1 INFO-VAX	Tue, 27 Jul 2004	Volume 2004 : Issue 413       Contents:, Re: =?ISO-8859-1?Q?OT:_Sun=B4s_AMD_Opteron?=( Re: =?iso-8859-1?Q?Sun=B4s?= AMD Opteron( Re: =?ISO-8859-1?Q?Sun=B4s_AMD_Opteron?=1 Re: Attaching tk50 to simh emulator running on pc  ES40 NIC's inactive state  Re: ES40 NIC's inactive state  Re: Getting stuck * Help! VAXServer 4000/500 Self-Test Failure# Re: HP World, OpenVMS Bootcamp, etc # Re: HP World, OpenVMS Bootcamp, etc # Re: HP World, OpenVMS Bootcamp, etc " Konvertierung OpenVMS 6.2 -- Linux& Re: Konvertierung OpenVMS 6.2 -- Linux& Re: Konvertierung OpenVMS 6.2 -- Linux& Re: Konvertierung OpenVMS 6.2 -- Linux& Re: Konvertierung OpenVMS 6.2 -- Linux Re: Microvax 3100-30 Re: Microvax 3100-30. Re: MSL5000 tape Library in a hetrogeneous SAN3 Need to recreate production system on backup system 7 RE: Need to recreate production system on backup system 7 Re: Need to recreate production system on backup system 7 Re: Need to recreate production system on backup system 7 Re: Need to recreate production system on backup system  Re: OpenVMS Backup Failure" Re: OpenVMS restarts while install OpenVMS/Sybase certification  Re: OpenVMS/Sybase certification  Re: OpenVMS/Sybase certification, Re: OT: =?iso-8859-1?Q?Sun=B4s?= AMD Opteron Re: Rdb 7.0.5 on VMS 7.3-2 RE: Solaris to Itanium...  Re: Spam storage on VMS Mail Re: Spam storage on VMS Mail Re: Spam storage on VMS Mail Re: Structure of SMTP Receiver1 Re: Turn-key OpenVMS E-mail, web server solution? O Re: [TCPware, V73_MGMTAGENTS] Does anyone have a working installation running ? O Re: [TCPware, V73_MGMTAGENTS] Does anyone have a working installation running ? O Re: [TCPware, V73_MGMTAGENTS] Does anyone have a working installation running ? O Re: [TCPware, V73_MGMTAGENTS] Does anyone have a working installation running ?   F ----------------------------------------------------------------------    Date: 27 Jul 2004 03:52:13 -0700. From: fabiopenvms@yahoo.com.br (Fabio Cardoso)5 Subject: Re: =?ISO-8859-1?Q?OT:_Sun=B4s_AMD_Opteron?= = Message-ID: <f30679fb.0407270252.3e1f341a@posting.google.com>   a JF Mezei <jfmezei.spamnot@teksavvy.com> wrote in message news:<410538EE.A89394EA@teksavvy.com>...  > Fabio Cardoso wrote:> > > If we have a massive production of 64 bits processors onlyC > > Intel and AMD can do that !  It will become a bi-polar industry 
 > > (choice).  > > F > > In fact Apple would try an Itanium or Opteron in the near future ! >  > P > When OS-X came out, there were rumours that Apple would start to make it cross- > platform and make it available on the 8086.   C I am saying to run OS-X (for Integrity) like a Scientific Computing 5 Operating System. Like those X-Serve are used for....   P  > However, at this point in time, you should remind yourself that IA64 is stillN > a very low volume proprietary chip and that Power has a greater market shareK > than IA64. So your statement above should be pitting Power against AMD.    > M > Now, when the 64 bit 8086 comes out from Intel to compete against AMD's,  I L > suspect that the success story will be for personal computers purchased byO > parents on behalf of their teenagers who want to play the latest and greatest S > games.  The server market is a much lower volume than the consumer retail market.     F So its time to rethink the model to develop servers. May be in slicesE or bladening them instead of a huge piece of metal. So you can buy a  J specific number of parts to mount your server. Example: 1 blade, 4 blades,K 16 blades, 1024 blades to mount a supercomputer. Connected through specific G inter-"blade" switches. Etc...Would be a much more generic way to mount F systems. So you can buy HP, Dell, Sun "components" and combine them in9 a "system". Its a kind of "distributed computing model".     @ > It is the comsumer market that drives the "industry standard". > O > And I suspect that it won't be long before the "industry standard" will be 64 O > bit 8086s and that Intel will stop making 32 bit ones in a matter of a couple  > of years.    ------------------------------  % Date: Tue, 27 Jul 2004 13:12:14 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> 1 Subject: Re: =?iso-8859-1?Q?Sun=B4s?= AMD Opteron , Message-ID: <41068CEB.72B727A6@teksavvy.com>   Bill Todd wrote:C > > So what are HP and Intel doing with these Itanium servers now ?  > I > Not all that much, according to publicly-available shipment statistics.     H In all fairness, IA64 is showing tremendous growth this year. (Just likeL Digital was at one point the fastest growing PC manufacturer from the simpleB math that growing from 0 to 10 gives you an infinite growth rate).  N I think that the telling number is that HP's IA64 revenus account for only 16%K of total high end server sales, which means that Alpha, PA-Risc, Mips still  outsel IA64 by a large margin.  J The problem with IA64 is that Intel and HP have bragged about so much thatL there is no way that they can meet the expectations they had originally set,J so the chip will always look like a loser. Had Intel delivered the chip onI time, or only begun to brag about it when Merced was unveiled (with clear H affirmnations that merced was just the "beta", thus no surprise that itsK performance would be dismal), perhaps the reception of IA64 might have been  more positive.  N And had Intel/HP begun to brag about IA64 only once it became a relaity, AlphaG would not have been killed prematurely and EV7 would probably have been K delivered earlier at its designed-for speed which would have made Merced or  McKinley look fairly shabby.  K So it was a lose lose situation for Intel/HP: they needed to pre-empt chips K like Alpha and Power by bragging a lot about the IA64 vapourware and how it N would take over the world, but in doing so, they set themselves up for failure3 because that chip won't achieve its promised goals.   I And had they been more realistic about IA64, the chip wouldn't have had a E chance in hell because Alpha wouldn't have been preemtively murdered.    ------------------------------    Date: 27 Jul 2004 03:45:47 -0700. From: fabiopenvms@yahoo.com.br (Fabio Cardoso)1 Subject: Re: =?ISO-8859-1?Q?Sun=B4s_AMD_Opteron?= = Message-ID: <f30679fb.0407270245.4a97b9ad@posting.google.com>   d "Bill Todd" <billtodd@metrocast.net> wrote in message news:<3NednVhzTPOw7JjcRVn-pg@metrocast.net>...= > "Fabio Cardoso" <fabiopenvms@yahoo.com.br> wrote in message 8 > news:f30679fb.0407260514.32e3e8a@posting.google.com... >  > ...  > F > > A 64 bits processor is a 64 bits processor. What is the difference( > > of benchmark between them ? 1%, 2% ? >  > Well, since you asked... > K > The new 4-processor Sun Opteron systems out-performs the best 4-processor G > Itanic by about 25% on SPECweb99_SSL (and is likely considerably less 
 > expensive).  > K > The new 16-processor IBM POWER5 system out-performs the best 16-processor K > Itanic by about 160% (yes, a factor of 2.6x) on TPC-C, and at a far lower 4 > price than anything else in its performance class. > L > Perhaps you need to develop a somewhat more detailed acquaintance with the > industry.   I Really ? So what are HP and Intel doing with these Itanium servers now ?  B By the way... in fact I never trusted in these numbers. You cannotD confirm if the numbers are really related to the processor itself orC the "server architecture" : bus, cache, memory etc ...I  dont know E if there is a "overall measurement" explaining the percentage of each  component of the computer.     > - bill   ------------------------------  % Date: Tue, 27 Jul 2004 11:53:47 +0200 - From: "Martin Vorlaender" <mv@pdv-systeme.de> : Subject: Re: Attaching tk50 to simh emulator running on pc* Message-ID: <2mmn1cFoou42U1@uni-berlin.de>  3 "Robert A.M. van Lopik" <lopik@mail.telepac.pt> ... / > "Roy C Wood" <rcwood0000@yahoo.com> wrote ... B > > I know the tk50 is scsi device and in theory can be hung off aB > >  pc. If I install the hardware, can a pc running vax-vms under= > >  simh actually read tapes? If it can, how do I set it up. 1 > >  I already have vax-vms 7.3 running on my pc.  > F > AFAIK simh doesn't allow to attach ANY REAL devices. All controllersF > are simulated and the attached media are simply disk files. The onlyF > exception being the ethernet interface and, to some extent, the DZ11C > terminal interface which is emulated through a TELNET connection.   > Another exception is the CD-ROM device - but only under Linux.  D That said, I'm not sure whether a tape drive would work under Linux.   cu,    Martin --  ;   OpenVMS DCL        | Martin Vorlaender  |  OpenVMS rules! .                      | work: mv@pdv-systeme.deA   The Original .COM  |   http://www.pdv-systeme.de/users/martinv/ 5                      | home: martin@radiogaga.harz.de    ------------------------------    Date: 27 Jul 2004 03:59:08 -0700, From: jefflanka@volcanomail.com (Jeff Lanka)" Subject: ES40 NIC's inactive state= Message-ID: <8201ded0.0407270259.664fe993@posting.google.com>   A I have a number of ES40's running 7.3-1 or 2.  All of them have 2  NIC's EWA-0 and EWA-1   E We are trying to develop a com file which will switch from one NIC to D the other (DECnet phase 4) in the case of a problem with the switch.   We are trying the command:   ncp>set circuit ewa-0 state off  ncp>set line ewa-0 state off ncp>set line ewa-1 state on  ncp>set circuit ewa-1 state on  A on most of the machines we will get the error INVADDR and the new A (EWA-1) circuit will be "synchronizing" forever.  It won't switch D unless we reboot.  These machines have steady green (link) lights onD both NIC's and the data (amber) lights are both blinking (indicating data)   C But on 2 of the machines, the 2nd NIC comes up with a slow blinking A green light and no amber.  When we try to switch (to EWA-1) those E lights change to solid green and flashing amber.  Then when we switch 4 back to EWA-0 it goes back to slowly flashing green.  E Also, if we define EWA-1 active in the permanent database and reboot, C both NIC's have steady green and flashing amber and we can't switch   (like the majority of machines).  C Here's the question:  What is the configuration option which causes B the inactive NIC to come up in the slowly blinking green state?  I1 couldn't find anything in LANCP or SYSGEN or NCP.    Thanks in advance.   ------------------------------    Date: 27 Jul 2004 08:19:10 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) & Subject: Re: ES40 NIC's inactive state3 Message-ID: <$mmhiAqkfySS@eisner.encompasserve.org>   l In article <8201ded0.0407270259.664fe993@posting.google.com>, jefflanka@volcanomail.com (Jeff Lanka) writes:C > I have a number of ES40's running 7.3-1 or 2.  All of them have 2  > NIC's EWA-0 and EWA-1  > G > We are trying to develop a com file which will switch from one NIC to F > the other (DECnet phase 4) in the case of a problem with the switch. >  > We are trying the command: > ! > ncp>set circuit ewa-0 state off  > ncp>set line ewa-0 state off > ncp>set line ewa-1 state on   > ncp>set circuit ewa-1 state on > C > on most of the machines we will get the error INVADDR and the new C > (EWA-1) circuit will be "synchronizing" forever.  It won't switch F > unless we reboot.  These machines have steady green (link) lights onF > both NIC's and the data (amber) lights are both blinking (indicating > data)   I    If you aleady have other protocols, such as TCP/IP, running on the NIC D    DECnet probably won't be able to start because it needs to change    the MAC address.    ------------------------------  # Date: Tue, 27 Jul 2004 13:57:16 GMT 9 From: "Fred Kleinsorge" <my-last-name@stardotzko.dec.com>  Subject: Re: Getting stuck2 Message-ID: <0btNc.6636$a82.5150@news.cpqcorp.net>  : "Keith A. Lewis" <lewis@spyder.mitre.org> wrote in message( news:ce330c$q79$1@newslocal.mitre.org...G > "Fred Kleinsorge" <my-last-name@stardotzko.dec.com> writes in article K <Jz9Lc.6123$cc1.4041@news.cpqcorp.net> dated Tue, 20 Jul 2004 14:00:09 GMT:  > H > The real problem is that the X application running on the OpenVMS side can't D > send the overlay to any display unless I retro-install DECW 1.2-5. >   J Hmmm.  Interesting.  "Overlay Planes" in this particular case, is just the@ presence of a specific pseudocolor visual, and hardware with theB capability...  mostly this is just server side stuff.  It would beI interesting to know what changed.  The changes between those two versions % are for the most part just bug fixes.    ------------------------------  # Date: Tue, 27 Jul 2004 17:24:04 GMT ' From: Rick Dyson <rick-dyson@uiowa.edu> 3 Subject: Help! VAXServer 4000/500 Self-Test Failure 1 Message-ID: <I1Ito6.JJs@sysadm.physics.uiowa.edu>   N I am trying to help someone with a self-test failure at powerup on this model  VAX.  It reports:    KA680-A v4.1  VMB 2.12 66...  ... clipped countdown ...  ..04..  K The "03.." is not presented as well as the string "Tests Complete" and then D handing over the control from the console to the O/S does not occur.  J I am told there was a power failure earlier in the day for his system.  ItI is sitting behind a huge, ancient DIGITAL(!) power conditioning system (I I have never seen one of these before).  The users claim the VAX should not J have seen any power surges or spikes because the conditioner is present... Famous last words!  N Of course, these folks don't have a maint contract and I am not very good with VAXen...  N Does anyone know where I can find a list of what the self-tests are doing?  ToL see if we can figure out what is failing the self-test and then maybe see ifM we can swap out the part?  I am told they have lots of spares (though I don't M know if they have the main board or not...).  I figure we'll find out if they  have the one we need, right?   Thanks, 
 Rick Dyson   ------------------------------  # Date: Tue, 27 Jul 2004 13:46:42 GMT & From: John Reagan <john.reagan@hp.com>, Subject: Re: HP World, OpenVMS Bootcamp, etc2 Message-ID: <61tNc.6635$A52.2748@news.cpqcorp.net>   David J Dachtera wrote: H > As the so-called head of the ersatz VMSUS, I would consider sponsoringI > the bootcamp at the hotel near Nashua on a break-even basis if hp would D > commit OpenVMS engineers and resources to the bootcamp, and foregoG > profits on the attendees in the form registration fees over and above G > the cost of the meetings rooms, AV gear, etc. That might make it more  > affordable, if only slightly.  >  > D.J.D.  I I'm confused.  Are you implying that Sue runs the bootcamp for a profit?  G   I think we certainly loose money on the operation.  The bootcamp was  0 never intended to make money for HP and OpenVMS.  I HP already commits LOTS of resources and engineers to the bootcamp.  Ask   anybody who attended.   K Who people like less food or lower services in exchange for a cheaper rate?      --   John Reagan / HP Pascal/{A|I}MACRO for OpenVMS Project Leader  Hewlett-Packard Company    ------------------------------  % Date: Tue, 27 Jul 2004 12:51:23 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> , Subject: Re: HP World, OpenVMS Bootcamp, etc, Message-ID: <41068809.6DD07BEB@teksavvy.com>   John Reagan wrote:J > I'm confused.  Are you implying that Sue runs the bootcamp for a profit?H >   I think we certainly loose money on the operation.  The bootcamp was2 > never intended to make money for HP and OpenVMS.    D It depends on what is included in the costs for accounting purposes.  F You see, in a DECUS mindset, because it is/was operated as a volunteerK organisation, none of the organisation manhours counted as a "cost".  You'd K have the rental of rooms (and generally, if you purchased meals, those were K faily cheap at a hotel), meals, and whatever gifts/trinkets you handed out.   I In Sue's case would the engineer's times be charged to the event ? Or are  there there "for free" ?  G Would the time spent organising the even be charged to the event, or be L considered "free" ?  That would make a huge difference in the costs and thus" to the proce charged to attendees.   ------------------------------  # Date: Tue, 27 Jul 2004 17:18:48 GMT / From: brooks@cuebid.zko.dec.nospam (Rob Brooks) , Subject: Re: HP World, OpenVMS Bootcamp, etc- Message-ID: <LIatspcCeuvf@cuebid.zko.dec.com>   ( John Reagan <john.reagan@hp.com> writes: > David J Dachtera wrote: I >> As the so-called head of the ersatz VMSUS, I would consider sponsoring J >> the bootcamp at the hotel near Nashua on a break-even basis if hp wouldE >> commit OpenVMS engineers and resources to the bootcamp, and forego H >> profits on the attendees in the form registration fees over and aboveH >> the cost of the meetings rooms, AV gear, etc. That might make it more  >> affordable, if only slightly. > K > I'm confused.  Are you implying that Sue runs the bootcamp for a profit?  I >   I think we certainly loose money on the operation.  The bootcamp was  2 > never intended to make money for HP and OpenVMS. > K > HP already commits LOTS of resources and engineers to the bootcamp.  Ask   > anybody who attended.  > M > Who people like less food or lower services in exchange for a cheaper rate?   F . . . And for those who are watching a budget, there is a Red Roof InnC located a half-mile from the Sheraton (where the bootcamp is held). I Rooms can be had for about $45 a night.  I've spent many a night there --  it's clean.   @ Since Southwest Airlines flies into MHT (Manchester, NH -- aboutH 30 minutes from Nashua for a slow driver), round trip prices from places@ like MDW can be had for about $155 with enough advance planning.   --    M Rob Brooks    VMS Engineering -- I/O Exec Group     brooks!cuebid.zko.dec.com    ------------------------------  % Date: Tue, 27 Jul 2004 12:13:58 +0200 # From: Jan Wilm <jan.wilm@ch.tum.de> + Subject: Konvertierung OpenVMS 6.2 -- Linux 0 Message-ID: <ce57r1$g3c$1@wsc10.lrz-muenchen.de>   Hi all,   : I have a dec3000 with tons of scasi harddrives to attach. I Now i am looking for a way to get the files onto a linux /windows system.   " The DEC is connected to a network.  K Is there a way to transfer the files and directories? Doing ftp is possible F but the problem is that it is a deep directory structure and i want to conserve it.  J My try was using nfs, but i could not mount the nfs directory that i opend> on the dec with bind to a linux system, it refused to read it.  I Anyone knows a way? Either archiving the files so i can unarchive it (can C transfer single files), transfering all files with dir structure or # mounting the dec drives to linux ??  ciao jan   ------------------------------    Date: 27 Jul 2004 12:17:38 +0200C From: vaxinf@chclu.chemie.uni-konstanz.de (Eberhard Heuser-Hofmann) / Subject: Re: Konvertierung OpenVMS 6.2 -- Linux - Message-ID: <41062bc2$1@news.uni-konstanz.de>   9 In article <ce57r1$g3c$1@wsc10.lrz-muenchen.de>, Jan Wilm  <jan.wilm@ch.tum.de> writes: >Hi all, > ; >I have a dec3000 with tons of scasi harddrives to attach.  J >Now i am looking for a way to get the files onto a linux /windows system. > # >The DEC is connected to a network.  > C >Is there a way to transfer the files and directories? Doing ftp is 	 >possible G >but the problem is that it is a deep directory structure and i want to 
 >conserve it.  > E >My try was using nfs, but i could not mount the nfs directory that i  >opend? >on the dec with bind to a linux system, it refused to read it.  > J >Anyone knows a way? Either archiving the files so i can unarchive it (canD >transfer single files), transfering all files with dir structure or$ >mounting the dec drives to linux ??  > 1. zip the file incl. directory trees with "-V" under OpenVMS.< 2. transfer the zip-files or burn them as an ISO-FS on a DVD   Eberhard   ------------------------------    Date: 27 Jul 2004 08:16:04 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) / Subject: Re: Konvertierung OpenVMS 6.2 -- Linux 3 Message-ID: <ki7X3xN9qMWU@eisner.encompasserve.org>   V In article <ce57r1$g3c$1@wsc10.lrz-muenchen.de>, Jan Wilm <jan.wilm@ch.tum.de> writes:	 > Hi all,  > < > I have a dec3000 with tons of scasi harddrives to attach. K > Now i am looking for a way to get the files onto a linux /windows system.  > $ > The DEC is connected to a network.  A    It's going to depend on what kind of software your DEC 3000 is <    running.  Since you post to comp.os.vms, I'll assume VMS.  B    Probably the easiest thing for you to do is get one of the freeD    tar programs for VMS, tar up the directory tree, transfer the tarF    file over the network, and untar it there.  You could even compress4    it first with a free copy of gzip if you want to.  E    If you tell us what's going on:  what software, what version, what D    error messages, ... (see the FAQ) we could help you with your NFSB    problem.  Be aware that there is more than one popular IP stackF    in use on VMS, so you'll need to tell us which one and what version    for that, too.    ------------------------------  % Date: Tue, 27 Jul 2004 07:52:44 -0700 + From: "Barry Treahy, Jr." <Treahy@MMaz.com> / Subject: Re: Konvertierung OpenVMS 6.2 -- Linux ' Message-ID: <41066C3C.7060902@MMaz.com>    Jan Wilm wrote:    >Hi all, > ; >I have a dec3000 with tons of scasi harddrives to attach.  J >Now i am looking for a way to get the files onto a linux /windows system. > # >The DEC is connected to a network.  > L >Is there a way to transfer the files and directories? Doing ftp is possibleG >but the problem is that it is a deep directory structure and i want to 
 >conserve it.  > K >My try was using nfs, but i could not mount the nfs directory that i opend ? >on the dec with bind to a linux system, it refused to read it.  > J >Anyone knows a way? Either archiving the files so i can unarchive it (canD >transfer single files), transfering all files with dir structure or$ >mounting the dec drives to linux ?? >    > F Just a word of caution, if the files are not plain vanilla text files @ that you are trying to preserve, you will most likely end of up A directories of soon to be useless data because RMS file formats,  " especially INDEXED, will be lost.   C That said, if you want to do it free, getting a copy of VMS tar is  I probably your best bet.  If you just want to 'archive' the data, I would  C put it into a VMS Backup Saveset and just move that saveset across  F because there are both commercial, and free, readers that can extract H individual files from the savesets and if at a later date you just have D the data, in its original VMS form, you'll still have the option of K pushing that saveset back to a VMS system and extracting the data intact...    Barry    --    > Barry Treahy, Jr                       E-mail: Treahy@MMaz.com> Midwest Microwave                          Phone: 480/314-1320> Vice President & CIO                         FAX: 480/661-7028                            ------------------------------  # Date: Tue, 27 Jul 2004 16:03:11 GMT ! From: Nigel Barker <nigel@hp.com> / Subject: Re: Konvertierung OpenVMS 6.2 -- Linux 8 Message-ID: <1cscg01qn3567kjui49hd4npeedmmin3k2@4ax.com>  H On Tue, 27 Jul 2004 12:13:58 +0200, Jan Wilm <jan.wilm@ch.tum.de> wrote:   >Hi all, > ; >I have a dec3000 with tons of scasi harddrives to attach.  J >Now i am looking for a way to get the files onto a linux /windows system. > # >The DEC is connected to a network.  > L >Is there a way to transfer the files and directories? Doing ftp is possibleG >but the problem is that it is a deep directory structure and i want to 
 >conserve it.  > K >My try was using nfs, but i could not mount the nfs directory that i opend ? >on the dec with bind to a linux system, it refused to read it.  > J >Anyone knows a way? Either archiving the files so i can unarchive it (canD >transfer single files), transfering all files with dir structure or$ >mounting the dec drives to linux ??	 >ciao jan   N The best way of preserving the directory structure is to use BACKUP & create aO save set of the required files. Then use zip "-V" to create a zip file. Now ftp L the zipped save set to the other machine & use unzip to extract the save setM from the zip archive. Now use BACKUP to restore the save set. Zip & unzip are  available on the freeware CD.    -- Nigel Barker Live from the sunny Cote d'Azur    ------------------------------  % Date: Tue, 27 Jul 2004 12:45:52 -0500 / From: Chris Scheers <chris@applied-synergy.com>  Subject: Re: Microvax 3100-30 3 Message-ID: <410694D0.9A3697B6@applied-synergy.com>    JF Mezei wrote:  >  > Antonio Carlini wrote:E > > Model 38. Late in its product cycle, the VAXstation 3100 Model 30 A > > mainboard was switched to use the Model 38 one but with a few @ > > differences (e.g. crystal) to make it operate identically to > > the "original" Model 30. > P > Now that the cat is out of the bag, do you mind telling us how one can tell if6 > a VS3100-30 is a vintage or a model-38-in-disguise ? > N > Can't wait to go to Radio Shack and buy a faster crystal and maket my -30 to > magically become a -38 :-)  H It's been a while since I've done this, but, IIRC, you change a 44.44MHzE crystal to a 66.66MHz one.  You also need to change one jumper on the  motherboard (wait state?).  H There aren't that many jumpers on the board.  So just change them one at9 a time until you find the one that makes it run reliably.   H I wasn't sure if my motherboard would take the change or not, so I put aG socket in place of the crystal to make it easy to swap back.  (I didn't 	 need to.)   
 Good luck!  G ----------------------------------------------------------------------- $ Chris Scheers, Applied Synergy, Inc.  C Voice: 817-237-3360            Internet: chris@applied-synergy.com     Fax: 817-237-3074    ------------------------------  % Date: Tue, 27 Jul 2004 12:42:07 -0500 / From: Chris Scheers <chris@applied-synergy.com>  Subject: Re: Microvax 3100-30 3 Message-ID: <410693EF.9DCA60D2@applied-synergy.com>    Antonio Carlini wrote: > " > Peter 'EPLAN' LANGSTOEGER wrote:` > > In article <41044453.371666E2@teksavvy.com>, JF Mezei <jfmezei.spamnot@teksavvy.com> writes: > > O > >>Also just a comment: the Microvax 3100 is a totally differeant beast from a Q > >>Vaxstation 3100. (different chip, different motherboard, different connectors  > >>in the back).  > >  > > + > > Make this, it depends what model it is.  >  > So far, so good. > F > > A MicroVAX 3100 Model 10 is the same as a VAXstation 3000 Model 30R > > and a MicroVAX 3100 Model 20 is the same as a VAXstation 3000 Model 40 (IIRC). > ; > The MicroVAX 3100 Model 10 and MicroVAX 3100 Model 20 are = > identical apart from the size of the enclosure. Neither use C > the same mainboard as any VAXstation 3100. The (somewhat earlier) ; > MicroVAX 2000 and VAXstation 2000 are identical (a jumper   > determines the "personality"). > = > There is no VAXstation 3100 Model 40 - although there was a C > Model 38. Late in its product cycle, the VAXstation 3100 Model 30 ? > mainboard was switched to use the Model 38 one but with a few > > differences (e.g. crystal) to make it operate identically to > the "original" Model 30. > 8 > The MicroVAX 3100 Model 30 (which started this off) is6 > based on the SOC chip (as is the VAXstation VLC, but4 > they bear no other resemblance). The MicroVAX 31001 > Model 40 is identical to the Model 30 but lives  > in a bigger enclosure. > 	 > Antonio     E Actually, there is indeed a VAXstation 3100 Model 40.  It is the "big F box" Model 30.  Likewise, the Model 48 is the "big box" Model 38.  TheE 30 and 40 use the same electronics.  The only difference is more room ) for drives.  Likewise with the 38 and 48.   G ----------------------------------------------------------------------- $ Chris Scheers, Applied Synergy, Inc.  C Voice: 817-237-3360            Internet: chris@applied-synergy.com     Fax: 817-237-3074    ------------------------------  # Date: Tue, 27 Jul 2004 17:16:22 GMT ' From: Rick Dyson <rick-dyson@uiowa.edu> 7 Subject: Re: MSL5000 tape Library in a hetrogeneous SAN 1 Message-ID: <I1ItBB.I5s@sysadm.physics.uiowa.edu>    Chris C. wrote: H > We have a heterogeneous SAN environment with the following components: >  > 	Dual redundant HSG80's  > 	2 x SAN Switch 2/8 " > 	2 x clustered ES40's VMS V7.3-1* > 	3 x "Windows 2003 Server" Intel servers > 	1 x MSL5026 Tape Library   A 	What is the tape bridge you are using?  I am assuming the MSL is I in the SAN?  Maybe an MDR or the newer NSRs?  I have an ES40 with OpenVMS 6 v7.3 a SAN Switch 8EL, an HSG80, an MDR and a MSL5026.  B 	The firmware on the SAN is v2.6.1x and the MDR has something likeE 1175.  The HSG80 is running ACS v8.7F-4.  If you are keeping the ES40 G F/W updated, it should be updating the FC HBA card, too.  These devices E all need to be in sync with each other.  I am always told this when I F call in a SAN issue to support. :)  The MSL's current F/W is v4.23 and the SDLT1 should be at v75...   @ 	I have an MSL5026 in a SAN with a WinPC server.  I believe theyE are now using Veritas backup solution there.  We have just started to E coordinate shared use of the MSL.  It has never been a problem in the H past.  I backup daily on the OpenVMS side and have never had any issues. The SAN is flat and non-zoned.  G > The issue we have is that in order to get the VMS backups to function D > correctly we had to zone the VMS and the Windows separately on theH > switches and add the tape library to the VMS zone, therefore making it" > unavailable to the Windows zone.E > Prior to zoning, the VMS backups would keep going into mount verify H > state then falling over. At the time I came to the conclusion that theD > tape drives were receiving a "SCSI reset" (?) originating from the4 > Wintel Boxes, so I zoned and alleviated the issue. > E > The Backups on the Wintel Servers are currently handled by 2 direct D > attach SCSI LTO drives using the Veritas BackupEXEC V9.1 software.F > However due to the expanding nature of this environment the use of a= > tape Library would now be beneficial, so my query is; do I:  > H > a.)     Purchase a second MSL5026 tape library and place it on the SAN >         in the Windows zone E > b.)	Find a way to use the existing Library on both Windows and VMS  ; >          I have capacity (tape slots) and a time window.   
 	Choose b. :)   - > c.)	None of the above  suggestions please?   I 	I guess you have to figure out what is causing the negative interaction. # I realize that is the hard part. :)    Rick   ------------------------------    Date: 27 Jul 2004 06:53:33 -0700' From: rcwood0000@yahoo.com (Roy C Wood) < Subject: Need to recreate production system on backup system< Message-ID: <22e43e71.0407270553.f0975c9@posting.google.com>  E We plan on getting a backup system to our vax. I need to recreate the E production environment on the new vax. I have backup images (includes E system disk) from vax1 on tape that I need to port to vax2. Whats the E simplest way to recreate the mirror environment on the new vax? Oh, I B can't create a cluster because nobody wants the the production vax changed in any way.      Thanks   RC Wood...   ------------------------------  % Date: Tue, 27 Jul 2004 12:19:18 -0400 $ From: Mike Duffy <Duffy@process.com>@ Subject: RE: Need to recreate production system on backup systemJ Message-ID: <63D30D6E10CFD11190A90000F805FE8605AED19D@lespaul.process.com>   > -----Original Message-----: > From: rcwood0000@yahoo.com [mailto:rcwood0000@yahoo.com]& > Sent: Tuesday, July 27, 2004 9:54 AM > To: Info-VAX@Mvb.Saic.Com > > Subject: Need to recreate production system on backup system >  > G > We plan on getting a backup system to our vax. I need to recreate the G > production environment on the new vax. I have backup images (includes G > system disk) from vax1 on tape that I need to port to vax2. Whats the G > simplest way to recreate the mirror environment on the new vax? Oh, I D > can't create a cluster because nobody wants the the production vax > changed in any way.  > 
 >   Thanks >   RC Wood... >   1 That's kind of a broad question, but I'll try it.   9 If the production VAX has been configured with the proper < level of abstraction (explained below), it should be fairly  easy to make VMS itself happy.  7 I'm assuming your configuration is sufficiently complex 5 to start by restoring your production backups to your 7 new machine.  If you have a simple environment, you may 9 actually have less to do by configuring your new machine  = from scratch, and you could avoid importing any long-obsolete 3 parts of your old configuration this way.  But youru2 intention is probably to make the backup system as6 close as possibly to production to make the transition as smooth as possible.  7 Make sure the backup system is configured with at leasti8 as much disk space as the original.  You can restore the6 backups with less, but you might be asking for trouble7 when some application temporarily needs a large amount M7 of space you take for granted on the production system.e  ? Isolate the new machine from the network, restore your backups,n  and deal with software licenses.  9 Then, you should only need to change a few things to makei> VMS go, like the nodename, DECnet address, and your IP address9 and host name and probably a few other things that escape  me at the moment.   ? It will probably be appropriate to tune performance parameters  G separately, unless the machines are nearly identical hardware models.  eC Be sure you're not advertising any LAT services you don't want to, u: or network printers, etc. (although you should be ready to8 do so when the day comes; have a procedure ready to go.): At this point, it's probably ok to hook it to the network.  ; Now comes the test to determine whether the original systemi9 was properly configured.  Do you have command procedures  8 or application parameters that refer to specific device ? names instead of logicals (e.g. DKA0: instead of SYS$SYSDEVICE o: or MKB600: instead of TAPE?) If so, you could have a long > period of tedium ahead fixing those things.  It could be made ; easier by ensuring device names on the second system match e? the first as closely as possible (switching SCSI IDs and such.)o= It's those items that should have been abstracted but weren'tw< that can make your job difficult.  Those you find, you'll be: better off in the long run by fixing the right way (addingB identical abstrat names on both systems) than simply substituting  the new physical reference.7  ; After that, it's a matter of making the applications happy,:4 which can't really be addressed without knowing much more about your system.1   -Mike Duffyo      m   ------------------------------    Date: 27 Jul 2004 10:22:48 -0500- From: Kilgallen@SpamCop.net (Larry Kilgallen)d@ Subject: Re: Need to recreate production system on backup system3 Message-ID: <OMaeGMsIjGFB@eisner.encompasserve.org>c  f In article <22e43e71.0407270553.f0975c9@posting.google.com>, rcwood0000@yahoo.com (Roy C Wood) writes:G > We plan on getting a backup system to our vax. I need to recreate the G > production environment on the new vax. I have backup images (includes6G > system disk) from vax1 on tape that I need to port to vax2. Whats the G > simplest way to recreate the mirror environment on the new vax? Oh, ItD > can't create a cluster because nobody wants the the production vax > changed in any way.i  D Presuming the tapes were created with BACKUP/IMAGE, just restore the tapes with BACKUP/IMAGE.  G Since the two systems will have the same network addresses, ensure they  are not networked together.    ------------------------------  % Date: Tue, 27 Jul 2004 13:03:18 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com>a@ Subject: Re: Need to recreate production system on backup system, Message-ID: <41068AD3.D2D771C6@teksavvy.com>   Roy C Wood wrote: G > system disk) from vax1 on tape that I need to port to vax2. Whats thesG > simplest way to recreate the mirror environment on the new vax? Oh, I D > can't create a cluster because nobody wants the the production vax > changed in any way.l  H Restoring the backup on the new vax will usually work fine, but the work doesn't stop there.e  M If the backup vax does not have identical hardware, you will need to identifyeE any dependancies on hardware that exists on vax1 but not on vax2 (fortH instance, one vax may have TX serial ports while the new vax may have TT serial ports).  J If the 2 vaxes are to coexist on the same LAN, then vax2 should be given aI different identity. You need to change the sysgen SCSNODE, as well as thecK SCSSYSTEMID, and in DECNET, you need to change the node name as well as the0, node number (stay in same area as the vax1).  A the SCSSYSTEMID is generally derived from the decnet node number.o  R If vax2 has different amounts of memory, you may also want to do an autogen on it.  J Once you have gotten vax2 in a good working state, you can backup just thea files that are specific to that vax. (vaxvmssys.par, and the NET* files in SYS$SPECIFIC:[SYSEXE])d  J Then, at regular intervals, you can copy vax1 to vax2, and then reload the vax2 specific files.  J Also, you may wish to consider splitting your systartup_vms.com on vax1 toH have any vax1 specific commands in a separare command file (for instanceX systartup_vax1.com) and have systartup_vms call up @systartup_''f$getsyi("SCSNODE")'.com  L This way, you would have a totally vanilla systartup_vms.com which calls theL right command procedure depending on which vax it runs on. This will make itM possible for you to just copy systartup_vms from one vax to the other withoutl. worrying about having to do any updates to it.  K You vax-specific command precure can then setup the logocal names which areaM hardware speficic, allowing systartup_vms.com to then use those logical namesh4 totally oblivious to what vaxc they are actually on.    J Also, do not forget that a backup vax is of no use if it can't perform itsN tasks, so you will also need to carefully study how any telecom infrastructureO will need to adapt to allow you to perform your duties from any of the 2 vaxes.c   ------------------------------    Date: 27 Jul 2004 12:28:07 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) @ Subject: Re: Need to recreate production system on backup system3 Message-ID: <IKh7SQnKdTlV@eisner.encompasserve.org>p  f In article <22e43e71.0407270553.f0975c9@posting.google.com>, rcwood0000@yahoo.com (Roy C Wood) writes:G > We plan on getting a backup system to our vax. I need to recreate thedG > production environment on the new vax. I have backup images (includes G > system disk) from vax1 on tape that I need to port to vax2. Whats theaG > simplest way to recreate the mirror environment on the new vax? Oh, IfD > can't create a cluster because nobody wants the the production vax > changed in any way.e  >    1) make sure the new system is not connected to the network.    2) boot standalone backup on the new systemE    3) restore the backup images from the old system to the new system '       (make sure they were made /image)6F    4) boot VMS on the new system, remove the old licenses and install        the new licenses3    5) update the network settings on the new systeme+    6) connect the new system to the network   H   You do not need to re-install VMS to get it to boot on the new system,D   just make sure the model of the new system is listed in the SPD of    the VMS version you are using.   ------------------------------    Date: 27 Jul 2004 03:41:07 -0700. From: fabiopenvms@yahoo.com.br (Fabio Cardoso)# Subject: Re: OpenVMS Backup Failuree= Message-ID: <f30679fb.0407270241.38acab94@posting.google.com>   ] "Yong Boon, Lim" <y0ngb00n@yahoo.com> wrote in message news:<ce4tvd$rh7$1@news4.jaring.my>...f > Dear Friend, > L >         l've recently done the image backup of my SCSI harddisk to tape by > the following command,
 >        i.e.c1 >             $ BACKUP/IMAGE/LOG/VERIFY DKA200: -f0 >             _$ MKB800:S1AP040726.BCK/LABEL=SMC > G >         When the backup almost closed to the end, the following error- > message occurs, i.e.H >         %BACKUP-F-CLOSEOUT, error closing nkb800:[]S1AP040726.BCK;0 as > output9 >         %SYSTEM-F-INSFARG, insufficient call arguments.  > 3 >         Where's gone wrong? Could anyone know it?.  F The message shows "nkb800"  instead of "mkb800" of the backup command.B Try a $ SHOW LOGICAL MKB800* to see if the tape name is redefined.   Regardso   FC   >         Thank you. > 
 > Regards, > Yong Boonv   ------------------------------    Date: 27 Jul 2004 08:09:17 -07002 From: williamwebb@openvms-rocks.com (William Webb)+ Subject: Re: OpenVMS restarts while installa= Message-ID: <bf98c417.0407270709.6208c86b@posting.google.com>t  a JF Mezei <jfmezei.spamnot@teksavvy.com> wrote in message news:<4101C769.376938D4@teksavvy.com>...x > Ghazan Haider wrote:4 > > I cant read them, they zip by fast, and reboots./ > > > It might help if you posted the messages.- > K > On Alphas, is it possible to have the console on a serial port (with a VT  > terminal connected to it ?)  > N > If so, you could then boot from a VT terminal emulator and log all messages.  9 You never seem to have an LA120 around when you need one.:   :^)d   WWWebb   ------------------------------  # Date: Tue, 27 Jul 2004 07:06:24 GMT . From: "Steve Shadbolt" <steve.shadbolt@hp.com>% Subject: OpenVMS/Sybase certification@2 Message-ID: <Q9nNc.6624$IV1.6392@news.cpqcorp.net>  G Sybase have now retired their product set on OpenVMS.  According to the ( Sybase Support Certifications web page :J http://www.sybase.com/detail/1,6904,1010821,00.html the last certificationF performed was SQL Server 11.0.3.2 on OpenVMS 7.3.  Does anybody know :  K 1) Have any certifications been done against later versions of the OS (e.g.  7.3-2) ?1 2) Are any more certifications likely to happen ?oJ 3) Are any customers out there successfully using Sybase 11.0.3.2 on later versions of OVMS than 7.3 ?h   Thanks for any help.   SteveA   ------------------------------    Date: 27 Jul 2004 07:35:45 -0700. From: fabiopenvms@yahoo.com.br (Fabio Cardoso)) Subject: Re: OpenVMS/Sybase certificationb= Message-ID: <f30679fb.0407270635.63ee6032@posting.google.com>   h "Steve Shadbolt" <steve.shadbolt@hp.com> wrote in message news:<Q9nNc.6624$IV1.6392@news.cpqcorp.net>...I > Sybase have now retired their product set on OpenVMS.  According to thee* > Sybase Support Certifications web page :L > http://www.sybase.com/detail/1,6904,1010821,00.html the last certificationH > performed was SQL Server 11.0.3.2 on OpenVMS 7.3.  Does anybody know : > M > 1) Have any certifications been done against later versions of the OS (e.g.h
 > 7.3-2) ?3 > 2) Are any more certifications likely to happen ? L > 3) Are any customers out there successfully using Sybase 11.0.3.2 on later > versions of OVMS than 7.3 ?c >  > Thanks for any help. >  > Stevea  H Migrate your database from Sybase to Oracle RDB ! Its the better choice for you to have support !    []s  
 Fabio Cardosoa   ------------------------------  % Date: Tue, 27 Jul 2004 10:16:55 -04001# From: "John Smith" <a@nonymous.com>l) Subject: Re: OpenVMS/Sybase certificationk, Message-ID: <oJadnVZ2XpRF_pvcRVn-rQ@igs.net>   Steve Shadbolt wrote:mE > Sybase have now retired their product set on OpenVMS.  According tom. > the Sybase Support Certifications web page :> > http://www.sybase.com/detail/1,6904,1010821,00.html the lastG > certification performed was SQL Server 11.0.3.2 on OpenVMS 7.3.  Does  > anybody know : >0G > 1) Have any certifications been done against later versions of the OS  > (e.g. 
 > 7.3-2) ?3 > 2) Are any more certifications likely to happen ? F > 3) Are any customers out there successfully using Sybase 11.0.3.2 on# > later versions of OVMS than 7.3 ?a >s > Thanks for any help. >n > SteveN      J All the 'certs' shown on that URL are 'paper' certs. It is highly unlikelyK that any further certs will be issued. Sybase stopped active development oneE VMS about 5 years ago because they saw the Sybase market on VMS goingC nowhere.  I Alpha AXP OpenVMS 7.3 11.0.3 This is a paper cert, based on analysis fromr the Compaq engineering team -n  K Alpha AXP OpenVMS 7.2-1 11.0.3 This is a paper cert, based on analysis from8/ the DEC engineering team. Covers 7.2 & 7.2-1. -i  I Alpha AXP OpenVMS 7.2 11.0.x This is a paper cert, based on analysis from./ the DEC engineering team. Covers 7.2 & 7.2-1. -c  L Alpha AXP OpenVMS 7.1 11.0.x This is a paper certification based on a Binary* compatibility letter received from Digital   ------------------------------  % Date: Tue, 27 Jul 2004 12:44:06 -0400e- From: JF Mezei <jfmezei.spamnot@teksavvy.com>D5 Subject: Re: OT: =?iso-8859-1?Q?Sun=B4s?= AMD Opteron , Message-ID: <41068655.F98CE3BB@teksavvy.com>   Fabio Cardoso wrote:E > I am saying to run OS-X (for Integrity) like a Scientific Computingr7 > Operating System. Like those X-Serve are used for....a  M IA64 is ill suited for blades and large numbers of processors in a huge room.xN It eats too much electriity and generates too much heat. Even that Apple superG cluster required special cooling equipment to handle such a large load.0  1 Google refused IA64 due to power and heat issues.p  N Yes, I know, Intel has managed to get its IA64 in a few "supercomputer" sites.J But I wouldn't be surprised if those sites were given a deal they couldn'tD refuse by Intel/HP/whatever in exchange for bragging right by Intel.  H > So its time to rethink the model to develop servers. May be in slices6 > or bladening them instead of a huge piece of metal.   J Blades are a wintel thing due to windows inability to scale up to have one" instance handle many applications.  J For real operating systems, you are far better off having one large systemM with multiple CPUs since it is far easier to manage and the OS automatically,rF thourgh the magic of pre-emtptive multitasking, will give allocate CPU resources on demand.  I And then, when you get into Galaxy class machines, you can also move CPUsoM between OS instances to give one instance more power if needed. This is stufftN "blades" can't handle. And blades don't have stuff such as shared memory which/ makes intel-instance communications far easier.h   ------------------------------  % Date: Tue, 27 Jul 2004 07:32:09 -0600o4 From: Norman Lastovica <norman.lastovica@oracle.com># Subject: Re: Rdb 7.0.5 on VMS 7.3-22* Message-ID: <41065959.8D091191@oracle.com>  7 Rdb 705 is not certified on vms 732.  I would encouragem9 you to contact oracle rdb support directly for additional  information and assistance.    Robert Trawinski wrote:e > = > Does anybody run Rdb 7.0.5 (7.0.x in general) on VMS 7.3.2?o; > We need to upgrade production environment from VMS 7.2-1.O& > Do we (client) need upgrade Rdb too? > 
 > Regards, >  > Robert   -- a	 - - - - - 0  opinions expressed here are mine and mine alone.  and certainly are not intended in any way to 0  express or represent any opinions or commitment  of oracle corporation.o  *  norman lastovica / oracle rdb engineering   ------------------------------  % Date: Tue, 27 Jul 2004 05:15:09 -0700.# From: "Tom Linden" <tom@kednos.com>t" Subject: RE: Solaris to Itanium...9 Message-ID: <NDEMLKKEBOIFBMJLCECIEEFKDJAA.tom@kednos.com>o   < -----Original Message-----> < From: Robert Deininger [mailto:rdeininger@mindspringdot.com]% < Sent: Monday, July 26, 2004 7:28 PMR < To: Info-VAX@Mvb.Saic.Comf$ < Subject: Re: Solaris to Itanium... <r <o7 < In article <4102DFCA.2F81932F@teksavvy.com>, JF Mezei ' < <jfmezei.spamnot@teksavvy.com> wrote:< <e < >Robert Deininger wrote:? < >> This is a lie. I'm tired of refuting it in detail.  Anyone0 < interested can# < >> likely find my previous posts.o < >dB < >Do you deny that Carly has publicly stated that one reason they < cancelled EV79B < >was that they realised that EV7 batches already produced showed < that EV7 was( < >capable of running at higher speeds ? <nK < I don't deny anything Carly might have said.  I pay very little attentionbC < to Carly's public statements.  And on technical matters, I pay norI < attention at all. I have much better sources for technical information,(F < including technical information about Alpha CPUs, past, present, andK < future, real and unreal.  Why on Earth would you use Carly slideware as alD < source for technical information?  Carly doesn't make pesky littleF < decisions such as the best way to persue an Alpha CPU speed-up.  She/ < leaves that to people who are more qualified.  <eK < EV7 has shipped in quantity at 2 speeds that I know of.  The earlier ones L < were not as fast as the later ones.  My understanding is that the yield of> < fast chips from the early days was simply too low to supportH < productization at the fast speeds.  I don't have all the details at myA < fingertips, and they aren't for public consumption in any case.o < G < EV7z will run faster than EV7.  There are real engineering changes innE < EV7z, not just yield improvements.  And not just changing the clockp1 < chips.  You seem unwilling to grasp this point.o <eJ < When EV79 development was stopped, there were serious problems remainingJ < to overcome.  The prototype chips DID NOT WORK, i.e. they did not behaveF < like the Alpha spec says they should, and they did not pass the testJ < suites.  They were not useable in alphaservers, because the alphaserversK < would not have worked right.  Fixing the problems was expected to take sodK < long that EV79 would have been too late to market to be worthwhile.  EV7zcK < was a less aggressive compromise design that could ship soon enough to be * < interesting from a business perspective. <aA < Advance estimates from engineers will always predict a range ofbG < performance, features, and schedule, for any new CPU in the pipeline.eI < (I'm not just talking about Alpha CPUs.)  NOBODY knows what will reallypF < happen until significant numbers of chips have been manufactured andL < tested.  When marketing/publicity folks take the engineering estimates and6 < turn them into slideware, they ALWAYS do two things:G < 1.  They gloss over the uncertainties and give the entire developmentn% < process a veneer of predictability.cG < 2.  They emphasize the most optimistic scenarios out of the estimates  < provided by the engineers.K < That's just what publicity people do.  I don't think it is helpful, but Ie) < don't expect it to change anytime soon.d < J < Anyone paying attention to new IT products ought to realize that advanceG < product hype is not reliable, and should never build plans around the I < details of un-announced products.  The word "vaporware" wasn't inventeds < for nothing.  & I thought it was ivented for "nothing"   <a < --- ( < Incoming mail is certified Virus Free.< < Checked by AVG anti-virus system (http://www.grisoft.com).B < Version: 6.0.726 / Virus Database: 481 - Release Date: 7/22/2004 <t ---s& Outgoing mail is certified Virus Free.: Checked by AVG anti-virus system (http://www.grisoft.com).@ Version: 6.0.726 / Virus Database: 481 - Release Date: 7/22/2004   ------------------------------  + Date: Tue, 27 Jul 2004 10:40:54 +0000 (UTC)  From: david20@alpha2.mdx.ac.uk% Subject: Re: Spam storage on VMS Mailz) Message-ID: <ce5bfm$cej$1@news.mdx.ac.uk>:  \ In article <41053738.DBD40050@teksavvy.com>, JF Mezei <jfmezei.spamnot@teksavvy.com> writes:  >david20@alpha2.mdx.ac.uk wrote:Q >> Alternatively it can be setup to quarantine such mail on the gateway where thehR >> user can access quarantined mail via a GUI interface and review and release it. >tO >Nice idea, but that essentally requires a totally separate email system/clientr >to hold the quarantined mail. >i  H Yes the quarantining system obviously needs enough storage space for theK quarantined mail. It would often be a different system to that on which the,0 mail is normally stored but doesn't have to be.   . I have precisemail set up on two VMS clusters.  I 1 VMS cluster has VMS users on it who receive their mail on that cluster.oM Accessing it via logging in and using local clients PMDF PINE, PMDF MAIL, VMSi/ MAIL or accessing it via POP , IMAP or YAHMAIL. K Mail can be quarantined on a separate disk on that cluster for those users.s  J 1 VMS cluster acts as a central mailhub directing mail to Novell, ExchangeN systems etc. There are no users whose mail is normally stored on that cluster.M Mail can be quarantined for those Novell, Exchange etc users on that cluster.-  K Twice a day a mail message is sent to users who have turned on quarantiningeN detailing what mail has been quarantined for each user since the last message.K This means the user can just allow messages to be automatically deleted and J only needs to use the GUI to release a message if they think something has wrongly been quarantined. H The "client" to access these quarantined messages is just a web browser.  I A lot of users seem to prefer this mode of operation since the spam nevera reaches their inbox.      E >On ALL-IN-1, there is a file (pending.dat) which contains unreceivedeO >documents. When a new message is received, a pointer is added to a record withc# >a key such as "MAIL   <username>".a >eL >Then, when the user wants his emails, the application automatically fetchesN >the document headers pointed to by the "MAIL <username" record and adds those0 >to the user's file cabinet as unread messages.  > F >It would then be possible to have spam messages be stored under "SPAMI ><username>" record in pending.dat and then the user could fetch all SPAM-H >messages and have them stpred in the WASTEBASKET folder where they'd be, >automatically deleted at regular intervals. >gN >On VMS mail however, the callable routines lack the ability to store messagesO >in specific folders to begin with, so it is rather hard to perform such tasks.   G Thats rubbish. The callable mail routines allow copying and moving mail.L messages between folders in the same or different mail files. The routine is MAIL$MESSAGE_COPYr   seei  W http://h71000.www.hp.com/doc/731FINAL/4493/4493pro_042.html#36_copyingandmovingmessages   M As I said in the previous reply as long as your anti-spam product can tag theoI spam messages in someway so that they can be distinguished after they areaO delivered then DELIVER can be setup to move them automatically into a different  folder.a  4 PMDF includes it's own integrated version of DELIVER  / The non-PMDF version of DELIVER is available at   < http://vms.process.com/ftp/vms-freeware/FILESERV/DELIVER.ZIP    
 David Webb Security team Leader CCSS Middlesex University   ------------------------------  % Date: Tue, 27 Jul 2004 12:37:29 -0400t- From: JF Mezei <jfmezei.spamnot@teksavvy.com>t% Subject: Re: Spam storage on VMS Mailn, Message-ID: <410684C8.4FEAF46B@teksavvy.com>   david20@alpha2.mdx.ac.uk wrote:tM > Twice a day a mail message is sent to users who have turned on quarantining P > detailing what mail has been quarantined for each user since the last message.M > This means the user can just allow messages to be automatically deleted andaL > only needs to use the GUI to release a message if they think something has > wrongly been quarantined.n  N Yes, that is a great way to have spam handled. But I wonder if it can scale toB enterprise-wide though. You'd have a hell of a spam message store.  N For as much as I don't really trust yahoo mail, I like the way they set it up.$ spam is stored in a separate folder.  K However, for me, if a message is considered SPAM, I would still want to seemM the headers (as well as the SMTP dialogue information) not only to judge if I G am blocking legit messages, but also to see what tricks are being used.J  I > Thats rubbish. The callable mail routines allow copying and moving mail N > messages between folders in the same or different mail files. The routine is > MAIL$MESSAGE_COPYf  L Well, last I looked, it was impossible to create a MAIL document and have itF stored in the "OUTBOX" folder (or any folder for that matter) with theN documented callable routines. In fact, in VMS mail, if you you set the "keep aN copy of messages you send", all it can do is include your name as a CC and you) get a new message in your NEWMAIL folder.o  H I realise that there are probably tricks that may allow a document to be created in any folder.J I guess you can always deliver a new email message and then refile it. ButK that means you need to scan the inbox for the documeht you just created andw then refile that one.r   ------------------------------  + Date: Tue, 27 Jul 2004 17:17:02 +0000 (UTC)n From: david20@alpha2.mdx.ac.uk% Subject: Re: Spam storage on VMS Maile) Message-ID: <ce62md$jlm$1@news.mdx.ac.uk>a  \ In article <410684C8.4FEAF46B@teksavvy.com>, JF Mezei <jfmezei.spamnot@teksavvy.com> writes:  >david20@alpha2.mdx.ac.uk wrote:N >> Twice a day a mail message is sent to users who have turned on quarantiningQ >> detailing what mail has been quarantined for each user since the last message.cN >> This means the user can just allow messages to be automatically deleted andM >> only needs to use the GUI to release a message if they think something hase >> wrongly been quarantined. >tO >Yes, that is a great way to have spam handled. But I wonder if it can scale to C >enterprise-wide though. You'd have a hell of a spam message store.o >   ; Well we will see - we have 30,000 users - staff + students.eB So far the majority are just using the default tagging rather thanM quarantining. We'll really see in late september/ early october when the new   students arrive.    d
 David Webb Security Team Leader CCSS Middlesex University  O >For as much as I don't really trust yahoo mail, I like the way they set it up. % >spam is stored in a separate folder.r > L >However, for me, if a message is considered SPAM, I would still want to seeN >the headers (as well as the SMTP dialogue information) not only to judge if IH >am blocking legit messages, but also to see what tricks are being used. >oJ >> Thats rubbish. The callable mail routines allow copying and moving mailO >> messages between folders in the same or different mail files. The routine is- >> MAIL$MESSAGE_COPY > M >Well, last I looked, it was impossible to create a MAIL document and have iteG >stored in the "OUTBOX" folder (or any folder for that matter) with the-O >documented callable routines. In fact, in VMS mail, if you you set the "keep asO >copy of messages you send", all it can do is include your name as a CC and youA* >get a new message in your NEWMAIL folder. >rI >I realise that there are probably tricks that may allow a document to be6 >created in any folder.eK >I guess you can always deliver a new email message and then refile it. ButML >that means you need to scan the inbox for the documeht you just created and >then refile that one.   ------------------------------  + Date: Tue, 27 Jul 2004 10:58:06 +0000 (UTC)i From: david20@alpha2.mdx.ac.uk' Subject: Re: Structure of SMTP Receiver3) Message-ID: <ce5cfu$cej$2@news.mdx.ac.uk>0  \ In article <41054DFA.D427FBAE@teksavvy.com>, JF Mezei <jfmezei.spamnot@teksavvy.com> writes:L >Since the TCPIP Services SMTP receiver lacks the hooks to scan the contentsL >and has very very poor logging, I was thinkking if it were possible to justI >rewrite one (possibly using portions of postfix which have the anti spam2 >filter capabilities). >   M I've only used the mailserver integrated (in my case PMDF integrated version)aE of PreciseMail Anti-Spam. However there is a standalone version whicht) supposedly works with any email server ond   OpenVMS Vax 6.1 or higherw and  OpenVMS Alpha 6.2 or higher.   It acts as an SMTP proxy.i     see   . http://www.process.com/precisemail/newin2.html    
 David Webb Security team leader CCSS Middlesex University      J >Could the home grown receiver simply use the Send-From-File to submit theN >received documents to the SMTP queues for actual deliveries ? Is that how the >current software runs ? > L >Or does SFF lack certain functionality for the receiver to perform its task? >once the message has been received and needs to be processed ?    ------------------------------  # Date: Tue, 27 Jul 2004 09:45:41 GMT ! From: Nigel Barker <nigel@hp.com>-: Subject: Re: Turn-key OpenVMS E-mail, web server solution?8 Message-ID: <jn8cg01aspvq2b6fhmt9u5fedanslno5ci@4ax.com>  K On Mon, 26 Jul 2004 14:27:37 -0400, JF Mezei <jfmezei.spamnot@teksavvy.com>3 wrote:M >If large companies are not interested in the shrinking VMS marketplace, thatiM >leaves the hobbyists who port open source software to VMS. But to do so, youe5 >need modern features of VMS on their platform (VAX).a  L Why do you assume that hobbyists only have VAX hardware? All of the ports ofO open source software for VMS that I have seen have Alpha versions. In fact many2N only have Alpha versions & in fact I would go so far as to say that I have not% seen VAX only applications for years.e   -- Nigel Barker Live from the sunny Cote d'Azur    ------------------------------  % Date: Tue, 27 Jul 2004 10:04:45 +0200 - From: "Martin Vorlaender" <mv@pdv-systeme.de> X Subject: Re: [TCPware, V73_MGMTAGENTS] Does anyone have a working installation running ?* Message-ID: <2mmgkuFodjqcU1@uni-berlin.de>  ; "Peter 'EPLAN' LANGSTOEGER" <peter@langstoeger.at> wrote... A > I tried to run V73_MGMTAGENTS V3.0-36 on TCPware V5.6-2 but the 
 > agents die.l   Not on my system.i  @ > It seems that the eSNMP service isn't running or accepting the	 > agents.u >b- > Does anyone have this combination working ?h  6 Did you follow the specific installation instructions?K http://h71000.www7.hp.com/openvms/products/mgmt_agents/install.html#tcpwareM   cu,    Martin -- a>                        |  Martin Vorlaender  |  OpenVMS rules!1   OpenVMS: When you    |  work: mv@pdv-systeme.de-D   KNOW where you want  |    http://www.pdv-systeme.de/users/martinv/8   to go today.         |  home: martin@radiogaga.harz.de   ------------------------------  % Date: Tue, 27 Jul 2004 10:15:42 +0200n- From: "Martin Vorlaender" <mv@pdv-systeme.de>nX Subject: Re: [TCPware, V73_MGMTAGENTS] Does anyone have a working installation running ?* Message-ID: <2mmh9fFo72pcU1@uni-berlin.de>  + "Martin Vorlaender" <mv@pdv-systeme.de> ...i= > "Peter 'EPLAN' LANGSTOEGER" <peter@langstoeger.at> wrote...oC > > I tried to run V73_MGMTAGENTS V3.0-36 on TCPware V5.6-2 but thee > > agents die.l >a > Not on my system.   . OK, I'll have to clarify that statement a bit:5 - I'm running V72_MGMTAGENTS V3.0-22 under VMS V7.2-2 F   (sorry. When posting I thought we had already upgraded that machine)/ - it doesn't work with Opera on the client sidea4   (it does with Internet Exploder and Win32 Mozilla); - I've yet to make the thresholds work. For some reason thex9   java program responsible for those doesn't accept me ase
   authorized.    cu,a   Martin -- w;    OpenVMS @ 25      | Martin Vorlaender  |  OpenVMS rules! .                      | work: mv@pdv-systeme.deA    Still exceeding   |   http://www.pdv-systeme.de/users/martinv/p5    expectations      | home: martin@radiogaga.harz.dea   ------------------------------  % Date: Tue, 27 Jul 2004 08:42:06 -0400u, From: "Richard Whalen" <WhalenR@process.com>X Subject: Re: [TCPware, V73_MGMTAGENTS] Does anyone have a working installation running ?+ Message-ID: <ce5ij6$h5g$1@news.process.com>r  D As Martin pointed out, getting the configuration set up correctly is
 essential.3 The web page that is correct, but missing one step:t $ @TCPWARE:CNFNET SNMP' and answer YES for SNMP AgentX service.n     ---------------------------r Richard Whalen Process Software   ------------------------------  % Date: Tue, 27 Jul 2004 15:08:01 +0200 - From: "Martin Vorlaender" <mv@pdv-systeme.de> X Subject: Re: [TCPware, V73_MGMTAGENTS] Does anyone have a working installation running ?* Message-ID: <2mn2diFoaf98U1@uni-berlin.de>  
 I wrote...? > > "Peter 'EPLAN' LANGSTOEGER" <peter@langstoeger.at> wrote...-E > > > I tried to run V73_MGMTAGENTS V3.0-36 on TCPware V5.6-2 but the@ > > > agents die.  > >a > > Not on my system.M > 0 > OK, I'll have to clarify that statement a bit:7 > - I'm running V72_MGMTAGENTS V3.0-22 under VMS V7.2-2aH >   (sorry. When posting I thought we had already upgraded that machine)   Another machine:   $ netcu show version5 TCPware(R) V5.6-2 Copyright (c) 2002 Process Softwaret  ; OpenVMS version V7.3-2   booted on 27-JUL-2004 13:43:33.00,a,         running on a AlphaStation 250 4/266.    $ product show product v73_mgmt*< ----------------------------------- ----------- ------------5 PRODUCT                             KIT TYPE    STATEb< ----------------------------------- ----------- ------------9 HP AXPVMS V73_MGMTAGENTS V3.0-30    Full LP     Installedl< ----------------------------------- ----------- ------------  A While I wouldn't say that this configuration works - I do see thes@ frames but they don't don't contain any data - the agent doesn't? crash. Did you follow the special installation instructions for 
 TCPware onK http://h71000.www7.hp.com/openvms/products/mgmt_agents/install.html#tcpwareg ?s  = > - I've yet to make the thresholds work. For some reason thee; >   java program responsible for those doesn't accept me asi >   authorized.   ? Funny enough: the thresholds don't work with TCP/IP Services onk that machine either.   cu,a   Martin -- c>                     | Martin Vorlaender    |    OpenVMS rules!-   Smiert Spamionem  | work: mv@pdv-systeme.den@                     |   http://www.pdv-systeme.de/users/martinv/4                     | home: martin@radiogaga.harz.de   ------------------------------   End of INFO-VAX 2004.413 ************************