1 INFO-VAX	Sun, 29 Apr 2001	Volume 2001 : Issue 239       Contents: Re: Compaq and SSH Re: Compaq and SSH Re: DFWDAYS a Big Success G Re: Hackers crack challenge within 24 hours, and Solaris gets the blame G Re: Hackers crack challenge within 24 hours, and Solaris gets the blame G Re: Hackers crack challenge within 24 hours, and Solaris gets the blame  Re: Limit on telnet sessions.  Problem in submit ?  Re: Process memory-usage on VMS  Re: Process memory-usage on VMS 6 Re: Question: QLogic QLA1080 SCSI Controller & OpenVMS> Re: Ripping ods-2 cd on AlphaOpenVMS burning ods-2 cd in Win98N Re: seeking performance numbers for distributed lock manager traffic via  FDDIM Re: seeking performance numbers for distributed lock manager traffic via FDDI  Re: Sync Comm - PCI Card Windows-XP on Alpha ?? Re: Windows-XP on Alpha ?? Re: Windows-XP on Alpha ?? Re: Windows-XP on Alpha ?? Re: Windows-XP on Alpha ?? Re: Windows-XP on Alpha ??  F ----------------------------------------------------------------------  # Date: Sun, 29 Apr 2001 10:58:24 GMT ( From: David Webb <nospam@newsranger.com> Subject: Re: Compaq and SSH 6 Message-ID: <ktSG6.2770$SZ5.211329@www.newsranger.com>  B In article <3AE9563C.613DE6A6@uiowa.edu>, Richard L. Dyson says... >  >Mitchell, David R. wrote: >>  M >> It looks like Compaq has licensed secure shell for integration with Tru64.  >>  7 >> http://www.ssh.com/about/press/2001/2001-04-10B.html  >>  	 >> <snip>  > @ >	Durning a recent teleconference sponsored by Compaq/OpenVMS (aF >playback is available from the OpenVMS web page) I asked specifically >aboutJ >SSH at the server and client level.  IMO, their answer was that they did D >not think it was important to customers, but would be interested inH >hearing from customers if they were interested in getting SSH available* >with the network products (DECnet/TCPIP). >  >	Let them know. :)  >   M I asked the VMS Engineers on the stand at Infosec and they said SSH version 2  was in VMS 7.3. P I haven't  seen the list of new features for VMS 7.3 but trust they were telling
 me the truth.   M I had been asking for this at every opportunity for the last couple of years. : Dave Jones' SSH server with OpenSSL is great but we reallyI need a fully supported SSH server. Plus Dave Jones' server is only an SSH  version 1 server.   L I also asked about Secure FTP and was assured that this was also in VMS 7.3.  M (Note. SSH version 3 has only just been released so if SSH is in VMS 7.3 then 1 they would not have had time to release a version 
 3 server).  M It is possible they may not have provided an SSH client for VMS (I didn't ask ; explicitly about an SSH client) as that might be perceived  K as less important to most customers. (Though I do not share that position).   
 David Webb VMS and Unix  team leader  CCSS Middlesex University         >Rick  >-- I >Richard L. Dyson                                    rick-dyson@uiowa.edu I > _   _      _____                http://www-pi.physics.uiowa.edu/~dyson/ I >| | | |    |_   _|   Systems Analyst                     O: 319/335-1879 I >| | | | of   | |     The University of Iowa            FAX: 319/335-1753 7 >| \_/ |     _| |_    Department of Physics & Astronomy . > \___/     |_____|   Iowa City, IA 52242-1479   ------------------------------  # Date: Sun, 29 Apr 2001 15:03:29 GMT - From: goathunter@goatley.com (Hunter Goatley)  Subject: Re: Compaq and SSH 1 Message-ID: <3aec2bea.222824003@swen.process.com>   K On Sun, 29 Apr 2001 10:58:24 GMT, David Webb <nospam@newsranger.com> wrote:  > N >I asked the VMS Engineers on the stand at Infosec and they said SSH version 2 >was in VMS 7.3.Q >I haven't  seen the list of new features for VMS 7.3 but trust they were telling  >me the truth. > F SSH is not in VMS V7.3.  Even if Compaq added it, it would be added to! TCP/IP Services, not to base VMS.   M >I also asked about Secure FTP and was assured that this was also in VMS 7.3.  > 	 It's not.   H I haven't seen the final V7.3 release yet, but was a field test site forF it.  You can also check out the V7.3 feature list.  No mention of SSH.  9 http://www.openvms.compaq.com/openvms/os/v73features.html   C As has been stated before in this thread, both MultiNet and TCPware D include SSH V1 servers and clients.  Work is underway for SSH V2 and Secure Copy.   Hunter ------9 Hunter Goatley, Process Software, http://www.process.com/ 9 goathunter@goatley.com     http://www.goatley.com/hunter/    ------------------------------    Date: 29 Apr 2001 17:05:21 +0800, From: Paul Repacholi <prep@prep.synonet.com>" Subject: Re: DFWDAYS a Big Success- Message-ID: <87g0es2i0e.fsf@prep.synonet.com>   ' John Wisniewski <user@user.com> writes:   > > And I hope to have some of the censored Pictures and the PPTC > Presentations up on the web by the end of the weekend... Look for  > them on the CUG homepage...   B Puka Puke?!! On a DECUS page? Who do you think you are? Compaq? ;)   Goodbye level 44 -more-    --  < Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda. @                                              West Australia 6076. Raw, Cooked or Well-done, it's all half baked.H Spam-To: uce@ftc.gov,enforcement@sec.gov,sness@fcc.gov,hfurchtg@fcc.gov,#   mpowell@fcc.gov,gtristan@fcc.gov     ------------------------------  # Date: Sun, 29 Apr 2001 10:39:59 GMT ( From: David Webb <nospam@newsranger.com>P Subject: Re: Hackers crack challenge within 24 hours, and Solaris gets the blame6 Message-ID: <3cSG6.2767$SZ5.210736@www.newsranger.com>  K In article <P5DfWF8WteVq@eisner.encompasserve.org>, Larry Kilgallen says...  > l >In article <OFD3901F94.6BD59F74-ON80256A3B.0059FE6C@qedi.quintiles.com>, steven.reece@quintiles.com writes: >>   >>   > F >Perhaps someone from this group should contact Argus and suggest some, >different operating system for them to use. >     E Not much point. The Argus Pitbull products are designed to reduce the ' insecurities in Solaris, AIX and Linux.    They like to quote        @ "The threats posed by the modern computing environment cannot beE addressed without secure operating systems. Any security effort which D ignores this fact can only result in a 'fortress built upon sand'."   6 The Inevitability of Failure: The Flawed Assumption of) Security in Modern Computing Environments  The National Security Agency        in their advertising literature.  M Although some parts of the Pitbull product set might be applicable to VMS (eg . they provide B1 as opposed to C2 security withK features such as Mandatory access Control) much of the rest eg Extensive OS O source code auditing, measures to eliminate/reduce the security implications of L buffer overflow problems, replacement of the traditional all or nothing Unix  privilege paradigm with multipleN privileges  etc are already done by VMS Engineering. The B1 level security can$ also of course be provided by SEVMS.  M Like any product it will only sell if there is a perceived need. I can easily < see the need for a Solaris version of this product but would. not see nearly as much need for a VMS version.    
 David Webb VMS and Unix team leader CCSS Middlesex University   ------------------------------    Date: 29 Apr 2001 07:31:51 -05009 From: Kilgallen@eisner.decus.org.nospam (Larry Kilgallen) P Subject: Re: Hackers crack challenge within 24 hours, and Solaris gets the blame3 Message-ID: <2j+ky6kTrcxx@eisner.encompasserve.org>   a In article <3cSG6.2767$SZ5.210736@www.newsranger.com>, David Webb <nospam@newsranger.com> writes:   O > Although some parts of the Pitbull product set might be applicable to VMS (eg 0 > they provide B1 as opposed to C2 security withM > features such as Mandatory access Control) much of the rest eg Extensive OS Q > source code auditing, measures to eliminate/reduce the security implications of N > buffer overflow problems, replacement of the traditional all or nothing Unix" > privilege paradigm with multipleP > privileges  etc are already done by VMS Engineering. The B1 level security can& > also of course be provided by SEVMS. > O > Like any product it will only sell if there is a perceived need. I can easily > > see the need for a Solaris version of this product but would0 > not see nearly as much need for a VMS version.  G So what is wrong with "Trusted Solaris" (evaluated for MAC) in place of 	 Pitbull ?   N ==============================================================================N Great Inventors of our time: Al Gore -> Internet; Sun Microsystems -> ClustersN ==============================================================================   ------------------------------  # Date: Sun, 29 Apr 2001 16:08:35 GMT ( From: David Webb <nospam@newsranger.com>P Subject: Re: Hackers crack challenge within 24 hours, and Solaris gets the blame6 Message-ID: <70XG6.2973$SZ5.228517@www.newsranger.com>  K In article <2j+ky6kTrcxx@eisner.encompasserve.org>, Larry Kilgallen says...  > b >In article <3cSG6.2767$SZ5.210736@www.newsranger.com>, David Webb <nospam@newsranger.com> writes: > P >> Although some parts of the Pitbull product set might be applicable to VMS (eg1 >> they provide B1 as opposed to C2 security with N >> features such as Mandatory access Control) much of the rest eg Extensive OSR >> source code auditing, measures to eliminate/reduce the security implications ofO >> buffer overflow problems, replacement of the traditional all or nothing Unix # >> privilege paradigm with multiple Q >> privileges  etc are already done by VMS Engineering. The B1 level security can ' >> also of course be provided by SEVMS.  >>  P >> Like any product it will only sell if there is a perceived need. I can easily? >> see the need for a Solaris version of this product but would 1 >> not see nearly as much need for a VMS version.  > H >So what is wrong with "Trusted Solaris" (evaluated for MAC) in place of
 >Pitbull ? >   : I believe Trusted Solaris just provides B1 level security.O The fact that an OS can gain a C2 or B1 (or equivalent) evaluation only implies 2 that the OS possesses the appropriate auditing andL access control mechanisms. It does not have any implications for how it will( handle buffer overflow or other attacks.  L This is where Pitbull and similar products (I saw at least one other company9 with a similar product at Infosec) come in. They attempt  M to increase the resilience of the OS above that provided by the OS supplier.      P In an ideal world all OS suppliers should be equally concerned over the security6 quality of their products. They should all be writing P their OS's  to incorporate the very things that Pitbull provides.  Unfortunately# we live in a far from ideal world.      
 David Webb VMS and Unix team leader CCSS Middlesex University   ------------------------------  % Date: Sun, 29 Apr 2001 09:32:35 +0200 2 From: martin@radiogaga.harz.de (Martin Vorlaender)& Subject: Re: Limit on telnet sessions.; Message-ID: <3aebc393.524144494f47414741@radiogaga.harz.de>   $ jester (joedutro@hotmail.com) wrote:J > Does anyone know where I can limit the number of telnet sessions in VMS?  H It depends on the TCP/IP stack you're running. In Compaq TCP/IP Services. (nee Digital TCP/IP Services, nee UCX) you can   $ TCPIP  ! or $ UCX if pre-V5 1 TCPIP|UCX> SET SERVICE TELNET /LIMIT=<max number> ! TCPIP|UCX> DISABLE SERVICE TELNET   TCPIP|UCX> ENABLE SERVICE TELNET  < Note that the DISABLE command will kill all telnet sessions!  # You can see the current value using   $ TCPIP|UCX> SHOW SERVICE TELNET /FULL   cu,    Martin --  D                         | Martin Vorlaender  |  VMS & WNT programmer1  VMS is today what      | work: mv@pdv-systeme.de E  Microsoft wants        |    http://www.pdv-systeme.de/users/martinv/ 8  Windows NT 8.0 to be!  | home: martin@radiogaga.harz.de   ------------------------------  % Date: Sun, 29 Apr 2001 13:39:37 +0200 5 From: Christophe Evrard <christophe.evrard@skynet.be>  Subject: Problem in submit ?( Message-ID: <3AEBFD79.3080409@skynet.be>   Hello,  I We've recently had a weird problem, unfortunately not reproductible, but  2 which reproduces periodically (2-3 times on 1000).I The problem occurs in a submit instruction with 2 jobs submitted (submit   job_a, job_b).F The log shows that job_a completely finishes (last line of job_a is a I display - showed in the log), and that job_b didn't start (first line is   a display). L Instead of showing the execution of job_b, the log contains a register dump.I That makes me think that the problem is located at system level, and not  ( so much depending of the jobs submitted.   Any good explanation ?   Thanks.    Christophe.    ------------------------------  % Date: Sun, 29 Apr 2001 17:18:06 +0200 , From: "Martie Ressing" <m.ressing@planet.nl>( Subject: Re: Process memory-usage on VMS, Message-ID: <9chb82$5sfh1$1@reader02.wxs.nl>  	 Valentin,   > when the program stops, we don't get to see any error-message.G So it's a guess that the problem is concerning memory (that's the first  thing I thougt about).A It could also be a problem concerning other process quota (as you  mentioned).    Martie.     C "Valentin Likoum" <valentin.likoum@ncc.volga.ru> schreef in bericht . news:7110051732.20010428105100@ncc.volga.ru...; > On 28.04.2001 Martie Ressing <m.ressing@planet.nl> wrote:  > 
 > > Hello, > G > > I've got a problem concerning jobs seemingly running out of memory. K > > We are running a program (on Alpha/VMS) developed in PowerHouse (a 4GL) * > > which call's, repeatedly, a C-program.A > > We think the C-program doesn't free the memory it uses, after A > > returning to the calling program, because after some time the K > > program just quits (without some message). I would expect the C-program A > > just give back all the memory it used to the calling process. < > > Changing process-quota just delays the time of crashing. >  > > My question's:F > > - how can I monitor memory usage for a specific process on VMS, so- > >   I could perhaps track down the problem. C > > - Does anyone have suggestions how we could tackle this problem  > C >   AFAIK, Powerhouse runs external programms in subprocesses and I C > can't imagine how possibly unallocated memory can survive process B > rundown. So your first assumption seems to me not very likely toD > occur. But why do you think that there are insufficient memory for5 > process? What are the simptoms? Any error messages? C >   Your process can run out of VIRTUALPAGECNT, or out of page file E > quota, or GBLPAGES or GBLSECTIONS can be exhausted if you choose to ' > map screen sections to global memory. G >   You can monitor memory usage with SHOW PROC /CONT /ID=xxx (see line F > "Virtual pages"). You can track down dynamic memory related problemsC > with CMEM package from Freeware CD, we use it with great success.  >  > -- >   Valentin Likoum   >   valentin.likoum@ncc.volga.ru >  >    ------------------------------  % Date: Sun, 29 Apr 2001 17:26:38 +0200 , From: "Martie Ressing" <m.ressing@planet.nl>( Subject: Re: Process memory-usage on VMS, Message-ID: <9chbo2$5s11a$1@reader02.wxs.nl>   Joe,  J the 4GL we use offers a 'do external' command to call a routine written in C.B I don't know the lib$spawn, and I'm not sure if that's being used.   Martie  B "cstranslations" <cstranslations@email.msn.com> schreef in bericht" news:O6g$A13zAHA.83@cpmsnbbsa07...I > You could start with show process/accounting or show process/continuous  (on B > 2nd typing a "V" might be of some interest - HELP SHOW PROCESS). > J > How is the 4GL language "calling" the C program? Are you using lib$spawn toH > call a program or are you calling a routine written in C (your wording seems 8 > to indicate something along the lines of a lib$spawn). > K > Unless the program (or routine) getting called is doing something unusual F > (such as inner mode stuff) resources are typically released when the program F > terminates. With something along the lines of a lib$spawn I wouldn't expectE > to see behavior you are describing. I would expect to see the below  problemsJ > from an (improperly) coded function. There are exceptions to every rule.F > Calling something like $GETQUI and negelecting to release the "queue9 > context" can get you into problems even in user mode...  > I > If you search through past "OpenVMS Wizard" articles you can find a few B > rather indepth ones on memory leaks (common causes and debugging9 > techniques). The wizard is a quite a knowledgable chap.h >l1 > http://www.openvms.compaq.com/wizard/index.htmla >S > Joeg >  >  >m9 > "Martie Ressing" <m.ressing@planet.nl> wrote in messages( > news:9ccmnm$5qe65$1@reader02.wxs.nl...
 > > Hello, > >nG > > I've got a problem concerning jobs seemingly running out of memory. K > > We are running a program (on Alpha/VMS) developed in PowerHouse (a 4GL)-* > > which call's, repeatedly, a C-program.A > > We think the C-program doesn't free the memory it uses, afteraA > > returning to the calling program, because after some time thekK > > program just quits (without some message). I would expect the C-program A > > just give back all the memory it used to the calling process.-< > > Changing process-quota just delays the time of crashing. > >  > > My question's:F > > - how can I monitor memory usage for a specific process on VMS, so- > >   I could perhaps track down the problem.pC > > - Does anyone have suggestions how we could tackle this problem2 > >Z > >3 > > Martie Ressing > > Software engineer6, > > Inter Access, Deventer (The Netherlands) > >  > >t >l >D   ------------------------------  # Date: Sun, 29 Apr 2001 07:58:55 GMTo From: Dirk Munk <munk@home.nl>? Subject: Re: Question: QLogic QLA1080 SCSI Controller & OpenVMS ' Message-ID: <3AEBC9BD.35CA701A@home.nl>n  L Carefull: Compaq does use the QLA1040 for alpha systems, but AFAIK it is notL a standard QLA1040. The bios (V5.57) for the QLA1040 is part of the firmwareG of a alpha system and is loaded into the controller by the alpha.  So Io! doubt if you can use the QLA1080.h   Robert Alan Byer wrote:e   > Question.e >nI > Does anyone know if a QLogic QLA1080 Ultra2 SCSI controller can be usede! > in an Alpha system for OpenVMS?  > J > I know that the QLogic QLA1040 is OpenVMS Alpha compatiable, but I can't' > seem to find any info on the QLA1080.t >s > Thanks in advance. >e > -- >lB >  +------------------+--------------------------+---------------+B >  | Robert Alan Byer | byer@mail.ourservers.net | ICQ #65926579 |B >  +------------------+--------------------------+---------------+B >  | Send an E-mail request to obtain a copy of my PGP key.      |B >  +-------------------------------------------------------------+B >  | "It is by caffeine alone I set my mind in motion.  It is by |B >  |  cans of cola the thoughts aquire speed, the hands aquire   |B >  |  shakes, the shakes become a warning.  It is by caffeine    |B >  |  alone I set my mind in motion."                            |B >  +-------------------------------------------------------------+   ------------------------------    Date: 29 Apr 2001 18:30:41 +0800, From: Paul Repacholi <prep@prep.synonet.com>G Subject: Re: Ripping ods-2 cd on AlphaOpenVMS burning ods-2 cd in Win98.- Message-ID: <877l042e26.fsf@prep.synonet.com>w  ? wayne@tachysoft.xxx.065234.killspam.015d (Wayne Sewell) writes:l  @ > I've never done this on a pc, but I've used basically the sameC > technique on a cd-burner connected directly to an alpha.  The raw A > copy of the foreign-mounted cdrom to a file works fine.  I havetB > found, however, that cdrecord does not like arbitrary numbers ofF > blocks.  Your pc program may not either.  For me, it works better toD > use copy/alloc=<count>, where count is rounded up to a multiple ofF > 100.  You would need to do the copy twice to see how many blocks are2 > present, then use the /alloc on the second copy.  J It will need to be a modulo 8 blocks; 2048 byte physical sectors. A larger multiple would not hurt...   -- o< Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda.b@                                              West Australia 6076. Raw, Cooked or Well-done, it's all half baked.H Spam-To: uce@ftc.gov,enforcement@sec.gov,sness@fcc.gov,hfurchtg@fcc.gov,#   mpowell@fcc.gov,gtristan@fcc.gov     ------------------------------    Date: 29 Apr 2001 18:55:45 +0800, From: Paul Repacholi <prep@prep.synonet.com>W Subject: Re: seeking performance numbers for distributed lock manager traffic via  FDDIl- Message-ID: <873das2cwe.fsf@prep.synonet.com>a  4 Andreas Stiller <Andreas.Stiller@netsurf.de> writes:  H > The "bad feeling" inside the application starts with 9000 ENQ/DEQ rateG > shown with MON CLUSTER and is strongly correlated with stalls on this  > lock (SEQBLK)   A Are you having any deadlock searches? If a lock request times out:D because it is not granted in time, this can cause a deadlock search,C with no deadlock found. (I think this is the worst case in deadlockaG stall time) I've seen this happen under high load locking, esp if theree$ are a few locks being serialized on.  < > B) With ANAL/SYS, SHO PORT, SHO PORT/ADDR=PE_PDT, SHO PORT? > /VC=VC_'node' one can measure roundtrip times for the clustereD > communication channel (here FDDI) to a remote node. I can see fastF > changing numbers of 1 - 10 ms here. Is this a good value ? What does7 > this number mean exactly ? It seems very large to me.a  G AMDS will show trip time and Standard Dev in the NICSA display. You caneF get to it from the cluster transition display. I see 'normally' around8 50-70 ms on a 10base2, so 1-10 on fddi seems reasonable.  @ Perhaps a word with Oricle and ask if there is some way to avoidE serializing on one lock. Or MC, or point-to-point 100baseT as clustert coms.e   -- g< Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda. @                                              West Australia 6076. Raw, Cooked or Well-done, it's all half baked.H Spam-To: uce@ftc.gov,enforcement@sec.gov,sness@fcc.gov,hfurchtg@fcc.gov,#   mpowell@fcc.gov,gtristan@fcc.gov e   ------------------------------  % Date: Sun, 29 Apr 2001 11:24:16 +0200r2 From: "Thomas H. Pauli" <thomaspauli@arcormail.de>V Subject: Re: seeking performance numbers for distributed lock manager traffic via FDDI+ Message-ID: <3AEBDDC0.8020003@arcormail.de>i   Andreas,  E not that we employ any GS, but we've got for ES40 in two data centersyF connected solely via FDDI, also running Oracle V7 (.3.4). I was fairlyD astonished seeing round trip times as of around one second. It looksF like you discovered something! We plan to have a close eye on our con-A figuration and the troughput through our DEChub 900s. If you are b6 interested in our results, we'd like for forward them.   Thomas.t   -- c9 Thomas H. Pauli, Hammersteinstr.19, 14199 Berlin, Germanye   ------------------------------  # Date: Sun, 29 Apr 2001 08:11:15 GMTl From: Dirk Munk <munk@home.nl>! Subject: Re: Sync Comm - PCI Cardn' Message-ID: <3AEBCCA2.D3F7942F@home.nl>m   john nixon wrote:l  M > If I can get the eight port model  here in the States, and if it works with N > the latest software revs, it may be a little more expensive than a four portN > board, or maybe even more than two 4 port boards, but it will most certainlyJ > be cheaper than another system to put it in, with vms licensese, cluster- > licenses, volume shadowing licenses etc....v >i3 > Thanks.  I will poke around Compaq a little more.2  K And if you can't get them from Compaq, you can always get them from Emulex.   ! http://www.emulex.com/dcp/xp.htmlw   Successu   Dirk     >  >s- > "Dirk Munk" <munk@home.nl> wrote in messagec# > news:3AEB3BF7.76E0C32A@home.nl.... > snip > >3M > > The AC model is available. I just ordered 2 of them over here in Holland.h > They; > > are just a little bit more expensive then the AB model.  > >h
 > end snip   ------------------------------  # Date: Sun, 29 Apr 2001 10:26:18 GMTt From: Dirk Munk <munk@home.nl> Subject: Windows-XP on Alpha ??e' Message-ID: <3AEBEC48.4674D23B@home.nl>m  H I've heard several rumours there will be a Windows-XP version for Alpha.- Does anyone have any information about this ?u  H The return of Windows to Alpha can be a very good thing for Compaq & the= Alpha processor, if only to break the hegemony of Intel classe@ processors. But of course there has to be a sufficient amount of3 software for Win-XP on Alpha to be able to do this.s  B And Compaq can offer upgrade licenses to real operating systems of course :-)).   ------------------------------    Date: 29 Apr 2001 07:30:06 -05009 From: Kilgallen@eisner.decus.org.nospam (Larry Kilgallen)=# Subject: Re: Windows-XP on Alpha ??t3 Message-ID: <grmEJntZG5aR@eisner.encompasserve.org>s  H In article <3AEBEC48.4674D23B@home.nl>, Dirk Munk <munk@home.nl> writes:J > I've heard several rumours there will be a Windows-XP version for Alpha./ > Does anyone have any information about this ?e > J > The return of Windows to Alpha can be a very good thing for Compaq & the? > Alpha processor, if only to break the hegemony of Intel class B > processors. But of course there has to be a sufficient amount of5 > software for Win-XP on Alpha to be able to do this.n  C Therefore the idea of a Microsoft operating system running on Alphao@ (like before) is totally worthless, even if you _like_ Microsoft operating systems.  E The only thing that would please _any_ of the Alpha fans who are alsoeD Microsoft fans (probably none in this group) is if Microsoft were toD keep all their applications current on Alpha.  Since on the previousF go-round Microsoft did not even try, I think people should just forget% about the idea of Microsoft on Alpha.>  B Remember, Microsoft on Alpha brought those special hardware models% _designed_ so they could not run VMS.e  N ==============================================================================N Great Inventors of our time: Al Gore -> Internet; Sun Microsystems -> ClustersN ==============================================================================   ------------------------------  % Date: Sun, 29 Apr 2001 12:35:15 +0100 5 From: "Robert A.M. van Lopik" <lopik@mail.telepac.pt>.# Subject: Re: Windows-XP on Alpha ??r+ Message-ID: <9cgula$du0$1@venus.telepac.pt>   F "Larry Kilgallen" <Kilgallen@eisner.decus.org.nospam> wrote in message- news:grmEJntZG5aR@eisner.encompasserve.org...iJ > In article <3AEBEC48.4674D23B@home.nl>, Dirk Munk <munk@home.nl> writes:L > > I've heard several rumours there will be a Windows-XP version for Alpha.1 > > Does anyone have any information about this ?  > >iL > > The return of Windows to Alpha can be a very good thing for Compaq & theA > > Alpha processor, if only to break the hegemony of Intel classuD > > processors. But of course there has to be a sufficient amount of7 > > software for Win-XP on Alpha to be able to do this.u >eE > Therefore the idea of a Microsoft operating system running on AlphatB > (like before) is totally worthless, even if you _like_ Microsoft > operating systems. > G > The only thing that would please _any_ of the Alpha fans who are also F > Microsoft fans (probably none in this group) is if Microsoft were toF > keep all their applications current on Alpha.  Since on the previousH > go-round Microsoft did not even try, I think people should just forget' > about the idea of Microsoft on Alpha.u  ; Not if they were to port the .NET virtual machine to alpha!   
 rob van lopik    ------------------------------  % Date: Sun, 29 Apr 2001 11:20:03 -0400 ' From: "Bill Todd" <billtodd@foo.mv.com>E# Subject: Re: Windows-XP on Alpha ?? ( Message-ID: <9chbdd$od6$1@pyrite.mv.net>  F "Larry Kilgallen" <Kilgallen@eisner.decus.org.nospam> wrote in message- news:grmEJntZG5aR@eisner.encompasserve.org...-   ...   G > The only thing that would please _any_ of the Alpha fans who are alsooF > Microsoft fans (probably none in this group) is if Microsoft were to/ > keep all their applications current on Alpha.s  L It might please even non-Microsoft-fan Alpha fans if Alpha were thereby able- to penetrate a market otherwise closed to it.r  H And if that market were something like a server market (database server,J some specialized server - anything that would really benefit from a 64-bitH address space) that didn't require much in the way of standard MicrosoftK application software, lack of same would not seem that big a deal:  I don'tcJ think too many people believe that even with all that application softwareK there'd be very much penetration into the general desktop market, though it)J might be nice to be able to offer it for the few who would find it useful.   - bill   ------------------------------  # Date: Sun, 29 Apr 2001 15:29:22 GMT 4 From: "Terry C. Shannon" <terryshannon@mediaone.net># Subject: Re: Windows-XP on Alpha ??t< Message-ID: <mrWG6.4367$e85.1750461@typhoon.ne.mediaone.net>  + "Dirk Munk" <munk@home.nl> wrote in message ! news:3AEBEC48.4674D23B@home.nl...rJ > I've heard several rumours there will be a Windows-XP version for Alpha./ > Does anyone have any information about this ?   H That rumour seems to surface about once a month. I'm told that there areK Alpha directories on the XP CDROM, I don't know for sure. I also understandhF that Microsoft continues to do its 64-bit development on Alpha systems8 (apparently because they are faster than IA-64 systems).   >mJ > The return of Windows to Alpha can be a very good thing for Compaq & the? > Alpha processor, if only to break the hegemony of Intel classnB > processors. But of course there has to be a sufficient amount of5 > software for Win-XP on Alpha to be able to do this.a >r  I The lack of applications software is one of the main reasons AlphaNT wentaI teats up. Even if Microsoft was to go ahead and do the port, I doubt that J CPQ has sufficient cash in the bank to go out and pay all the ISVs to portI their apps to the platform. Microsoft alone would likely demand the wholecG $2.6B Compaq bank account just to provide MS-Office, etc, on Alpha ;-}.g  K On the bright side, there *is* Build 2128 of Windows 2000 for Alpha. Copies C have been sighted more frequently than Elvis has been seem of late.    ------------------------------  # Date: Sun, 29 Apr 2001 17:10:06 GMT / From: StevenU@POBoxes.com (Steven P. Underwood)t# Subject: Re: Windows-XP on Alpha ??s2 Message-ID: <3aec4979.229245249@news.telocity.com>  4 On Sun, 29 Apr 2001 15:29:22 GMT, "Terry C. Shannon"" <terryshannon@mediaone.net> wrote:   >m, >"Dirk Munk" <munk@home.nl> wrote in message" >news:3AEBEC48.4674D23B@home.nl...K >> I've heard several rumours there will be a Windows-XP version for Alpha.s0 >> Does anyone have any information about this ? >rI >That rumour seems to surface about once a month. I'm told that there are L >Alpha directories on the XP CDROM, I don't know for sure. I also understandG >that Microsoft continues to do its 64-bit development on Alpha systems 9 >(apparently because they are faster than IA-64 systems).C >  >>K >> The return of Windows to Alpha can be a very good thing for Compaq & theL@ >> Alpha processor, if only to break the hegemony of Intel classC >> processors. But of course there has to be a sufficient amount ofr6 >> software for Win-XP on Alpha to be able to do this. >> >-J >The lack of applications software is one of the main reasons AlphaNT wentJ >teats up. Even if Microsoft was to go ahead and do the port, I doubt thatK >CPQ has sufficient cash in the bank to go out and pay all the ISVs to port J >their apps to the platform. Microsoft alone would likely demand the wholeH >$2.6B Compaq bank account just to provide MS-Office, etc, on Alpha ;-}. > L >On the bright side, there *is* Build 2128 of Windows 2000 for Alpha. CopiesD >have been sighted more frequently than Elvis has been seem of late. >a >A >yE Where are these sightings? I asked about a month ago trying to locaterD a CD with the same to run on a newly acquired Alpha XL266 but got no positive responses.    Stevee Steven P. Underwood,DNRC Whitinsville,MAi StevenU@POBoxes.coms   ------------------------------   End of INFO-VAX 2001.239 ************************terminates. With something along the lines of a lib$spawn I wouldn't expectE > to see behavior you are describing. I would expect to see the below  problemsJ > from an (improperly) coded function. There are exceptions to every rule.F > Calling something like $GETQUI and negelecting to release the "queue9 > context" can get you into problems even in user mode...  > I 