanced Server V7.2 and/or PATHWORKS V6.0> 	PWLMXXXCA07.03 for Advanced Server V7.3 and/or PATHWORKS V6.1  5 >Our University computer center has a campus license.  > J >The license administrator tells me that there is no PAK for the Advanced 5 >Server and that Pathworks 6.1 requires an extra fee.   L If you do not want to run SAMBA, the PWLM update license is worth the money.G ASOVMS is still under development and brings bug fixes and new features E with every (maintenance) version. So, if you don't freeze your WINTEL H boxes on eg. NT4, then upgrade also ASOVMS to the latest and greatest...  . 	http://www.openvms.compaq.com/advancedserver/   --   Peter "EPLAN" LANGSTOEGER % Network and OpenVMS system specialist  E-mail  peter@langstoeger.atF A-1030 VIENNA  AUSTRIA              I'm not a pessimist, I'm a realist   ------------------------------  % Date: Fri, 29 Nov 2002 16:37:43 +0100 ) From: Hans Aus <aus@vim.uni-wuerzburg.de> ? Subject: Re: ??== PAKS, Campus License, Samba, Adavnced Server. B Message-ID: <aus-83D08A.16374029112002@wrzx08.rz.uni-wuerzburg.de>   Peter and Jan-Erik,   I Does this mean that I don't need either Pathworks or Advanced Server? My  F understanding of the Samba-VMS Faq is that Samba is only a client and  requires Pathworks.    Our objective:  ! Setup a Open VMS directory where  & 1) our Alpha writes a text file, which= 2) the PCs detect and read as input to their PC program, and   3) delete the VMS text file.  5 I would prefer to use only Samba, if that's possible.     4 In article <b4IF9.159765$up.1689567@news.chello.at>,0  peter@langstoeger.at (Peter LANGSTOEGER) wrote:  N > In article <aus-B48B4A.12045329112002@wrzx08.rz.uni-wuerzburg.de>, "Hans M. ) > Aus" <aus@vim.uni-wuerzburg.de> writes: = > >Which PAKS do I need to use the Advanced Server and Samba?  > $ > Nothing for SAMBA. It is freeware. > $ > 	PWLMXXXCA06.00 for PATHWORKS V6.0@ > 	PWLMXXXCA07.02 for Advanced Server V7.2 and/or PATHWORKS V6.0@ > 	PWLMXXXCA07.03 for Advanced Server V7.3 and/or PATHWORKS V6.1 > 7 > >Our University computer center has a campus license.  > > L > >The license administrator tells me that there is no PAK for the Advanced 7 > >Server and that Pathworks 6.1 requires an extra fee.  > N > If you do not want to run SAMBA, the PWLM update license is worth the money.I > ASOVMS is still under development and brings bug fixes and new features G > with every (maintenance) version. So, if you don't freeze your WINTEL J > boxes on eg. NT4, then upgrade also ASOVMS to the latest and greatest... > 0 > 	http://www.openvms.compaq.com/advancedserver/   ------------------------------  % Date: Fri, 29 Nov 2002 11:37:59 +0100 9 From: Jan-Erik =?iso-8859-1?Q?S=F6derholm?= <aaa@aaa.com> 1 Subject: Alternatives to "DEC LMF PAK GEN V1.3" ? ' Message-ID: <3DE74387.90E52A69@aaa.com>    Hi. 0 A client is running "DEC LMF PAK GEN V1.3" on an6 old VAX 4000. What are the alternatives in this case ?( They have a number of Alpha VMS systems.' Have anyone tried to VEST the PAK GEN ?   1 Or is there a native Alpha version of this tool ?    Jan-Erik Sderholm.    ------------------------------  % Date: Fri, 29 Nov 2002 07:59:35 -0500 2 From: rdeininger@mindspring.com (Robert Deininger)5 Subject: Re: Alternatives to "DEC LMF PAK GEN V1.3" ? K Message-ID: <rdeininger-2911020759360001@1cust212.tnt1.nashua.nh.da.uu.net>   0 In article <3DE74387.90E52A69@aaa.com>, Jan-Erik1 =?iso-8859-1?Q?S=F6derholm?= <aaa@aaa.com> wrote:    >Hi.1 >A client is running "DEC LMF PAK GEN V1.3" on an 7 >old VAX 4000. What are the alternatives in this case ? ) >They have a number of Alpha VMS systems. ( >Have anyone tried to VEST the PAK GEN ? > 2 >Or is there a native Alpha version of this tool ?  J Starting with VMS 7.3 (IIRC), there is a PAKGEN facility included with VMSG on alpha.  You need a PAKGEN PAK to use it.  Customers can get a PAKGEN H PAK which allows generation of PAKs with a specific, registered PRODUCER* (and/or ISSUER, I don't remember exactly).  H Customers who join the former CSA program, now DSPP, can register to useG PAKGEN.  I don't know about the costs, but I believe it is MUCH cheaper  than the old VAX PAKGEN.  P http://www.hp.com/dspp  -- seems to require javascript, so it's invisible to me.       -- Robert    ------------------------------  # Date: Fri, 29 Nov 2002 13:29:51 GMT " From:   VAXman-  @SendSpamHere.ORG5 Subject: Re: Alternatives to "DEC LMF PAK GEN V1.3" ? 0 Message-ID: <00A17B09.F5E93686@SendSpamHere.ORG>   In article <rdeininger-2911020759360001@1cust212.tnt1.nashua.nh.da.uu.net>, rdeininger@mindspring.com (Robert Deininger) writes:1 >In article <3DE74387.90E52A69@aaa.com>, Jan-Erik 2 >=?iso-8859-1?Q?S=F6derholm?= <aaa@aaa.com> wrote: >  >>Hi. 2 >>A client is running "DEC LMF PAK GEN V1.3" on an8 >>old VAX 4000. What are the alternatives in this case ?* >>They have a number of Alpha VMS systems.) >>Have anyone tried to VEST the PAK GEN ?  >>3 >>Or is there a native Alpha version of this tool ?  > K >Starting with VMS 7.3 (IIRC), there is a PAKGEN facility included with VMS H >on alpha.  You need a PAKGEN PAK to use it.  Customers can get a PAKGENI >PAK which allows generation of PAKs with a specific, registered PRODUCER + >(and/or ISSUER, I don't remember exactly).  > I >Customers who join the former CSA program, now DSPP, can register to use H >PAKGEN.  I don't know about the costs, but I believe it is MUCH cheaper >than the old VAX PAKGEN.   J For CSA members, the price of the PAKGEN license was quite reasonable and,I in fact, was free.  The PRODUCER and ISSUER must be cleared with the PAK- : GEN "police" before the generator license key is provided.    Q >http://www.hp.com/dspp  -- seems to require javascript, so it's invisible to me.   I CSA cost $600/yr.  The DSPP program, I am told, is free but, sadly, their I registration site is so littered with troubled javascript and problematic & background CGIs that I can't register.   --O VAXman- OpenVMS APE certification number: AAA-0001     VAXman(at)TMESIS(dot)COM              5   "Well my son, life is like a beanstalk, isn't it?"     ------------------------------    Date: 29 Nov 2002 10:40:08 -0600- From: Kilgallen@SpamCop.net (Larry Kilgallen) 5 Subject: Re: Alternatives to "DEC LMF PAK GEN V1.3" ? 3 Message-ID: <hBGSa24XJxly@eisner.encompasserve.org>   c In article <3DE79BD9.1B4A65BE@aaa.com>, Jan-Erik =?iso-8859-1?Q?S=F6derholm?= <aaa@aaa.com> writes:  > OK. ? > I have a 7.2-1 system but found nothing called *LMF* or *PAC*  > that looked promising. > & > Any pointers to what to search for ?  F Since VMS Development has chosen to have documentation of this featureE be distributed only through CSA, I don't think it is proper for me to  distribute it feature.  H I have heard at least one person from VMS Development discuss the chanceI that this documentation might be made more generally available, but until 9 that happens, you should contact your CSA representative.    ------------------------------  % Date: Fri, 29 Nov 2002 12:06:24 +0000 ' From: Andrew Harrison SUNUK Consultancy < Subject: Re: Cracking OpenVMS passwords with John the Ripper. Message-ID: <3DE75840.1060005@nospamn.sun.com>   JF Mezei wrote:  > Jean-loup Gailly wrote:  > J >>By using John, a system administrator can effectively eliminate all weakN >>passwords. I'm surprised that many VMS fans cannot understand the usefulnessG >>of such a tool. (I wrote the John VMS patch for a security audit of a  >>large company.)  >  >  > O > Suppose you do find that your boss has a weak password. What do you do then ? P > Do you go and tell him that you extracted his password and decided that it wasP > weak ? Wouldn't that jeoperdize the integrity of the system manager who is not/ > supposed to have access to user's passwords ?  >   ; You tell him, but in private so as not to make him/her look  idiotic in public.  = Most major organsations have standards that they expect their 9 employees to follow when setting their passwords, if your < bosses password is crackable then it probably doesn't follow your security standards.  < If you don't tell him and someone breas into his account you7 will be worse off then not telling him in the the first  place.  8 We run crack internally in Sun, written by Alex Muffett.  8 If you have a poorly constructed password then you get a< warning email from the security team. It doesn't distinguish8 between job titles and doesn't have a CEO userid filter.  = This discussion is a re-run of discussions about crack, COP's  etc.  / Have a look at (kindly forwarded to me by Alex)   W http://groups.google.com/groups?hl=en&lr=&ie=UTF-8&safe=off&selm=0gcHpabz1%40cs.psu.edu   9 The thread that follows and the views expressed there are ? is pretty much reflected by the views expressed in this thread. @ You will however note that the origional discussion was 11 yearsD ago. (Was the Apollo DN10000 arround 11 years ago, it seems longer).   Regards  Andrew Harrison    ------------------------------  + Date: Fri, 29 Nov 2002 12:13:22 +0000 (UTC) + From: david20@alpha2.mdx.ac.uk (David Webb) < Subject: Re: Cracking OpenVMS passwords with John the Ripper+ Message-ID: <as7ll2$q9e$1@aquila.mdx.ac.uk>   b In article <3DE667B9.F8374923@vl.videotron.ca>, JF Mezei <jfmezei.spamnot@vl.videotron.ca> writes: >Jean-loup Gailly wrote:K >> By using John, a system administrator can effectively eliminate all weak O >> passwords. I'm surprised that many VMS fans cannot understand the usefulness H >> of such a tool. (I wrote the John VMS patch for a security audit of a >> large company.) > I Although such tools have there uses it is better to have them runnable on C the system itself (from a suitably privileged and secured account). K The idea of copying the sysuaf onto a PC to do the analysis circumvents all ; the VMS mechanisms designed to protect this important file.   J I seem to recall that there were some public domain native programs (from F before VMS had password dictionaries etc ie quite some time ago) whichJ did this. I haven't looked at them in years and hence have no idea whether they still work.  G Problems with weak passwords can be reduced by using VMS mechanisms ie     minimum password length,  7 password history list - to prevent reuse of passwords,   password dictionary,  M inbuilt triviality checks - to stop users choosing trivial passwords such as  .                             the account name    M Together with setting appropriate LGI values so that the system takes evasive H action after the user gets the password wrong a certain number of times I ie for a certain period even if the hacker got the password correct they  I would not be allowed to login - this drastically reduces the chance of a   successful brute force attack.  K Increasing the characterset from which VMS passwords can be chosen probably H wouldn't really increase VMS security when these mechanisms are properlyG deployed. However it would increase usability when users (as they will) K use the same password on multiple systems - this of course reduces security # to that of the least secure system. I Many companies have policies on what makes a strong password - in such an J environment either VMS is left out or the company policy is reduced to theH characterset allowed in VMS which may be entirely inappropriate for some other less protected systems.   
 David Webb VMS and Unix team leader CCSS Middlesex University       > N >Suppose you do find that your boss has a weak password. What do you do then ?O >Do you go and tell him that you extracted his password and decided that it was O >weak ? Wouldn't that jeoperdize the integrity of the system manager who is not . >supposed to have access to user's passwords ? > M >He may have access to his own sysuaf.dat on that small departmental VAX, but L >getting his boss's password on the vax would probably also get his passwordO >for the IBM mainframe and email servers etc etc etc. Revealing that you can in [ >fact extract passowrds would jeoperdize the security of other systems in the organisation.  > L >One would have to be very diplomatic in trying to educate the person on theP >use of stronger passowrds without giving any hints that you saw their password.   ------------------------------   Date: 29 Nov 2002 15:24:02 GMT( From: bill@cs.uofs.edu (Bill Gunshannon)< Subject: Re: Cracking OpenVMS passwords with John the Ripper5 Message-ID: <as80qi$p0e3t$4@ID-135708.news.dfncis.de>   / In article <3DE667B9.F8374923@vl.videotron.ca>, 3 	JF Mezei <jfmezei.spamnot@vl.videotron.ca> writes:  > Jean-loup Gailly wrote: K >> By using John, a system administrator can effectively eliminate all weak O >> passwords. I'm surprised that many VMS fans cannot understand the usefulness H >> of such a tool. (I wrote the John VMS patch for a security audit of a >> large company.) >  > O > Suppose you do find that your boss has a weak password. What do you do then ? P > Do you go and tell him that you extracted his password and decided that it wasP > weak ? Wouldn't that jeoperdize the integrity of the system manager who is not/ > supposed to have access to user's passwords ?  > N > He may have access to his own sysuaf.dat on that small departmental VAX, butM > getting his boss's password on the vax would probably also get his password P > for the IBM mainframe and email servers etc etc etc. Revealing that you can in\ > fact extract passowrds would jeoperdize the security of other systems in the organisation. > M > One would have to be very diplomatic in trying to educate the person on the Q > use of stronger passowrds without giving any hints that you saw their password.   E I used to regularly run password cracking programs against the master D password files in the department usually then informing the the userD that their password was easily determinable (often telling them whatC it was, including their little tricks like "h311o" for "hello").  I D stopped doing it when common Unix eliminated the ability of the noneE priveledged user to actually read the encrypted passwords and because F I never found a user who cared.  They usually just changed their pass-1 word to something else just as easily determined.   E As to the idea that the number of possible combinations prevents this E kind of attack, while that was true when I first ran "crack" (it took H a week and broke only one or two passwords) now days, I can do the wholeG password file overnight and frequently break all the passwords that are H not pure random strings of characters (which is what I issue for initialD passwords.)  Remember also, on most systems there really is only oneG password you need to break to compromise the system.  Now imagine a lab C full of 2GHZ P3's spending the 4 day Thanksgiving weekend running a H distributed password cracking program against the SYSTEM password.  Even+ A totally random string is not a guarantee.   
 All the best.    bill   --  J Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>       ------------------------------   Date: 29 Nov 2002 15:31:16 GMT, From: bill@gw5.cs.uofs.edu (Bill Gunshannon)< Subject: Re: Cracking OpenVMS passwords with John the Ripper5 Message-ID: <as8184$p0e3t$5@ID-135708.news.dfncis.de>   8 In article <20021127082748.7a021426.mathog@caltech.edu>,* 	David Mathog <mathog@caltech.edu> writes: > L > Which is nice but of little relevance to VMS security.  Only a priv'd userI > should be able to read SYSUAF.DAT, and if that user has evil intentions    ^^^^^^^^^^^^^^^^^ C > then the system is already compromised - there's no need to crack J > the password file.  This contrasts with many (most?) Unix systems, where1 > everybody can read the encrypted password file.   E Once again, we compare current VMS to 20 year old Unix.  I know of no E "current" version of Unix (or Linux for that matter) that allows non- B priveledged access to the encrypted passwords without the explicitG allowance of that by the Administrator.  And shadow passwords have been H available as a graft on addition as far back as SunOS4.  I expect I willG even put that high on the list of improvements I am making to Ultrix-11 : even though I doubt anyone using it will really care.  :-)   bill   --  J Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>       ------------------------------  % Date: Fri, 29 Nov 2002 09:38:07 +0100 6 From: =?iso-8859-1?Q?Jean=2DFran=E7ois=20PI=C9RONNE?= < Subject: Re: Cracking OpenVMS passwords with John the Ripper+ Message-ID: <3DE7276F.464C9194@laposte.net>    Salut Jean-Loup,  L First I would like not to argument pro/cons security based on password, only about password guessable.   O First, you agree that on a normally configure VMS system you can't read the UAF  file.   M Very bad passwords is much less a problem on VMS than on many others systems,  because there is other security ! which prevent you to do many try: O for example if you have a direct access, then you terminal will become suspect, P then intruder,then if I have enablethis feature the account will become disuser,# and not including dome audit alarm. N Why did you think that credit card have only a four digits code, because after  three attempts the game is over.O So what is the chance of a, for example, 6 letters passwords, to be be found if  you have only five attempt? H On some other systems, you can made as many attempt as you want, this is fundamental different.  K I don't say it is impossible, just that the probability is very very small.   P The thread http://www.openvms.compaq.com/wizard/wiz_4612.html is much more aboutJ general password security versus other methods (Kerberos, smart cards, andN biometric hardware, etc...) and not about some problem into the implementation# of the password mechanism into VMS.   D I have only seen in movies people finding password after 3 attempts.  I Any way, a good password is better than a bad, but a too good password is C sometime written on some paper because people can't remember it :-)   M On most of VMS system I have seen, there are much important security problem.   
 Jean-Franois    > F > It seems that many people have misunderstood my intent. It should beG > obvious that if a bad guy has access to your sysuaf.dat you're in big K > trouble already. I have added a sentence on http://gailly.net/security/ : F > "This tool is designed for system administrators to detect users who= > too often select very bad passwords, too easily guessable".  > ? > I advise system administrators who do not believe this to try K > John at least once. You will be amazed to see how bad most passwords are, F > and how quickly you can find most of them. Of course, make sure thatE > the working directory of John and all files within it are correctly * > protected against read access by others. >  > Nic Clews writes:  > L > > Secondly, a password list you hit a single user account with in the mostI > > part cannot exist in those that are in the password dictionary (where  > > implemented!)  > F > I had trouble parsing your sentence. John does much more than tryingF > all words from a dictionary. It has elaborate rules to generate manyF > variations of dictionary words, and a sophisticated incremental modeE > to try non-dictionary words in optimum order. This is what makes it " > the best password cracking tool. > G > > and for those of us that use the password policy module forcing the @ > > presence of non alphabetic characters and therefore _really_K > > 'unreal' words, you'd have to slow that 150,000 guesses per second down  > G > My patch is designed for the builtin OpenVMS password algorithms, not A > for external modules. HP (Hoff Hoffman) indeed recommends using 8 > algorithms other than their own, in this thread and in4 > http://www.openvms.compaq.com/wizard/wiz_4612.html > E > > I don't really believe that anyone has the storage or the time to G > > pregenerate such a colossal list, moving each character through all J > > possible combinations starting at some arbitrary length, and extendingD > > to the maximum allowable or expected (and risking missing longerF > > ones). The sheer time to perform the IO alone to this list makes a > > mockery of the technique.  > K > There is no point in generating a colossal list on disk. The VMS password L > algorithms correctly take into account a salt and the user name to preventH > such attacks. But at 150,000 guesses per second it is very easy to tryL > *all* legal VMS passwords of 7 characters or less, working in memory only.E > You can also easily try all simple variations of a small dictionary J > (1 million words). John is good at generating the variations that people  > actually use most of the time. > J > By using John, a system administrator can effectively eliminate all weakN > passwords. I'm surprised that many VMS fans cannot understand the usefulnessG > of such a tool. (I wrote the John VMS patch for a security audit of a  > large company.)  >  > Jean-loup Gailly > http://gailly.net/security/    ------------------------------  % Date: Fri, 29 Nov 2002 19:07:24 +0100 + From: "Juan C. Blanco" <jcblanco@fi.upm.es> ) Subject: DS10 VMS 7.3-1 DECW$SERVER crash ) Message-ID: <3DE7ACDC.A919FD4C@fi.upm.es>   > Hello, I've installed VMS v7.3-1 on a DS10 test system and theG DECW$SERVER crash when a user logs out (using the exit button) from the  CDE session.  D I've applied all patches to this version of VMS, I can reproduce theC crash if start some X applications from other systems displaying on E the DS10 server, the DECW$SERVER_0_ERROR.LOG reports an ACCVIO error.   9 We are using also Multinet V4.4 with all current patches.   B I've revised the protection of the [.DT] directory of the user and all seems right.  E Using "decw$startup restart" starts a new server and the login screen - but don't accept any keystroke on the window.    Any ideas ?    Thanks Juan C. Blanco   The error from de LOG is:   ; %SYSTEM-F-ACCVIO, access violation, reason mask=00, virtual : address=0000000045524940, PC=0000000000004048, PS=00000000  H Unrecoverable server internal error (error code = 12) found, terminating all connections. Mapped Images...  -   START       END        LENGTH    IMAGE NAME -   -----       ---        ------    ---------- 2        0      301ff      301ff    DECW$SERVER_MAIN1    94000     245bff     1b1bff    DECW$SERVER_DIX 3 7bbb6000   7bbe7fff      31fff    DECW$SECURITY_VMS + 7b884000   7b8f5fff      71fff    SECURESHR , 7b36c000   7b3edfff      81fff    SECURESHRP7    32000     246410     214410    DECW$TRANSPORT_COMMON 0 82020120   82032bf0      12ad0    SYS$BASE_IMAGE/ 7bf9c000   7c02ffff      93fff    DECC$SHR_EV56 * 7bc72000   7bcb7fff      45fff    DPML$SHR- 7b926000   7b935fff       ffff    CMA$TIS_SHR ( 7b5ba000   7b60bfff      51fff    LIBRTL( 7b60c000   7b613fff       7fff    LIBOTS4 82003f58   82005f88       2030    SYS$PUBLIC_VECTORS0   34a000     41b3ff      d13ff    DECW$SVEXT_XIE6   41c000     45c3ff      403ff    DECW$SVEXT_DEC_XTRAP<   45e000     49e3ff      403ff    DECW$SVEXT_MULTI_BUFFERING7   4a2000     5e37ff     1417ff    DECW$TRANSPORT_DECNET *   5e4000     5e7fff       3fff    DECC$MSG+   5e8000     5ee9ff       69ff    SHRIMGMSG 3   5f0000     6001ff      101ff    DECW$TRANSPORTMSG 6   602000     6423ff      403ff    DECW$TRANSPORT_LOCAL6   644000     6843ff      403ff    DECW$TRANSPORT_TCPIP4   686000     6f89ff      729ff    DECW$SERVER_DDX_P27   820000     8809ff      609ff    DECW$SERVER_DDX_CFB16 7   6fa000     74a9ff      509ff    DECW$SERVER_DDX_CFB32 5   74c000     7bc9ff      709ff    DECW$SERVER_DDX_CFB 5   7be000     81e7ff      607ff    DECW$SERVER_DDX_MFB   " Exception Call stack dump follows:         PC     IMAGE+offset of call       --     --------------------%   1f8bb4     DECW$SERVER_DIX + 164bb4 %   1fca5c     DECW$SERVER_DIX + 168a5c $   13cfd8     DECW$SERVER_DIX + a8fd8$   13d2d4     DECW$SERVER_DIX + a92d4$   1665b8     DECW$SERVER_DIX + d25b8  8 ********** marking the end of call stack dump **********8 ******************************************************** --  B +----------------------------------------------------------------+B | Juan C. Blanco                                                 |B |                                                                |B |  Centro de Calculo              |                              |B |  Facultad de Informatica U.P.M. |  E-mail: jcblanco@fi.upm.es  |B |  Campus de Montegancedo         |                              |B |  Boadilla del Monte             |  Tel.:    (+34) 91 336 7466  |B |  28660 MADRID (Spain)           |  Fax :    (+34) 91 336 6913  |B +----------------------------------------------------------------+   ------------------------------  % Date: Fri, 29 Nov 2002 09:04:22 -0000  From: Martin.Walker@csf.co.uk ( Subject: RE: Galaxy and ES47 vs ES80....A Message-ID: <C3521D5A224C344284C3666E831BF680768FE3@london_exch2>   I I would like to see something like the mixed version migration support in J clustering.  There is no guarantee to fix a problem but support assistanceL will be available.  If the problem can't be fixed - the solution would be to
 "upgrade".   -----Original Message-----/ From: Nic Clews [mailto:sendspamhere@127.0.0.1]  Sent: 27 November 2002 12:45 To: Info-VAX@Mvb.Saic.Com ( Subject: Re: Galaxy and ES47 vs ES80....     Dave Weatherall wrote: >  > ... B > If HP/Compaq mean they won't qualify tri-architecture clusteringD > unless there is explicit 'customer demand', well I think they'd beE > shooting themselves in the foot (it must be a centipede). I believe F > most users would take it as read. To 'compile and run and _test_' inH > many environments still includes VAX. Not just at Compaq. That is very2 > much easier with 'normal' clustered disk access. > / > Maybe someone could explain the difficulties.   A The problem is the word "supported", that requires qualification. C Qualification means not just A VAX with AN Alpha and AN Itanium, it 8 means all possible combinations with supported hardware.  C VAX and Alpha has a two dimensional matrix, it goes 3D when you add B Itanium, so you double that VAX/Alpha testing matrix every Itanium system you also support.  D Nothing is being coded to _prevent_ tri clustering, but in trying toF support that, they could spend inordinate amounts of time testing, and# not releasing _supported_ software.    This is my understanding of it.  --  ? Regards, Nic Clews a.k.a. Mr. CP Charges, CSC Computer Sciences  nclews at csc dot com   A This e-mail including any attachments is confidential and may be  F legally privileged. If you have received it in error please advise the@ sender immediately by return email and then delete it from your  system. ? All electronic communications with the Company may be monitored A in accordance with the UK Regulation of Investigatory Powers Act, , Lawful Business Practice Regulations, 2000. = If you do not consent to such monitoring, you should contact   the sender of the e-mail. B The unauthorised use, distribution, copying or alteration of this C email is strictly forbidden. If you need assistance please contact  C Head Office on (+44)(0)870 749 9000 between 9:00 am and 5:30 pm GMT    ------------------------------  % Date: Fri, 29 Nov 2002 12:10:11 +0000 ( From: Nic Clews <sendspamhere@127.0.0.1>A Subject: Re: Gigabit Ethernet DEGPA-TA as second network problems ( Message-ID: <3DE75923.8C66BC1@127.0.0.1>   Alan Fay wrote:  >   S > In our Lab the DEGPA-TA NIC is used on two V7.3-1 833MHz DS20E (single processor) M > Alpha's and in tests we get a backup throughput of about 40 MBytes/sec to a  > SunFire Media Server.  > K > This is a very fast NIC -- the network almost disappears as a bottleneck. / > We are using a NETGEAR GS508T Gigabit Switch.  > M > I must say that the 833MHz DS20E Alpha is a very well built machine, when I K > installed the NIC I was very impressed with the layout and quality of the J > hardware (much better than Sun), it is also the fasted Alpha I have ever  > had the pleasure to work with. > 
 > Alan Fay > VERITAS Software Corporation > Roseville Engineering    Alan,   H Thank you for your confirmation, seems we now have the issue resolved, IC called HP services who pushed the diagnosis beyond the card itself. 1 Turned out that I'd been misquoted an IP address!   H I have to agree with your comments on the DS20, and we are using VERITASH as part of our BACKUP solution. I installed the machine and upgraded and< installed options and agree about how nice it is to work on.   --  ? Regards, Nic Clews a.k.a. Mr. CP Charges, CSC Computer Sciences  nclews at csc dot com    ------------------------------    Date: 28 Nov 2002 23:27:16 -0800' From: roose_chua@yahoo.com (Roose Chua) ' Subject: HELP: DCL and quotation marks! = Message-ID: <127ddcff.0211282327.61a4c428@posting.google.com>   	 Hi folks,   C I've been searching for days now on how to go about with my problemFC and it seems that I have been stucked without the correct solution.C  E What I am trying to do is to create a DCL procedure using another DCLmE procedure. What I would like to be the final file would be like this:r   $ testsym = "test"/ $ write sys$output "This is just a ''testsym'."v $ exit  0 What my initial procedure that did this is like:  " $ open/write testfile testfile.com' $ write testfile "$ testsym = ""test"""s/ $ write testfile "$ This is just a ''testsym'."S $ write testfile "$ exit"p $ close testfile $ exit  @ However, this does not produce the result that I wanted to have.    Thanks in advance for your help! Rooses   ------------------------------    Date: 29 Nov 2002 09:13:47 +0100' From: huber@mppmu.mpg.de (Joseph Huber)c+ Subject: Re: HELP: DCL and quotation marks!i+ Message-ID: <vbzYKNoQjs9g@vms.mppmu.mpg.de>b  g In article <127ddcff.0211282327.61a4c428@posting.google.com>, roose_chua@yahoo.com (Roose Chua) writes:! > Hi folks,a > E > I've been searching for days now on how to go about with my problemaE > and it seems that I have been stucked without the correct solution.e > G > What I am trying to do is to create a DCL procedure using another DCLrG > procedure. What I would like to be the final file would be like this:n >  > $ testsym = "test"1 > $ write sys$output "This is just a ''testsym'."  > $ exit > 2 > What my initial procedure that did this is like: > $ > $ open/write testfile testfile.com) > $ write testfile "$ testsym = ""test""" 1 > $ write testfile "$ This is just a ''testsym'."- > $ write testfile "$ exit"- > $ close testfile > $ exit > B > However, this does not produce the result that I wanted to have.  v  ) No, it produces the result as written :-)p The following:   $ q="'" 	 $ dq=""" e" $ open/write testfile testfile.com' $ write testfile "$ testsym = ""test"""SK $ write testfile "$ write sys$output ",dq,"This is just a ",q,q,"test",q,dqe $ close testfile   produces testfile.com: $ testsym = "test"+ $ write sys$output "This is just a ''test'"v    What is probably what You want .   --  N Joseph "Sepp" Huber   mailto:joseph.huber@web.de   http://www.huber-joseph.de/   ------------------------------  % Date: Fri, 29 Nov 2002 09:09:21 -0500 ! From: VAXVMS <bounce@notmail.com>6+ Subject: RE: HELP: DCL and quotation marks!AK Message-ID: <BA52530E3149734A9BAABDBBFA808E4903027BA6@rlghncst964.usps.gov>a   Roose:  7 Here's an alternative approach which uses the plus signd7 as a string concatenation operator (instead of creatinge7 symbols for the single and double quote characters) to r achieve your desired result:  9 Personally, I like the ability to use + (and also - ) forh: string manipulation- sometimes it avoids the need to use a8 combination of f$locate and f$extract to perform certain string-related tasks.    $ type test.comr  " $ open/write testfile testfile.com' $ write testfile "$ testsym = ""test""" H $ str = "$ write sys$output ""This is just a " + "'" + "'" + "testsym" + "'.""" $ write testfile str $ write testfile "$ exit"9 $ close testfile $ exit   $ type testfile.com<   $ testsym = "test"/ $ write sys$output "This is just a ''testsym'."u $ exit   HTH,   WWWebb    	 Hi folks,P  C I've been searching for days now on how to go about with my problemBC and it seems that I have been stucked without the correct solution.7  E What I am trying to do is to create a DCL procedure using another DCLdE procedure. What I would like to be the final file would be like this:n   $ testsym = "test"/ $ write sys$output "This is just a ''testsym'."t $ exit  0 What my initial procedure that did this is like:  " $ open/write testfile testfile.com' $ write testfile "$ testsym = ""test"""a/ $ write testfile "$ This is just a ''testsym'."n $ write testfile "$ exit"b $ close testfile $ exit  @ However, this does not produce the result that I wanted to have.    Thanks in advance for your help! RooseS   ========================  William W. Webb / DSSC/RLM, USPS OpenVMS Support Services& 4924 Green Road Raleigh, NC 27616-2800: 919.874.3043 <FirstInitialDotLastNameAtEmailDotUSPSDotGov>   ------------------------------  % Date: Fri, 29 Nov 2002 12:07:52 -0500 " From: "Hal Kuff" <Kuff@Tessco.Com>G Subject: INCONSTATE at SYS$FGEDRIVER+04558 with VMS731_FIBRE_SCSI-V0100bO Message-ID: <4CA4C7004F08C1AB.DC35DD8FC64811B8.47D56A20803C1261@lp.airnews.net>g  L We took a DS10 system running 7.2-2 and upgraded to 7.3-1 via the boot CD...L We applied no patches.  When the startup issued the IO Autoconfigure All (orK equiv) the system crashed as above.  We put all of the 7.3-1 patches in and H got the same result.  When we took the same system and did a clean 7.3-1G install everything was fine with or without the patches, so we know the  hardware and SAN are OK.  L Support has the call now for a week or so, they suggested one or two things:G         * Check the system for *.exe in the sys$specificn path that the  install might have overlookedN<         * Make sure that Sys$Library:Sys$Icbm.exe is present  4 No word yet, wondering if anyone else has seen this?   ------------------------------  % Date: Fri, 29 Nov 2002 10:24:13 -0500r' From: "Main, Kerry" <Kerry.Main@hp.com> 2 Subject: RE: Independent Consultants + OpenVMS.orgT Message-ID: <BE56C50EA024184DAF48F0B9A47F5CF402660BAA@kaoexc01.americas.cpqcorp.net>   Bill,a  F >>> Competing with the rest of the world, for the first time since theF Industrial Revolution really took hold here (and in many cases for the first time ever).<<f  G Yep, everyone is having to compete globally and for many, that is a new G experience. And individual Govt's are becoming much less of a factor inh$ trying to control what is happening.  = A few examples that are a bit closer to home (at least for usn Northerners) -  F - recent software lumber dispute between US and Canada appears to be aG logging issue. However, while both sides likely have valid points, homeqF builders in US have stated that they estimate is that it is costing USH citizens something like an additional $2k for every new house built. So,B all US citizens are paying more to protect special interest lumber industries in the US.A  ? - recent US TV show (20/20 ?? Can't remember) interviewed a few F previously US based candy companies and asked them why they had pickedH up and relocated lock, stock and barrel to Canada. Number one reason wasB cost of sugar in the US and the fact that 60% of the cost of theirD products was sugar. Because of import duties on sugar and high sugarA cost in the US, these companies were finding they could no longeruC compete with global competitors. So, US citizens are paying more toeH protect special interest Sugar groups in the US and now it is costing US jobs.3  E Right or wrong, imho, the globalization and competitive nature of all>C future businesses is only going to increase. The more Govt's try to F coddle specific groups with protectionist policies, the more they willH hurt other citizens. I would certainly not depend on any Govt to protect my job in the future.=20  F One only has to look at the recent events in Europe i.e.. establishing@ common currency, eliminating some of the trade bureaucracy    ofG individual countries to see that they are very much in tune to becomingK$ much more competitive in the future.     RegardsU  
 Kerry Main Senior Consultant  Hewlett-Packard (Canada) Co.! Consulting & Integration Services  Voice: 613-592-4660n Fax   : 613-591-4477 Email: kerryDOTmain@hpDOTcom-     (remove the DOT's and replace with "."'s)h     -----Original Message-----2 From: Bill Todd [mailto:billtodd@metrocast.net]=20 Sent: November 26, 2002 9:29 PM- To: Info-VAX@Mvb.Saic.Com22 Subject: Re: Independent Consultants + OpenVMS.org      < "David J. Dachtera" <djesys.nospam@fsi.net> wrote in message! news:3DE2E130.DC640550@fsi.net...> > John Vottero wrote:g   ..  J > > Certainly there are many oppressed people in the world but we won't=20/ > > help them by withholding job opportunities.0 >AG > Must we sacrifice our own workforce so they can have one? ...or is=20:1 > there an equitable way for the two to co-exist?N  F What we're seeing is the emergence of exactly that equitable way:  the@ problem is that U.S. workers find it less comfortable than theirE previous coddled existence where they didn't have to compete on equaloC terms with people elsewhere who were willing to do the same job forn considerably less money.  G (This of course ignores the question of whether the quality of the workhG is in fact equal, but that's not at all the same issue:  companies thatrE save money by sacrificing quality will likely get what they deserve -r+ and this will vary situation by situation.)    >E' > > Instead of trying to erect barrierse* > > that prevent jobs from leaving the US, >G' > Whole-heartedly disagree. Dead wrong.   D Disagree all you want:  it's happening, for very simple and powerfulF economic reasons, and there's no reason to expect it to stop until the) playing fields are a lot closer to equal.:   >." > > we should be trying to enforce! > > standards for workers rights,  >SC > What about our own people? Is not our loyalty first to our own=20i
 > countrymen?n  A Clearly it's not:  if it were, Americans would refuse to purchaser= foreign products that cost less and demand domestic products.t   >t > > environmental controls >(G > What about here at home? Why not provide opportunities for our own=20k3 > people, our own parents, siblings, progeny, ... ?a  G Why not indeed?  Incompetent corporate leadership lacking any technicalnE vision is a prime reason why the kinds of jobs that *can't* easily be E exported aren't being created to replace those jobs that are leaving:BE when new products and even entire industries are born, for at least a H while they're much easier to grow locally because of the difficulties of? arm's-length management and coordination - it's the more maturefH industries that have the time to sit back and figure out how to optimizeH costs (because they are no longer expending all their efforts developing their technology).   >. > > and human rights.s >nI > Maybe we need a world-wide government to whom all rulers must answer=20i > for their crimes ... ?  H Sounds like a good goal to me, though a major challenge to make work.  IB certainly have a great deal more respect for the attitudes of manyB Western governments than I have for our those of our own (though IG wouldn't have said that a few years ago), and since such attitudes seem.D to be on the ascendent world wide such a higher authority could be a major improvement.   >a > > WeE > > shouldn't erect trade barriers to ensure that an American high=20r
 > > school dropH > > out doesn't lose their job, we should do it to ensure that a foreign childaF > > is able to finish high school without being forced into a sweat=20	 > > shop.n >h4 > ...and that leaves U.S. workers exactly ... where?  B Competing with the rest of the world, for the first time since theF Industrial Revolution really took hold here (and in many cases for the first time ever).s   - bill   ------------------------------    Date: 29 Nov 2002 00:27:56 -0800" From: nipun_java@yahoo.com (nipun)8 Subject: Information Needed -Cross Compiler Availability= Message-ID: <a52b9f26.0211290027.2745938f@posting.google.com>t   Any input would be appreciated.s   Domain : Embedded Systems    The problem statement :   A Software Development Environment on VAX-VMS has to be ported to anE suitable target development environment.The target processor is Inteli 80186 Processor.  C The code is written in ASM186 & PLM186 assembly languages and Intel * cross compilers are used to generate code.  C 1)Can anyone give any pointers with respect to availability of suchoE cross compilers/assembler for ASM186 & PLM186 on other platforms liker- windows,linux,sun solaris?.If so, any links..   F 2)Does GCC support such cross compilation for 80186 processors and are6 gcc ports available for different target environments?   Nipund   ------------------------------  # Date: Fri, 29 Nov 2002 10:20:42 GMTt. From: peter@langstoeger.at (Peter LANGSTOEGER)< Subject: Re: Information Needed -Cross Compiler Availability4 Message-ID: <_bHF9.159238$up.1683594@news.chello.at>  b In article <a52b9f26.0211290027.2745938f@posting.google.com>, nipun_java@yahoo.com (nipun) writes:  >Any input would be appreciated.   Any ? No problem ;-)   >Domain : Embedded Systems >c >The problem statement : >oB >Software Development Environment on VAX-VMS has to be ported to aF >suitable target development environment.The target processor is Intel >80186 Processor.n >>D >The code is written in ASM186 & PLM186 assembly languages and Intel+ >cross compilers are used to generate code.>  7 I've done this for some years but about 18 years ago...a  D >1)Can anyone give any pointers with respect to availability of suchF >cross compilers/assembler for ASM186 & PLM186 on other platforms like. >windows,linux,sun solaris?.If so, any links.. >hG >2)Does GCC support such cross compilation for 80186 processors and ared7 >gcc ports available for different target environments?c  8 How about VESTing the cross compilers to OpenVMS Alpha ?K The cross compilers I know have been pure (and not very bad) usermode code.p; So, look at the OpenVMS Migration Software for VAX to Alphau  6 	http://www.openvms.compaq.com/openvms/products/omsva/  C and try to make a crosscompiler image running on OpenVMS Alpha. YousF will be impressed of the performance then. You might need to reiterate9 the migration process more than once for optimal results.r   --   Peter "EPLAN" LANGSTOEGERi% Network and OpenVMS system specialisto E-mail  peter@langstoeger.atF A-1030 VIENNA  AUSTRIA              I'm not a pessimist, I'm a realist   ------------------------------  % Date: Fri, 29 Nov 2002 08:30:31 -0500 ' From: "Main, Kerry" <Kerry.Main@hp.com> < Subject: RE: Information Needed -Cross Compiler AvailabilityT Message-ID: <BE56C50EA024184DAF48F0B9A47F5CF4023D9974@kaoexc01.americas.cpqcorp.net>  G >>> and try to make a crosscompiler image running on OpenVMS Alpha. Your- will be impressed of the performance then.<<<s  F As well as not having to worry about the virus and/or hacker attack of( the day associated with other platforms.  E If looking to move to a more open (i.e.. not tied to MS) environment,s/ you might want to consider NetBeans on OpenVMS:s  
 Reference:< http://www.openvms.compaq.com/openvms/products/ips/netbeans/   :-)h   Regardss  
 Kerry Main Senior Consultanti Hewlett-Packard (Canada) Co.! Consulting & Integration Serviceso Voice: 613-592-4660l Fax   : 613-591-4477 Email: kerryDOTmain@hpDOTcom-     (remove the DOT's and replace with "."'s)e     -----Original Message-----8 From: Peter LANGSTOEGER [mailto:peter@langstoeger.at]=20 Sent: November 29, 2002 5:21 AM. To: Info-VAX@Mvb.Saic.Com < Subject: Re: Information Needed -Cross Compiler Availability    = In article <a52b9f26.0211290027.2745938f@posting.google.com>,h$ nipun_java@yahoo.com (nipun) writes:  >Any input would be appreciated.   Any ? No problem ;-)   >Domain : Embedded Systems >a >The problem statement : > E >Software Development Environment on VAX-VMS has to be ported to a=20aI >suitable target development environment.The target processor is Intel=20t >80186 Processor.  >rG >The code is written in ASM186 & PLM186 assembly languages and Intel=20i+ >cross compilers are used to generate code.o  7 I've done this for some years but about 18 years ago...g  G >1)Can anyone give any pointers with respect to availability of such=20-I >cross compilers/assembler for ASM186 & PLM186 on other platforms like=20o. >windows,linux,sun solaris?.If so, any links.. >nJ >2)Does GCC support such cross compilation for 80186 processors and are=207 >gcc ports available for different target environments?r  8 How about VESTing the cross compilers to OpenVMS Alpha ?E The cross compilers I know have been pure (and not very bad) usermoderA code. So, look at the OpenVMS Migration Software for VAX to Alphah  6 	http://www.openvms.compaq.com/openvms/products/omsva/  H and try to make a crosscompiler image running on OpenVMS Alpha. You willE be impressed of the performance then. You might need to reiterate thes5 migration process more than once for optimal results.t   --=20x Peter "EPLAN" LANGSTOEGERc% Network and OpenVMS system specialistt E-mail  peter@langstoeger.atF A-1030 VIENNA  AUSTRIA              I'm not a pessimist, I'm a realist   ------------------------------  + Date: Fri, 29 Nov 2002 10:00:41 +0100 (MET) & From: Rudolf Wingert <win@fom.fgan.de>. Subject: Re: poor Gigabit Ethernet performance6 Message-ID: <200211290900.KAA04857@sinet1.fom.fgan.de>   Hello,   Matt Muggeridge wrotes:t   >>>sG Since you are wanting to test max throughput then you need to tune your.L application and your system, and as Rick noted, verify you are not receiving errors.   L There has been lots of good suggestions from the community here, but I don'tD see you trying them.  You need to be methodical and thorough in yourB approach and take the time to understand all the factors involved. <<<r  F Sorry, but I do not understand what you wrote. TTCP is not my own testD tool. It is a tool from HP. Other then DTSEND looks it it like a nonA optimized tool. Also the user do not use optimized tools. He uses F normal system utilities and stamdard file open. So the the result with. the copy is the best test for the performance.D Now a few facts: there are no network errors. Also it is possible toE get 77% of the FastEthernet performance with the COPY utility. DTSENDsG meassures 95% for FastEthernet, but only 27% for GigabitEthernet. I didtE measure the disk read (COPY CONT128MB NL:) for that disk, which I dideB use for the test. Result: 25.2MB/s. If Gigabit would tick right, I6 should be able to transfer 25.2MB/s to the NL: device.D Years ago, I did see a performance slowdown from 95% to 11% if I didC made a transfer from a FastEthernet to the 10Base Ethernet. (11% ofiC 10Base). The reason where a lot of retransmission in case of switchdC buffer overflow. The switch did send a buffer near full to OpenVMS,gG but OpenVMS did ignore this messages.  Decreasing the PIPELINE QUOTA togE 7490 did speed up the network between the two nodes and also the rest|C of the world. In case of this I don't like to use JUMBO frames. MayuB be you can me explain why JUMBO frames won't make the same problem? (not enaugh buffer space on the switch --> destroyed pakets -->i retransmissions).-B At last, I did made a call to TSC Muenchen. They did say, that theE DEGPA-SA do have a bad performance and that there is a new generationsE of GigabitEthernet adapters. But test within HP did measure with thisd% new one only a performance of 57MB/s.8F I am realy interisted to tune the system, but I do need time to do so.   TIA and best regards R. Wingertl   ------------------------------  % Date: Fri, 29 Nov 2002 11:34:37 +0100S9 From: Jan-Erik =?iso-8859-1?Q?S=F6derholm?= <aaa@aaa.com>^+ Subject: Reason to use MMS instead of MMK ?r' Message-ID: <3DE742BD.C96E75CB@aaa.com>    Hi.d- Is there any reason to use MMS from Compaq/HPy instead of the free MMK ?a- Is there anything MMS can do that MMK can't ?a   Jan-Erik Sderholm.V   ------------------------------  + Date: Fri, 29 Nov 2002 17:18:40 +0000 (UTC)h, From: lewis@mazda.mitre.org (Keith A. Lewis)/ Subject: Re: Reason to use MMS instead of MMK ?t. Message-ID: <as87hg$66f$2@newslocal.mitre.org>   Jan-Erik =?iso-8859-1?Q?S=F6derholm?= <aaa@aaa.com> writes in article <3DE742BD.C96E75CB@aaa.com> dated Fri, 29 Nov 2002 11:34:37 +0100: >Hi.. >Is there any reason to use MMS from Compaq/HP >instead of the free MMK ?  % MMS is contractually supported by HP.n  . >Is there anything MMS can do that MMK can't ?  L Although I have stumbled across differences, they behave identically for the vast majority of projects.  F Compaq added some new features to MMS as recently as Dec 2001 (V3.4) II think.  The .ACTION_STATUS directive would be good if you're using javac,rE for example.  I do not know if the MMK crew is attempting to keep up.k  + --Keith Lewis              klewis$mitre.orgs> The above may not (yet) represent the opinions of my employer.   ------------------------------    Date: 29 Nov 2002 08:05:30 -0800! From: soterro@yahoo.com (Soterro)  Subject: SRM for multiboot= Message-ID: <d5440555.0211290805.6f6d8d63@posting.google.com>    Hello,  = I looked around for information on this stuff, but I found noe5 reference whatsoever... (so I'm quite pessimist here)o  D I have an AlphaServer which boots multiple OS's. Each time I want toB change the OS, I have to go to SRM, change some variables by hand,
 then boot.B Would there be a way to access from the operating system those SRMF resources so the next OS will boot automatically? One could write thenA some scripts/programs in those OS like 'boot_vms' under tru64 andf 'boot64' under vms etc...p  
 Thank you, Sorinu   ------------------------------  % Date: Fri, 29 Nov 2002 08:10:45 -0800s# From: "Tom Linden" <tom@kednos.com>u Subject: RE: SRM for multiboot9 Message-ID: <CIEJLCMNHNNDLLOOGNJIKEHCGCAA.tom@kednos.com>y  - I think only from later version of Tru64, 5.1c   >-----Original Message----- ) >From: Soterro [mailto:soterro@yahoo.com]s( >Sent: Friday, November 29, 2002 8:06 AM >To: Info-VAX@Mvb.Saic.Com >Subject: SRM for multibooti >  >  >Hello,a >n> >I looked around for information on this stuff, but I found no6 >reference whatsoever... (so I'm quite pessimist here) >aE >I have an AlphaServer which boots multiple OS's. Each time I want toiC >change the OS, I have to go to SRM, change some variables by hand,m >then boot.sC >Would there be a way to access from the operating system those SRMFG >resources so the next OS will boot automatically? One could write then B >some scripts/programs in those OS like 'boot_vms' under tru64 and >'boot64' under vms etc... >l >Thank you,  >Sorin >t >---' >Incoming mail is certified Virus Free. ; >Checked by AVG anti-virus system (http://www.grisoft.com).oB >Version: 6.0.410 / Virus Database: 231 - Release Date: 10/31/2002 >  ---i& Outgoing mail is certified Virus Free.: Checked by AVG anti-virus system (http://www.grisoft.com).A Version: 6.0.410 / Virus Database: 231 - Release Date: 10/31/2002m   ------------------------------  % Date: Fri, 29 Nov 2002 09:01:48 +0100 < From: "Martin Vorlaender" <martin.vorlaender@pdv-systeme.de>0 Subject: Re: System uptime - The Uptimes Project4 Message-ID: <as76tm$ojd0b$1@ID-56200.news.dfncis.de>   Larry Kilgallen wrote:, > Mark E. Levy <levy@sysman-inc.com> writes:8 >> Robert Deininger <rdeininger@mindspring.com> wrote...A >>> For the Uptimes Project, this would really give VMS an unfairl >>> advantage. >> >> And your point is?  >>C >> If we're talking about application availability, which is reallyaE >> what's it's all about, then the overall cluster availability wouldhD >> give a truer picture. Who cares about the uptime of an individual4 >> node as long as the cluster is never unavailable? >n> > The people running the competition might see it differently. >aC > I would suggest the cluster version identify the operating system C > as VMScluster, and the node version identify the operating systemw	 > as VMS.h  2 It seems they took a different approach: There's a> "OpenVMSUpClientCluster-1.1" which reports the OS as "OpenVMS"7 (see http://uptimes.wonko.com/stats.php?op=osbyclient).w= But that client is not available for "official" download frome  wonko's site. Where do I get it?   John Vottero wrote: - > "Jan-Erik Sderholm" <aaa@aaa.com> wrote...o@ > > When I looked at that site, I thought that many of the sites. > > with long uptime had very low CPU work (or- > > whatever it sax called). Some even 0.0 %.t > >o; > > I think they should multiply wall-time uptime with thise= > > percentage and get an "effective" working-time. That saysl6 > > more about such things as memory leaks and similar
 > > problems.t >sD > I would tend to agree but I think we should wait until the OpenVMS9 > client reports the load before lobbying for the change.s  3 Anything new on that? There's a discussion going on : (http://uptimes.wonko.com/forums/viewtopic.php?t=16) whose> result seems to be to bin the "load average" values (which are4 specific to Unix anyway) and use "CPU load" instead.   cu,    Martin --I One OS to rule them all       | Martin Vorlaender  | VMS & WNT programmerI7 One OS to find them           | work: mv@pdv-systeme.defI One OS to bring them all      |  http://www.pdv-systeme.de/users/martinv/o> And in the Darkness bind them.| home: martin@radiogaga.harz.de   ------------------------------  # Date: Fri, 29 Nov 2002 09:37:30 GMT03 From: "Jerome" <Jerome.Forissier@removethis.hp.com>f0 Subject: Re: System uptime - The Uptimes Project1 Message-ID: <uzGF9.3$6U1.193690@news.cpqcorp.net>   G "Martin Vorlaender" <martin.vorlaender@pdv-systeme.de> wrote in message . news:as76tm$ojd0b$1@ID-56200.news.dfncis.de... > John Vottero wrote:o/ > > "Jan-Erik Sderholm" <aaa@aaa.com> wrote... B > > > When I looked at that site, I thought that many of the sites0 > > > with long uptime had very low CPU work (or/ > > > whatever it sax called). Some even 0.0 %.  > > >e= > > > I think they should multiply wall-time uptime with thisv? > > > percentage and get an "effective" working-time. That sayse8 > > > more about such things as memory leaks and similar > > > problems.n > >eF > > I would tend to agree but I think we should wait until the OpenVMS; > > client reports the load before lobbying for the change.- >C5 > Anything new on that? There's a discussion going ons< > (http://uptimes.wonko.com/forums/viewtopic.php?t=16) whose@ > result seems to be to bin the "load average" values (which are6 > specific to Unix anyway) and use "CPU load" instead. > J FYI, I've added support for idle & load reporting in OpenVMSUpClient v1.2.J I've just sent it to Ryan Grove for upload to the uptimes server. The code reports a CPU load in %.   -- Jerome   ------------------------------  % Date: Fri, 29 Nov 2002 11:41:15 +0100 . From: "Gorazd Kikelj" <gorazd.kikelj@aster.si>0 Subject: Re: System uptime - The Uptimes Project0 Message-ID: <2sHF9.1071$tQ1.74275@news.siol.net>  
 Hi, Martin  H I just replaced SYI$_BOOTTIME with SYI$_CLUSTER_FTIME and altered Client name stringP (CLIENT_NAME) in UPCLIENT.C.  C I think it's fair to distinguish Cluster uptime from system uptime.0  K and here is the command procedure to start uptime client as detach process.L  . --> cut here <----------------------- lines my7 wrap --------------------------------------------------D $  verify := 'f$verify("No")'  $ status = %x00000001M $ on error then goto error $!++ $!; $!TITLE:          HP OpenVMS Client for teh Uptimes Project  (http://uptimes.wonko.com/)  $! $!VERSION:    1.1H $!4 $!ABSTRACT: Start Uptimes Client as detached process $! $!AUTHOR:     Gorazd Kikelj3 $! $!CREATION DATE: 25-Nov-2002 $! $!MODIFICATION HISTORY:E $!# $! 1.0  - Gorazd Kikelj 25-Nov-2002   $!              Original version# $! 1.1  - Gorazd Kikelj 29-Nov-2002S& $!               Modified for wide use $!
 $!PARAMETERS:R $!; $! P1 = "DETACH"             - run client as detach process_G $! P2 = Exacutable Directory - Directory where UPCLIENT*.EXE is located  (default LOC$EXE)lH $! P3 = Log directory        - Output directory  (default NET$:{DECNET])< $! P4 = UIC                  - Client UIC (default [DECNET])2 $! P5 = Process name         - Process name string $! $!NOTES:@ $! Replace defaults with meaningfull values for your instalation/ $! Defaults are: Executable directory = LOC$EXE 4 $!                     Log directory = NET$:[DECNET]% $!                     UIC = [DECNET]-5 $!                     Process name = "Uptime Client"n $!+ $! Replace !**node... with your nodename(s)  $! $!CALLING SEQUENCE:  $!B $! @UPCLIENT ["DETACH"] [exe_dir_spec] [log_dir_spec] [unpriv_uic] [process_name] $! $!-- $ & $  client_arch = F$GETSYI("ARCH_NAME")F $  client_exed = f$elem(f$edit(p2,"TRIM").nes."","!","loc$exe:!''p2'")L $  client_cmd = "$" + ''client_exed' + "UPCLIENT-" + ''client_arch' + ".EXE"I $ client_dir = f$elem(f$edit(p3,"TRIM").nes."","!","net$:[decnet]!''p3'"):) $ client_com = f$environment("PROCEDURE")-D $ client_uic = f$elem(f$edit(p4,"TRIM").nes."","!","[DECNET]!''p4'")> $ client_process = f$elem(f$edit(p5,"TRIM").nes."","!","Uptime Client!''p5'")! $  write sys$output ''client_cmd'7 $  upclient :== 'client_cmd5 $5 $49 $! spawn /nowait /input=nl: /process=upclient upclient -ht6 $ if (f$mode().eqs."INTERACTIVE").or.(p1.eqs."DETACH") $ then( $   run/detach sys$system:loginout.exe -        /input='client_com' -,        /error='client_dir'upclient_err.log -)        /output='client_dir'upclient.log -.        /uic='client_uic' -"        /proces="''client_process'"
 $   goto exit  $ endif7 $v $  on warning then goto errorM, $  node = f$edit(f$getsy("NODENAME"),"TRIM") $  goto 'node' $W" $!**node1**: key = "your key here" $  goto run_proces $t) $!**node n**: key = "your other key here"e $  goto run_proces $u $run_proces: $!***I8 $! Customize the client command line here. For instance: $! $! upclient -k <key> $! upclient -k <key> -p http% $! upclient -k <key> -p http -x proxys $! $!***a $   upclient -k 'key' -p httpV $  $error:a $        status = $statusu $exit:- $ exit 0*'f$verify(verify)'+%x10000000+statusa   ------------------------------  % Date: Fri, 29 Nov 2002 11:01:11 +0000p( From: Nic Clews <sendspamhere@127.0.0.1>+ Subject: Re: Wanted: Second hand Alpha DS10a) Message-ID: <3DE748F7.1F6C5519@127.0.0.1>m   Mike Knowles wrote:d  H > Can anybody point me in the right direction for finding second hand or? > refurbished Compaq Alpha kit? I am based in the UK. We want ayB > redundant server for restoration and verification of DLT backups  , http://www.nexpress.co.uk competent and able  G (To me their website seems to have gone bonkers too, try 01509 501100 )n   -- o? Regards, Nic Clews a.k.a. Mr. CP Charges, CSC Computer Sciencesd nclews at csc dot com    ------------------------------   End of INFO-VAX 2002.660 ************************tachments is confidential and may be  F legally privileged. If you have received it in error please advise the@ sender immediately by return email and then delete it from your  system. ? All electronic communications with the Company may be monitored A in accordance with the UK Regulation of Investigatory Powers Act, , Lawful Business Practice Regul ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    	ɪ    
ɪ    ɪ    ɪ    
ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ     ɪ    !ɪ    "ɪ    #ɪ    $ɪ    %ɪ    &ɪ    'ɪ    (ɪ    )ɪ    *ɪ    +ɪ    ,ɪ    -ɪ    .ɪ    /ɪ    0ɪ    1ɪ    2ɪ    3ɪ    4ɪ    5ɪ    6ɪ    7ɪ    8ɪ    9ɪ    :ɪ    ;ɪ    <ɪ    =ɪ    >ɪ    ?ɪ    @ɪ    Aɪ    Bɪ    Cɪ    Dɪ    Eɪ    Fɪ    Gɪ    Hɪ    Iɪ    Jɪ    Kɪ    Lɪ    Mɪ    Nɪ    Oɪ    Pɪ    Qɪ    Rɪ    Sɪ    Tɪ    Uɪ    Vɪ    Wɪ    Xɪ    Yɪ    Zɪ    [ɪ    \ɪ    ]ɪ    ^ɪ    _ɪ    `ɪ    aɪ    bɪ    cɪ    dɪ    eɪ    fɪ    gɪ    hɪ    iɪ    jɪ    kɪ    lɪ    mɪ    nɪ    oɪ    pɪ    qɪ    rɪ    sɪ    tɪ    uɪ    vɪ    wɪ    xɪ    yɪ    zɪ    {ɪ    |ɪ    }ɪ    ~ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ½ɪ    ýɪ    Ľɪ    Žɪ    ƽɪ    ǽɪ    Ƚɪ    ɽɪ    ʽɪ    ˽ɪ    ̽ɪ    ͽɪ    νɪ    Ͻɪ    нɪ    ѽɪ    ҽɪ    ӽɪ    Խɪ    սɪ    ֽɪ    ׽ɪ    ؽɪ    ٽɪ    ڽɪ    ۽ɪ    ܽɪ    ݽɪ    ޽ɪ    ߽ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    ɪ    