1 INFO-VAX	Fri, 28 Jun 2002	Volume 2002 : Issue 352       Contents:" Re: 7.3-1 new backplane controller" Re: 7.3-1 new backplane controller8 Re: Amdrew wants numbers, here they are!  Blow out baby!' Re: Andrew wan'ts the numbers, here ... G Re: BACKUP/INCREMENTAL not parsing directories properly during restores G Re: BACKUP/INCREMENTAL not parsing directories properly during restores G Re: BACKUP/INCREMENTAL not parsing directories properly during restores G Re: BACKUP/INCREMENTAL not parsing directories properly during restores G Re: BACKUP/INCREMENTAL not parsing directories properly during restores & Re: Capellas resigns from Dynegy board@ Re: Deutsche Bank would like to outsource there IT to IBM or CSC@ Re: Deutsche Bank would like to outsource there IT to IBM or CSC@ Re: Deutsche Bank would like to outsource there IT to IBM or CSC  Re: Encompass - DECUS VMS survey( Re: Has SMTP RBL list worked for anyone?( Re: Has SMTP RBL list worked for anyone?, Re: My conversation with Linus about VMS ..., Re: My conversation with Linus about VMS ..., Re: My conversation with Linus about VMS ..., Re: My conversation with Linus about VMS ... RE: New VMS Marketing  Re: New VMS Marketing  Re: New VMS Marketing  Re: Open Letter to HP  Re: OVMS 7.3 BACKUP problem  Re: Quorum discussion/questions 3 Re: Reuters test - Itanium II blows away Sparky III 3 Re: Reuters test - Itanium II blows away Sparky III 3 Re: Reuters test - Itanium II blows away Sparky III ) Re: RMS : preserving a RAB across $OPEN ? ) Re: RMS : preserving a RAB across $OPEN ? 2 Re: SSH v2 (was Re: OpenVMS, Volume Desktop OS...)2 Re: SSH v2 (was Re: OpenVMS, Volume Desktop OS...)" Re: Sun benchmarketeering campaign" Re: Sun benchmarketeering campaign" Re: Sun benchmarketeering campaign
 RE: TCO study  TCP/IP
 Re: TCP/IP
 Re: TCP/IP
 Re: TCP/IP
 Re: TCP/IP Re: Trouble with BACKUP/RECORD Re: Trouble with BACKUP/RECORD Re: Trouble with BACKUP/RECORD Re: Trouble with BACKUP/RECORD Re: Trouble with Samba 2.0.3 Re: Trouble with Samba 2.0.3 Xwindows: XDM mode  F ----------------------------------------------------------------------  % Date: Thu, 27 Jun 2002 17:35:32 -0400 6 From: "John.Malmberg" <Malmberg@dskwld.zko.dec.compaq>+ Subject: Re: 7.3-1 new backplane controller 2 Message-ID: <3D1B8524.30007@dskwld.zko.dec.compaq>   Dave Gudewicz wrote:6  > Will _all_ the SmartArray controllers be supported?  * No, the older one's will not be supported.  ; And only support for direct attached SCSI disks is planned.   H I would expect that initially a 2 channel and 4 channel model will show  up on the supported list.   %  > And whats the word on the MSA1000.    It seems to be quite fast.  H As I understand it support for it under OpenVMS has not officially been  announced yet.  D If you have an interest in unanounced products, please contact your D local sales person or OpenVMS Amabassador for information.  It will 3 probably require the standard non-disclosure stuff.    -John ! malmberg@dskwld.zko.dec.compaq.hp  Personal Opinion Only    ------------------------------  % Date: Thu, 27 Jun 2002 20:35:17 -0500 + From: Shael Richmond <ksrich@bellsouth.net> + Subject: Re: 7.3-1 new backplane controller - Message-ID: <3D1BBD55.24DD09D8@bellsouth.net>    Dave Gudewicz wrote: > L > Will _all_ the SmartArray controllers be supported?  And whats the word on > the MSA1000. >   @ I was at a SAN presentation today and they said VMS support willE be available by September for the MSA1000.  It looks like a real nice  box.   Shael    ------------------------------  # Date: Fri, 28 Jun 2002 02:19:00 GMT   From: cjt <cheljuba@prodigy.net>A Subject: Re: Amdrew wants numbers, here they are!  Blow out baby! + Message-ID: <3D1BC796.466F6326@prodigy.net>    Bob Ceculski wrote:  > g > "rob kas" <rob@paychoice.nospam.com> wrote in message news:<3d18c13d$0$1421$8e9e3842@news.atx.net>...  > > Bob  > > 9 > >  Do you ever think about what you read or do you just  > >  regurgitate Web Pages ? > > ; > > You do know that is comparing the lasest GREATEST Alpha M > >  againest a older SUN chip. And all thing considered not much better then 
 > > IBM's P4.  > > < > > By the way I am looking foward to meeting you at the VMS
 > > Roadshow.  > > & > >                                Rob > >  > B > and you think the newer sparky is going to overtake the EV68 ... > I don't think so ...  ' Your loyalty is touching, if misplaced.    ------------------------------    Date: 28 Jun 2002 03:08:33 +0800, From: Paul Repacholi <prep@prep.synonet.com>0 Subject: Re: Andrew wan'ts the numbers, here ...- Message-ID: <87wuskilou.fsf@prep.synonet.com>   ( Rudolf Wingert <win@fom.fgan.de> writes:    C > Andrew did wrote a lot of number, but did not write, where we can E > find this number to check the conditions. Anybody could write a lot D > os arguments. But I would like to see verifyable statements. We doE > have an AlphaStation ES40 with four 600MHz EV6 CPUs and 16GB memory E > and a SunFire 880 with the 900MHz UltraSparc III. Now I can compare E > performance/price. Sorry, but we do not have an application running > > on both. The only one I could test is I/O. There was a small; > difference between both. Alpha a little bit faster (under F > OpenVMS!!!!). Also the boot sequence was much faster (I know this isC > meaningless because OpenVMS have to be booted only once). But the B > big difference is the price. We did pay the double value for the
 > SunFire.  D You can run STREAM no problems, John has all the code etc on his web1 site, and you are free to grab a copy and run it.   A Spec is restricted to Spec consortium members. The TPC benchmarks B require a large (humungous??) cost in set up and the like. IE, youF need another system to drive the `user input' into the SUT, and for it= to be fast enough to never be a constraint. And so it goes...   A If you could get Spec or Bonnie, they are reasonably small to run ? and report. I have also run FFTW from MIT as a test. 2D FFTs of ) large arrays are a *good* test of paging!    --  < Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda. @                                              West Australia 6076. Raw, Cooked or Well-done, it's all half baked.F EPIC, The Architecture of the future, always has been, always will be.   ------------------------------    Date: 27 Jun 2002 12:30:40 -0700. From: SPAMSINK2001@YAHOO.COM (Alan E. Feldman)P Subject: Re: BACKUP/INCREMENTAL not parsing directories properly during restores= Message-ID: <343f30ae.0206271130.3bc3502a@posting.google.com>    CORRECTION BELOW!   s SPAMSINK2001@YAHOO.COM (Alan E. Feldman) wrote in message news:<343f30ae.0206261138.1179f251@posting.google.com>... i > skulker@easynews.com.yourpants (skulker) wrote in message news:<3d19b8ff.55246228@news.easynews.com>...  > > Greetings! > > D > > I have encountered a problem with the BACKUP/INCREMENTAL command( > > during a disk restore. (VMS 7.2-1H1) > >  > > My comand (example): > > < > > $ BACKUP/INCREMENTAL/DENS=TK87 $9$MKA80:saveset.bak/SAVE > > $1$DUA400:/LOG >    [SNIP] > B > Another poster mentioned order of tapes. I'm pretty sure this isE > irrelevant except that doing the tapes in the wrong order increases G > the chances of filling up the disk. Other than that, order should not B > matter as far as success goes. At least that's what the VMS v6.2D > manual says. Doing the tapes in reverse chronological order is theD > most efficient way to go, but is not necessary for success. (It isH > most efficient because in that case, the first incremental will deleteC > all directory entries for files that should not be present in the D > final outcome and that gives BACKUP/INCR the clues it needs duringE > subseuqnet incremental restore operations to not restore files that C > were ultimately deleted by the time of the final incremental save C > operation -- and at least some of those files would be needlessly D > restored and deleted if the incrementals were applied in any other	 > order.)     F CORRECTION: Reverse chronological order is most efficient because whenC you restore from the latest incremental first, it adjusts the files C listed in the .DIR files to their final forms. Subsequent BACK/INCR C operations using the next oldest tape check the .DIR files and skip E restoring files that are not listed there because the older tapes are E based on out-of-date directories. (BACKUP/INCR knows to do this based F on the backup dates of the .DIR files: BACKUP/INCR compares the backupF date of the current tape with the backup date of the .DIR file that isE being restored to and then goes by the one that is more recent.) When E you run the tapes in the opposite order, the newer tape is run second B and has to delete any files that were restored from the older saveB set, but deleted between the save operations that produced those 2F save sets and add files and their directory entries for any files that are in the newer save set.  D That is why reverse chronological order is most efficient. But it isB not necessary for a successful final outcome. You are more likely,C however, when not using reverse chronological order, to fill up the A disk since restoration from the intermediate incrementals may add F files that would need to be deleted by restoration of files from laterE tapes. I believe that if you fill up the disk, you can just rerun the 9 incremental(s) as needed and eventually all will be well.       Disclaimer: JMHO Alan E. Feldman  afeldman gfigroup com 0 "It'll work fine until the next time it breaks."   ------------------------------    Date: 27 Jun 2002 17:59:48 -0600- From: Kilgallen@SpamCop.net (Larry Kilgallen) P Subject: Re: BACKUP/INCREMENTAL not parsing directories properly during restores3 Message-ID: <z$WnXVtpzKZy@eisner.encompasserve.org>   n In article <343f30ae.0206271130.3bc3502a@posting.google.com>, SPAMSINK2001@YAHOO.COM (Alan E. Feldman) writes:  F > That is why reverse chronological order is most efficient. But it isD > not necessary for a successful final outcome. You are more likely,E > however, when not using reverse chronological order, to fill up the C > disk since restoration from the intermediate incrementals may add H > files that would need to be deleted by restoration of files from later > tapes.  E But going in forward chronological order each set of directories does G represent a point in time where all the files fit on disk.  That should G be no more likely to overflow that starting with the final incremental. E (I have just a slight hesitiation about new files on an early tape of E a given incremental using space from files deleted by a later tape of  the same incremental.)   ------------------------------  # Date: Fri, 28 Jun 2002 01:27:24 GMT . From: skulker@easynews.com.yourpants (skulker)P Subject: Re: BACKUP/INCREMENTAL not parsing directories properly during restores2 Message-ID: <3d1a69f7.187031171@news.easynews.com>   Thanks to all who replied!  E Compaq has a patch for the BACKUP image. After applying the patch all  is well.  
 Thanks again!        skulker    skulker@easynews.com.yourpants   email replies remove .yourpants    ------------------------------    Date: 27 Jun 2002 19:58:59 -0700. From: spamsink2001@yahoo.com (Alan E. Feldman)P Subject: Re: BACKUP/INCREMENTAL not parsing directories properly during restores= Message-ID: <b096a4ee.0206271858.2a63e4d4@posting.google.com>   h Kilgallen@SpamCop.net (Larry Kilgallen) wrote in message news:<z$WnXVtpzKZy@eisner.encompasserve.org>...p > In article <343f30ae.0206271130.3bc3502a@posting.google.com>, SPAMSINK2001@YAHOO.COM (Alan E. Feldman) writes: > H > > That is why reverse chronological order is most efficient. But it isF > > not necessary for a successful final outcome. You are more likely,G > > however, when not using reverse chronological order, to fill up the E > > disk since restoration from the intermediate incrementals may add J > > files that would need to be deleted by restoration of files from later
 > > tapes. > G > But going in forward chronological order each set of directories does I > represent a point in time where all the files fit on disk.  That should I > be no more likely to overflow that starting with the final incremental. G > (I have just a slight hesitiation about new files on an early tape of G > a given incremental using space from files deleted by a later tape of  > the same incremental.)    C I think you're right except if the top of a large directory tree is F renamed to something earlier in the alphabet than its parent directoryF and "shallower" in the MFD directory tree (a sufficient condition, butE perhaps not a necessary one for this example, and I do mean perhaps).   @ Anyway, suppose you did some incremental saves. Then you renamed< [FLINT]FRED.DIR to [000000]BARNIE.DIR. Also assume no or fewF modifications to any files in this tree. Then you did more incrementalA saves. Now, upon restoring, if you restored the incrementals from D oldest to newest, when the first save set containing BARNIE.DIR cameD around, the files from FRED.DIR will still be there while files from= BARNIE.DIR are being copied to disk. This can cause, at least C temporarliy, more files to be on the disk at that time than were on D the disk during the time that that save set was created because whenE that save set was created, only the files from BARNIE.DIR were on the F disk. And now you still have to wait for the files from FRED.DIR to beC deleted since [000000]FLINT.DIR is alphabetically "larger" than the F BARNIE files and therefore later in the save set. And only when BACKUP: gets to FLINT.DIR will it delete the [FLINT]FRED.DIR tree.  A [NOTE: I'm assuming that the incremental restore would check .DIR F files in [000000] first and delete any that should be deleted based on> the date of the save set being applied and the backup dates ofA relevant files on the disk. (That's why I put FRED.DIR in [FLINT] E instead of [000000]). Then, the BARNIE files would be restored. Then, F eventually the restore operation would come upon [000000]FLINT.DIR and@ see that on tape, [FLINT]FRED.DIR is gone, and BACKUP would thenF proceed to delete that tree. But just prior to that you have the filesD pretty much duplicated in both trees, only one of which will survive= in the end. The extra copy of this tree could fill the disk.]   B If you instead use reverse chronologial order, [FLINT] would neverF have FRED.DIR placed in it and so the older copy of the directory treeD would never appear on the disk, thereby reducing the odds of filling up the disk.  @ But I admit I am not sure about this because I have not tried itE (well, at least not recently), but based on my experience with BACKUP A and on what I've read in the manuals, that's what I would expect.   ? It's late, this is from memory and a quick look in the Web site 9 manual, so ICWBWYK, so one would have to actually try it.   F And then there's alias directories to worry about!!! But not now! I'll think about that later.      Disclaimer: JMHO (or OFICBWYK) Alan E. Feldman  afeldman gfigroup com    ------------------------------    Date: 27 Jun 2002 20:03:24 -0700. From: spamsink2001@yahoo.com (Alan E. Feldman)P Subject: Re: BACKUP/INCREMENTAL not parsing directories properly during restores= Message-ID: <b096a4ee.0206271903.29ec221b@posting.google.com>   D I inadvertently posted before I finished composing. My apologies and6 please disregard that post. Here is the complete copy:  h Kilgallen@SpamCop.net (Larry Kilgallen) wrote in message news:<z$WnXVtpzKZy@eisner.encompasserve.org>...p > In article <343f30ae.0206271130.3bc3502a@posting.google.com>, SPAMSINK2001@YAHOO.COM (Alan E. Feldman) writes: > H > > That is why reverse chronological order is most efficient. But it isF > > not necessary for a successful final outcome. You are more likely,G > > however, when not using reverse chronological order, to fill up the E > > disk since restoration from the intermediate incrementals may add J > > files that would need to be deleted by restoration of files from later
 > > tapes. > G > But going in forward chronological order each set of directories does I > represent a point in time where all the files fit on disk.  That should I > be no more likely to overflow that starting with the final incremental.sG > (I have just a slight hesitiation about new files on an early tape ofeG > a given incremental using space from files deleted by a later tape oft > the same incremental.)    C I think you're right except if the top of a large directory tree isoF renamed to something earlier in the alphabet than its parent directoryF and "shallower" in the MFD directory tree (a sufficient condition, butE perhaps not a necessary one for this example, and I do mean perhaps).i  @ Anyway, suppose you did some incremental saves. Then you renamed< [FLINT]FRED.DIR to [000000]BARNIE.DIR. Also assume no or fewF modifications to any files in this tree. Then you did more incrementalA saves. Now, upon restoring, if you restored the incrementals fromrD oldest to newest, when the first save set containing BARNIE.DIR cameD around, the files from FRED.DIR will still be there while files from= BARNIE.DIR are being copied to disk. This can cause, at leastsC temporarliy, more files to be on the disk at that time than were onVD the disk during the time that that save set was created because whenE that save set was created, only the files from BARNIE.DIR were on thetF disk. And now you still have to wait for the files from FRED.DIR to beC deleted since [000000]FLINT.DIR is alphabetically "larger" than theiF BARNIE files and therefore later in the save set. And only when BACKUP: gets to FLINT.DIR will it delete the [FLINT]FRED.DIR tree.  A [NOTE: I'm assuming that the incremental restore would check .DIRgF files in [000000] first and delete any that should be deleted based on> the date of the save set being applied and the backup dates ofA relevant files on the disk. (That's why I put FRED.DIR in [FLINT] E instead of [000000]). Then, the BARNIE files would be restored. Then, F eventually the restore operation would come upon [000000]FLINT.DIR and@ see that on tape, [FLINT]FRED.DIR is gone, and BACKUP would thenF proceed to delete that tree. But just prior to that you have the filesD pretty much duplicated in both trees, only one of which will survive= in the end. The extra copy of this tree could fill the disk.]5  B If you instead use reverse chronologial order, [FLINT] would neverF have FRED.DIR placed in it and so the older copy of the directory treeD would never appear on the disk, thereby reducing the odds of filling up the disk.  @ But I admit I am not sure about this because I have not tried itE (well, at least not recently), but based on my experience with BACKUP A and on what I've read in the manuals, that's what I would expect.p  ? It's late, this is from memory and a quick look in the Web site 9 manual, so ICWBWYK, so one would have to actually try it.f  C And then there are alias directories to worry about!!! But not now!h I'll think about that later.     Disclaimer: JMHO (or OFICBWYK) Alan E. Feldman  afeldman gfigroup com    ------------------------------  # Date: Fri, 28 Jun 2002 02:31:28 GMTf  From: cjt <cheljuba@prodigy.net>/ Subject: Re: Capellas resigns from Dynegy board-* Message-ID: <3D1BCA82.43643DE@prodigy.net>  E Maybe the job at the combined company is a lot tougher than expected.c   JF Mezei wrote:c > M > Interesting on how, as CEO and president of Compaq, he had time to do this, O > but now, as a mere civil servant within HP, he doesn't... My guess is that hewL > doesn't want to be associated with those Houston energy firms. It is a badN > omen though because I would have prefered for him to keep that job and leave > HP :-) > < > Dynegy Says Hewlett-Packard's Capellas Resigned From Board >  >   6/25/02 9:08am > L >   WASHINGTON -(Dow Jones)- Dynegy Inc. (DYN) said that Hewlett-Packard Co. > (HPQ) President I >   Michael Capellas told it June 18 that he was resigning from its board  > effective immediately. > F >   According to an 8-K filed Tuesday with the Securities and Exchange > Commission, CapellasN >   resigned "in light of the demands on his time and his new responsibilities > at Hewlett-Packard." > K >   Dynegy said that Capellas' resignation wasn't based on any disagreementn > between it and
 >   Capellas.e > R >   Capellas, formerly the chief executive at Compaq Computer, become president of9 >   Hewlett-Packard when the two companies merged in May.    ------------------------------    Date: 27 Jun 2002 13:25:31 -0600+ From: young_r@encompasserve.org (Rob Young)oI Subject: Re: Deutsche Bank would like to outsource there IT to IBM or CSC 3 Message-ID: <nk2F2hgRpDcG@eisner.encompasserve.org>i  \ In article <3D1B51FA.9B396F7D@videotron.ca>, JF Mezei <jfmezei.spamnot@videotron.ca> writes: > Sue Skonetski wrote:N >> JF there are a number of major banks/stock exchanges/financial institutionsO >> that use OpenVMS.  There are several reasons that you may not see the actualo	 >> names.  >>  F >> 1.  Many customers view their computer environment as a competitiveE >> advantage and if they disclosed what they used they would lose the6
 >> advantage.n > K > Then Sue, perhaps you should find a way for senior execs outside of IT in/O > those banks to be aware of the importance VMS plays in their own bank instead P > of allowing them to make departmental decisions with the idea that VMS is dead > and should be gotten rid of.   	Tell me . . . e 	n@ 	Am I reading what you wrote correctly?  Are you suggesting that? 	senior management within a large corporation is concerned withr? 	what OS they are running?  I'm talking CEO.  And then by that,u1 	are you suggesting CEOs make platform decisions?h  E 	Converseley, if a platform gives a corporation a competive advantagetB 	how many CEOs know or are concerned with what OS said platform is	 	running?o   				Robi   ------------------------------  % Date: Thu, 27 Jun 2002 15:16:54 -0400t- From: JF Mezei <jfmezei.spamnot@videotron.ca>oI Subject: Re: Deutsche Bank would like to outsource there IT to IBM or CSCi, Message-ID: <3D1B64A0.4BFAC38E@videotron.ca>   Rob Young wrote:I >         Am I reading what you wrote correctly?  Are you suggesting thatiH >         senior management within a large corporation is concerned withH >         what OS they are running?  I'm talking CEO.  And then by that,: >         are you suggesting CEOs make platform decisions?  I Yes. I had bid a VMS solution to one non-IT department (a service I would K operate) and was told a flat "NO" and given a chance to bid again, provided7N nothing Digital was included in the second bid. This was at a time where LotusK Notes was being rolled out in banks bih time, and yes, it involved a senioreM exec at BT. Ironically, this happened at a day when Citibank had just orderedWJ a huge amount of DEC alpha gear, and ironically, there was also a CitibankM senior exex involved in that decision to prohibit me from picthing a solutions that involved DEC gear.g  N So yes, my message still stands. Senior folks inside banks had a very nagativeL image of Digital and VMS and that influenced their department's IT decisions	 big time.M  L The citibank guy was under the impression that rolling out Lotus Notes was aL simple matter that took 15 minutes, because that is how long it took for theN tech to "install" it on his desktop. He had no clue of the amount of work thatM would have gone on downstairs by the grunts before they were ready to add the  icon to his desktopt  K This is why I feel very strongly that is crucial to advertise VMS in non-IThM stuff so that it reaches those senior execs and changes their opinion of VMS.eJ Banks are very large and departments outside of IT may not be aware of the! hidden VMS boxes somewhere in IT.t  K I was also at another bank where the IT execs were IBM-blue-brainwashed and N tried very hard to get rid of anything not sanctioned by their friends at IBM.L You see, IBM had reps showing up and wining  ,dining and golfing those execsM while I was unable to get our DEC rep to even show up with proper folks to do M a demo of the DECNET-SNA gateway. What impression do you think this leaves ona
 those execs ?h   ------------------------------    Date: 28 Jun 2002 03:02:36 +0800, From: Paul Repacholi <prep@prep.synonet.com>I Subject: Re: Deutsche Bank would like to outsource there IT to IBM or CSC - Message-ID: <871yask0j7.fsf@prep.synonet.com>   / Kilgallen@SpamCop.net (Larry Kilgallen) writes:n  > > Both IBM and CSC have quite competent VMS people among their; > employees.  One of the CSC folk is a major contributor ton > comp.os.vms !c  8 And several of .au's best VMS people now have ibm.com in their mail addresses...B   -- N< Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda.h@                                              West Australia 6076. Raw, Cooked or Well-done, it's all half baked.F EPIC, The Architecture of the future, always has been, always will be.   ------------------------------  # Date: Thu, 27 Jun 2002 21:26:55 GMTe# From: "John Smith" <a@nonymous.com>r) Subject: Re: Encompass - DECUS VMS surveywF Message-ID: <zqLS8.5001$2er.4868@news01.bloor.is.net.cable.rogers.com>  @ Surveys like this always raise the hairs on the back of my neck.  F These is almost always a SET of dependencies that result in an overallL conclusion, and this survey with its simplistic choose-one mode won't arriveH at a result that even approximates the reasons/rationales VMS shops will have.b  I This survey should have bee set up more along the line of asking the sameaK questions and then giving those participating a range of responses, perhapseH along the lines of  "I completely agree, I completely disagree, somewhatK agree/disagree, it depends, don't care, don't know, we will continue to buy L as many Alpha-based machines as we can get on eBay for the next 20 years andD never migrate to IA-64 garbage, I use Microsoft so eliminate me fromG tabulation, and other responses that allow some reasonable answers that3! capture the expectations better."0    < "Dave Gudewicz" <david.gudewicz@abbott.com> wrote in message* news:affiqa$hk$1@fizban.pprd.abbott.com..., > There is a VMS / Itanium survey posted at: >e > http://www.encompassus.org/e >w) > Part of the online advocacy initiative.t >i	 > Dave...  >i >s   ------------------------------  % Date: Fri, 28 Jun 2002 01:24:36 +0200w) From: "Steven Thompson" <steven@omga.biz>i1 Subject: Re: Has SMTP RBL list worked for anyone?rB Message-ID: <afg6r5$opn$1@nsnmrro2-gest.nuria.telefonica-data.net>   Hi Lawrence   K I tested RBL on my system since your first posting on this subject, which In hadn't done to now.eF I added two entries to the RBL line in my smtp.config file and my mailF "receive logs" are now nearly all "550 Closing transmissions channel". Something which I rarely see.)
 To explain...-L I follow the SMTP of my system quite closely, even allowing a certain laxity0 in the configuration "just to see what happens".I I  receive a fair quantity of junk on purpose which doesn't bother me. My32 server is for my own purposes, and for learning...L But I have added a protection in that if the system detects a suddenly largeJ quantity of files in the SMTP directory it closes down the SMTP queues and sends me a SMS on the mobile.e  I There wasn't anything in the "SMTP log's". BUT this change in the receive:, logs would indicate that the RBL is working.K I can't as yet be more objective and talk of efficiency, except that one ofdG the IP's that was refused I traced to a site that I know that sends outc "adverts" to buy stuff..  * The second RBL I located at www.spews.org.   Steven  = "Lawrence Bleau" <bleau@umtof.umd.edu> escribi en el mensaje.* news:afd791$n8v$1@grapevine.wam.umd.edu...G > I haven't received much help on my previous post, so I'm changing thei > question and bit.i >n? > If you are running TCPIP V5.1 (AXP) and you are using the RBLoH > (realtime blockhole list) feature, have you ever seen evidence that it > works? > ; > Evidence would be something in TCPIP$SMTP_RECV_RUN.LOG oreH > TCPIP$SMTP_LOGFILE.LOG that says a given host name (or client, as it'sF > also called) was found on one of the RBL sites you specified and wasH > therefore rejected.  A lack of such a log entry would not be evidence. > E > If you have such evidence, what RBL site are you using?  I am usingsG > bl.spamcop.net, and we still receive email from IP addresses that are F > listed on spamcop.  I don't know if the problem is at spamcop or theB > SMTP software or something else.  If anyone can give me positiveG > evidence that the RBL feature works, that eliminates one source of myeD > problem.  Also, please save the log file, it may help me.  Thanks. > 6 > P.S. You may have to set the log level (logical nameE > TCPIP$SMTP_LOG_LEVEL) to 2 or more to get the required detail level  > recorded.  >w > Lawrence Bleau > University of Maryland$ > Physics Dept., Space Physics Group > 301-405-6223 > bleau@umtof.umd.eduw   ------------------------------  % Date: Thu, 27 Jun 2002 21:40:47 -0400a1 From: Michael Austin <maustin@firstdbasource.com>m1 Subject: Re: Has SMTP RBL list worked for anyone? 2 Message-ID: <3D1BBE9F.F82C2614@firstdbasource.com>   Steven Thompson wrote: > 
 > Hi Lawrencer > M > I tested RBL on my system since your first posting on this subject, which Ir > hadn't done to now.sH > I added two entries to the RBL line in my smtp.config file and my mailH > "receive logs" are now nearly all "550 Closing transmissions channel". > Something which I rarely see.  > To explain...eN > I follow the SMTP of my system quite closely, even allowing a certain laxity2 > in the configuration "just to see what happens".K > I  receive a fair quantity of junk on purpose which doesn't bother me. Myi4 > server is for my own purposes, and for learning...N > But I have added a protection in that if the system detects a suddenly largeL > quantity of files in the SMTP directory it closes down the SMTP queues and > sends me a SMS on the mobile.e  : would you care to share your RBL list in your SMTP config? -- t Regards,  6 Michael Austin            OpenVMS User since June 19847 First DBA Source, Inc.    Registered Linux User #261163F7 Sr. Consultant            http://www.firstdbasource.comaE                           http://www.firstdbasource.com/donation.htmla/ 704-947-1089 (Office)     704-236-4377 (Mobile)o   ------------------------------  % Date: Thu, 27 Jun 2002 19:32:39 +0200s2 From: martin@radiogaga.harz.de (Martin Vorlaender)5 Subject: Re: My conversation with Linus about VMS ...E; Message-ID: <3d1b4c37.524144494f47414741@radiogaga.harz.de>e  3 Atlant Schmidt (atlantnospam@mindspring.com) wrote:o > Lyndon Bartels wrote:sK > > I said that no matter what, as long as it's alpha, I can take my system,K > > disk and move it from one box to another, and it'll work. I may have to ? > > retune, but I can get a basic OS running with no changes...t > >i; > > The person's response.... "Wow,  that's *REALLY* nice."o >W( > It'd be even better if it were *TRUE*. > 'Shame it's not, ehh?i  I What exactly (besides trying to troll around) do you mean to say by that?   C With the possible exception of the network configuration, the above  statement *IS* true.   cu,-   Martin -- eB                         | Martin Vorlaender | VMS & WNT programmer1  OpenVMS: Where do you  | work: mv@pdv-systeme.denD  want to BE today?      |   http://www.pdv-systeme.de/users/martinv/8                         | home: martin@radiogaga.harz.de   ------------------------------  # Date: Thu, 27 Jun 2002 19:58:53 GMTt( From: Don Sykes <annonymous@pacbell.net>5 Subject: Re: My conversation with Linus about VMS ...W+ Message-ID: <3D1B6EF9.15E5314E@pacbell.net>a   Alfred Falk wrote: > , > Brass Christof <welcome@spam.not> wrote in" > news:3D1AA5EF.AC8B80D6@spam.not: >  > > Lyndon Bartels wrote:o > >>E > >> It's been my experience, it seems that many people are biggottedfJ > >> towards whatever operating system they "grew up on". Same seems to goF > >> for vehicles (Ford, versus Chevy, versus Dodge, versus whatever.)I > >> People build an emotional bond with it, and there's nothing that cane% > >> be said or done to disuade them.  >  > Hmmm.... Lemme see:n >         Monitor I (IBM 1620) >         KM-90 >         OS/360 (nothing like proficient, ever) >         TOPS-10s/ >         KRONOS (is that right?  CDC Cyber 70)  >         DOS-153 >         MTS (Michigan Terminal System) - UGGGGHH!i >         Unix (V3, I think)$ >         RDOS (DG Nova and Eclipse) >         RSX-11Ml >         RSTS-11e >         AOS (DG Eclipse)
 >         VMS  >         RT-11t$ >         AOS/VS (DG 32-bit Eclipse) >         CP/M >         MS-DOS# >         Windows 3.1 and its heirst > E > No, I don't think I long to return to Monitor I or KM-9, not in theoF > least.  The only OSes I might get nostalgic for are TOPS-10 and AOS,A > and they're certainly not where I began, nor where I'll end up.r > + Wow. You sound like me. In my case it was :s 	DOS (IBM 360) 	Honeywell GECOS 	Univac Exec-8
 	FourPhase 	DG EOSi 	VM/OS1c 	RSTS/E  	VMS (starting at 3.0) 	Unix (various flavors)n 	Mac (6x I think)l 	NTt 	Tru-64   M And the best designed and engineered of course was, and is, VMS. Second placeIL was probably Honeywell GECOS (way ahead of its time). The worst was probably1 DG-EOS, which didn't even include a sort utility.o  N The Unix's are lower on the list, because only the most inner kernel seemed toN be engineered at all. Instead they rely on many facilities that look like theyL were created as school projects (VI for example). Each with its own look and feel.D 	t --     Have VMS. Will Travel. Wire Paladin (@alphase.com)e
 San FranciscoA   ------------------------------  % Date: Thu, 27 Jun 2002 16:41:49 -0400t- From: JF Mezei <jfmezei.spamnot@videotron.ca>t5 Subject: Re: My conversation with Linus about VMS ...e, Message-ID: <3D1B7882.D4003090@videotron.ca>  G > > >> It's been my experience, it seems that many people are biggottedg; > > >> towards whatever operating system they "grew up on".   G To me, what makes me comfortable or not on a system is its text editor.NL Learning the commands to start/stop a queue is easy. But becoming proficient4 on the text editor, that takes some getting used to.  K I have never liked the text editors on the MAC or PC. I have liked the 3270 L based editors from IBM (SPF/ISPF, EDGAR of VM/CMS and even the MUSIC editor,C all of which were fine once you realise the limitations of the 3270lN terminals). I despised the editor on the DG AOS/VS (a long cobol program wouldL be broken up into multiple "pages" and you couldn't search across pages, andL it was like VI with some mode that isn't full screen and some mode where you& are allowed to type over the contents)  J Put TPU on Unix and I think I would get to like UNIX and become proficientL with its line commands very quickly. But remove TPU and I feel naked and notN productive when the time comes to edit files since I would be battling against/ the editor instead of focusing on the contents.o  H Now, when writing documents (as opposed to programs or system managementN stuff), then TPU is out the door and I use my MAC with page maker etc, or evenD WPSPLUS, unless I do stuff like VAXdocument where TPU is again used.  M It was an interesting challenge to try to port VMS applications to AOS-VS and J find ways around the lack of logical names. But it was a real pain to deal with its text editor.b  M Heck, even since I moved to decwindows TPU, when I get to character cell TPU,o I find It is inferior !n  L I think TPU is one of VMS' biggest assets and, like ALL-IN-1, its true power isn't fully realised.i   ------------------------------  % Date: Thu, 27 Jun 2002 20:57:04 -0500i& From: jlsue <jlsuexxxz@screaminet.com>5 Subject: Re: My conversation with Linus about VMS ... 8 Message-ID: <33enhugm7novfr8cjh11avnqtklq01b2f7@4ax.com>  2 On Thu, 27 Jun 2002 08:15:14 -0400, Atlant Schmidt$ <atlantnospam@mindspring.com> wrote:   >Lyndon Bartels wrote: >sJ >> I said that no matter what, as long as it's alpha, I can take my systemJ >> disk and move it from one box to another, and it'll work. I may have to> >> retune, but I can get a basic OS running with no changes... >>: >> The person's response.... "Wow,  that's *REALLY* nice." >e' >It'd be even better if it were *TRUE*.i >'Shame it's not, ehh? >" >Atlant  >e   Why do you make this statement?wA I KNOW that I can boot my system disk on any VMS-supported Alpha. @ I've done it and even performed off-line VMS upgrades for entire% VMSclusters by using this capability.r  1 Not speaking for anyone, certainly not DEC/Compaq-- (get rid of the xxxz in my address to e-mail):   ------------------------------  % Date: Thu, 27 Jun 2002 15:45:55 -0500h/ From: "Stuart, Ed" <Ed.Stuart@austinenergy.com>  Subject: RE: New VMS MarketingT Message-ID: <92EFB80E551BD511B39500D0B7B0CDCC0642C5A2@ohms.electric.ci.austin.tx.us>  I I've got some new OpenVMS info for you folks.  The new HP is setting up =  someH free OpenVMS system manager training courses.  I received the document = belowhH from our VAR.  I asked the VAR's contact in the new HP if I could post = it to C the group and their response was the wider the audience the better.aF Unfortunately they also informed me that the first class is probably = fullE already, but OpenVMS marketing is trying to get more classes in the =t future' and may even take the show on the road.i   EdE **Please apply a generous amount of all the usual disclaimers here.**p    3 Pilot Training for the New OpenVMS Immersion Course + 			(Also called OpenVMS Boot Camp Training)   A You are invited to attend the pilot delivery of the new OpenVMS = 	 ImmersionaA Course, an intensive, 2-week training that has two primary goals:   C =B7	To prepare students to pass the three new OpenVMS certification   exams (Exams 651, 450, and 436).C =B7	To provide useful system administration accelerated learning=20t  G Students are expected to have at least one year of OpenVMS support or =i systemF administration experience and will be expected to meet prerequisites = and to  complete post-course self-study.  I The pilot will be held at the Nashua, New Hampshire facility of Compaq, =I partE of the new HP from Monday, August 12 through Friday, August 23. The =r hours H for the course are 8:30-5:00, every day. There will be no registration = fee C to attend, but individuals are responsible for their own travel and 	 expenses.t  E The target audiences for this pilot are: OpenVMS business partners, =  OpenVMSsG customers, OpenVMS presales and service internal employees, and OpenVMSeH training instructors. We will reserve 5 slots for each target audience = soE prompt registration is important. Additional registrations beyond the E initial 20 slots will be put on a wait list and will be notified of =o theira attendance by July 31st.  I To register for this pilot, please send confirmation of your attendance =a by July 15 to:e   Kenneth BienstockD
 The New HP 110 Spit Brook Rdt Nashua, NH 03062 603-884-1151 Kenneth.Bienstock@HP.COM  B Please register only after obtaining travel approval. Additional = information ? on hotel, airline, and training logistics will be provided upona
 registration.D    + Details for the OpenVMS Immersion Course=20n  G This is a new course that is designed to provide customers and presaleshE system engineers at our selling partners with a short but intensive =e trainingI experience to help them achieve OpenVMS certifications. Students likely =o toE benefit from this course are those with previous OpenVMS training andtE experience and who need training in preparation for certification.=20h  F This training course will not have the same breadth of coverage as theF OpenVMS system management curriculum, but will instead emphasize the = parts F of the curriculum that directly support the certification exams. The = course
 will balance:e  B =B7	Enough coverage of the certification competency model for most  students to pass the three examsD =B7	Information needed to completely perform some OpenVMS system and network management functions0 =B7	Information that can be covered in two weeks  > Students taking this course are expected to meet the following prerequisites:  ? =B7	One or more years of "hands on" OpenVMS support or system =o
 management experience.-= =B7	The ability to perform the tasks covered in the OpenVMS =  Fundamentals courseB =B7	The ability to write OpenVMS command procedures to the level = coverede* in the current OpenVMS Fundamentals course@ =B7	The ability and time to complete the recommended post-courseG self-study. There are no labs in this course. However, there are labs =e that' are part of the post-course self-study.r   The goals of the course are:  $ =B7	Describe the OpenVMS environment: =B7	Understand OpenVMS operating system technical conceptsF =B7	Choose and configure OpenVMS and network configurations (LAN, LAT, TCP/IP, DECnet Plus)+ =B7	Install OpenVMS and layered products=20 D =B7	Perform day to day OpenVMS and network management tasks (TCP/IP)@ =B7	Choose, configure, and manage OpenVMS Cluster configurations( =B7	Choose and configure OpenVMS storage' =B7	Choose and configure OpenVMS Galaxy06 =B7	Trouble shoot OpenVMS (including OpenVMS Clusters)! =B7	Trouble shoot TCP/IP Servicess     Non-goals of the course are:  ) =B7	OpenVMS system performance management & =B7	OpenVMS operating system internals =B7	Programming on OpenVMSE =B7	Complete coverage of the OpenVMS system management curriculum andl" the certification competency modelH =B7	Practice taking exams that mirror the OpenVMS certification exams=20   ------------------------------  # Date: Thu, 27 Jun 2002 21:36:48 GMTh# From: "John Smith" <a@nonymous.com>  Subject: Re: New VMS MarketingF Message-ID: <QzLS8.5015$2er.1039@news01.bloor.is.net.cable.rogers.com>  ' Sounds like you have a gold-plated VAR.l  K I put the following search terms in on both the HP and Compaq web sites andl came up with zero matches:   "OpenVMS Immersion Course" "OpenVMS Boot Camp Training"          : "Stuart, Ed" <Ed.Stuart@austinenergy.com> wrote in messageL news:92EFB80E551BD511B39500D0B7B0CDCC0642C5A2@ohms.electric.ci.austin.tx.us. ..L I've got some new OpenVMS info for you folks.  The new HP is setting up someL free OpenVMS system manager training courses.  I received the document belowL from our VAR.  I asked the VAR's contact in the new HP if I could post it toC the group and their response was the wider the audience the better. I Unfortunately they also informed me that the first class is probably fullrJ already, but OpenVMS marketing is trying to get more classes in the future' and may even take the show on the road.$   EdE **Please apply a generous amount of all the usual disclaimers here.**     3 Pilot Training for the New OpenVMS Immersion Course ( (Also called OpenVMS Boot Camp Training)  I You are invited to attend the pilot delivery of the new OpenVMS ImmersionwA Course, an intensive, 2-week training that has two primary goals:   A  To prepare students to pass the three new OpenVMS certificationo  exams (Exams 651, 450, and 436).>  To provide useful system administration accelerated learning  L Students are expected to have at least one year of OpenVMS support or systemK administration experience and will be expected to meet prerequisites and tof  complete post-course self-study.  L The pilot will be held at the Nashua, New Hampshire facility of Compaq, partI of the new HP from Monday, August 12 through Friday, August 23. The hours.J for the course are 8:30-5:00, every day. There will be no registration feeC to attend, but individuals are responsible for their own travel andh	 expenses.   K The target audiences for this pilot are: OpenVMS business partners, OpenVMSrG customers, OpenVMS presales and service internal employees, and OpenVMSnI training instructors. We will reserve 5 slots for each target audience sodE prompt registration is important. Additional registrations beyond theyI initial 20 slots will be put on a wait list and will be notified of theire attendance by July 31st.  J To register for this pilot, please send confirmation of your attendance by July 15 to:e   Kenneth Bienstock 
 The New HP 110 Spit Brook Rd  Nashua, NH 03062 603-884-1151 Kenneth.Bienstock@HP.COM  L Please register only after obtaining travel approval. Additional information? on hotel, airline, and training logistics will be provided uponl
 registration.o    ( Details for the OpenVMS Immersion Course  G This is a new course that is designed to provide customers and presalesnL system engineers at our selling partners with a short but intensive trainingJ experience to help them achieve OpenVMS certifications. Students likely toE benefit from this course are those with previous OpenVMS training andeB experience and who need training in preparation for certification.  F This training course will not have the same breadth of coverage as theJ OpenVMS system management curriculum, but will instead emphasize the partsK of the curriculum that directly support the certification exams. The coursei
 will balance:   @  Enough coverage of the certification competency model for most  students to pass the three examsB  Information needed to completely perform some OpenVMS system and network management functions.  Information that can be covered in two weeks  > Students taking this course are expected to meet the following prerequisites:  F  One or more years of "hands on" OpenVMS support or system management experience.hF  The ability to perform the tasks covered in the OpenVMS Fundamentals courseF  The ability to write OpenVMS command procedures to the level covered* in the current OpenVMS Fundamentals course>  The ability and time to complete the recommended post-courseJ self-study. There are no labs in this course. However, there are labs that' are part of the post-course self-study.    The goals of the course are:  "  Describe the OpenVMS environment8  Understand OpenVMS operating system technical conceptsD  Choose and configure OpenVMS and network configurations (LAN, LAT, TCP/IP, DECnet Plus)&  Install OpenVMS and layered productsB  Perform day to day OpenVMS and network management tasks (TCP/IP)>  Choose, configure, and manage OpenVMS Cluster configurations&  Choose and configure OpenVMS storage%  Choose and configure OpenVMS Galaxyo4  Trouble shoot OpenVMS (including OpenVMS Clusters)  Trouble shoot TCP/IP Servicest     Non-goals of the course are:  '  OpenVMS system performance managementi$  OpenVMS operating system internals  Programming on OpenVMSC  Complete coverage of the OpenVMS system management curriculum ands" the certification competency modelC  Practice taking exams that mirror the OpenVMS certification exams.   ------------------------------  # Date: Fri, 28 Jun 2002 02:43:29 GMTt1 From: "Terry C. Shannon" <terryshannon@attbi.com>y Subject: Re: New VMS Marketing. Message-ID: <l3QS8.331105$352.37727@sccrnsc02>  . "John Smith" <a@nonymous.com> wrote in message@ news:QzLS8.5015$2er.1039@news01.bloor.is.net.cable.rogers.com...) > Sounds like you have a gold-plated VAR.. > I > I put the following search terms in on both the HP and Compaq web sitest and  > came up with zero matches: >  > "OpenVMS Immersion Course" > "OpenVMS Boot Camp Training" >r    Well, it's up on www.openvms.org   ------------------------------  # Date: Thu, 27 Jun 2002 21:11:43 GMTt# From: "John Smith" <a@nonymous.com>  Subject: Re: Open Letter to HPF Message-ID: <jcLS8.4912$2er.3819@news01.bloor.is.net.cable.rogers.com>  J And right now it is Intel & MS that are arguably the only ones that can do this.t  I Everyone else, including IBM, Sun, and HP are just along for the ride. (I-* know this will stir up the hornets' nest).        < "Terry C. Shannon" <terryshannon@attbi.com> wrote in message( news:KFGS8.326486$352.35599@sccrnsc02... >e% > "Andrew Harrison SUNUK Consultancy"t@ > <andrew_nospam.harrison_remove_this@sun#.com> wrote in message, > news:affc1l$b1h$1@new-usenet.uk.sun.com... >  > <snip> > >e< > > But then thats just one of the joys of Industry Standard > > Computing. >gH > There is only one real joy in "Industry Standard" Computing: Being big: > enough and wielding enough clout to set the standards... >v >u   ------------------------------    Date: 27 Jun 2002 19:08:43 -0700. From: spamsink2001@yahoo.com (Alan E. Feldman)$ Subject: Re: OVMS 7.3 BACKUP problem= Message-ID: <b096a4ee.0206271808.5f5ec7cd@posting.google.com>3  k "Ruslan R. Laishev" <Laishev@SMTP.DeltaTel.RU> wrote in message news:<3D109615.9010704@SMTP.DeltaTel.RU>... 	 > Hi All! C > I have a problem with BACKUP behaviour after upgrade to OVMS 7.3,(0 > 	/CONFIRMATION qualifier does not work anymore( > 	/SINCE=BACKUP as well does not effect >  > Just for demonstration: ; > $ back/reco/since=backup *.* sys$login:zz.bck /save /confiL > DTV3::LAISHEV 18:21:12 BACKUP    CPU=00:00:55.31 PF=24088 IO=80754 MEM=305@ >   Last file scanned: $1$DUA123:[AR_ORA4]T0001S0000009834.ARC;1: >   Saveset volume:1, saveset block:15 (32256 byte blocks)L > DTV3::LAISHEV 18:21:14 BACKUP    CPU=00:00:55.36 PF=24126 IO=80855 MEM=311@ >   Last file scanned: $1$DUA123:[AR_ORA4]T0001S0000009834.ARC;1: >   Saveset volume:1, saveset block:24 (32256 byte blocks)L > DTV3::LAISHEV 18:21:14 BACKUP    CPU=00:00:55.40 PF=24130 IO=80914 MEM=312@ >   Last file scanned: $1$DUA123:[AR_ORA4]T0001S0000009834.ARC;1: >   Saveset volume:1, saveset block:31 (32256 byte blocks) > " > $ dir /full T0001S0000009834.ARC >  > Directory $1$DUA123:[AR_ORA4]  > ? > T0001S0000009834.ARC;1                    File ID:  (220,5,0) : > Size:       137522/137550     Owner:    [ORACLE,ORACLE7]$ > Created:    7-JUN-2002 16:07:17.68( > Revised:    7-JUN-2002 16:08:03.91 (2) > Expires:   <None specified> $ > Backup:    17-JUN-2002 12:03:20.56 [snip]   2 Well, it's too late to do this directory command.   a/ > 	Is there someone who have the same problem ?- > 	Is there any secret ECOs ?0 > 	: > 	TIA.0    C I checked for this on one of my VMS v6.2 systems and my test showedrA the same problem. (I haven't applied any patches specifically fordF BACKUP.) I updated the backup date of the .DIR file and that fixed theF problem. I renamed the directory to itself and the problem recurred. IE suspect the cause of the problem is the same for you, but you'll have E to check by trying one of the following all-exciting recommendations:m  C Just before the incremental save operation, perform a "pre-emptive"  strike withl  3     $ BACKUP/RECORD [000000]AR_ORA4.DIR NL:A.B/SAVEe  A or try your original command using the /NOINCREMENTAL qualifier. e  C BTW, /SINCE=BACKUP and /CONFIRM are input file-selection qualifiers ; and should be placed immediately after the input specifier./  ? Also, I don't know why /CONF suddenly breaks in this instance. a  B Also, if you ever try to use BACKUP/INCR to restore this directoryD from this backup, it won't work right in some scenarios. You instead> have to use the method I perfected for incremental backups andE restores of a single directory. Method available upon request. If youBC instead want just a backup of files in this directory for possible,s/ later individual file restores, it's just fine.      Disclaimer: JMHO (or OCICBWYK.)  Alan E. Feldman, afeldman gfigroup com 0 "It'll work fine until the next time it breaks."   ------------------------------    Date: 28 Jun 2002 03:16:47 +0800, From: Paul Repacholi <prep@prep.synonet.com>( Subject: Re: Quorum discussion/questions- Message-ID: <87sn38ilb4.fsf@prep.synonet.com>a  0 Lyndon Bartels <lbartels@pressenter.com> writes:  F > I have some questions about possibly changing my quorum config on my
 > cluster.   > Here's the current config:  D > Two ES40s (currently v7.2-1, soon v7.3) The system disk and quroum > disk are CI based.  F > Both ES40s are voting members. So with ES40s and Qdisk, I have three > expected votes.-  -E > I'm probably going to be adding an ES45 (after v7.3) that will alsos > be a voting member.:  / > The way I see it I have two possible options:s  @ > 1. I can remove the quorom disk from the config, and leave the > expected votes at three (3).  iC > 2. I can retain the quorum disk, add the ES45 as a voting member,s< > and increase the expected votes from three (3) to four (4)   I would look at:  - Move the ES40s to 7.2-2, or leave them alone.l  F Install the ES45 with 7.3 and test etc. Boot the 40s into the new(ish)B 7.3 cluster with no QD. Or set and wait for them to fall over, but+ that tends to be a really long wait now. :)T   --  < Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda.a@                                              West Australia 6076. Raw, Cooked or Well-done, it's all half baked.F EPIC, The Architecture of the future, always has been, always will be.   ------------------------------    Date: 28 Jun 2002 02:07:44 +0800, From: Paul Repacholi <prep@prep.synonet.com>< Subject: Re: Reuters test - Itanium II blows away Sparky III- Message-ID: <87bs9wk32n.fsf@prep.synonet.com>0  / JF Mezei <jfmezei.spamnot@videotron.ca> writes:a  * > Andrew Harrison SUNUK Consultancy wrote:  ? > > Its replacement was WildFire, it is slower, it has less I/OsC > > bandwidth supports less CPU's less memory, has longer local andmE > > remote memory latency and a lower bisectional (and local now mostt; > > are NUMA) bandwidth and worse RAS than its competition.    E > Had wildwires been launched on time (under Digital, before merger),rC > wouldn't they have been much more competitive with their peers ofcC > the day ? Delaying a system/chip while your competitors are still D > advancing theirs will cause that delay to make your systems appear% > slower than originally anticipated.e  F Yep, late == slow, see `Computer Engineering' or Bell and Newell.  But? the real killer as I understand it was that the original design F outline was for everything in a Firebox was to have been on one board!C (no hint if you got a free Charles Atlas to lift it!) I did not getfE there, so the design `fell over the cliff' as it expanded with longerm# wires, more delay, more drivers,...2  C The GS and DS/ES systems seemed to be very weak in the support chiptB area for some reason. Well, perhaps not weak, but certainly out of balance with the CPU.    F > And as far as Digital having adverttised Alpha as the fastest systemC > in the world, this claim was not credible because they never suedaB > Apple or Intel for making claims that their own systems were the@ > fastest in the world. Because everyone is allowed to make that  > claim, the claim is worthless.  4 So, no news there. Only make the lawers even richer.   -- a< Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda.e@                                              West Australia 6076. Raw, Cooked or Well-done, it's all half baked.F EPIC, The Architecture of the future, always has been, always will be.   ------------------------------  # Date: Fri, 28 Jun 2002 02:23:33 GMTd  From: cjt <cheljuba@prodigy.net>< Subject: Re: Reuters test - Itanium II blows away Sparky III+ Message-ID: <3D1BC8A7.9D58ADB6@prodigy.net>    Fred Kleinsorge wrote: > $ > Nick Maclaren wrote in message ... > >t > >PI > >|> So even accepting your figures, it would seem that an Itanium-2 box 
 > could beN > >|> priced 20% higher than an Itanium-1 box, and *still* beat a Sun box that > has M > >|> twice the CPU count.  Exactly what is the upside for Sun you are tryings > to > >|> point out? > >dG > >Look, tell me any two systems selling for the same price at the sameeG > >time, and a preference, and I can produce a plausible benchmark thatsH > >will make your preferred system outperform the other by 2:1 and quite
 > >often 5:1.n > >tG > >The bias can occur in the choice of benchmark just as much as in the  > >running of it.r > >l > M > Sure.  Of course, we aren't really talking about a "benchmark" here are we?mL > But testing by a customer.  Perhaps people are excited about the fact thatN > it was significantly better than Sun - I can't really blame them.  Does thatN > mean that Sun won't be able to find some benchmark that either is slower, orL > more likely where there is no information - to claim otherwise.  Yup, I am > absolutely sure of it. > K > But when you get down to it, is it faster on *your* application workload, ? > and is it cheaper are really the only two metrics that count.m  H I'm sure if you thought about it you could add to that list of relevant @ metrics.  Things like reliability, availability, serviceability.   ------------------------------  # Date: Fri, 28 Jun 2002 03:17:52 GMTe1 From: "Terry C. Shannon" <terryshannon@attbi.com>F< Subject: Re: Reuters test - Itanium II blows away Sparky III, Message-ID: <AzQS8.17888$Uu2.3337@sccrnsc03>  9 "Paul Repacholi" <prep@prep.synonet.com> wrote in message ' news:87bs9wk32n.fsf@prep.synonet.com...h1 > JF Mezei <jfmezei.spamnot@videotron.ca> writes:V >e, > > Andrew Harrison SUNUK Consultancy wrote: >eA > > > Its replacement was WildFire, it is slower, it has less I/OcE > > > bandwidth supports less CPU's less memory, has longer local and G > > > remote memory latency and a lower bisectional (and local now mostn= > > > are NUMA) bandwidth and worse RAS than its competition.s > G > > Had wildwires been launched on time (under Digital, before merger),cE > > wouldn't they have been much more competitive with their peers ofCE > > the day ? Delaying a system/chip while your competitors are stillsF > > advancing theirs will cause that delay to make your systems appear' > > slower than originally anticipated.O >VH > Yep, late == slow, see `Computer Engineering' or Bell and Newell.  ButA > the real killer as I understand it was that the original designtH > outline was for everything in a Firebox was to have been on one board!E > (no hint if you got a free Charles Atlas to lift it!) I did not getoG > there, so the design `fell over the cliff' as it expanded with longer9% > wires, more delay, more drivers,...   G You'd have to ask the DECpaq engineers if the original goal was to cramrI everything that resides in a Firebox onto a single board. The earliest WF J stuff I saw (around 1996 IIRC) showed configs of up to 120 CPUS, though. IL no longer have the PPT files, but they showed 24-way systems lashed togetherH in trios (via extended global busses, I suppose) to form 72-way systems, etc.  L Whatever, the scheme didn't work and ASIC design proved to be a tough nut to4 crack. WF was at least 17 months late, perhaps more.   ------------------------------  # Date: Thu, 27 Jun 2002 18:16:56 GMT * From: "Bill Todd" <billtodd@metrocast.net>2 Subject: Re: RMS : preserving a RAB across $OPEN ?C Message-ID: <sEIS8.469432$Oa1.32393398@bin8.nnrp.aus1.giganews.com>e  : "JF Mezei" <jfmezei.spamnot@videotron.ca> wrote in message& news:3D1B52C0.5619AF08@videotron.ca... > danco@pebble.org wrote:rF > > If, before you closed the file, you had done a FIND by KEY using aA > > secondary key to establish a key of reference, then after youTD > > close, reopen, reconnect, and do your FIND by RFA, I don't think> > > your current key of reference will not have been restored. >oH > But if my RAB contains the former key of reference, wouldn't the first $FINDt< > after I have reopened the file set that key of reference ?   No.m   >rK > When I do a sequential $GET after a $FIND, doesn't RMS look at the key of H > reference inside the RAB or does it use only the internal hidden one ?   The latter.e   - bill   ------------------------------  # Date: Thu, 27 Jun 2002 19:50:58 GMT  From: danco@pebble.org2 Subject: Re: RMS : preserving a RAB across $OPEN ?- Message-ID: <slrnahms4l.em1.danco@pebble.org>   < In article <3D1B52C0.5619AF08@videotron.ca>, JF Mezei wrote:  N > But if my RAB contains the former key of reference, wouldn't the first $FIND< > after I have reopened the file set that key of reference ?  C The RMS Reference Manual specifies exactly what RAB fields are usednB as input for each operation.  Correct me if I'm wrong, but the keyB of reference field is not an input to a find by rfa operation.  IfC you had established a key of reference and sequentially retreived ap@ few records in that alternate key order, in order to reestablishB your context, I think you'd need to repeat the original operations@ that got you there starting from the point where you established@ your key of reference.  I suspect that in order to do that, yourC datafile access API would need to be able to 'journal' and 'replay'   some of those operations.  Yuck.  B I could be all wet on this whole topic of course!  It's been about> 12 years since I was that deeply involved in working with RMS.  ? But hey, if you're not using the key of reference feature or ife@ you won't be closing/reopening the file in the middle of reading> records sequentially using an alternate key of reference, then# this is not a problem for you.  :-)r  @ I will say that this "losing and re-establishing record context"C problem is the whole reason why, when we converted over from RSTS/ErD to VMS decades ago, we were not able to port a feature that unlocked> records after an inactivity timeout, then re-locked them later= when further activity occured.  Once we unlocked a record, weeD couldn't always get it back locked again without destroying whatever@ record context got us the record in the first place.  Just doingA a find by rfa to reestablish record context doesn't cut it in ALL = POSSIBLE cases, so a generic it always works" solution in ourw< datafile access API just wasn't possible.  So far as I know,@ the basic issues involved have not changed between then and now.   - Danr   ------------------------------  # Date: Thu, 27 Jun 2002 21:02:38 GMTc# From: "John Smith" <a@nonymous.com>o; Subject: Re: SSH v2 (was Re: OpenVMS, Volume Desktop OS...)eG Message-ID: <O3LS8.14948$MIK.6517@news01.bloor.is.net.cable.rogers.com>   2 http://www.eweek.com/article2/0,3959,284469,00.asp   Fix Is In for OpenSSH Flaw     By  Chris Gonsalvese    E A vulnerability in a popular, free implementation of the Secure ShelleF protocols that prompted a warning from the suite's developers has been quickly capped. L The vulnerability in OpenSSH versions 2.9.9 through 3.3 was the result of anE input validation error that enabled an integer overflow and privilegesC escalation, according to developers. OpenSSH, a free set of network J connectivity tools developed by the OpenBSD Project, is frequently used inI place of telnet, rlogin and ftp access and comes bundled with OpenBSD andlJ many other Unix operating systems, including the recently released Solaris 9.  K The vulnerability was first disclosed on the OpenSSH Web site Tuesday, with L a warning that users should enable privilege separation features and prepareD to upgrade to OpenSSH 3.4 on Monday, July 1. The security threat wasG detailed by Internet Security Systems researchers on Wednesday morning,f9 however, prompting an early release on the new SSH suite.r  B According to the ISS advisory, the vulnerability exists within theL "challenge-response" authentication mechanism in the OpenSSH daemon or sshd.  I "This mechanism, part of the SSH2 protocol, verifies a user's identity by1L generating a challenge and forcing the user to supply a number of responses.K It is possible for a remote attacker to send a specially-crafted reply thatdJ triggers an overflow," ISS researchers wrote. "This can result in a remoteC denial of service attack on the OpenSSH daemon or a complete remotewG compromise. The OpenSSH daemon runs with superuser privilege, so remote2F attackers can gain superuser access by exploiting this vulnerability."  L ISS researchers said they are aware of active development efforts to exploit the vulnerability.  E The OpenSSH advisory and patch is at www.openssh.org/txt/preauth.adv.   L The initial vulnerability disclosure came just days after the release of the Version 3.3 of the SSH package.e  H "We believe we have the information contained. It is after all in 27,000K lines of code," developer Theo de Raadt, founder of the OpenBSD and OpenSSHtL projects said late Tuesday. "If it does leak out, or a parallel discovery of6 it happens, we will be ready with an immediate patch."  K Even before the latest vulnerability was disclosed, OpenSSH developers havenH consistently suggested that users employ the tool's privilege separationI feature. The feature safeguards against any corruption in the sshd, which ? could lead to root compromise, according to OpenSSH developers.7  K OpenSSH encrypts all traffic, including passwords, to thwart eavesdropping,PB connection hijacking and other network-level attacks, according toK developers. In addition, OpenSSH provides secure tunneling capabilities and<$ a variety of authentication methods.  J In addition to OpenBSD and FreeBSD, OpenSSH works with dozens of operatingK systems including most flavors of Linux; NetBSD; Computone; Stallion; MacOSeE X Version 10.1; HP Procurve Switch 4108GL and 2524/2512; and IBM AIX.=   ------------------------------   Date: 27 Jun 2002 17:41:01 GMT7 From: sy18889@rabmbit.famrp.cosm (Bradford J. Hamilton)a; Subject: Re: SSH v2 (was Re: OpenVMS, Volume Desktop OS...) ! Message-ID: <UOBsAohAfXcq@rabbit>5  L The CERT article states that Process Software's implementation of SSH is not* vulnerable to the information shown below.  m In article <O3LS8.14948$MIK.6517@news01.bloor.is.net.cable.rogers.com>, "John Smith" <a@nonymous.com> writes:S4 > http://www.eweek.com/article2/0,3959,284469,00.asp >  > Fix Is In for OpenSSH Flaw >  >  > By  Chris Gonsalves0 >  > G > A vulnerability in a popular, free implementation of the Secure ShelliH > protocols that prompted a warning from the suite's developers has been > quickly capped.CN > The vulnerability in OpenSSH versions 2.9.9 through 3.3 was the result of anG > input validation error that enabled an integer overflow and privilege E > escalation, according to developers. OpenSSH, a free set of network L > connectivity tools developed by the OpenBSD Project, is frequently used inK > place of telnet, rlogin and ftp access and comes bundled with OpenBSD andrL > many other Unix operating systems, including the recently released Solaris > 9. > M > The vulnerability was first disclosed on the OpenSSH Web site Tuesday, withiN > a warning that users should enable privilege separation features and prepareF > to upgrade to OpenSSH 3.4 on Monday, July 1. The security threat wasI > detailed by Internet Security Systems researchers on Wednesday morning, ; > however, prompting an early release on the new SSH suite.d > D > According to the ISS advisory, the vulnerability exists within theN > "challenge-response" authentication mechanism in the OpenSSH daemon or sshd. > K > "This mechanism, part of the SSH2 protocol, verifies a user's identity by=N > generating a challenge and forcing the user to supply a number of responses.M > It is possible for a remote attacker to send a specially-crafted reply that L > triggers an overflow," ISS researchers wrote. "This can result in a remoteE > denial of service attack on the OpenSSH daemon or a complete remoterI > compromise. The OpenSSH daemon runs with superuser privilege, so remotesH > attackers can gain superuser access by exploiting this vulnerability." > N > ISS researchers said they are aware of active development efforts to exploit > the vulnerability. > G > The OpenSSH advisory and patch is at www.openssh.org/txt/preauth.adv.l > N > The initial vulnerability disclosure came just days after the release of the! > Version 3.3 of the SSH package.2 > J > "We believe we have the information contained. It is after all in 27,000M > lines of code," developer Theo de Raadt, founder of the OpenBSD and OpenSSHyN > projects said late Tuesday. "If it does leak out, or a parallel discovery of8 > it happens, we will be ready with an immediate patch." > M > Even before the latest vulnerability was disclosed, OpenSSH developers havedJ > consistently suggested that users employ the tool's privilege separationK > feature. The feature safeguards against any corruption in the sshd, whichiA > could lead to root compromise, according to OpenSSH developers.  > M > OpenSSH encrypts all traffic, including passwords, to thwart eavesdropping,hD > connection hijacking and other network-level attacks, according toM > developers. In addition, OpenSSH provides secure tunneling capabilities ands& > a variety of authentication methods. > L > In addition to OpenBSD and FreeBSD, OpenSSH works with dozens of operatingM > systems including most flavors of Linux; NetBSD; Computone; Stallion; MacOS(G > X Version 10.1; HP Procurve Switch 4108GL and 2524/2512; and IBM AIX.  >  >  >  -- a Bradford J. Hamilton& braMdhamAilPtoSn@aMtAtPbi.cSom		(home)& sMy1A88P89S@rabMbit.fAmPr.coSm		(work)  ; "All opinions that I express are my own, not my employer's"  "Lose the MAPS"    ------------------------------  # Date: Thu, 27 Jun 2002 18:19:11 GMTh* From: "Bill Todd" <billtodd@metrocast.net>+ Subject: Re: Sun benchmarketeering campaign-C Message-ID: <zGIS8.469445$Oa1.32395003@bin8.nnrp.aus1.giganews.com>   @ "Fred Kleinsorge" <kleinsorge@star.zko.dec.com> wrote in message, news:PBHS8.28$1U6.518303@news.cpqcorp.net...   ...k  J > The Sparc problem is that even playing compiler optimization tricks, the% > SPEC numbers for them are not good,   J A SPECint2K of 610 is 'not good'?  That's 90% of the best value a shippingI Alpha has yet managed (679), so it's hardly chopped liver - and since theoJ only suggestions I've heard about 'compiler tricks' have been in regard toK SPECfp (not SPECint) numbers, it sounds like an apples-to-apples comparisonuB of a benchmark that's considerably more applicable to server-style applications than SPECfp is.   > getting worse,  H SPARC's numbers of course have been improving, just as most others have.F Perhaps you just misspoke, and meant to say that the USIII numbers areC *about to* become less impressive due to advances by other vendors.g  K For example, Itanic seems poised to move from a SPECint2K value of 379 (now J there's a number that's *truly* 'not good':  it's not only at the absoluteE bottom of the heap but so far behind the next-worst performer that it E establishes an entire class by itself) to something comparable to theeH current Alpha number, while Alpha is poised to jump first to a SPECint2KF value of 850 with the 1.25 GHz EV68 (in August?) and then to somethingL likely in the vicinity of 1000 when EV7 appears (yeah, the officially-leakedF EV7 number was 804:  given the EV68 value just released, that seems to3 correspond to an EV7 running at 1.0, not 1.2, GHz).y  I In other words, Itanic seems likely to attain a *slight* (~10%) advantagetC over USIII, while Alpha leaves both in the dust.  Hardly a stinging L indictment of USIII unless you'd care to indict Itanic along with it, but it. does underscore Compaq's idiocy of a year ago.    and unlikely to evern > catch up to the competition.  J Oh, I don't know about that.  With Alpha development frozen at EV7, 'ever'K isn't really all that long a time.  And since USIII is already far ahead ofhL Itanic in SPECint, with the only immediate prospect being that they'll reachC rough parity, 'catching up' isn't really applicable in that context I (especially since we know that Itanic's future through the end of 2004 ispD limited to process-shrinks and additional on-chip cache, whereas SunF reportedly expects to release USIV next year with at least *some* real enhancements).  H Just trying to keep things honest, Fred.  Perhaps you really didn't knowD what the numbers were and were just regurgitating equally-uninformed' propaganda you had picked up elsewhere.l   - bill   ------------------------------    Date: 27 Jun 2002 18:37:47 -0700( From: bob@instantwhip.com (Bob Ceculski)+ Subject: Re: Sun benchmarketeering campaigne= Message-ID: <d7791aa1.0206271737.17f26e0e@posting.google.com>   u "Bill Todd" <billtodd@metrocast.net> wrote in message news:<zGIS8.469445$Oa1.32395003@bin8.nnrp.aus1.giganews.com>...cB > "Fred Kleinsorge" <kleinsorge@star.zko.dec.com> wrote in message. > news:PBHS8.28$1U6.518303@news.cpqcorp.net... >  > ...c > L > > The Sparc problem is that even playing compiler optimization tricks, the' > > SPEC numbers for them are not good,b > L > A SPECint2K of 610 is 'not good'?  That's 90% of the best value a shippingK > Alpha has yet managed (679), so it's hardly chopped liver - and since therL > only suggestions I've heard about 'compiler tricks' have been in regard toM > SPECfp (not SPECint) numbers, it sounds like an apples-to-apples comparisondD > of a benchmark that's considerably more applicable to server-style > applications than SPECfp is. >  > > getting worse, > J > SPARC's numbers of course have been improving, just as most others have.H > Perhaps you just misspoke, and meant to say that the USIII numbers areE > *about to* become less impressive due to advances by other vendors.R > M > For example, Itanic seems poised to move from a SPECint2K value of 379 (nowcL > there's a number that's *truly* 'not good':  it's not only at the absoluteG > bottom of the heap but so far behind the next-worst performer that it.G > establishes an entire class by itself) to something comparable to the J > current Alpha number, while Alpha is poised to jump first to a SPECint2KH > value of 850 with the 1.25 GHz EV68 (in August?) and then to somethingN > likely in the vicinity of 1000 when EV7 appears (yeah, the officially-leakedH > EV7 number was 804:  given the EV68 value just released, that seems to5 > correspond to an EV7 running at 1.0, not 1.2, GHz).n > K > In other words, Itanic seems likely to attain a *slight* (~10%) advantageiE > over USIII, while Alpha leaves both in the dust.  Hardly a stingingsN > indictment of USIII unless you'd care to indict Itanic along with it, but it0 > does underscore Compaq's idiocy of a year ago. >  >  and unlikely to evere  > > catch up to the competition. > L > Oh, I don't know about that.  With Alpha development frozen at EV7, 'ever'M > isn't really all that long a time.  And since USIII is already far ahead ofsN > Itanic in SPECint, with the only immediate prospect being that they'll reachE > rough parity, 'catching up' isn't really applicable in that context K > (especially since we know that Itanic's future through the end of 2004 isgF > limited to process-shrinks and additional on-chip cache, whereas SunH > reportedly expects to release USIV next year with at least *some* real > enhancements). > J > Just trying to keep things honest, Fred.  Perhaps you really didn't knowF > what the numbers were and were just regurgitating equally-uninformed) > propaganda you had picked up elsewhere.o >  > - bill  D well Bill, if you would have just read the Inquirer story, the AlphaF teams EV8-9 designs were a little late to make it onto Itanium II, butE after that look out ... then Itanium/EV8-9 will make you look bad ...c   ------------------------------  # Date: Fri, 28 Jun 2002 03:15:01 GMTe  From: cjt <cheljuba@prodigy.net>+ Subject: Re: Sun benchmarketeering campaign1+ Message-ID: <3D1BD4B7.5C9FFBEC@prodigy.net>e   Bob Ceculski wrote:e > w > "Bill Todd" <billtodd@metrocast.net> wrote in message news:<zGIS8.469445$Oa1.32395003@bin8.nnrp.aus1.giganews.com>...oD > > "Fred Kleinsorge" <kleinsorge@star.zko.dec.com> wrote in message0 > > news:PBHS8.28$1U6.518303@news.cpqcorp.net... > >n > > .... > > N > > > The Sparc problem is that even playing compiler optimization tricks, the) > > > SPEC numbers for them are not good,  > >fN > > A SPECint2K of 610 is 'not good'?  That's 90% of the best value a shippingM > > Alpha has yet managed (679), so it's hardly chopped liver - and since thetN > > only suggestions I've heard about 'compiler tricks' have been in regard toO > > SPECfp (not SPECint) numbers, it sounds like an apples-to-apples comparison:F > > of a benchmark that's considerably more applicable to server-style  > > applications than SPECfp is. > >0 > > > getting worse, > >rL > > SPARC's numbers of course have been improving, just as most others have.J > > Perhaps you just misspoke, and meant to say that the USIII numbers areG > > *about to* become less impressive due to advances by other vendors.p > >rO > > For example, Itanic seems poised to move from a SPECint2K value of 379 (now N > > there's a number that's *truly* 'not good':  it's not only at the absoluteI > > bottom of the heap but so far behind the next-worst performer that it I > > establishes an entire class by itself) to something comparable to theaL > > current Alpha number, while Alpha is poised to jump first to a SPECint2KJ > > value of 850 with the 1.25 GHz EV68 (in August?) and then to somethingP > > likely in the vicinity of 1000 when EV7 appears (yeah, the officially-leakedJ > > EV7 number was 804:  given the EV68 value just released, that seems to7 > > correspond to an EV7 running at 1.0, not 1.2, GHz).i > >tM > > In other words, Itanic seems likely to attain a *slight* (~10%) advantage G > > over USIII, while Alpha leaves both in the dust.  Hardly a stinginguP > > indictment of USIII unless you'd care to indict Itanic along with it, but it2 > > does underscore Compaq's idiocy of a year ago. > >g > >  and unlikely to evers" > > > catch up to the competition. > >oN > > Oh, I don't know about that.  With Alpha development frozen at EV7, 'ever'O > > isn't really all that long a time.  And since USIII is already far ahead ofhP > > Itanic in SPECint, with the only immediate prospect being that they'll reachG > > rough parity, 'catching up' isn't really applicable in that contextdM > > (especially since we know that Itanic's future through the end of 2004 isoH > > limited to process-shrinks and additional on-chip cache, whereas SunJ > > reportedly expects to release USIV next year with at least *some* real > > enhancements). > >aL > > Just trying to keep things honest, Fred.  Perhaps you really didn't knowH > > what the numbers were and were just regurgitating equally-uninformed+ > > propaganda you had picked up elsewhere.  > >e
 > > - bill > F > well Bill, if you would have just read the Inquirer story, the AlphaH > teams EV8-9 designs were a little late to make it onto Itanium II, butG > after that look out ... then Itanium/EV8-9 will make you look bad ...   ( Of course, by then pigs might be flying.   ------------------------------  % Date: Thu, 27 Jun 2002 21:42:15 -0400l' From: "Main, Kerry" <Kerry.Main@hp.com>v Subject: RE: TCO studyT Message-ID: <BE56C50EA024184DAF48F0B9A47F5CF4026607CF@kaoexc01.americas.cpqcorp.net>   Andrew,s  B >>> so why would Sun need to provide a result for a benchmark that" everyone accepts is very flawed.<<  4 By "everyone", I suspect you mean "everyone at Sun"?  C Lets see now, you don't like Spec benchmarks because they are a cpui measurement only.=20  F Ok, but now you don't like TPC benchmarks because you don't agree withC the way they measure system throughput even though IBM, HP, Compaq,s- Bull, Fujitsu, etc are officially ok with it?d  D >>> I guess if you think that having the worst benchmark numbers for+ TPC-C when compared with current systems<<<   D Andrew, do your homework - GS320's are number 4 and number 10 in the current TPC standings.; http://www.ideasinternational.com/benchmark/tpc/tpccnc.htmlh   :-)   D As far as declining market share goes, I keep emphasizng to you thatE those in glass houses should not throw rocks. A few examples from CNNd and VAR Business magazines:l  H http://money.cnn.com/2002/04/29/commentary/bottomline/lashinsky/index.ht maD More trouble for Sun Micro - Sun will lose out if it keeps trying to
 battle Intel.r  J http://www.varbusiness.com/sections/columns/columns.asp?ArticleID=3D35611=  / "It's Time For You to Get Out of the Sun Market E Sun's channel is in serious trouble; many loyalists switch platforms,  focus"   Regardsa  
 Kerry Main Senior Consultanti Hewlett-Packard Canada! Consulting & Integration Servicese Voice: 613-592-4660p Fax   : 613-591-4477 Email: Kerry.Main@hp.com     -----Original Message-----' From: Andrew Harrison SUNUK Consultancyb7 [mailto:andrew_nospam.harrison_remove_this@sun#.com]=20  Sent: June 27, 2002 8:25 AMe To: Info-VAX@Mvb.Saic.Comh Subject: Re: TCO study         Main, Kerry wrote:  E >>>>Of course they will, but to customers buying your systems now itsS >>>>3 > what they are capable of now thats important ..<<p >=205 > Yep, and that's why Customers continue to buy them.m >=20 >=20    4 Really, so declining from 11% of the Market to 5% is3 a victory because 5% still means that customers are  still buying systems !  4 I guess if you think that having the worst benchmark4 numbers for TPC-C when compared with current systems7 that you categorise as being the same size as the GS320p6 is acceptable then a mere halving of your market share is also great.    H >>>>No they don't sorry, the GS320's numbers are quite frankly terrible. >>>>J > Worse than Fujitsu and more expensive, worse than IBM (P690) and more=20H > expensive, worse than SuperDome and more expensive, almost the same as  # > a IBM P680 and more expensive.<<<  >=20? > Yada, yada, yada - let me know when Sun does a TPC benchmark.  >=20  9 I don't need to I have 3 very acceptable benchmarks and 6=  E very acceptable benchmark results, so why would Sun need to provide am< result for a benchmark that everyone accepts is very flawed.   Regards  Andrew Harrison    ------------------------------  # Date: Fri, 28 Jun 2002 00:45:56 GMTt+ From: "Christian C Ekstam" <md0@berusad.nu>a Subject: TCP/IPt5 Message-ID: <8lOS8.13745$p56.4387877@newsb.telia.net>c  I Is there any free TCP/IP stacks around to bee used on a VMS 5.5-2 system?<  	 Christianm   ------------------------------  % Date: Thu, 27 Jun 2002 18:37:13 -0700 + From: "Barry Treahy, Jr." <Treahy@mmaz.com>o Subject: Re: TCP/IP ' Message-ID: <3D1BBDC9.2000607@mmaz.com>t   Christian C Ekstam wrote:   J >Is there any free TCP/IP stacks around to bee used on a VMS 5.5-2 system? >  a >dC Yes, there is and we used it for some time but note that it is not  ' without its own problems.   The URL is:a4     http://www.agh.cc.kcl.ac.uk/files/vms/cmu-tcpip/D but consider that if you are asking for hobbyist reasons, there are H commercial stacks that are available for hobbyists.  If you are seeking C a stack for commercial reasons, CMU is a short term bridge, but be  B prepared for some hassles and headaches.  We ultimately went with E Process's TCPware and do not regret it in the least.  Not only is it e stable, but affordable.,   Regards,   Barryh   -- s  @ Barry Treahy, Jr  *  Midwest Microwave  *  Vice President & CIO   A E-mail: Treahy@mmaz.com * Phone: 480/314-1320 * FAX: 480/661-7028l   ------------------------------  % Date: Thu, 27 Jun 2002 22:35:24 -0400e- From: JF Mezei <jfmezei.spamnot@videotron.ca>i Subject: Re: TCP/IP , Message-ID: <3D1BCB6B.352F60F1@videotron.ca>   Christian C Ekstam wrote:o > K > Is there any free TCP/IP stacks around to bee used on a VMS 5.5-2 system?s  N CMU-IP works fine on VAX-VMS. I think it is available on the freeware CD whichL is on the vms web site. (www.openvms.com)  It is however a bit old and newer' software isn't compiled for it anymore.c  L If you are using this system at home, look for the VMS hobbysist licences atM www.montagar.com and tyhen you can get VMS 7.2 with the TCPIP V5.0 stack thateN is somewhat more complete than CMU-IP (although it lacks WHOIS which CMU has).   ------------------------------  # Date: Fri, 28 Jun 2002 03:14:49 GMTi+ From: "Christian C Ekstam" <md0@berusad.nu>r Subject: Re: TCP/IPn5 Message-ID: <JwQS8.13749$p56.4390387@newsb.telia.net>p  E Thank you very much for your post, do you have any URL to a page thathL describes the basics of running a VMS system? Something in the style of "VMS for dummies" =)e   Cinda new to this.    	 Christianh? (My friend just got his hands on a VAX 4000-300 and a VLC 4000)    ------------------------------   Date: 28 Jun 2002 04:30:07 GMT2 From: "Zane H. Healy" <healyzh@shell1.aracnet.com> Subject: Re: TCP/IPc, Message-ID: <afgoof01q9d@enews4.newsguy.com>  * Christian C Ekstam <md0@berusad.nu> wrote:G > Thank you very much for your post, do you have any URL to a page thatyN > describes the basics of running a VMS system? Something in the style of "VMS > for dummies" =)i  # http://www.openvms.compaq.com:8000/   L While intended for people using emulators, I've got pointers that are likely to be useful to you at:n+ http://www.aracnet.com/~healyzh/decemu.html ( http://www.aracnet.com/~healyzh/vax.html   			Zanes   ------------------------------   Date: 27 JUN 2002 16:13:16 GMT4 From: karcher@thuria.waisman.wisc.edu (Carl Karcher)' Subject: Re: Trouble with BACKUP/RECORDD6 Message-ID: <27JUN02.16131630@thuria.waisman.wisc.edu>  C In a previous article, Roland Barmettler <rob@bbp.ch.remove> wrote:a   ->OK, here's the log (7.3):u -> w1 ->NABOO:$ BACKUP/RECORD/LOG *.TXT;* TEST.BCK/SAVE C ->%BACKUP-S-COPIED, copied DSA0:[SYS0.][SYSMGR]ANNOUNCE_ALCOR.TXT;2 D ->%BACKUP-S-COPIED, copied DSA0:[SYS0.][SYSMGR]ANNOUNCE_NABOO.TXT;12 ->%BACKUP-S-COPIED, copied= ->DSA0:[SYS0.][SYSMGR]DECNET_REGISTER_IMPORT_FILE_NABOO.TXT;1e ->%BACKUP-S-COPIED, copied: ->DSA0:[SYS0.SYSCOMMON.][SYSMGR]TCPIP$BINDSETUP_HELP.TXT;1F ->%BACKUP-S-COPIED, copied DSA0:[SYS0.SYSCOMMON.][SYSMGR]WELCOME.TXT;1< ->%BACKUP-I-STARTRECORD, starting backup date recording pass -> ... zA ->NABOO:$ BACKUP/NORECORD/LOG *.TXT;*/SINCE=BACKUP TEST1.BCK/SAVE.C ->%BACKUP-S-COPIED, copied DSA0:[SYS0.][SYSMGR]ANNOUNCE_ALCOR.TXT;2pD ->%BACKUP-S-COPIED, copied DSA0:[SYS0.][SYSMGR]ANNOUNCE_NABOO.TXT;12 ->%BACKUP-S-COPIED, copied= ->DSA0:[SYS0.][SYSMGR]DECNET_REGISTER_IMPORT_FILE_NABOO.TXT;1e ->%BACKUP-S-COPIED, copied: ->DSA0:[SYS0.SYSCOMMON.][SYSMGR]TCPIP$BINDSETUP_HELP.TXT;1F ->%BACKUP-S-COPIED, copied DSA0:[SYS0.SYSCOMMON.][SYSMGR]WELCOME.TXT;1 -> s
 ->*** BUT: -> l@ ->NABOO:$ BACKUP/NORECORD/LOG *.TXT;*/SINCE=27-JUN-2002:09:28:18 ->TEST1.BCK/SAVEG ->%BACKUP-W-NOFILES, no files selected from DSA0:[SYS0.][SYSMGR]*.TXT;*b+ ->%BACKUP-W-NOFILES, no files selected froms' ->DSA0:[SYS0.SYSCOMMON.][SYSMGR]*.TXT;*, -> m2 ->I still don't get it...  what am I doing wrong ? ->    F I see the same behavior on VMS 7.2-1 with VMS721_BACKUP-V0100 applied.< If you use /NOINCREMENTAL with /SINCE=BACKUP you'll have theC expected behavior. I think it has to do with your default directoryeH being a search list. Try a SET DEFAULT SYS$COMMON:[SYSMGR] (not a search@ list) and repeat the test. Or if you specify "sys$manager:*.txt"9 as the input spec on the backup command it will work too.i  > More more info on why /NOINCREMENTAL works see (wrapped URL!):  N http://groups.google.com/groups?q=karcher+/noincremental+group:comp.os.vms&hl=M en&lr=&ie=UTF-8&oe=UTF8&selm=8MAR02.14395594%40thuria.waisman.wisc.edu&rnum=1t  G This doesn't fully explain it since the revision dates on the directoryhF files haven't changed here. It's something about the default directory being a search list.   --G -- Carl Karcher, Waisman Computing Services, Waisman Center, UW-Madisonc9 --                      karcher.nospam@waisman.wisc.edu      ------------------------------    Date: 27 Jun 2002 15:45:34 -0700. From: SPAMSINK2001@YAHOO.COM (Alan E. Feldman)' Subject: Re: Trouble with BACKUP/RECORD = Message-ID: <343f30ae.0206271445.487fb7ad@posting.google.com>a  j Roland Barmettler <rob@bbp.ch.remove> wrote in message news:<20020627111935.54fbd1da.rob@bbp.ch.remove>... > Rudolf Wingert schrieb:l
 > > Hello, > > E > > the show time command is irrelevant. Please look at the creation,eG > > record, etc. date and time. me be some in the future. So this won't, > > be backuped. > C > The backup date in the file header reads: 27-JUN-2002 09:27:22.22cE > which is clearly _before_ 27-JUN-2002 09:28:18 when I did the "SHOWy= > TIME" after the backup/record. That's what's puzzling me...     R This is normal. The date in the backup field is the time that the backup started.   + Just a guess: Try /MODIFIED/SINCE=BACKUP !.-   Disclaimer: JMHO Alan E. FeldmanC afeldman gfigroup comS0 "It'll work fine until the next time it breaks."   ------------------------------    Date: 27 Jun 2002 15:55:22 -0700. From: SPAMSINK2001@YAHOO.COM (Alan E. Feldman)' Subject: Re: Trouble with BACKUP/RECORDo= Message-ID: <343f30ae.0206271455.484b1fe6@posting.google.com>   j Roland Barmettler <rob@bbp.ch.remove> wrote in message news:<20020626114748.759b164d.rob@bbp.ch.remove>...	 > Hi Guysm > > > I have a problem with BACKUP/RECORD or rather /SINCE=BACKUP. >  > If I do a:+ > $ BACKUP/RECORD/LOG *.TXT;* TEST.BCK/SAVEe > C > and have a look at the backup date in the file header afterwards,p( > it's correctly recorded (as expected). >  > But if I then do a:i2 > $ BACKUP/LOG *.TXT;*/SINCE=BACKUP TEST1.BCK/SAVE > > > I still get all the files backed-up. /SINCE=yesterday works,7 > /SINCE=<timestamp> works, only /SINCE=BACKUP doesn't.r >   > This happens on 7.2-1 and 7.3. > What am I missing ?,    B Is this just a test case to illustrate the example? Are you really, having problems with a command sequence like  = $ BACKUP/LOG/RECORD [TOP...]*.TXT MKA0:ALLTXT.BCK/BLOCK=32256p   <wait one day>  6 $ BACKUP/LOG [TOP...]*.TXT MKA0:INCTXT.BCK/BLOCK=32256  F IIRC, this came up before in this newsgroup and the problem was tracedD to the fact that the first command *does not* update the backup dateD field of TOP.DIR. Then when you run the second command, it sees thatE TOP.DIR's backup date is too old and backs up everything in the tree.n" Could that be what you are seeing?   The cure would beS  1 $ BACKUP/LOG/RECORD [000000]TOP.DIR,[TOP...]*.TXTu MKA0:ALLTXT.BCK/BLOCK=32256    <wait one day>  > $ BACKUP/LOG/RECORD [000000]TOP.DIR,[TOP...]*.TXT/SINCE=BACKUP MKA0:INCTXT.BCK/BLOCK=32256E  A BTW, if you ever try to restore this tree, you must do it exactlygB right are you might not get the right results. When restoring, youF must let the first restore create the directory TOP.DIR. Otherwise theC backup date in a "leftover" TOP.DIR may confuse BACKUP/INCREMENTAL.y   Disclaimer: JMHO Alan E. Feldman  afeldman gfigroup comr0 "It'll work fine until the next time it breaks."   ------------------------------    Date: 27 Jun 2002 21:35:28 -0700. From: spamsink2001@yahoo.com (Alan E. Feldman)' Subject: Re: Trouble with BACKUP/RECORDo= Message-ID: <b096a4ee.0206272035.53e32813@posting.google.com>i  s SPAMSINK2001@YAHOO.COM (Alan E. Feldman) wrote in message news:<343f30ae.0206271455.484b1fe6@posting.google.com>...al > Roland Barmettler <rob@bbp.ch.remove> wrote in message news:<20020626114748.759b164d.rob@bbp.ch.remove>... > > Hi Guysr > > @ > > I have a problem with BACKUP/RECORD or rather /SINCE=BACKUP. > >  > > If I do a:- > > $ BACKUP/RECORD/LOG *.TXT;* TEST.BCK/SAVE  > > E > > and have a look at the backup date in the file header afterwards, * > > it's correctly recorded (as expected). > >  > > But if I then do a:r4 > > $ BACKUP/LOG *.TXT;*/SINCE=BACKUP TEST1.BCK/SAVE > > @ > > I still get all the files backed-up. /SINCE=yesterday works,9 > > /SINCE=<timestamp> works, only /SINCE=BACKUP doesn't.i > > " > > This happens on 7.2-1 and 7.3. > > What am I missing ?u >  > D > Is this just a test case to illustrate the example? Are you really. > having problems with a command sequence like > ? > $ BACKUP/LOG/RECORD [TOP...]*.TXT MKA0:ALLTXT.BCK/BLOCK=32256F >  > <wait one day> > 8 > $ BACKUP/LOG [TOP...]*.TXT MKA0:INCTXT.BCK/BLOCK=32256     Here I meant  E   $ BACKUP/LOG [TOP...]*.TXT/SINCE=BACKUP MKA0:INCTXT.BCK/BLOCK=32256F    H > IIRC, this came up before in this newsgroup and the problem was tracedF > to the fact that the first command *does not* update the backup dateF > field of TOP.DIR. Then when you run the second command, it sees thatG > TOP.DIR's backup date is too old and backs up everything in the tree.t$ > Could that be what you are seeing? >  > The cure would be  > 3 > $ BACKUP/LOG/RECORD [000000]TOP.DIR,[TOP...]*.TXTc > MKA0:ALLTXT.BCK/BLOCK=32256v >  > <wait one day> > @ > $ BACKUP/LOG/RECORD [000000]TOP.DIR,[TOP...]*.TXT/SINCE=BACKUP > MKA0:INCTXT.BCK/BLOCK=32256s > C > BTW, if you ever try to restore this tree, you must do it exactlyeD > right are you might not get the right results. When restoring, youH > must let the first restore create the directory TOP.DIR. Otherwise theE > backup date in a "leftover" TOP.DIR may confuse BACKUP/INCREMENTAL.U >  > Disclaimer: JMHO > Alan E. Feldmans > afeldman gfigroup come2 > "It'll work fine until the next time it breaks."    Disclaimer: JMHOf  Alan E. Feldman  afeldman gfigroup com1  "It'll work fine until the next time it breaks."    ------------------------------  % Date: Thu, 27 Jun 2002 21:24:53 -0400sH From: "Stephen Eickhoff (remove dash below)" <seickhoff@mind-spring.com>% Subject: Re: Trouble with Samba 2.0.3h. Message-ID: <3D1BBAE5.6080807@mind-spring.com>   John E. Malmberg wrote:8- > Stephen Eickhoff (remove dash below) wrote:  > J >> I've managed to get the client part working, but what i really need is H >> the smbd. When I run smbd_setup_tcpip.com, it starts up the service. E >> However, only a few seconds later the service terminates. The log ) >> looks like this:r >  > G > I am guessing at the version of OpenVMS, and what TCP/IP product and I > stack you are using: > K > See the SAMBA-VMS FAQ, posted on Encompasserve, also available searching i > http://www.google.comt >  > [Url is wrapped]( > http://eisner.encompasserve.org/htbin/U > dnqindexform?TEXT=R30328472-30368736-dra4%3A%5Bdecuserve_extracts%5Dvms.full%3B415   >  >  > 
 > SAMBA22. > : > SAMBA will not start and I am running Compaq TCP/IP 5.1. > F >  The following fix needs to be added to the SMBD_SETUP_TCPIP.COM and( >      possibly the SMBD_SETUP_TCPIP.COM > ! >      $ tcpip set service smbd -  >       /protocol=TCP -y >       /port=139 -o4 >       /flags=listen -   ! This is the change <<<<< >       /user=SYSTEM - >       /process=SMBD -p0 >       /file=SAMBA_ROOT:[BIN]SMBD_STARTUP.COM -: >       /log=(FILE:SAMBA_ROOT:[VAR]SMBD_STARTUP.LOG,ALL) - >       /limit=100  K Happens to be VMS 7.3, TCPIP 5.1. That fixed it, although I still get this:h  < %TCPIP-E-CONFIGERROR, error processing configuration request' -SYSTEM-F-BADPARAM, bad parameter value   E I also still can't log in, it takes forever to get the logon box and l( doesn't accept my username and password.F At least the service still keeps running. I'm sure I have an error in , smb.conf or something. Thanks for your help.   ------------------------------  # Date: Fri, 28 Jun 2002 02:10:45 GMTg- From: "John E. Malmberg" <wb8tyw@qsl.network>-% Subject: Re: Trouble with Samba 2.0.3 * Message-ID: <3D1BC198.9020902@qsl.network>  + Stephen Eickhoff (remove dash below) wrote:C >  2M > Happens to be VMS 7.3, TCPIP 5.1. That fixed it, although I still get this:  > > > %TCPIP-E-CONFIGERROR, error processing configuration request) > -SYSTEM-F-BADPARAM, bad parameter valueo > G > I also still can't log in, it takes forever to get the logon box and e* > doesn't accept my username and password.H > At least the service still keeps running. I'm sure I have an error in . > smb.conf or something. Thanks for your help.  6 I have never used SAMBA 2.0.3, so I am still guessing.  E I would recommend seeing if the SMBCLIENT program can connect to the r
 SMBD program.m  A Also be aware that you need to be at SAMBA 2.0.6 for OpenVMS for y encrypted passwords to work.  G For SAMBA 2.0.3 and earlier for OpenVMS, you need to disable encrypted t  passwords on your LANMAN client.   -John  wb8tyw@qsl.network Personal Opinion OnlyI   ------------------------------  % Date: Thu, 27 Jun 2002 23:19:56 -0400 - From: JF Mezei <jfmezei.spamnot@videotron.ca>z Subject: Xwindows: XDM modem, Message-ID: <3D1BD5D9.D7C804A6@videotron.ca>  L On my MAC, the X windows emulator has a setting for XDM mode of none, query, indirect or broadcast.   What does this mean ?c   ------------------------------   End of INFO-VAX 2002.352 ************************arket E Sun's channel is in serious trouble; many loyalists switch platforms,  focus"   Regardsa  
 Kerry Main Senior Consultanti Hewlett-Packard Canada! Consulting & Integration Servicese Voice: 613-592-4660p Fax   : 613-591-4477 EmU    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    U    