1 INFO-VAX	Fri, 14 May 2004	Volume 2004 : Issue 266       Contents:" Re: Alpha/VMS news on The Inquirer1 Re: Anyone here using products from this company? 1 Re: Anyone here using products from this company? 1 Re: Anyone here using products from this company? : Big Blue - Caught Red Handed, stealing KGPSA HBA info?????> Re: Big Blue - Caught Red Handed, stealing KGPSA HBA info?????> Re: Big Blue - Caught Red Handed, stealing KGPSA HBA info?????N Re: DCLER (DCL Enhancement Request) for DIRECTORY /SORT=[CREATED,MODIFIED,...]N Re: DCLER (DCL Enhancement Request) for DIRECTORY /SORT=[CREATED,MODIFIED,...] Re: DFO and SET FILE/NOMOVE  Re: Getting Hammered Re: Getting Hammered Re: Getting Hammered Re: Getting Hammered Re: Getting Hammered' Re: HP Enterprise organizational change ' Re: Icon (ico file) search for DCL lite  RE: INFO-VAX 2004.265  Re: installation boot failure  Re: longest uptime Re: longest uptime' Re: Parity errors when HSG80 restarted? ' Re: Parity errors when HSG80 restarted? > Re: Poor VMS file create performance compared to Linux/FreeBSD> Re: Poor VMS file create performance compared to Linux/FreeBSD> Re: Poor VMS file create performance compared to Linux/FreeBSD> Re: Poor VMS file create performance compared to Linux/FreeBSD< Re: Terry Shannon's "HP: Two Years Post-Merger Presentation"( Re: Timestamp format stored in RMS file? Re: VMS, tftp & ftp questions  Re: VMS, tftp & ftp questions  Re: XDMP and TCPIP 5.4) Re: You'll never guess what HP advertised ) Re: You'll never guess what HP advertised ) RE: You'll never guess what HP advertised   F ----------------------------------------------------------------------  % Date: Thu, 13 May 2004 20:21:52 +0200  From: Dirk Munk <munk@home.nl>+ Subject: Re: Alpha/VMS news on The Inquirer 2 Message-ID: <c80ecr$b00$1@news1.tilbu1.nb.home.nl>  M Very odd. To my knowledge the EV7z is already the standard cpu in the GS1280. O The remarks about the 8 cpu support only apply to Itanic systems, VMS on Alpha  M systems supports 32 cpus. Hopefully that will change too in the near future,  C after it is possible to get a 128 cpu GS1280 (if you ask politely).    Tim ffrench-Lynch wrote:, > At last a little possitive coverage on VMS > + > http://www.theinquirer.net/?article=15859  >  > Tim    ------------------------------  % Date: Thu, 13 May 2004 13:06:42 -0400 4 From: David R. Beatty <QWDavidER.TYBeattyUI@sas.com>: Subject: Re: Anyone here using products from this company?8 Message-ID: <hv97a0l2a0g9kcqu02b7922mr7r43nv813@4ax.com>  A On Thu, 13 May 2004 11:33:54 -0400, "John Smith" <a@nonymous.com>  wrote:   >http://www.netapp.com > 
 >Opinions? >  >   A     We use extensively.  If you have an environment that requires ; shared access between Unix and Windows, they work extremely = well.  If the usernames are the same, I think the Netapp will @ map from Unix to Windows automatically by default.  They aren't D terribly hard to manage, either.  Adding disks to an existing volumeE is very easy because of how their file system works; parity is always   on a single drive in a RAID set.  E     It takes some work to get VMS to talk to them, but it's possible  @ to do.  For Multinet, you have to create VMS username to UID/GID? mappings.  It defaults to any host, but you can restrict it to  F specific hosts if you want to.  The one problem on Multinet is the NFS= client through V4.4 supports ODS-2 only, so you see some very D strange filenames using Multinet.  I'm not sure about Multinet V5.0.< According to Process, the NFS server supports ODS-5, but I'm not sure about the client.  E     For TCP/IP Services, the best implementation is to create an NFS  C proxy from the VMS username to the appropriate UID/GID pair with a  E host of *, then define any hosts you want to connect to in the local   hosts table.  D     One of the issues you may run into is using mixed security for aA quota tree.  Whatever sets the security last wins.  The big issue 9 is if you have Windows ACLS and then perform a chmod from  Unix, you'll lose your ACLS.  F     I would probably not use a Netapp in a High Availability or a high? performance environment, but for general purpose NFS serving it  does it's job admirably.       David R. Beatty    ------------------------------  % Date: Thu, 13 May 2004 14:53:30 -0400 # From: "John Smith" <a@nonymous.com> : Subject: Re: Anyone here using products from this company?, Message-ID: <Y7ydnWr0qK6wWT7dRVn-vw@igs.net>  A "David R. Beatty" <QWDavidER.TYBeattyUI@sas.com> wrote in message 2 news:hv97a0l2a0g9kcqu02b7922mr7r43nv813@4ax.com...C > On Thu, 13 May 2004 11:33:54 -0400, "John Smith" <a@nonymous.com>  > wrote: >  > >http://www.netapp.com > >  > >Opinions? > >  > >  > C >     We use extensively.  If you have an environment that requires = > shared access between Unix and Windows, they work extremely ? > well.  If the usernames are the same, I think the Netapp will A > map from Unix to Windows automatically by default.  They aren't F > terribly hard to manage, either.  Adding disks to an existing volumeG > is very easy because of how their file system works; parity is always " > on a single drive in a RAID set. > F >     It takes some work to get VMS to talk to them, but it's possibleB > to do.  For Multinet, you have to create VMS username to UID/GID@ > mappings.  It defaults to any host, but you can restrict it toH > specific hosts if you want to.  The one problem on Multinet is the NFS? > client through V4.4 supports ODS-2 only, so you see some very F > strange filenames using Multinet.  I'm not sure about Multinet V5.0.> > According to Process, the NFS server supports ODS-5, but I'm > not sure about the client. > F >     For TCP/IP Services, the best implementation is to create an NFSD > proxy from the VMS username to the appropriate UID/GID pair with aF > host of *, then define any hosts you want to connect to in the local > hosts table. > F >     One of the issues you may run into is using mixed security for aC > quota tree.  Whatever sets the security last wins.  The big issue ; > is if you have Windows ACLS and then perform a chmod from  > Unix, you'll lose your ACLS. > H >     I would probably not use a Netapp in a High Availability or a highA > performance environment, but for general purpose NFS serving it  > does it's job admirably.     Thanks for the info.   ------------------------------  % Date: Thu, 13 May 2004 15:55:26 -0500 ( From: brandon@dalsemi.com (John Brandon): Subject: Re: Anyone here using products from this company?1 Message-ID: <04051315552652@dscis6-0.dalsemi.com>    >http://www.netapp.com > 
 >Opinions? >   3 For Win/client and UNIX file services - no problem.   5 For database (batch processing) - I do not recommend.            J*o*h*n B*r*a*n*d*o*n  VMS Systems Administrator * firstname.lastname.spam.me.not@dalsemi.com   ------------------------------  % Date: Fri, 14 May 2004 03:43:05 +0100 < From: "Alex Daniels" <AlexNOSPAMTHANKSDaniels@themail.co.uk>C Subject: Big Blue - Caught Red Handed, stealing KGPSA HBA info????? * Message-ID: <c81bn5$1o2k$1@news.wplus.net>  I Take a look at these two pages, one on digital.com and one ibm.com, to me  they look identical!!   K Are ibm taking DEC/Compaq/hp documents, and palming them off as their own??   R http://ftp.digital.com/pub/DEC/Alpha/firmware/readmes/archive/utility/identify.txt  P http://www-1.ibm.com/servers/storage/support/disk/64bitpcifibre/fc-hba_paid.html     Alex   ------------------------------  % Date: Fri, 14 May 2004 04:06:24 +0100 < From: "Alex Daniels" <AlexNOSPAMTHANKSDaniels@themail.co.uk>G Subject: Re: Big Blue - Caught Red Handed, stealing KGPSA HBA info????? * Message-ID: <c81d1k$1td6$1@news.wplus.net>  G "Alex Daniels" <AlexNOSPAMTHANKSDaniels@themail.co.uk> wrote in message $ news:c81bn5$1o2k$1@news.wplus.net...K > Take a look at these two pages, one on digital.com and one ibm.com, to me  > they look identical!!  > G > Are ibm taking DEC/Compaq/hp documents, and palming them off as their  own??  >  > R http://ftp.digital.com/pub/DEC/Alpha/firmware/readmes/archive/utility/identify.txt >  > P http://www-1.ibm.com/servers/storage/support/disk/64bitpcifibre/fc-hba_paid.html >  >  > Alex >   F And this page is quite funny too, it contains three references to "IBM Tru64", now thats news to me!!!   P http://www-1.ibm.com/servers/storage/support/disk/64bitpcifibre/fc-hba_paqa.html  E Serveral references to OpenVMS as well although they are not claiming  ownership, unlike Tru64.  5 Can't find a single mention of DEC/compaq/hp however.    Alex   ------------------------------    Date: 13 May 2004 22:51:43 -0600+ From: young_r@encompasserve.org (Rob Young) G Subject: Re: Big Blue - Caught Red Handed, stealing KGPSA HBA info????? 3 Message-ID: <mUSy5GhVjQuD@eisner.encompasserve.org>   i In article <c81bn5$1o2k$1@news.wplus.net>, "Alex Daniels" <AlexNOSPAMTHANKSDaniels@themail.co.uk> writes: K > Take a look at these two pages, one on digital.com and one ibm.com, to me  > they look identical!!  > M > Are ibm taking DEC/Compaq/hp documents, and palming them off as their own??  > T > http://ftp.digital.com/pub/DEC/Alpha/firmware/readmes/archive/utility/identify.txt > R > http://www-1.ibm.com/servers/storage/support/disk/64bitpcifibre/fc-hba_paid.html >  >   6 	No.  IBM was at one time selling compaq storage under? 	a joint venture/agreement.  The MSS became the rebranded 2106. E 	What you are pointing to there is technical support docs.  Something < 	that shouldn't disappear just because it is no longer being 	sold.   				Rob    ------------------------------    Date: 13 May 2004 16:05:35 -0700. From: spamsink2001@yahoo.com (Alan E. Feldman)W Subject: Re: DCLER (DCL Enhancement Request) for DIRECTORY /SORT=[CREATED,MODIFIED,...] = Message-ID: <b096a4ee.0405131505.67e10ff2@posting.google.com>   \ Paul Sture <nospam@sture.homeip.net> wrote in message news:<2g5uibF4o8fkU1@uni-berlin.de>... > Alan E. Feldman wrote:G > > How about directory listings sorted by date-time? Yes, you'd either H > > have to list a single directory or suppress the headers and trailersH > > for multi-directory listings, and you'd have to wait to see even the? > > first file listed. But so what -- it would still be useful.  > @ > Peter Weaver posted this neat trick with dates on 30-Apr-2004: > @ > Make sure you run it with SET TERM/WIDTH=132 to avoid wrapping >  > J > $ ! Needs @SYS$STARTUP:LIB$DT_STARTUP executing by privileged user first  > $ ASSIGN LIB$DATE_FORMAT_037,- >     LIB$TIME_FORMAT_001 -  >     LIB$DT_FORMAT/USER_MODE ' > $ directx -  ! Ignore any DIR symbols  >     /date=modified- # >     /width=(file:80,display:132)-  >     /out=out.txt > $ sort out.txt - >      /key=(pos:83,siz:22) tt:   ? OK. Cool, this works. However, what else does this trick affect C besides DIRECTORY/DATE output? I'm asking about the effects of both @ running the LIB$DT_STARTUP file and defining the LIB$DATE_FORMAT
 logical name.    Thanks.   C P.S. My apologies if this shows up twice; I already posted this via  Google but it didn't show up.    ------------------------------    Date: 13 May 2004 20:41:45 -0700. From: spamsink2001@yahoo.com (Alan E. Feldman)W Subject: Re: DCLER (DCL Enhancement Request) for DIRECTORY /SORT=[CREATED,MODIFIED,...] = Message-ID: <b096a4ee.0405131941.51022371@posting.google.com>   \ Paul Sture <nospam@sture.homeip.net> wrote in message news:<2g5uibF4o8fkU1@uni-berlin.de>... > Alan E. Feldman wrote:G > > How about directory listings sorted by date-time? Yes, you'd either H > > have to list a single directory or suppress the headers and trailersH > > for multi-directory listings, and you'd have to wait to see even the? > > first file listed. But so what -- it would still be useful.  > @ > Peter Weaver posted this neat trick with dates on 30-Apr-2004: > @ > Make sure you run it with SET TERM/WIDTH=132 to avoid wrapping >  > J > $ ! Needs @SYS$STARTUP:LIB$DT_STARTUP executing by privileged user first  > $ ASSIGN LIB$DATE_FORMAT_037,- >     LIB$TIME_FORMAT_001 -  >     LIB$DT_FORMAT/USER_MODE ' > $ directx -  ! Ignore any DIR symbols  >     /date=modified- # >     /width=(file:80,display:132)-  >     /out=out.txt > $ sort out.txt - >      /key=(pos:83,siz:22) tt:      Thanks! It worked.  C But I have a question: What else does running that startup file and @ defining the logical name affect? IOW, are there any "unintended consequences"? Thanks.   ------------------------------  % Date: Fri, 14 May 2004 08:45:19 +0800 , From: Paul Repacholi <prep@prep.synonet.com>$ Subject: Re: DFO and SET FILE/NOMOVE- Message-ID: <87isezu99c.fsf@prep.synonet.com>   R helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) writes:  3 > In article <3Rtoc.1324$Lf2.397@news.cpqcorp.net>, 8 > hammond@not@peek.ssr.hp.com (Charlie Hammond) writes:   = >> The OpenVMS Alpha and OpenVMS I64 installation and upgrade F >> procedure ensures that the files which require SET FILE/NOMOVE haveD >> this attribute set.  If they have not been changed, it should not) >> be necessary to run SETFILENOMOVE.COM.   C > In that case, it might be a good idea for the DFO installation to A > indicate that perhaps executing SETFILENOMOVE is not necessary.   D NOT a good idea! The clasic case is system that is on its own systemC disk. To the rest of the cluster, it is just a data disk, and files D opened with the primitive system do not have locks take out, so theyJ are not seen as open by other systems. Not only are they open, they are in8 active use. Well, until some random stuff gets paged in.   --  < Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda. @                                              West Australia 6076* comp.os.vms,- The Older, Grumpier Slashdot. 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, 13 May 2004 21:18:14 +0200 * From: Paul Sture <nospam@sture.homeip.net> Subject: Re: Getting Hammered * Message-ID: <2ghvvoF34sreU1@uni-berlin.de>   Kenneth Farmer wrote: C > Hammered is the word, or should I say Inquirered or maybe Inq'ed.  > C > I got up this morning and the network was dragging butt.  Terry's N > presentation being mention on The Inquirer coupled with Golden Egg downloadsM > is knocking the crap out of my bandwidth.  I had over 20 people downloading M > Terry's presentation at one time and over 20 people downloading Golden Eggs 7 > at the same time, not to mention the regular traffic.  >   E Understood. Since I posted my version of Terry's presentation I have  G been getting pretty constant hits, but nothing like the mix of traffic  
 you describe.   I > This business class Road Runner cable connection is becoming a problem. N > I've had them pump up the bandwidth and it seems to have helped.  At least I2 > can now pull up webpages and get my mail faster. >   B Out of interest, what kind of bandwidth do you have? FWIW, my own F business class connection has improved beyond all recognition since I , first started with it a couple of years ago.  > > Sorry for the slowness, should get better as the day passes. >   H Hopefully I have helped alleviate that, which was why I put some effort  into the exercise.   ------------------------------  % Date: Thu, 13 May 2004 17:03:31 -0400 # From: "John Smith" <a@nonymous.com>  Subject: Re: Getting Hammered , Message-ID: <rrGdnXYqdsIyfz7dRVn-uQ@igs.net>  7 "Paul Sture" <nospam@sture.homeip.net> wrote in message $ news:2ghvvoF34sreU1@uni-berlin.de... > Kenneth Farmer wrote: E > > Hammered is the word, or should I say Inquirered or maybe Inq'ed.  > > E > > I got up this morning and the network was dragging butt.  Terry's F > > presentation being mention on The Inquirer coupled with Golden Egg	 downloads C > > is knocking the crap out of my bandwidth.  I had over 20 people  downloading J > > Terry's presentation at one time and over 20 people downloading Golden Eggs9 > > at the same time, not to mention the regular traffic.  > >  > F > Understood. Since I posted my version of Terry's presentation I haveH > been getting pretty constant hits, but nothing like the mix of traffic > you describe.  > K > > This business class Road Runner cable connection is becoming a problem. H > > I've had them pump up the bandwidth and it seems to have helped.  At least I 4 > > can now pull up webpages and get my mail faster. > >  > C > Out of interest, what kind of bandwidth do you have? FWIW, my own G > business class connection has improved beyond all recognition since I . > first started with it a couple of years ago.    L Most cable access providers provide something that is effecively the same asK ADSL - slower uplink than downlink. Cable providers where I am claim 5Mbps, L but that's not the uplink. The best we can get on the uplink is something in the order of 720-1024 Kbps.   K The worst part about the cable providers is that the segment of cable which L snakes though the neighborhood is shared bandwidth to the head-end. At leastG with xDSL you have a dedicated pipe at your contracted bandwidth to the  central office.   L From that point, head-end or central office (call it what you want), you areJ at the mercy of whatever MAN's, NAP's,  and fiber connections your carrier has deigned to pay for.    ------------------------------  % Date: Fri, 14 May 2004 00:23:57 +0200 * From: Paul Sture <nospam@sture.homeip.net> Subject: Re: Getting Hammered * Message-ID: <2giarvF2urbrU1@uni-berlin.de>   John Smith wrote: 9 > "Paul Sture" <nospam@sture.homeip.net> wrote in message & > news:2ghvvoF34sreU1@uni-berlin.de... >  >>Kenneth Farmer wrote:  >>D >>>Hammered is the word, or should I say Inquirered or maybe Inq'ed. >>> D >>>I got up this morning and the network was dragging butt.  Terry'sE >>>presentation being mention on The Inquirer coupled with Golden Egg  >  > downloads  > B >>>is knocking the crap out of my bandwidth.  I had over 20 people > 
 > downloading  > I >>>Terry's presentation at one time and over 20 people downloading Golden  >  > Eggs > 8 >>>at the same time, not to mention the regular traffic. >>>  >>F >>Understood. Since I posted my version of Terry's presentation I haveH >>been getting pretty constant hits, but nothing like the mix of traffic >>you describe.  >> >>J >>>This business class Road Runner cable connection is becoming a problem.G >>>I've had them pump up the bandwidth and it seems to have helped.  At  > 	 > least I  > 3 >>>can now pull up webpages and get my mail faster.  >>>  >>C >>Out of interest, what kind of bandwidth do you have? FWIW, my own G >>business class connection has improved beyond all recognition since I . >>first started with it a couple of years ago. >  >  > N > Most cable access providers provide something that is effecively the same asM > ADSL - slower uplink than downlink. Cable providers where I am claim 5Mbps, N > but that's not the uplink. The best we can get on the uplink is something in > the order of 720-1024 Kbps.  >   F I'm on ADSL rather than cable.The blurb when I signed up said I might E have contention with my ISP , but definitely not with my neighbours,  = because my line is straight into to local telephone exchange.   @ The blurb also emphasised that on cable I would be sharing with H neighbours, but folks I know on cable have mixed responses - some good, 	 some bad,    ------------------------------  # Date: Thu, 13 May 2004 22:44:56 GMT 6 From: "Kenneth Farmer" <KFarmer@NOSPAM.SpyderByte.com> Subject: Re: Getting Hammered = Message-ID: <ITSoc.31319$V_.1240026@twister.southeast.rr.com>    ----- Original Message -----  , From: "Paul Sture" <nospam@sture.homeip.net> Newsgroups: comp.os.vms $ Sent: Thursday, May 13, 2004 3:18 PM Subject: Re: Getting Hammered      > Kenneth Farmer wrote: E > > Hammered is the word, or should I say Inquirered or maybe Inq'ed.  > > E > > I got up this morning and the network was dragging butt.  Terry's F > > presentation being mention on The Inquirer coupled with Golden Egg	 downloads C > > is knocking the crap out of my bandwidth.  I had over 20 people  downloading J > > Terry's presentation at one time and over 20 people downloading Golden Eggs9 > > at the same time, not to mention the regular traffic.  > F > Understood. Since I posted my version of Terry's presentation I haveH > been getting pretty constant hits, but nothing like the mix of traffic > you describe.  > K > > This business class Road Runner cable connection is becoming a problem. H > > I've had them pump up the bandwidth and it seems to have helped.  At least I 4 > > can now pull up webpages and get my mail faster. > >  > C > Out of interest, what kind of bandwidth do you have? FWIW, my own G > business class connection has improved beyond all recognition since I . > first started with it a couple of years ago.    H 1Mb up / 2Mb down.  I had them double it to 2Mb/2Mb.  Still sluggish but better.     @ > > Sorry for the slowness, should get better as the day passes. > >  > I > Hopefully I have helped alleviate that, which was why I put some effort  > into the exercise.       Thanks.  :)    Ken    -- Kenneth Farmer  <><  OpenVMS.org  |  dcl.OpenVMS.org    ------------------------------  % Date: Fri, 14 May 2004 01:18:01 +0200 * From: Paul Sture <nospam@sture.homeip.net> Subject: Re: Getting Hammered * Message-ID: <2gie1aF3a2ksU1@uni-berlin.de>   Kenneth Farmer wrote:   D >>>Hammered is the word, or should I say Inquirered or maybe Inq'ed. >>>  >>    > Paul Sture scribbled:  C >>Out of interest, what kind of bandwidth do you have? FWIW, my own G >>business class connection has improved beyond all recognition since I . >>first started with it a couple of years ago. > J > 1Mb up / 2Mb down.  I had them double it to 2Mb/2Mb.  Still sluggish but	 > better.  >   F 640Kb / 1536KB here, and reliably so. I don't consume much of that at > the moment, so please feel free to ask if you want my help in 8 distributing annything of interest to the VMS community.   ------------------------------  % Date: Thu, 13 May 2004 20:56:29 -0500 @ From: "David J. Dachtera" <djesys.nospam@NeOaSrPtAhMlNiOnWk.net>0 Subject: Re: HP Enterprise organizational change6 Message-ID: <40A4274D.EB87827C@NeOaSrPtAhMlNiOnWk.net>   John Smith wrote:  > * > This story is just making the press now. > N > http://story.news.yahoo.com/news?tmpl=story&ncid=1208&e=10&u=/cmp/20040513/t > c_cmp/20300676&sid=95573650  > 0 > Gonzalez to Oversee HP's Storage, Server Sales >  > Wed May 12, 2004 > Joseph F. Kovar, CRN > G > Hewlett-Packard has quietly moved to put its server and storage sales H > organizations under a unified command with the May 1 promotion of MarkM > Gonzalez to the position of vice president of the Americas for *enterprise*  > storage and server sales.  > L > Gonzalez said the move is the culmination of a shift in HP's thinking overK > the past year from a focus on product lines to a focus on customer needs.  > L > "The reality is, everything is about the customer," he said. "If you focus= > on what the customer asks you to do, nothing can go wrong."   F Well! How refreshing to see someone whose perspective is not as warped as the rest of them.   > [snip]I > "Customers complain when different people come in and say they work for K > different groups within HP," Gonzalez said. "Customers say, 'Whoa, you're M > confusing me. ... I just want to deal with one person carrying an HP card.'  > "   G Amazing! Has he been reading this newsgroup? Is that not what many here  have been asking for?    > [snip]M > As customers look to consolidate suppliers, it makes sense for them to also F > consolidate their server and storage purchases on a single supplier,I > Gonzalez said. "Over 50 percent of server dollars are spent on storage, J > according to IDC," he said. "If we don't offer storage with our ProLiantL > servers, we are selling half a loaf of bread. The easiest customer to deal  > with is an existing customer."  F Blimey! This guy has actually got his head on half-way straight! Maybe6 all the way, but we'll have to see if that proves out.  N > By combining storage and server sales into a single organization but keepingJ > specialists in specific areas such as Unix (news - web sites) servers orN > Superdome servers, HP and its partners are better able to focus specifically* > on customer requirements, said Gonzalez. > M > "Before, I ran storage," he said. "Now I run storage and servers. Before, I F > had to ask the server people, 'We gotta work together, we gotta work: > together.' Now, I tell them they have to work together." > L > Gonzalez said he also expects the unified sales organization to help driveJ > more predictability in the channel by showing partners how much they canH > benefit by doing business with HP, leading to a much more consultative: > relationship between the company and solution providers.  < Well, we can only hope that he makes sure the "unified sales/ organization" is totally familiar with OpenVMS.   L > "So instead of throwing things against the wall to see what sticks, we canL > now consult with partners," he said. "Think about it, from who can you getG > more G2 intelligence than channel partners? They touch thousands more  > customers than we do."  C Now: if the partners will listen to the customers, and the "unified 6 sales organization" will listen to the partners... Eh?  5 > Now we start writing him about OpenVMS efforts too.   H Hhmmm... Maybe best not to bother him too much soon. Sounds like winner.  * Hope he doesn't turn out to be a wiener...   --   David J. Dachtera  dba DJE Systems  http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/    ------------------------------  % Date: Thu, 13 May 2004 20:42:28 -0500 @ From: "David J. Dachtera" <djesys.nospam@NeOaSrPtAhMlNiOnWk.net>0 Subject: Re: Icon (ico file) search for DCL lite6 Message-ID: <40A42404.4B86000C@NeOaSrPtAhMlNiOnWk.net>   Nic Clews wrote: > H > I was searching for an icon for use with Acceler8's DCL lite, and fell > over this: > 3 > http://www.cs.jhu.edu/~fezzik/virtualboy/vms.html  > ' > Sega's Visual Memory System, aka VMS.  > G > Anyway, back to subject in hand, anyone got or using a reasonable ICO  > file for this? > % > You can get the software from here:  > " > http://www.accelr8.com/dcll.html  @ I take it the .EXE itself does not have any icons in it? (Make a4 shortcut on the desktop and try to change the icon.)   --   David J. Dachtera  dba DJE Systems  http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/    ------------------------------  % Date: Thu, 13 May 2004 12:34:13 -0700 1 From: "Wolf, Gerald J" <gerald.j.wolf@boeing.com>  Subject: RE: INFO-VAX 2004.265R Message-ID: <3BFEACE361F5BF429DD1DA593E3A7C09031E46DC@xch-nw-28.nw.nos.boeing.com>  F I need to install a 3rd party software package called Interchange on = another Alpha system. I The last time I installed this package, it somehow messed up DECNET. So =  I had to get into NCP G And figure out what was wrong and fix it. This was not a fun thing to =  do. :-))  H What files are associated with DECNET? I want to either copy or backup =G up those files to another location, just in case the next Interchange =  install does the same thing.   Please advise.   Warmest Regards,=20  =20  Gerald(Gerry) Wolf=20   Systems Administrator,=209  VAX/Alpha OpenVMS Dev/Prod Systems; SMARTS & Finish Cell #  HP3000 & HP Unix; Tooling Services   desk ph: (253)931-4306=20   email: gerald.j.wolf@boeing.com =20  -----Original Message-----% From: Info-VAX-Request@Mvb.Saic.Com = ) [mailto:Info-VAX-Request@Mvb.Saic.Com]=20 % Sent: Thursday, May 13, 2004 11:02 AM  To: Info-VAX@Mvb.Saic.Com  Subject: INFO-VAX 2004.265    1 INFO-VAX	Thu, 13 May 2004	Volume 2004 : Issue 265       Contents:- Anyone here using products from this company? 3 Re: Determining amount of  space used on a DLT tape  Re: DFO and SET FILE/NOMOVE % Disk errors using ANALYZE or DIAGNOSE ) Re: Disk errors using ANALYZE or DIAGNOSE  Re: EVE customizations Getting Hammered Re: Help required with CLD file # HP Enterprise organizational change # Icon (ico file) search for DCL lite  Re: installation boot failure  Re: installation boot failure  Re: installation boot failure  Re: installation boot failure  Re: installation boot failure  Re: Java v 1.4.2-1 RE: Java v 1.4.2-1 Re: longest uptime Re: longest uptime2 Oracle touts Linux Integity (sic.) and reliability, Re: OT You'll never guess what HP advertised, Re: OT You'll never guess what HP advertised' Re: Parity errors when HSG80 restarted? : Poor VMS file create performance compared to Linux/FreeBSD> Re: Poor VMS file create performance compared to Linux/FreeBSD> Re: Poor VMS file create performance compared to Linux/FreeBSD) Re: RZ2DD-LS performance specifications ? 8 Re: SCSI controllers for 4000 series microvax, for TZK62D Terry Shannon's "HP: Two Years Post-Merger Presentation" Timestamp = format stored in RMS file?( Re: Timestamp format stored in RMS file?( Re: Timestamp format stored in RMS file?( Re: Timestamp format stored in RMS file? VMS, tftp & ftp questions  Re: VMS, tftp & ftp questions ) Re: XFC experiences on OpenVMS Alpha V7.3 ) Re: You'll never guess what HP advertised ) Re: You'll never guess what HP advertised ) Re: You'll never guess what HP advertised ) Re: You'll never guess what HP advertised ) RE: You'll never guess what HP advertised ) Re: You'll never guess what HP advertised ) Re: You'll never guess what HP advertised ) Re: You'll never guess what HP advertised   F ----------------------------------------------------------------------  % Date: Thu, 13 May 2004 11:33:54 -0400 # From: "John Smith" <a@nonymous.com> 6 Subject: Anyone here using products from this company?, Message-ID: <qKqdnfKrm9j0CD7dRVn-vA@igs.net>   http://www.netapp.com   	 Opinions?     -----------------------------  # Date: Thu, 13 May 2004 12:09:45 GMT 2 From: Bob Willard <BobwBSGS@TrashThis.comcast.net>< Subject: Re: Determining amount of  space used on a DLT tape/ Message-ID: <dAJoc.39268$536.7113059@attbi_s03>    tr dorr wrote:  I > Is there a way to determine how much data is on a DLT tape? It would=20 J > be useful to know what the compation ratio is and how much of the tape =  % > is used? Is there a way to do this?  >=20 >=20	 > Thanks,  > Tom   I Compression ratio is highly data-dependent.  Tests run on the first DLT = H controller showed a c.r. range from <1:1 to 30:1.  We decided to use a =G de facto industry-standard c.r. of 2:1 as a simplification (OK, gross =  over-simplification).   C Data that has been pre-compressed does not compress well, and may = I actually expand when compressed, since the compression metadata must be = E written along with the data.  Text compresses pretty well; 2:1 is a = J reasonable ballpark estimate.  ISTR that the 30:1 c.r. was for blocks of =F all 0's or, probably, for blocks of any repeated byte. Surprisingly, =F executables sometimes compressed better than I expected, due IIRC to =( blocks of initialized data and the like. --=20  Cheers, Bob     -----------------------------  # Date: Thu, 13 May 2004 12:46:25 GMT 3 From: hammond@not@peek.ssr.hp.com (Charlie Hammond) $ Subject: Re: DFO and SET FILE/NOMOVE1 Message-ID: <B6Koc.1373$%U2.532@news.cpqcorp.net>   F In article <c7v12j$k5n$1@online.de>, helbig@astro.multiCLOTHESvax.de =2 (Phillip Helbig---remove CLOTHES to reply) writes:2 >In article <3Rtoc.1324$Lf2.397@news.cpqcorp.net>,6 >hammond@not@peek.ssr.hp.com (Charlie Hammond) writes: > J >> The OpenVMS Alpha and OpenVMS I64 installation and upgrade procedure=20D >> ensures that the files which require SET FILE/NOMOVE have this=20E >> attribute set.  If they have not been changed, it should not be=20 & >> necessary to run SETFILENOMOVE.COM. > B >In that case, it might be a good idea for the DFO installation to@ >indicate that perhaps executing SETFILENOMOVE is not necessary.   I will forward that suggestion.    --=20 H       Charlie Hammond -- Hewlett-Packard Company -- Ft Lauderdale  FL  = USA F           (hammond@not@peek.ssr.hp.com -- remove "@not" when replying)@       All opinions expressed are my own and not necessarily my = employer's.     -----------------------------    Date: 13 May 2004 01:13:02 -0700# From: sinobato@yahoo.com (Sinobato) . Subject: Disk errors using ANALYZE or DIAGNOSE= Message-ID: <ec657b64.0405130013.17b871e9@posting.google.com>   
 Hi VMS Gurus,4  H Is there a documentation out there that will give you some hints on 1. =I what particular section of an ANALYZE/ERROR or DIAGNOSE to look at, and =h> 2. the specific disk errors that are produced when you do an =' ANALYZE/ERROR or DIAGNOSE? For example:u  F  ERROR SEQUENCE 7297.                            LOGGED ON:        SID 13002602F  DATE/TIME 11-MAY-2004 00:31:13.74                            SYS_TYPE 04130002   SYSTEM UPTIME: 47 DAYS 15:08:22F  SCS NODE: MYNODE                                              VAX/VMS V5.5-2H4  ?  DEVICE ERROR KA49  CPU Microcode Rev # 2.  CONSOLE FW REV# 1.3 A                       Standard Microcode Patch    Patch Rev # 19.n  ,  GENERIC DK SUB-SYSTEM, UNIT _MYNODE$DKA100:          HW REVISION     34303131i;                                        HW REVISION =3D 1104N        ERROR TYPE            05nC                                        EXTENDED SENSE DATA RECEIVEDr        SCSI ID               01a5                                        SCSI ID =3D 1.r        SCSI LUN              00x6                                        SCSI LUN =3D 0.        SCSI SUBLUN           00l9                                        SCSI SUBLUN =3D 0.S        PORT STATUS     00000001 A                                        %SYSTEM-S-NORMAL, NORMAL =e
 SUCCESSFUL2                                         COMPLETION        SCSI CMD        3D000028t                        0000F9B6T                            0001w4                                        READ EXTENDED        SCSI STATUS           02w6                                        CHECK CONDITION    EXTENDED SENSE DATA          EXTENDED SENSE  000300F00                        0AF9B63Dn                        00000000p                        80EA0011s                            110013                                        MEDIUM ERROR C                                        UNRECOVERED READ ERR IN DATA         UCB$B_ERTCNT          00e;                                        0. RETRIES REMAININGw        UCB$B_ERTMAX          00,;                                        0. RETRIES ALLOWABLEp        ORB$L_OWNER     00010004r:                                        OWNER UIC [001,004]        UCB$L_CHAR      1C4D4008=;                                        DIRECTORY STRUCTURED 4                                        FILE ORIENTED/                                        SHARABLEy0                                        AVAILABLE.                                        MOUNTED4                                        ERROR LOGGING7                                        CAPABLE OF INPUT 8                                        CAPABLE OF OUTPUT4                                        RANDOM ACCESS        UCB$W_STS           0000m        UCB$L_OPCNT     02817364 @                                        42038116. QIO'S THIS UNIT        UCB$W_ERRCNT        000B ;                                        11. ERRORS THIS UNITn        IRP$W_BCNT          0200 A                                        TRANSFER SIZE 512. BYTE(S)n        IRP$W_BOFF          0000 <                                        TRANSFER PAGE ALIGNED        IRP$L_PID       0060008BV6                                        REQUESTOR "PID"        IRP$Q_IOSB      00000000 C                        00000000        IOSB, 0. BYTE(S) TRANSFERREDS  I This error seems to be critical due to the MEDIUM ERROR message, but it =eG also seems to be non-critical as the port status says it had a normal =o successful completion.  G I have already tried looking at the VMS manuals but there are no such =,E information there. Hopefully, by knowing at least some ideas on the =nJ criticality of these error messages, I will know when to escalate to the =' vendor or just ignore an error message.     Thanks in advance for your help. Sinobato    -----------------------------  % Date: Thu, 13 May 2004 10:28:15 +0100r* From: Nic Clews <sendspamhere@[127.0.0.1]>2 Subject: Re: Disk errors using ANALYZE or DIAGNOSE& Message-ID: <c7vf6g$jr$1@lore.csc.com>   Sinobato wrote:N  J > Is there a documentation out there that will give you some hints on 1. =  H > what particular section of an ANALYZE/ERROR or DIAGNOSE to look at,=20E > and 2. the specific disk errors that are produced when you do an=200) > ANALYZE/ERROR or DIAGNOSE? For example:      No - read on...e   >=20H >  ERROR SEQUENCE 7297.                            LOGGED ON:        SID
 > 13002602H >  DATE/TIME 11-MAY-2004 00:31:13.74                            SYS_TYPE
 > 04130002" >  SYSTEM UPTIME: 47 DAYS 15:08:22H >  SCS NODE: MYNODE                                              VAX/VMS
 > V5.5-2H4 >=20A >  DEVICE ERROR KA49  CPU Microcode Rev # 2.  CONSOLE FW REV# 1.3yC >                       Standard Microcode Patch    Patch Rev # 19.  >=20. >  GENERIC DK SUB-SYSTEM, UNIT _MYNODE$DKA100:    J This is the first issue, GENERIC DK, so not a recognized VMS device [for = this VMS version]...   >=20! >        HW REVISION     34303131,= >                                        HW REVISION =3D 1104c! >        ERROR TYPE            05;E >                                        EXTENDED SENSE DATA RECEIVEDk! >        SCSI ID               01t7 >                                        SCSI ID =3D 1.t! >        SCSI LUN              00 8 >                                        SCSI LUN =3D 0.! >        SCSI SUBLUN           00t; >                                        SCSI SUBLUN =3D 0.V! >        PORT STATUS     00000001 D >                                        %SYSTEM-S-NORMAL, NORMAL=20 > SUCCESSFUL4 >                                         COMPLETION! >        SCSI CMD        3D000028n! >                        0000F9B6t! >                            0001 6 >                                        READ EXTENDED! >        SCSI STATUS           02t8 >                                        CHECK CONDITION >=20 >  EXTENDED SENSE DATA >=20! >        EXTENDED SENSE  000300F0i! >                        0AF9B63DW! >                        00000000.! >                        80EA0011W! >                            1100d5 >                                        MEDIUM ERRORlE >                                        UNRECOVERED READ ERR IN DATAa  J It is contained in the above fields. You will need the manufacturer data =C for the drive you are using, and you should be able to decode the = . extended sense data into something meaningful.  A The text, well, I would be dubious of its accuracy as you've no =uC guarantee that a generic device reports the same sense codes as a =c6 "Digital Storage Architecture" compliant device would.      ! >        UCB$B_ERTCNT          00p= >                                        0. RETRIES REMAININGd! >        UCB$B_ERTMAX          00 = >                                        0. RETRIES ALLOWABLEl! >        ORB$L_OWNER     00010004-< >                                        OWNER UIC [001,004]! >        UCB$L_CHAR      1C4D4008h= >                                        DIRECTORY STRUCTUREDG6 >                                        FILE ORIENTED1 >                                        SHARABLEg2 >                                        AVAILABLE0 >                                        MOUNTED6 >                                        ERROR LOGGING9 >                                        CAPABLE OF INPUTy: >                                        CAPABLE OF OUTPUT6 >                                        RANDOM ACCESS! >        UCB$W_STS           0000a! >        UCB$L_OPCNT     02817364 B >                                        42038116. QIO'S THIS UNIT! >        UCB$W_ERRCNT        000BG= >                                        11. ERRORS THIS UNIT ! >        IRP$W_BCNT          0200aC >                                        TRANSFER SIZE 512. BYTE(S).! >        IRP$W_BOFF          0000p> >                                        TRANSFER PAGE ALIGNED! >        IRP$L_PID       0060008BC8 >                                        REQUESTOR "PID"! >        IRP$Q_IOSB      00000000pE >                        00000000        IOSB, 0. BYTE(S) TRANSFERREDp >=20I > This error seems to be critical due to the MEDIUM ERROR message, but=20 F > it also seems to be non-critical as the port status says it had a=20 > normal successful completion.t >=20J > I have already tried looking at the VMS manuals but there are no such=20H > information there. Hopefully, by knowing at least some ideas on the=20I > criticality of these error messages, I will know when to escalate to=20 - > the vendor or just ignore an error message.-  J But are you seeing the first attempt at the read failed logged, then the = second attempt was successful?   --=20gH Regards, Nic Clews a.k.a. Mr. CP Charges, CSC Computer Sciences nclews = at csc dot com    -----------------------------  + Date: Thu, 13 May 2004 17:17:39 +0000 (UTC) - From: lewis@spyder.mitre.org (Keith A. Lewis)  Subject: Re: EVE customizationss. Message-ID: <c80ajj$34v$1@newslocal.mitre.org>  A Rich Alderson <news@alderson.users.panix.com> writes in article = D <mddu0ysoyra.fsf@panix5.panix.com> dated 07 May 2004 14:52:57 -0400:> >koehler@eisner.nospam.encompasserve.org (Bob Koehler) writes: >pE >> In article <cs9u0yt3cnq.fsf@Tempo.Update.UU.SE>, Andreas Davour=20. >> <ante@update.uu.se> writes: >e? >>> I've just started using EVE for some VAX COBOL programming.T >M? >>> Having used emacs and its ability to recognize files and=20s >>> automaticallydH >>> start the right "code-mode", I was wondering if there was some way = to >>> make EVE do the same?-  I See LSE (Language Sensitive Editor), part of Decset.  It's built on top =k of TPU.o  J >As is Emacs, which does not *force* any coding standards on anyone, but =  H >rather simply allows those who wish to, to use them.  It's just more=20 >easily programmed than EVE.  H I can't imagine anything simpler than the learn-^R sequence, but maybe =D you're talking about the TPU language, which is sort of Pascal-like.  I >(This posting, for example, was written *and transmitted* from within=20r >Emacs.)  F This one was written and transmitted from within EVE.  Nyaah, nyaah, =@ nyaah! [I hardly ever say "nyaah, nyaah, nyaah" in tech groups.]  0 --Keith Lewis              klewis {at} mitre.org> The above may not (yet) represent the opinions of my employer.    -----------------------------  # Date: Thu, 13 May 2004 14:45:23 GMT 6 From: "Kenneth Farmer" <KFarmer@NOSPAM.SpyderByte.com> Subject: Getting Hammered_= Message-ID: <7SLoc.29235$V_.1203121@twister.southeast.rr.com>/  A Hammered is the word, or should I say Inquirered or maybe Inq'ed.i  C I got up this morning and the network was dragging butt.  Terry's =oD presentation being mention on The Inquirer coupled with Golden Egg =D downloads is knocking the crap out of my bandwidth.  I had over 20 =H people downloading Terry's presentation at one time and over 20 people =F downloading Golden Eggs at the same time, not to mention the regular = traffic.  I This business class Road Runner cable connection is becoming a problem. =sF I've had them pump up the bandwidth and it seems to have helped.  At =8 least I can now pull up webpages and get my mail faster.  < Sorry for the slowness, should get better as the day passes.   Ken,   -- Kenneth Farmer  <><  OpenVMS.org  |  dcl.OpenVMS.orgt              9 "Paul Sture" <nospam@sture.homeip.net> wrote in message = $ news:2gh8c7F2ppi0U1@uni-berlin.de...# > As reported by The Inquirer at=20b- > http://www.theinquirer.net/?article=3D15879,8 > Terry Shannon's has a presentation up in PDF format at >sE > http://www.shannonknowshpc.com/stories.php?story=3D04/05/12/7726404  >eH > Terry claims "It took me 150 hours out of a 168-hour week to render=20C > the slideware reasonably consistent with HP standards; you can=20sI > download the results in just a few minutes.", but I must assume that=20iJ > his server is either getting hammered or is somewhat sick, as it felt=20  > like 150 hours to download it. >TJ > The PDF pages containing graphics took ages to render too, so to save=20@ > others the frustration I've knocked up some quick web pages at > 0 > http://www.sture.homeip.net/itanium/index.html >t    -----------------------------  % Date: Thu, 13 May 2004 09:58:19 +0100e* From: Nic Clews <sendspamhere@[127.0.0.1]>( Subject: Re: Help required with CLD file& Message-ID: <c7vdeh$1n$1@lore.csc.com>   Ron Atkinson wrote:e >=20> > I want to be able to enter a command in the following format >=20! >         > APP CMD /QUAL=3D<txt>i >=20H > where the first two characters of the <txt> value are fixed (i.e. AB). >=20E > In my CLD file the value of QUAL is set to $quoted_string type i.e.l >=20 > DEFINE SYNTAX   QUAL_SYN > IMAGE           "QUAL.EXE"4 > QUALIFIER       QUAL VALUE (TYPE=3D$QUOTED_STRING) >=20I > I guess I want to be able to define a user-defined keyword where the=20lI > first two characters are set to AB and the rest of the string can be=20I > free text. >=20J > I cannot create a new QUAL.EXE file. So is this possible under VMS 6.2 =  % > purely within the CLD file itself ?t   Couple of things.   F Check out the Command Definition Utility manual (part of programming =E set, available online, later version will be OK). I'm not an expert =iI though so I don't know if it will address what you need. An alternative = 9 is to use a DCL front end and parsing the input that way.t  G Also I'd be tempted to replace the "QUAL.EXE" with a suitable logical =mI name, e.g. "QUALEXE" and you could DEFINE QUALEXE to the file location. = ; You would not then have to keep fiddling with the CLD file.    --=20 H Regards, Nic Clews a.k.a. Mr. CP Charges, CSC Computer Sciences nclews = at csc dot com    -----------------------------  % Date: Thu, 13 May 2004 11:29:36 -0400e# From: "John Smith" <a@nonymous.com>W, Subject: HP Enterprise organizational change, Message-ID: <1ZidnZ6Kcsb1CT7d4p2dnA@igs.net>  ( This story is just making the press now.    J http://story.news.yahoo.com/news?tmpl=3Dstory&ncid=3D1208&e=3D10&u=3D/cmp= /20040513/tc c_cmp/20300676&sid=3D95573650g    . Gonzalez to Oversee HP's Storage, Server Sales   Wed May 12, 2004 Joseph F. Kovar, CRN  G Hewlett-Packard has quietly moved to put its server and storage sales = H organizations under a unified command with the May 1 promotion of Mark =@ Gonzalez to the position of vice president of the Americas for =& *enterprise* storage and server sales.  G Gonzalez said the move is the culmination of a shift in HP's thinking =nI over the past year from a focus on product lines to a focus on customer =c needs.  F "The reality is, everything is about the customer," he said. "If you =A focus on what the customer asks you to do, nothing can go wrong."f  C HP, Palo Alto, Calif., currently serves three types of customers: =iG corporate and enterprise; public sector, including government, health =s? and education; and small and midsize businesses, Gonzalez said.g  D Many customers in the three categories are facing issues regarding =F server, storage and supplier consolidation. The latter is especially =E important as customers increasingly demand relationships with fewer = J vendors but expect more from partners, especially those that can provide = end-to-end solutions, he said.  I "Customers complain when different people come in and say they work for =-D different groups within HP," Gonzalez said. "Customers say, 'Whoa, =J you're confusing me. ... I just want to deal with one person carrying an = HP card.' "g  J The focus on customer needs and customer types has a particularly strong =H advantage for the channel, Gonzalez said. HP has about 12,000 accounts =I in the United States with more than 1,000 employees each. Of those, the = G company focuses its direct sales on about 750 customers, but they are =i3 still open to work with the channel if they choose.o  H "If HP focuses on 750 customers, then there are many accounts serviced =B exclusively by the channel where it's to HP's disadvantage to go =I direct," he said. "There is absolutely no way HP can lead direct in any =g more cases."  H As customers look to consolidate suppliers, it makes sense for them to =A also consolidate their server and storage purchases on a single =iJ supplier, Gonzalez said. "Over 50 percent of server dollars are spent on =J storage, according to IDC," he said. "If we don't offer storage with our =D ProLiant servers, we are selling half a loaf of bread. The easiest =/ customer to deal with is an existing customer."i  F By combining storage and server sales into a single organization but =G keeping specialists in specific areas such as Unix (news - web sites) =eF servers or Superdome servers, HP and its partners are better able to =; focus specifically on customer requirements, said Gonzalez.i  C "Before, I ran storage," he said. "Now I run storage and servers. =nE Before, I had to ask the server people, 'We gotta work together, we =rC gotta work together.' Now, I tell them they have to work together."   F Gonzalez said he also expects the unified sales organization to help =G drive more predictability in the channel by showing partners how much = D they can benefit by doing business with HP, leading to a much more =E consultative relationship between the company and solution providers.e  H "So instead of throwing things against the wall to see what sticks, we =H can now consult with partners," he said. "Think about it, from who can =J you get more G2 intelligence than channel partners? They touch thousands = more customers than we do."-     ----------------  3 Now we start writing him about OpenVMS efforts too.p    -----------------------------  % Date: Thu, 13 May 2004 11:48:02 +0100.* From: Nic Clews <sendspamhere@[127.0.0.1]>, Subject: Icon (ico file) search for DCL lite' Message-ID: <c7vjs6$23i$1@lore.csc.com>m  H I was searching for an icon for use with Acceler8's DCL lite, and fell =
 over this:  1 http://www.cs.jhu.edu/~fezzik/virtualboy/vms.html   % Sega's Visual Memory System, aka VMS.d  G Anyway, back to subject in hand, anyone got or using a reasonable ICO =o file for this?  # You can get the software from here:     http://www.accelr8.com/dcll.html   Thanks,o --=20eH Regards, Nic Clews a.k.a. Mr. CP Charges, CSC Computer Sciences nclews = at csc dot com    -----------------------------  % Date: Thu, 13 May 2004 09:26:03 +0100a* From: Nic Clews <sendspamhere@[127.0.0.1]>& Subject: Re: installation boot failure' Message-ID: <c7vbht$sp6$1@lore.csc.com>p   o-o-o wrote: >=20I > I have a Digital Server 5000 (same as AlphaServer 1200) on which I'm=20b+ > trying to install VMS. I checked here:=20u5 > http://h71000.www7.hp.com/openvms/supportchart.html,G > to see what version of VMS the system will run, and concluded that=20 J > 7.3-1 should work. I checked my firmware, and I believe it is current: =   > ... jumping to bootstrap codeeC > %SYSBOOT-I-FILENOTLOC, Unable to locate SYS$CPU_ROUTINES_EA05.EXEn@ > %SYSBOOT-E-LDFAIL, failed to load execlet, status =3D 00000910  ) $ EXIT %X910 is no such file of course...i  H I would check that your OS type is set to VMS. I tried looking for the =H name of that file at askq.compaq.com but I'm none the wiser as to what =J it is for. On our 7.3-2 Alpha, the highest number the CPU routines image =J is named is 270F. It is almost as if you need a very specific version of =F VMS for the box, or the operating system loader is getting something = wrong.  D You have some implicit support I believe, having paid for or being =E eligible to the update contractually, so I would consider raising a =P
 support call.o  I Another question, has this system ever run VMS? Can you find that named =e file in SYS$LOADABLE_IMAGES: ?  < I would also run up the firmware CD and do a verification...   --=20sH Regards, Nic Clews a.k.a. Mr. CP Charges, CSC Computer Sciences nclews = at csc dot com    -----------------------------  # Date: Thu, 13 May 2004 06:06:46 GMT % From: "Mike Naime" <mnaime@kc.rr.com> & Subject: Re: installation boot failure6 Message-ID: <WfEoc.7086$Lf.2734@twister.rdc-kc.rr.com>   Are you positive thatd a.) your VMS7.3-1 CD is good. H b.) Your CD drive is good. (this is your more likely source of why it is	 failing.)   H If you do a verbose boot  you should be able to see the exact error. B =- dka500 -fl 0,30001  (If I remember correctly)l  ' o-o-o <bkt@null.net> wrote in message =t6 news:DJzoc.8222$Fu7.6643@newssvr24.news.prodigy.com...F > I have a Digital Server 5000 (same as AlphaServer 1200) on which I'm trying$ > to install VMS. I checked here:=205 > http://h71000.www7.hp.com/openvms/supportchart.htmlhG > to see what version of VMS the system will run, and concluded that=20 H > 7.3-1 should work. I checked my firmware, and I believe it is current: >r > UPD> listd > H > Device         Current Revision    Filename            Update RevisionC > AlphaBIOS         V5.70-0          arcrom                 V5.70-0eB > srmflash          V6.0-4           srmrom                 V6.0-4> >                                    cipca_fw             A420A >                                    dfxaa_fw                3.20a> >                                    kzpsa_fw              A12 >t > UPD> >.H > Yet when I try to boot the 7.3-1 installation CD, I get the following: >.E > DIGITAL Server 5000 Model 5305 6533A Console V6.0-4, 10-MAY-2001=204
 > 10:11:42 >  > CPU 0 bootinge >o  > (boot dka500.5.0.1.1 -flags 0)1 > block 0 of dka500.5.0.1.1 is a valid boot blockn( > reading 969 blocks from dka500.5.0.1.1 > bootstrap code read in > Building FRU tableJ > base =3D 200000, image_start =3D 0, image_bytes =3D 79200 initializing = HWRPB=20+ > at 2000 initializing page table at 1f2000  > initializing machine state% > setting affinity to the primary CPU  > jumping to bootstrap code C > %SYSBOOT-I-FILENOTLOC, Unable to locate SYS$CPU_ROUTINES_EA05.EXEt@ > %SYSBOOT-E-LDFAIL, failed to load execlet, status =3D 00000910 >, > halted CPU 0 >  > halt code =3D 5o > HALT instruction executede > PC =3D 1f01c > boot failure > P00>>> > F > I am at a loss for what else to check. Any ideas of where to look=20 > next?- >- >-    -----------------------------  % Date: Thu, 13 May 2004 15:31:57 +0800y+ From: Tim E Sneddon <tesneddon@bigpond.com>:& Subject: Re: installation boot failure* Message-ID: <2ggmjmF2i62aU1@uni-berlin.de>   o-o-o wrote:I > I have a Digital Server 5000 (same as AlphaServer 1200) on which I'm=20/ > trying  J I've never heard of a Digital Server 5000 myself. Is this the "whitebox" = AlphaServer 1200?   $ > to install VMS. I checked here:=205 > http://h71000.www7.hp.com/openvms/supportchart.htmlNG > to see what version of VMS the system will run, and concluded that=20aH > 7.3-1 should work. I checked my firmware, and I believe it is current: >=20 > UPD> list  >=20H > Device         Current Revision    Filename            Update RevisionC > AlphaBIOS         V5.70-0          arcrom                 V5.70-0sB > srmflash          V6.0-4           srmrom                 V6.0-4> >                                    cipca_fw             A420A >                                    dfxaa_fw                3.20h> >                                    kzpsa_fw              A12 >=20 > UPD> >=20H > Yet when I try to boot the 7.3-1 installation CD, I get the following: >=20E > DIGITAL Server 5000 Model 5305 6533A Console V6.0-4, 10-MAY-2001=20 
 > 10:11:42 >=20 > CPU 0 bootinga >=20  > (boot dka500.5.0.1.1 -flags 0)1 > block 0 of dka500.5.0.1.1 is a valid boot blockp( > reading 969 blocks from dka500.5.0.1.1 > bootstrap code read in > Building FRU tableJ > base =3D 200000, image_start =3D 0, image_bytes =3D 79200 initializing = HWRPB=20+ > at 2000 initializing page table at 1f2000a > initializing machine state% > setting affinity to the primary CPUp > jumping to bootstrap coderC > %SYSBOOT-I-FILENOTLOC, Unable to locate SYS$CPU_ROUTINES_EA05.EXEI  ' As far as I can tell, it doesn't exist.t  E I don't know heaps about the OpenVMS Alpha boot process, but... I'm =sJ guessing that "EA05" is some sort of CPU identification that the OpenVMS =E bootstrap fetches from the firmware. This is then used to build the ='  filename for the CPU-specific=20G SYS$CPU_ROUTINES_*.EXE image. It can't load it because there isn't one.t  F If this is a "whitebox" chances are OpenVMS won't have the necessary = modules to boot.  @ > %SYSBOOT-E-LDFAIL, failed to load execlet, status =3D 00000910 >=20 > halted CPU 0 >=20 > halt code =3D 5. > HALT instruction executede > PC =3D 1f01c > boot failure > P00>>> >=20F > I am at a loss for what else to check. Any ideas of where to look=20 > next?   H You might no be able to do anything if this is a WindowsNT-only machine.  
 Regards, Tim.     -----------------------------    Date: 13 May 2004 05:43:56 -0700- From: martin.platts@cdl.co.uk (Martin Platts)c& Subject: Re: installation boot failure= Message-ID: <b367fb16.0405130443.289e5da5@posting.google.com>e   Google is your friend...  J http://groups.google.com/groups?hl=3Den&lr=3D&threadm=3D00256937.0045DD3E=I 00%40quegw01.btyp&rnum=3D3&prev=3D/groups%3Fhl%3Den%26lr%3D%26q%3DSYS%25=s1 24CPU_ROUTINES_EA05.EXE%26btnG%3DSearch%26meta%3D    :D    -----------------------------  # Date: Thu, 13 May 2004 15:05:37 GMTG# From: hoff@hp.nospam (Hoff Hoffman) & Subject: Re: installation boot failure1 Message-ID: <59Moc.1389$q53.642@news.cpqcorp.net>   F In article <DJzoc.8222$Fu7.6643@newssvr24.news.prodigy.com>, "o-o-o" =? <bkt@null.net> writes: :I have a Digital Server 5000 (same as =sH AlphaServer 1200) on which I'm trying :to install VMS. I checked here: =4 :http://h71000.www7.hp.com/openvms/supportchart.htmlE :to see what version of VMS the system will run, and concluded that =  7.3-1 :should work.=20  ,   Once more into the breach, dear friends...  I   The Whitebox Alpha systems are not supported by OpenVMS, as indicated =0 byI   the lack of a listing in the cited support chart, and as indicated by =R textH   in the FAQ, and as indicated by various topics in the Ask The Wizard = area   asking about this. =20  I   Again, the Microsoft Windows NT (only) Alpha systems -- often refered =e toF   as the "whitebox" Alpha systems -- are not supported by OpenVMS.   = ThisH   includes all members of the DIGITAL Server series boxes, all members = of?   the DIGITAL Professional Workstation (PWS) boxes with an -a =o designation,#   and the DECpc 150 AXP series. =20w  ?   These Windows NT (only) systems ARE NOT SUPPORTED by OpenVMS.n  G :Yet when I try to boot the 7.3-1 installation CD, I get the following:) : C :DIGITAL Server 5000 Model 5305 6533A Console V6.0-4, 10-MAY-2001 =  10:11:42 :f :CPU 0 booting :e :(boot dka500.5.0.1.1 -flags 0)/0 :block 0 of dka500.5.0.1.1 is a valid boot block ., :jumping to bootstrap codeD :%SYSBOOT-I-FILENOTLOC, Unable to locate SYS$CPU_ROUTINES_EA05.EXE =? :%SYSBOOT-E-LDFAIL, failed to load execlet, status =3D 000009104    D   That's a central indication of an unsupported processor -- OpenVMSA   has no code for the particular processor type -- your system istA   identifying itself as a DIGITAL Server series system, and these 3   Alpha systems are (wait for it :-) not supported.n  H :I am at a loss for what else to check. Any ideas of where to look next?     The OpenVMS FAQ:  $     http://www.hp.com/go/openvms/faq     Specifically:a  <     SYSBOOT-I-FILENOTLOC, Unable to locate SYS$CPU_ROUTINES?6     OpenVMS on the Whitebox Windows-Only series Alpha?  H   Support aside, you might be able to get this box to bootstrap OpenVMS.F   But you'll have to work at it some, you'll have to check the PCI andG   system configuration, you won't have support, and the result might or-)   might not be stable or even functional.q  F   Various of the Whitebox systems have integral motherboard hardware = thatD   is incompatible with the operation of OpenVMS.  (One of the more = commonF   problem areas involves the southbridge -- anyone that tells you thatE   *all* whitebox Alpha systems *will* run OpenVMS is *wrong*, as someuG   of the southbridges used on some of the whiteboxes are not compatiblerH   with OpenVMS.  We had no end of trouble with some of the southbridges.F   I'm here ignoring the usual and obvious PCI widget support, which is/   another area that you will have to evaluate.)c   	--   F   ps: as you've undoubtedly searched the FAQ for this information, andG   were clearly unable to locate it, please let me know what your search E   target(s) were and I'll re-word and/or re-index the information for1A   the next edition; I'll continue on my quest to exterminate thist"   Alpha Whitebox support question.  =20  2  ---------------------------- #include <rtfaq.h> = -----------------------------y3     For additional, please see the OpenVMS FAQ -- =p www.hp.com/go/openvms/faqx4  --------------------------- pure personal opinion = ---------------------------VE         Hoff (Stephen) Hoffman   OpenVMS Engineering   hoff[at]hp.como    -----------------------------    Date: 12 May 2004 23:02:25 -0700. From: dieter.rossbach@gmx.de (dieter rossbach) Subject: Re: Java v 1.4.2-1a= Message-ID: <e1d40caf.0405122202.7a4df489@posting.google.com>n   >=205 > Any known issues with Java v 1.4.2-1 under OpenVMS?t >=20  H Yes. We found some bugs in the IP-support with 1.4.2-1, 1.4.2-2 is the =I right version to install, all the bugs we reported to them are fixed in =s
 this version.i   Regards Dieter    -----------------------------  % Date: Thu, 13 May 2004 07:57:31 -0400 ' From: "Main, Kerry" <kerry.main@hp.com>  Subject: RE: Java v 1.4.2-1 
 Message-ID: = F <FD827B33AB0D9C4E92EACEEFEE2BA2FB314665@tayexc19.americas.cpqcorp.net>   > -----Original Message-----< > From: dieter rossbach [mailto:dieter.rossbach@gmx.de]=3D20 > Sent: May 13, 2004 2:02 AM > To: Info-VAX@Mvb.Saic.Com  > Subject: Re: Java v 1.4.2-1h >=3D20 > >=3D207 > > Any known issues with Java v 1.4.2-1 under OpenVMS?o > >=3D20 >=3D20J > Yes. We found some bugs in the IP-support with 1.4.2-1, 1.4.2-2 is the = =20-I >right version to install, all the bugs we reported to them are fixed =20o >in this version. =3D20T > Regards Dieter >=3D20  E As a fyi, here is the pointer for the V1.4.2-2 Java kit on OpenVMS: =3= http://h18012.www1.hp.com/java/download/ovms/1.4.2/index.htmlw   Regardsm  
 Kerry Main Senior Consultante HP Services Canada Voice: 613-592-4660t Fax: 613-591-4477i Email: kerryDOTmainAThpDOTcom 0 (remove the DOT's and AT for email address)=3D20    -----------------------------  % Date: Thu, 13 May 2004 09:27:37 -0400r* From: "Syltrem" <syltremzulu@videotron.ca> Subject: Re: longest uptime 4 Message-ID: <nFKoc.4048$Xy3.17904@tor-nn1.netcom.ca>  J > Well, I have a cluster that was established well over a year ago, but=20J > the longest uptime for any member is currently just over 30 days, due=20' > to monthly reboots, patches and such.m >d > -- > David J. Dachteran > dba DJE Systemse > http://www.djesys.com/     What is a monthly reboot?r# This is a VMS forum, not windows...w ;-)r   --=20f Syltrem'   OpenVMS 7.3-1 + Oracle 8.1.7.4@ http://pages.infinit.net/syltrem (OpenVMS related web site, en = fran=E7ais)e% ---zulu is not in my email address---a    -----------------------------  + Date: Thu, 13 May 2004 15:05:44 +0000 (UTC) 7 From: moroney@world.std.spaamtrap.com (Michael Moroney)@ Subject: Re: longest uptimeo( Message-ID: <c802s8$4a3$1@pcls4.std.com>  , "Syltrem" <syltremzulu@videotron.ca> writes:  H >> Well, I have a cluster that was established well over a year ago, butJ >> the longest uptime for any member is currently just over 30 days, due = to% >> monthly reboots, patches and such.t   >What is a monthly reboot?$ >This is a VMS forum, not windows... >;-)  - How do you keep Windows up for a whole month?  --=20r -Mike     -----------------------------  % Date: Thu, 13 May 2004 08:54:37 -0400 # From: "John Smith" <a@nonymous.com> ; Subject: Oracle touts Linux Integity (sic.) and reliabilityh, Message-ID: <tdKdnatK47SM7T7dRVn-ug@igs.net>   Ad in my local newspaper today:a    " Oracle Sets New Linux Speed Record
 1,000,000+ Transactions Per Minuten   The Oracle Grid on Linux It's fast...it's cheap...C and it never breaksl   cites www.tpc.org  December 8, 2003- 16 x 4-cpu HP Integrity rx5670 server cluster  cpus were Itanic II 1.5GHz 1,184,893.38 tmpC  $5.52 / tpmC  / Now what were the Rdb on VMS comparables again?     -----------------------------  % Date: Thu, 13 May 2004 08:09:40 +0200u* From: Paul Sture <nospam@sture.homeip.net>5 Subject: Re: OT You'll never guess what HP advertisedg* Message-ID: <2gghp6F2bnoaU1@uni-berlin.de>   Bob Koehler wrote:9 > In article <2gf6omF27vimU1@uni-berlin.de>, Paul Sture =o! <nospam@sture.homeip.net> writes:o >=20J >>Mind you, the US site for the VW Phaeton claims that it is pronounced=20C >>FayTon, whereas I automatically think something akin to "Feeton".  >> >=20E >    The many Phaetons produced by other auto makers have always been: >    fae-eh-tun, to me.  >=20  . OK. It just didn't feel right when I saw it...    -----------------------------  % Date: Thu, 13 May 2004 08:13:01 +0200V* From: Paul Sture <nospam@sture.homeip.net>5 Subject: Re: OT You'll never guess what HP advertised * Message-ID: <2gghvdF2bnoaU2@uni-berlin.de>   Dr. Dweeb wrote: > Paul Sture wrote:O >=20 >>Dr. Dweeb wrote: >> >>>Paul Sture wrote: >>>  >>>  >>>>Dr. Dweeb wrote: >>>> >>>>G >>>>>>>Just what is a MayBach ?? - I will need to resaerch this one :-)  >>>>>>C >>>>>>It is Daimler-Chryslers car in the Bentley/R&R class of cars.  >>>>>> >>>>>  >>>>> G >>>>>Out of my league I guess.  Are they actively sold in Europe, or is > >>>>>it a US only thing like a Cadillac from the Chrysler arm? >>>>>  >>>>>  >>>>/ >>>>The best description I've seen of it, is at0 >>>>- >>>>http://auto.howstuffworks.com/maybach.htm  >>>>E >>>>Lots of links to follow from which appears it is not just sold in  >>>>the US.  >>>  >>> D >>>Good link, thanks.  I thought that the name sounded familiar, andD >>>the site refreshes that blurry memory from my earlier interest in >>>old cars. Thanks Paul.  >>> G >>>By the way, they cite "my-bok" as the pronunciation.  Since the wordCD >>>is German, I would have expected the "bach" part to be pronouncedD >>>like Bach (of J.S. fame).  You maybe have better insight into the >>>German language than I. >>>  >>A >>More like My-Bach (guttural ch ending, depending on area) in myM
 >>opinion. >>G >>Mind you, the US site for the VW Phaeton claims that it is pronounced C >>FayTon, whereas I automatically think something akin to "Feeton".  >=20 >=20C > Well, we have the diphthong =E6 (as does archaic english, as in =  encyclop=E6diaJ > or onomatopoa) so I want to read something that cannot be expressed in = the F > modern english alphabet! :-) But I confess, it is closer to FeyTon = thanG > FeetOn.  Which reminds me that I must collect my new futon tomorrow =  :-)  >=20   I stand corrected.    -----------------------------  % Date: Thu, 13 May 2004 10:18:12 -0400  From: norm.raphael@metso.com0 Subject: Re: Parity errors when HSG80 restarted?
 Message-ID: = E <OFD686A1EB.8E68E4C1-ON85256E93.004DBDBA-85256E93.004EA1BB@metso.com>O  C "David J. Dachtera" <djesys.nospam@NeOaSrPtAhMlNiOnWk.net> wrote on  05/11/2004 10:53:50 PM:    [snip] >OA > *** WARNING *** WARNING *** WARNING *** WARNING *** WARNING ***0 >7C > There is a problem with the FC-SCSI patch V5 for V7.3-1. Do not =T installUF > this patch, and if you have installed it, back it out IMMEDIATELY!!! > H > Do not attempt to apply the replacement patch (V6) as it has also been > recalled.A >EA > *** WARNING *** WARNING *** WARNING *** WARNING *** WARNING ***  >   H I guess I was away when this happened.  Can you explain the problem withH the FC_SCSI V5 patch or point me to the documentation about the problem, please.0  F (I am still at V4 of this ECO, but I have some other Level 1 ECO's I = wantA to schedule, and I certainly do not want FC problems afterwards.)b  H > Also, contact your support rep. about an update to the HSG firmware, = buto$ > be ready for some unexpected news. >y > -- > David J. Dachteraa > dba DJE Systems  > http://www.djesys.com/ >.* > Unofficial Affordable OpenVMS Home Page:! > http://www.djesys.com/vms/soho/r    -----------------------------    Date: 13 May 2004 07:52:11 -0600B From: clubley@remove_me.eisner.decus.org-Earth.UFP (Simon Clubley)C Subject: Poor VMS file create performance compared to Linux/FreeBSD03 Message-ID: <ACNLDAui2WbG@eisner.encompasserve.org>j  H I am seeing very poor VMS I/O performance compared to Linux/FreeBSD whenH trying to perform the same task - unpacking a GCC tar archive - on these/ operating systems. A table of results is below.i  D I've gone into a lot of detail about my test environments, so that =	 hopefully F if I've missed something it will be noticed. I am also interested in = seeing8 the results on other platforms if anyone is so inclined.  J The tar archive in question is GCC 3.4.0 which contains 22735 files/dirs = andED is 191,467,520 bytes long uncompressed; as VMSTAR does not support =
 compressedH archives I uncompressed all archives first[1]. The times below are for =	 the untar H operation only. The archive is available from the usual GCC mirrors. I = used:O  J ftp://ftp.mirror.ac.uk/sites/sources.redhat.com/pub/gcc/releases/gcc-3.4.= 0/G The archive is: gcc-3.4.0.tar.bz2 (25.9MB) or gcc-3.4.0.tar.gz (34.3MB)   C I turned off Linux deferred writes on the filesystem in question bysE remounting the filesystem with "-o sync" on the mount command; I have = included the Linux deferred write performance for comparison.E  H In all cases, the input tar file was on the same disk as the extracted = files.  	 Machines: F                                                           Avg      AvgG Machine/OS	Filesystem  Disk/Interface        RPM   Seek(ms) Latency(ms) ! AlphaStation 200 4/166, VMS 7.3-13F                 ODS-5       IBM DORS-32160, SCSI  5400   8.5      5.56' Testdrive Alpha DS20, 500MHz, VMS 7.3-2PF                 ODS-5       RZ2DD-LS, SCSI        10000  5.4      2.99  Pentium 233Mhz, Redhat Linux 6.2F                 ext2        ST320410A, IDE        5400   8.9      5.55 Pentium 233Mhz, FreeBSD 4.5 E                 FreeBSD     ST34321A, IDE         5400   11.5     5.6  		native  F I choose the PC above so that the CPU performance would be as close as possible to my AlphaStation.  8 In all cases, the command line was:	[tar_program] -vxf = [archive_name].tar   Results:   Machine		OS		Unpack time	NotesL =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=L =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3 =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D A AS 200 4/166	VMS 7.3-1	1 hour 46 mins	Extract didn't complete [2]f8 Pentium 233	Redhat 6.2	2 mins 10 secs	Deferred writes ON9 Pentium 233	Redhat 6.2	4 mins 22 secs	Deferred writes OFFx+ Pentium 233	FreeBSD 4.5	15 mins 55 secs	[3] A Testdrive	VMS 7.3-2	55 mins 48 secs	Extract didn't complete [2,4]u  J For VMS, I used VMSTAR V3.4-1, obtained from Hunter's site and linked on = theoF target machine. All VMS measurements were done using this version. [5]  B Does anyone have any ideas on how I can increase VMS file create = performance 4 to bring it up to Linux/FreeBSD performance levels ?   Simon.  F [1] Interestingly, the uncompress operation completed in roughly the =	 same time E (allowing for hardware differences) on VMS as it did on Linux. This =4 would G seem to suggest that raw I/O write performance on VMS is ok, and it's =  the D actual creating of files that appears to be the performance problem.  G [2] The untar failed towards the end, because of a full path that was =  longerJ than 100 bytes. VMSTAR does not support LongLink tar blocks, which means = thatG VMSTAR will not fully unpack any tar archive with a full path of more =  thanJ 100 bytes. (I've already notified Hunter; I encountered the same problem = doing F some early testing with a GCC preview kit.) The last file successfully extracted was:   Jan  4 23:18:59 2004     2316 = 2 [C3.pvg.libstdc^+^+-v3.testsuite.27_io.basic_istre$ am.extractors_arithmetic.char]13.cc;%                               [C3.^.]H  tar: unexpected EOF on tar file.  B This is sufficiently close to the end of the archive that VMSTAR = shouldn't have= taken more than a few minutes to finish extracting the files.   J [3] The FreeBSD filesystem metadata is written using sync I/O; file data = isI written using async I/O. Even allowing for the slower disk that FreeBSD =0 is on F (you can't place FreeBSD on an extended partition), the native FreeBSDG filesystem appears to be slower than ext2 at writing large numbers of =2 files.% Any FreeBSD experts care to comment ?o  H [4] I made sure that VMSTAR was the only process doing any significant = I/Oa& by using MONITOR at regular intervals.  F [5] The VMSTAR binaries supplied as part of the GNV kits were slower = than, the VMSTAR that I linked. I do not know why.   --=20ID Simon Clubley, clubley@remove_me.eisner.decus.org-Earth.UFP      =20H SCO: Proudly pushing Microsoft down to #2 on the list of most disliked =	 companiest    -----------------------------    Date: 13 May 2004 11:58:12 -0600B From: clubley@remove_me.eisner.decus.org-Earth.UFP (Simon Clubley)G Subject: Re: Poor VMS file create performance compared to Linux/FreeBSD 3 Message-ID: <P3VGay4IXhcD@eisner.encompasserve.org>   H In article <urMoc.1392$q53.296@news.cpqcorp.net>, hoff@hp.nospam (Hoff = Hoffman) writes: >=20F >   Search for "mybenchmark"; a set of tools made available by David = Mathog.iC >   Around the discussions of the tools, you'll also find various =m previous' >   discussions of the I/O performance.l >=20  + Thanks for the pointer; I will take a look.r   >=20@ >   Y'all can get a hobbyist license for the HP C compiler, too. >=20  G Perfectly true, and I have one installed. Unfortunately, my interest isc because of Ada...s   Simon.   --=20dD Simon Clubley, clubley@remove_me.eisner.decus.org-Earth.UFP      =20H SCO: Proudly pushing Microsoft down to #2 on the list of most disliked =	 companiesD    -----------------------------  # Date: Thu, 13 May 2004 15:25:14 GMTo# From: hoff@hp.nospam (Hoff Hoffman)bG Subject: Re: Poor VMS file create performance compared to Linux/FreeBSDm1 Message-ID: <urMoc.1392$q53.296@news.cpqcorp.net>g  5 In article <ACNLDAui2WbG@eisner.encompasserve.org>, =eD clubley@remove_me.eisner.decus.org-Earth.UFP (Simon Clubley) writes:F :I am seeing very poor VMS I/O performance compared to Linux/FreeBSD = whenE :trying to perform the same task - unpacking a GCC tar archive - on =N thesey0 :operating systems. A table of results is below.  D   Search for "mybenchmark"; a set of tools made available by David = Mathog.bH   Around the discussions of the tools, you'll also find various previous%   discussions of the I/O performance.0  C :Does anyone have any ideas on how I can increase VMS file create => performancet5 :to bring it up to Linux/FreeBSD performance levels ?s  H   Given differences in the file systems of UNIX and OpenVMS, there are = andbE   will be differences in performance.  That written, the OpenVMS file F   system, XFC caching, and file system I/O performance are all areas = thatD   are all under investigation within OpenVMS Engineering; there is =	 typicallydE   on-going profiling and performance-related work within engineering.o  E   We have made various improvements in recent releases, and have been.G   actively working to increase the aggregate I/O performance -- this=20mC   includes changes made in V7.3-2, as well as work targeting future.   OpenVMS releases.o   	--u  >   Y'all can get a hobbyist license for the HP C compiler, too.    2  ---------------------------- #include <rtfaq.h> = -----------------------------c3     For additional, please see the OpenVMS FAQ -- =u www.hp.com/go/openvms/faqs4  --------------------------- pure personal opinion = --------------------------- E         Hoff (Stephen) Hoffman   OpenVMS Engineering   hoff[at]hp.com     -----------------------------    Date: 13 May 2004 07:51:07 -0600B From: clubley@remove_me.eisner.decus.org-Earth.UFP (Simon Clubley)2 Subject: Re: RZ2DD-LS performance specifications ?3 Message-ID: <k2W$ecBUu36s@eisner.encompasserve.org>;  9 In article <2gfa58F28o87U1@uni-berlin.de>, "Hans Vlems" = ! <hvlems.dotweg@zonnet.nl> writes:e >=205 > Apparently the DEC RZ2DD-LS is a Seagate ST39102LC.lE > Ref.: http://www.seagate.com/support/disc/specs/scsi/st39102lc.htmlg >=20  < Thank you. I didn't realise that it was a re-badged Seagate.   Simon.   --=20sD Simon Clubley, clubley@remove_me.eisner.decus.org-Earth.UFP      =20H SCO: Proudly pushing Microsoft down to #2 on the list of most disliked =	 companiese    -----------------------------  # Date: Thu, 13 May 2004 14:01:43 GMTwA From: "Colin Butcher" <colin_DOT.butcher_AT@xdelta_DOT.co_DOT.uk>nA Subject: Re: SCSI controllers for 4000 series microvax, for TZK62w= Message-ID: <bdLoc.2616$H23.21685278@news-text.cableinet.net>h  H Nemonix made the KZCCA series of devices for the StorageWorks group in = DEC.  C Nemonix still make such devices, now in both HVD (for HSZ and other 4 StorageWorks type applications) and LVD/SE variants.  E The boards come as SCSI only, Fast Ethernet only or Combo boards. For E smaller VAXes they generally plug into the CDAL bus on the CPU board.sH Regrettably they don't do one for the VAX 4000-200 or 4000-300. There is also an XMI bus version.   --=20    Hope this helps, Colin.i) colin DOT butcher AT xdelta DOT co DOT ukiJ Systems Archaeologist - Investigation & troubleshooting of older systems = andn	 networks.n    -----------------------------  % Date: Thu, 13 May 2004 14:35:17 +0200)* From: Paul Sture <nospam@sture.homeip.net>A Subject: Terry Shannon's "HP: Two Years Post-Merger Presentation" * Message-ID: <2gh8c7F2ppi0U1@uni-berlin.de>    As reported by The Inquirer at =+ http://www.theinquirer.net/?article=3D15879a6 Terry Shannon's has a presentation up in PDF format at  C http://www.shannonknowshpc.com/stories.php?story=3D04/05/12/7726404   J Terry claims "It took me 150 hours out of a 168-hour week to render the=20J slideware reasonably consistent with HP standards; you can download the=20H results in just a few minutes.", but I must assume that his server is=20I either getting hammered or is somewhat sick, as it felt like 150 hours=20@ to download it.s  H The PDF pages containing graphics took ages to render too, so to save=20> others the frustration I've knocked up some quick web pages at  . http://www.sture.homeip.net/itanium/index.html    -----------------------------  % Date: Thu, 13 May 2004 17:43:46 +0300-+ From: Jaan Kronberg <jaan.kronberg@mail.ee>n- Subject: Timestamp format stored in RMS file?e/ Message-ID: <opr7x0i8q6xckwek@diablo.uninet.ee>z  	 Hi there,o   8 bytes: 0000 36BD 04FB A200m  C Could anybody "translate" it to "readable" date without using any =h built-in=20  OpenVMS function?iI Or, alternatively, using Perl standard functions (let's assume I don't=20 9 have any modules and I don't know C - and I really don't)n  % Just tell me _how_, I'm kinda stuck..u  J I've read bunch of documents regarding Julian calendar, tried to compare =  J some Julian calendar information with what I've got in file - nothing :) = I=207 think I'm going crazy with that (too hard case for me).n  1 If question sounds very stupid - just ignore it..e   thx, jk> p.s. Actually those bytes _should_ represent April 1st, 2004 = 00:00:00[.00=20r <-- not sure about that]    -----------------------------  % Date: Thu, 13 May 2004 16:45:42 +0100d* From: "Richard Brodie" <R.Brodie@rl.ac.uk>1 Subject: Re: Timestamp format stored in RMS file?p, Message-ID: <c80578$1a3s@newton.cc.rl.ac.uk>  : "Jaan Kronberg" <jaan.kronberg@mail.ee> wrote in message =) news:opr7x0i8q6xckwek@diablo.uninet.ee...n  
 > 8 bytes: > 0000 36BD 04FB A200  >'E > Could anybody "translate" it to "readable" date without using any =  built-in > OpenVMS function?s  D Yes. Time in 100 nanosecond intervals since November 17th 1858, as a 64-bit word.   Some Python:  " >>> base =3D '1-jan-1970 00:00:00' >>> vms.starlet.bintim(base)[1]0 35067168003000000L@ >>> seconds =3D ( 0xA2FB04BDBDCDA0L - 35067168002100000L ) / 1E7& >>> time.asctime(time.gmtime(seconds)) 'Thu Apr  1 00:00:00 2004'  
 Close enough:P, >>> vms.starlet.asctim(0xA2FB04BDBDCDA0L)[1] ' 1-APR-2004 00:00:00.89's    -----------------------------  + Date: Thu, 13 May 2004 16:14:07 +0000 (UTC)T. From: Dale Dellutri <ddelQQQlutr@panQQQix.com>1 Subject: Re: Timestamp format stored in RMS file?i, Message-ID: <c806sf$gha$1@reader2.panix.com>  3 On Thu, 13 May 2004 17:43:46 +0300, Jaan Kronberg =a <jaan.kronberg@mail.ee> wrote: > Hi there,   
 > 8 bytes: > 0000 36BD 04FB A200n  E > Could anybody "translate" it to "readable" date without using any =o built-in=20t > OpenVMS function?iJ > Or, alternatively, using Perl standard functions (let's assume I don't =  ; > have any modules and I don't know C - and I really don't)n  ' > Just tell me _how_, I'm kinda stuck..   D > I've read bunch of documents regarding Julian calendar, tried to =
 compare=20I > some Julian calendar information with what I've got in file - nothing =q :) I=20,9 > think I'm going crazy with that (too hard case for me).I  3 > If question sounds very stupid - just ignore it..s   > thx, > jk@ > p.s. Actually those bytes _should_ represent April 1st, 2004 = 00:00:00[.00=20o > <-- not sure about that]  B What you've got is a 64 bit time in 100 nanosecond intervals sinceC 1858-11-17 00:00, the base of the VMS time system (also the base ofy5 the modified Julian day number, see the calendar FAQ:c-   http://www.tondering.dk/claus/calendar.html  )p  @ First reconstruct the 64 bit hex value (reverse the byte order):   Hex value? 00A2FB04BD360000  Convert to base ten:   (base 10): 45874944000000000E So it's 4,587,494,400,000,000,000 nanoseconds since the base time =3Dt0   4,587,494,400 seconds / 86,400 seconds/day =3D   53,096 days (no remainder)  B From here, getting to the correct date requires a somewhat complex@ procedure which takes into account the leap days.  So the entire8 process requires 64 bit arithmetic and a date procedure.  C However, if you just use SYS$ASCTIM with the 64 bit value given, it  says:    1-APR-2004 00:00:00.00 which is a lot more direct.    --=20=7 Dale Dellutri <ddelQQQlutr@panQQQix.com> (lose the Q's)v    -----------------------------  % Date: Thu, 13 May 2004 13:23:13 -0400c( From: "Hein" <hein.nomail@hp.nomail.com>1 Subject: Re: Timestamp format stored in RMS file?e* Message-ID: <40a3afe4@usenet01.boi.hp.com>  I I know you said no VMS build-ins (btw... why? going to use that data on =t aS PC?)F Just to understand the problem I would encourage you to use VMS tools, notably the debugger. 1 Take an image, any image, run it with debugger...  $RUN/DEB TMP DBG> set radix hex* DBG> dep/date 10000=3D"1-apr-2004 0:0:0.0" DBG> ex/date .' TMP\main:        1-APR-2004 00:00:00.00( DBG> ex/quad .  TMP\main:       00A2FB04BD360000  ( see... there are your bytes... reversed.  + Now for grins, ooops: better understanding:m DBG> ex/qua/dec  .! TMP\main:       45874944000000000n* DBG> dep/date 10000=3D"1-apr-2004 0:0:1.0" DBG> ex/qua/dec  .! TMP\main:       45874944010000000   F See those 10,000,000 clunks of 100 nano-seconds being added to add a = second?_   Same second added in Hex:T DBG> ex/quad .  TMP\main:       00A2FB04BDCE9680  B If you look around (Wizard!, OpenVMS Forum) you will also fidn DCL( constructs to pretty print binary dates.    8 "Jaan Kronberg" <jaan.kronberg@mail.ee> wrote in message) news:opr7x0i8q6xckwek@diablo.uninet.ee...u > Hi there,, >e
 > 8 bytes: > 0000 36BD 04FB A200a >=E > Could anybody "translate" it to "readable" date without using any =  built-in > OpenVMS function?pH > Or, alternatively, using Perl standard functions (let's assume I don't; > have any modules and I don't know C - and I really don't)t >h' > Just tell me _how_, I'm kinda stuck..e >aD > I've read bunch of documents regarding Julian calendar, tried to = compareaI > some Julian calendar information with what I've got in file - nothing =n :) I9 > think I'm going crazy with that (too hard case for me).N >C3 > If question sounds very stupid - just ignore it..e >= > thx, > jk@ > p.s. Actually those bytes _should_ represent April 1st, 2004 = 00:00:00[.00 > <-- not sure about that]    -----------------------------  % Date: Thu, 13 May 2004 08:53:47 -0500f% From: Neil Cherry <njc@wolfgang.uucp> " Subject: VMS, tftp & ftp questions. Message-ID: <slrnca6va1.mm4.njc@wolfgang.uucp>  C I need to use my Alpha (VMS 7.3) as a tftp server for my LAN. Hen I D ftp the files to the Alpha were should I put them so tftp can accessB them. Also how do I 'touch' a file (create an empty file) so I can? save my equipment binaries and config files to the tftp server?   1 Any pointers would be greatly appreciated. Thanks0   --=20hD Linux Home Automation         Neil Cherry        ncherry@comcast.net; http://home.comcast.net/~ncherry/               (Text only)f= http://linuxha.sourceforge.net/                 (SourceForge)d8 http://hcs.sourceforge.net/                     (HCS II)    -----------------------------  # Date: Thu, 13 May 2004 15:16:16 GMT # From: hoff@hp.nospam (Hoff Hoffman) & Subject: Re: VMS, tftp & ftp questions1 Message-ID: <4jMoc.1390$q53.420@news.cpqcorp.net>   < In article <slrnca6va1.mm4.njc@wolfgang.uucp>, Neil Cherry = <njc@wolfgang.uucp> writes:0D :I need to use my Alpha (VMS 7.3) as a tftp server for my LAN. Hen IE :ftp the files to the Alpha were should I put them so tftp can accessPC :them. Also how do I 'touch' a file (create an empty file) so I can @ :save my equipment binaries and config files to the tftp server?  E   For configuring an FTP or TFTP server, please install and configure0E   the TCP/IP Services for OpenVMS package, or one of the available=20    third-party TCP/IP Stacks.  9   For "touch", please search the OpenVMS FAQ for "touch".   >   To create an empty file, you can use COPY NLA0: filename, or'   a text editor, or the CREATE command.p  F   Available documentation resources include the OpenVMS user's manual,E   the OpenVMS system manager's manual, the TCP/IP services management,G   documentation (FTP and TFTP have chapters in this manual), and the=20    OpenVMS FAQ. =20  G   For programming, there exists the OpenVMS Programming Concepts manualcC   and various language-specific manuals -- since you are probably =a workingfC   in C or C++, there is a section on C and C++ programming in the =- OpenVMS-F   FAQ, there are C programming manuals, and (somewhat confusingly) theC   most recent C library documentation now ships with the OpenVMS=20oG   documentation set (as the run-time library itself ships with OpenVMS, G   and not with the compiler) and not with the C language documentation.   C   You'll want to upgrade to OpenVMS Alpha V7.3-2 and (whether you =h upgradeoG   OpenVMS or not) you'll want to install the mandatory ECO kits for the/   OpenVMS release in use./  2  ---------------------------- #include <rtfaq.h> = -----------------------------h3     For additional, please see the OpenVMS FAQ -- =n www.hp.com/go/openvms/faq 4  --------------------------- pure personal opinion = ---------------------------UE         Hoff (Stephen) Hoffman   OpenVMS Engineering   hoff[at]hp.com     -----------------------------    Date: 12 May 2004 23:04:09 -0700. From: dieter.rossbach@gmx.de (dieter rossbach)2 Subject: Re: XFC experiences on OpenVMS Alpha V7.3= Message-ID: <e1d40caf.0405122204.7700970d@posting.google.com>   D We use XFC with 7.3, 7.3-1 and 7.3-2, latest patch levels, without = problems   Regards7   Dieter    -----------------------------   Date: 13 May 2004 06:05:29 GMT2 From: "Dave Weatherall" <djw-nothere@nospam.nohow>2 Subject: Re: You'll never guess what HP advertised? Message-ID: <DTiotGxQ0bj6-pn2-yw6UEf4Vbdx4@dave2_os2.home.ours>p  $ On Wed, 12 May 2004 13:14:25 UTC,=20< koehler@eisner.nospam.encompasserve.org (Bob Koehler) wrote:  B > In article <c7scnd$30tn$1@news.cybercity.dk>, "Karsten Nyblad" = <nospam@nospam.com> writes:p > >=20I > > True.  But most bugs can be fixed without understanding more than a =r fewp > > hundred lines of code. >=20 >    ROTFLOL >=20; >    You've been doing maintenance programing for how long?   2 Actually I tend agree with him. He did say 'most'.  E Its a function of the bug. If it's a crash, you've a good place to=20 G start the trace from, divide by zero, invalid pointer etc.. The more=20mH obscure the behaviour, the wider the knowledge of the code/system you=20H need. I'm sure we've all helped colleagues to trace bugs without even=20A seeing a line of their code. We draw on our experience and our=20t abilities in logic/deduction..  4 It all depends, as my economics teacher used to say.   --=209 Cheers - Dave.    -----------------------------   Date: 13 May 2004 06:05:18 GMT2 From: "Dave Weatherall" <djw-nothere@nospam.nohow>2 Subject: Re: You'll never guess what HP advertised? Message-ID: <DTiotGxQ0bj6-pn2-WGgd8gdyZU10@dave2_os2.home.ours>t  F On Thu, 13 May 2004 03:22:18 UTC, "Karsten Nyblad" <nospam@nospam.com> wrote: >=209 > No, I am careful not to put to much faith into roadmapso  H > David wrote that clustering in VMS is much better than clustering in = current4F > HP-UX.  But that does not seem to be true 1-2 years from now.  So inJ > principle HP could try actively selling VMS in the 1-2 years time span = untilwF > HP-UX gets better clustering.  I just do not think that would be a = good > idea.X  E Do you not see the contradiction in these two statements? Unless I=20aF misread it and the 'not good idea' is believing the roadmap. I take it7 to be a comment on how/whether HP should market(?) VMS.M   --=205 Cheers - Dave.    -----------------------------  % Date: Thu, 13 May 2004 11:45:45 +0200-  From: "Dr. Dweeb" <dr@dweeb.com>2 Subject: Re: You'll never guess what HP advertised, Message-ID: <c7vg49$rac$1@news.cybercity.dk>   Karsten Nyblad wrote:74 > "Main, Kerry" <kerry.main@hp.com> wrote in message >vJ news:FD827B33AB0D9C4E92EACEEFEE2BA2FB3145A9@tayexc19.americas.cpqcorp.net= ..@ >>> True.  But most bugs can be fixed without understanding more >>> than a few >>> hundred lines of code. > F >> Please, this is getting silly - you do not fix OS kernel, security,B >> cluster and driver problems by only understanding a few hundredB >> lines of code. Especially ones which deal with loading, clusterD >> interaction, race problems caused by faster servers, intermittent2 >> crashes (hw?/os?/isv? software issues?) etc ... >sB > OK, you get a longer answer.  I would avoid using an unsupportedD > operating system when ever I can.  Unfortunately there is always a@ > risk ending up running an important application on unsupported > hardware and software. >hH > Even if you buy operating systems large market share from big softwareD > vendors, there is a chance that the operating system becomes EOLedG > some time in the future.  Just look at VMS.  It was a very successful:D > operating system for many years.  HP might stop supporting it someB > times in the future.  Let us assume that happens.  There will beG > organizations that will want to use VMS after HP has stopped support.=D > What do you think these organization would like if given these two? > options:  1)  HP opens the sources of VMS and makes it freelye/ > available.  2)  HP shreds the sources of VMS.t > D > You are right that many of the problems you list, are difficult to= > solve, but do you really think, e.g., security problems areoG > necessarily difficult to solve once the problems have been disclosed?oD > E.g., do you think you need to understand an entire application toC > fix a buffer over round?  In that case I know one consultant I amS > never going to hire. >    I can name one as well.y  E > Try thinking about the bugs fixes you have made during your career.oE > Do you agree with me in that the main part of those have been fixedc > just a few lines?y >m  @ Some, Yes.  Most, no. And in any case, the number of lines fixedH (added/changed/deleted) is not some general function of the complexity = of the problem.  G I recall spending 5 days understanding/fixing an 8000 line program to =M solveeG a trivial programming error that required less than 10 lines of code toeI correct.  Most of the time was spent constructing a logically exhausive =* testC data set to test all possible inputs in order to isolate under whatnE circumstances the code produced incorrect results.  In other words, =t figuringH out where in the code to even look.  Which category is this in ?  This = is a' pretty normal state of affairs I think.u  ? Sometimes you have to throw it all away because it is so brokentJ structurally, that it cannot be patched.  Like the classic 80% solution, = the H 7-10 split, there is no way to reach the goal (10 pins) from the current; position. This is a pretty normal state of affairs I think.e  $ Like someone else wrote, it depends.  D However, statistically speaking, the number of people qualified to =	 modify OS-C code at client sites is really small and for a host of reasons, I =- suspect., few sites actually want that responsibility.  I Anyhow, this discussion is tangential to the original point.  The right =t and)D technical possibility to modify OS code is, in general, a specious = argument in support of an OpenSource OS.-    5 Finally, as an observation on this particular matter,c  F I have been hearing hotshoe CS grads still wet behind the ears spout = this@ crap for more than 20 years.  Same, Same.  Only time and a few = spectacular / budget overuns will teaches them any different.4  A You believe your position to be true, and as a matter of simple =o psychology,aG have invested part of your ego in taking what is technically known as a-I *position*.  Nothing anyone says will cause you to alter this position, =r onlyC *events* which falsify it, and thus create the cognitive dissonance0H necessary for a realignment of your mind - and perhaps not even then - = areeH likely to have affect.  Its just the way most people are wired together.D Some minds are sufficiently well trained to avoid the burned fingers0 requirement, but they are sadly in the minority.    
 Dr. Dweeb.   >  Karsten NybladO! > ibpit1202 at sneakemail dot comW    -----------------------------  + Date: Thu, 13 May 2004 11:50:15 +0000 (UTC)v From: david20@alpha2.mdx.ac.uk2 Subject: Re: You'll never guess what HP advertised) Message-ID: <c7vndn$ob6$1@news.mdx.ac.uk>f  ? In article <c7uple$34h$1@news.cybercity.dk>, "Karsten Nyblad" =t <nospam@nospam.com> writes:d >l, >"Dr. Dweeb" <dr@dweeb.com> wrote in message( >news:c7ukge$302j$1@news.cybercity.dk...@ >> If you are willing to pay for a promised future, then you are >permanantaly,6 >> irrevocabley placed in J.P.Barnums famous category.E >> If you you ar unaware of what this is, the popular vernacular is =m "idiot",E >> though in your obvious case I might concede "sprog" or "neophyte".  >>D >> I listened to a similar clown make this argument 10 years befor a >particulareG >> feature was even partially available in Unix, as a an argument for =r MOVINGG >> AWAY from VMS to Unix.  Sadly, the moron collected his exhorbitant =  salaryF >> and the corporation suffered severely for their folly for quite a = long >time  >> afterwards. >>H >> Og det var et stort dansk firma, og "guruen" var en idiot dansker.  = S=E5 >dansk- >> at det g=F8r ondt.  Kort sagt, en "moron".e >eH >Gee, what a language.  I am feeling like back in the good old days of = Carl >Lydick. >-J >No, I am careful not to put to much faith into roadmaps.  Once I bought = a.J >type wheel printer, and the Danish letter =E6 on one of the type wheels = wassJ >not printed on the line.  It was a type wheel we did not need, so I did = not,H >care that much.  I was told that the problem would be fixed, but when I+ >asked a year later, I got the same answer.t >eH >Second time:  Digital told me that there would most likely be support = forhJ >SMTP in UCX in version 1.3.  When version 13a showed it was not there.  = Not 8 >a big problem since MX was free and you could buy PMDF. >-C >But I learned not to trust roadmaps.  I would not buy a software =t
 product onH >promises, I cannot get in writing.  An oral promise tends to be worth = the-G >paper it is written on.  When I can, I avoid products were all I can =  get is >written promises. > E >> > You do not revive a product like an operating system because youx, >> > intend to sell it a little over a year. >> > >>$ >> I do not understand this comment. >4G >David wrote that clustering in VMS is much better than clustering in =8 currenttE >HP-UX.  But that does not seem to be true 1-2 years from now.  So in I >principle HP could try actively selling VMS in the 1-2 years time span =  untilaH >HP-UX gets better clustering.  I just do not think that would be a good >idea. >iH No what I wrote was that clustering in VMS is much better than that in = Tru64.H Hence you are positing a future in which HP-UX both gets everything in = Tru64sE (which in itself will not be a simple job) and then improves on that.p4 Nothing is impossible but I wouldn't hold my breath.  G You might as well posit that HP-UX will add in Fault tolerance Tandem =o	 style anda' hence HP should also not advertise NSK.tD (after all one of the VAX systems produced in the past was a fault = tolerant system).  G There are many other reasons apart from clustering or fault tolerance =s whyhB certain VMS or NSK customers would probably not be happy at that = prospect not least of which is security.     
 David Webb VMS and Unix team leader CCSS Middlesex University       >Karsten Nyblad   >ibpit1202 at sneakemail dot com >> >>    -----------------------------  % Date: Thu, 13 May 2004 07:48:17 -0400 ' From: "Main, Kerry" <kerry.main@hp.com>a2 Subject: RE: You'll never guess what HP advertised
 Message-ID: =OF <FD827B33AB0D9C4E92EACEEFEE2BA2FB314663@tayexc19.americas.cpqcorp.net>   > -----Original Message-----6 > From: Karsten Nyblad [mailto:nospam@nospam.com]=3D20 > Sent: May 13, 2004 12:27 AM  > To: Info-VAX@Mvb.Saic.Coma4 > Subject: Re: You'll never guess what HP advertised >=3D20 >=3D204 > "Main, Kerry" <kerry.main@hp.com> wrote in message@ > news:FD827B33AB0D9C4E92EACEEFEE2BA2FB3145A9@tayexc19.americas. > cpqcorp.net...A > >> True.  But most bugs can be fixed without understanding more  > >> than a fewT > >> hundred lines of code.  >=3D20F > >Please, this is getting silly - you do not fix OS kernel, security,? > >cluster and driver problems by only understanding a few=3D20h > hundred lines of> > >code. Especially ones which deal with loading, cluster=3D20 > interaction, racekH > >problems caused by faster servers, intermittent crashes (hw?/os?/isv? > >software issues?) etc ... >=3D20; > OK, you get a longer answer.  I would avoid using an=3D20a > unsupported operatingkD > system when ever I can.  Unfortunately there is always a risk=3D20 > ending upgH > running an important application on unsupported hardware and software. >=3D20H > Even if you buy operating systems large market share from big softwareC > vendors, there is a chance that the operating system becomes=3D20I > EOLed some timewB > in the future.  Just look at VMS.  It was a very successful=3D20 > operating systemA > for many years.  HP might stop supporting it some times in=3D20c > the future.  Let@ > us assume that happens.  There will be organizations that=3D20 > will want to useA > VMS after HP has stopped support.  What do you think these=3D20o > organization? > would like if given these two options:  1)  HP opens the=3D20  > sources of VMS and? > makes it freely available.  2)  HP shreds the sources of VMS.l >=3D20  H Most Customers I deal with would move to another OS platform before theyD would decide to do OS kernel/clustering/driver support on their own.F Especially since even if Sun/HP/IBM did decide to drop one of their OSH platforms, they would typically give something like at least a 5-10 year window.s  G And that applies to any OS platform from one of the major players.=3D20S  B The open source movement has some good idea's, but the "we want toH maintain our own code in case the vendor goes out of business.." is sortG of ok(?) for applications, but is political mumbo jumbo when discussingd OS support.=3D20  C The number of open source folks in the wild (i.e. not working for a C commercial vendor) that have the expertise to maintain kernel code, 7 cluster and drivers is extremely, extremely small.=3D20i  B > Try thinking about the bugs fixes you have made during your=3D20 > career.  Do you-C > agree with me in that the main part of those have been fixed=3D20e > just a few > lines? >=3D20 >  Karsten Nybladj! > ibpit1202 at sneakemail dot comc >=3D20  G I am sure VMS Engineering would state some percentage (likely small) of H bugs are easily fixed when they are pointed out. However, I am sure thatE they would also state the bugs that are the big time killers and showeF stoppers are the ones that are typically very complex and that require: an understanding of the entire system and/or architecture.  ) And that applies to any OS platform.=3D20i   Regardsl  
 Kerry Main Senior Consultante HP Services Canada Voice: 613-592-4660i Fax: 613-591-4477  Email: kerryDOTmainAThpDOTcom00 (remove the DOT's and AT for email address)=3D20    -----------------------------    Date: 13 May 2004 07:02:50 -0600; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler)b2 Subject: Re: You'll never guess what HP advertised3 Message-ID: <tRkaW$1d3J+M@eisner.encompasserve.org>d  E In article <283e357c10ee63b4ffa97fb790d13ebb@news.teranews.com>, JF =c, Mezei <jfmezei.spamnot@teksavvy.com> writes: > Bob Koehler wrote: >=20G > My 370 assembler card says the opposite. It has 360 instruction set =s	 with someeG > added instructions/opcode present only on 370s. Just like Intel addedl; > instructions on Pentiums (such as stuff for games/video).   H    Although the 370 instruction set is a superset of the 360 instructionD    set, the 370 had and used a compatability mode when executing 360
    images.    -----------------------------  % Date: Thu, 13 May 2004 17:46:02 +0200c* From: "Karsten Nyblad" <nospam@nospam.com>2 Subject: Re: You'll never guess what HP advertised- Message-ID: <c8058o$1k13$1@news.cybercity.dk>z  2 "Main, Kerry" <kerry.main@hp.com> wrote in messageJ news:FD827B33AB0D9C4E92EACEEFEE2BA2FB314663@tayexc19.americas.cpqcorp.net= ..F >Most Customers I deal with would move to another OS platform before = theyE >would decide to do OS kernel/clustering/driver support on their own. G >Especially since even if Sun/HP/IBM did decide to drop one of their OSoF >platforms, they would typically give something like at least a 5-10 = year >window.  C >And that applies to any OS platform from one of the major players.p  F Very true.  However, there are always some companies with relatively = stableI applications that chose not to move on.  They are betting on not runing == intoJ problems because they rarely change anything.  That is what we have seen = in	 the past.   D >The number of open source folks in the wild (i.e. not working for aD >commercial vendor) that have the expertise to maintain kernel code,3 >cluster and drivers is extremely, extremely small.	  
 Kernel : Yes.M
 Cluster : Yes D Drivers : I think many of the drivers are developed by hobby people.    -----------------------------  % Date: Thu, 13 May 2004 18:15:44 +0200M* From: "Karsten Nyblad" <nospam@nospam.com>2 Subject: Re: You'll never guess what HP advertised- Message-ID: <c806vr$1m1a$1@news.cybercity.dk>i  + <david20@alpha2.mdx.ac.uk> wrote in messagei# news:c7vndn$ob6$1@news.mdx.ac.uk...r > H > No what I wrote was that clustering in VMS is much better than that in Tru64.H > Hence you are positing a future in which HP-UX both gets everything in Tru64eG > (which in itself will not be a simple job) and then improves on that.a6 > Nothing is impossible but I wouldn't hold my breath. >aG > You might as well posit that HP-UX will add in Fault tolerance Tandem0	 style andT) > hence HP should also not advertise NSK.iD > (after all one of the VAX systems produced in the past was a fault tolerant
 > system). >nI > There are many other reasons apart from clustering or fault tolerance == whynD > certain VMS or NSK customers would probably not be happy at that = prospect not  > least of which is security.    Sorry, my fault.  G So you would like to tell the customers that the security of HP-UX is =c not asJ good as that in VMS.  The customers will hear that the security in HP-UX = isE not good enough.  Rule number A1 while selling is never ever tell a =a customerJ that there is something that your product is not good at something, even = ifJ it is obvious.  Been there.  Tried it.  The customer looked funny, and I = got/< a lecture from the sales person I was supporting afterwards.  F That is also one of the reasons why it will be difficult to actively = sell, VMS clustering currently with selling HP-UX.  - I think it is time we agree that we disagree.a   Karsten Nyblad ibpit1202 at sneakemail dot com     -----------------------------   End of INFO-VAX 2004.265 ************************   ------------------------------  # Date: Fri, 14 May 2004 05:14:08 GMT  From: "o-o-o" <bkt@null.net>& Subject: Re: installation boot failure> Message-ID: <AAYoc.26967$e%1.18687@newssvr22.news.prodigy.com>  I Thanks for the help. This box was indeed running NT4.0 SP6 when I got it.nK But I blew it away, switched the console to serial, switched the os type tor" VMS, and thought I'd give it a go.L the next question is, does anyone want an alpha that can't run VMS? I don't.0 It has some nice parts that could be salvaged...    0 "Hoff Hoffman" <hoff@hp.nospam> wrote in message+ news:59Moc.1389$q53.642@news.cpqcorp.net...oF > In article <DJzoc.8222$Fu7.6643@newssvr24.news.prodigy.com>, "o-o-o" <bkt@null.net> writes:G > :I have a Digital Server 5000 (same as AlphaServer 1200) on which I'm  trying" > :to install VMS. I checked here:6 > :http://h71000.www7.hp.com/openvms/supportchart.htmlK > :to see what version of VMS the system will run, and concluded that 7.3-1o > :should work.0 > . >   Once more into the breach, dear friends... >tL >   The Whitebox Alpha systems are not supported by OpenVMS, as indicated byI >   the lack of a listing in the cited support chart, and as indicated byM textH >   in the FAQ, and as indicated by various topics in the Ask The Wizard area >   asking about this. > L >   Again, the Microsoft Windows NT (only) Alpha systems -- often refered toK >   as the "whitebox" Alpha systems -- are not supported by OpenVMS.   This1K >   includes all members of the DIGITAL Server series boxes, all members ofvL >   the DIGITAL Professional Workstation (PWS) boxes with an -a designation,! >   and the DECpc 150 AXP series.p >oA >   These Windows NT (only) systems ARE NOT SUPPORTED by OpenVMS.m >aI > :Yet when I try to boot the 7.3-1 installation CD, I get the following:l > :rL > :DIGITAL Server 5000 Model 5305 6533A Console V6.0-4, 10-MAY-2001 10:11:42 > :v > :CPU 0 booting > :d! > :(boot dka500.5.0.1.1 -flags 0)l2 > :block 0 of dka500.5.0.1.1 is a valid boot block > .. > :jumping to bootstrap codeD > :%SYSBOOT-I-FILENOTLOC, Unable to locate SYS$CPU_ROUTINES_EA05.EXE? > :%SYSBOOT-E-LDFAIL, failed to load execlet, status = 00000910i >r >eF >   That's a central indication of an unsupported processor -- OpenVMSC >   has no code for the particular processor type -- your system ismC >   identifying itself as a DIGITAL Server series system, and these 5 >   Alpha systems are (wait for it :-) not supported.g >hJ > :I am at a loss for what else to check. Any ideas of where to look next? >e >   The OpenVMS FAQ: > & >     http://www.hp.com/go/openvms/faq >s >   Specifically:  >i> >     SYSBOOT-I-FILENOTLOC, Unable to locate SYS$CPU_ROUTINES?8 >     OpenVMS on the Whitebox Windows-Only series Alpha? >iJ >   Support aside, you might be able to get this box to bootstrap OpenVMS.H >   But you'll have to work at it some, you'll have to check the PCI andI >   system configuration, you won't have support, and the result might ore+ >   might not be stable or even functional./ >sK >   Various of the Whitebox systems have integral motherboard hardware that-K >   is incompatible with the operation of OpenVMS.  (One of the more common H >   problem areas involves the southbridge -- anyone that tells you thatG >   *all* whitebox Alpha systems *will* run OpenVMS is *wrong*, as some:I >   of the southbridges used on some of the whiteboxes are not compatiblerJ >   with OpenVMS.  We had no end of trouble with some of the southbridges.H >   I'm here ignoring the usual and obvious PCI widget support, which is1 >   another area that you will have to evaluate.)s >a > -- >pH >   ps: as you've undoubtedly searched the FAQ for this information, andI >   were clearly unable to locate it, please let me know what your searchyG >   target(s) were and I'll re-word and/or re-index the information foroC >   the next edition; I'll continue on my quest to exterminate this-$ >   Alpha Whitebox support question. >a >0( >  ---------------------------- #include' <rtfaq.h> -----------------------------n4 >     For additional, please see the OpenVMS FAQ --  www.hp.com/go/openvms/faq , >  --------------------------- pure personal# opinion --------------------------- G >         Hoff (Stephen) Hoffman   OpenVMS Engineering   hoff[at]hp.comc >e   ------------------------------  % Date: Fri, 14 May 2004 00:12:28 +0100 & From: Elliott Roper <nospam@yrl.co.uk> Subject: Re: longest uptimen1 Message-ID: <140520040012283554%nospam@yrl.co.uk>l  ? In article <40A2D4DC.6AFBC9C7@NeOaSrPtAhMlNiOnWk.net>, David J.t6 Dachtera <djesys.nospam@NeOaSrPtAhMlNiOnWk.net> wrote:   > "Klaus-D. Bohn" wrote: > >  > > Hello all VMS lovers,o > > B > > im searching for the longest uptime from a VMS System/Cluster. > > C > > If have anyone of you a screen shot "show sys/noproc" and "show-" > > cluster" please send it to me. >  > G > Well, I have a cluster that was established well over a year ago, butrJ > the longest uptime for any member is currently just over 30 days, due to$ > monthly reboots, patches and such.  C One of our customers' machines is showing 393 days. Sadly it may berF rebooted early on Saturday when they take the power off the UPS to fixB a chiller. 393 days ago was their 5 year mandatory building wiring@ inspection power outage. The main application transaction routerB process's show proc/accounting has a connect time of a few secondsD less, and the slave machine, which will take over if the master everF fails, is showing the same uptime give or take a couple of minutes. ItD has between 2 and 30 interactive users at any time, doing everythingA associated with a 24*7 heavy vehicle breakdown service, includingcF micro-managing the telephones. It has been running since 1991, more orE less continuosly, give or take a few hardware upgrades and a building:F move or two. Currently on MicroVax 3100/96 VMS 6.2. It started life on a MicroVax II.  If it ain't broke. Don't fix it!  E Klaus, I took a screen shot if you'd like one. My sig explains how to4 ask (in teco of course)    -- aC I thought I would be the last on earth to mangle my e-mail address.  fsnospam$elliott$$   ------------------------------  % Date: Thu, 13 May 2004 20:29:30 -0500y@ From: "David J. Dachtera" <djesys.nospam@NeOaSrPtAhMlNiOnWk.net> Subject: Re: longest uptime 6 Message-ID: <40A420FA.B799A859@NeOaSrPtAhMlNiOnWk.net>   Syltrem wrote: > I > > Well, I have a cluster that was established well over a year ago, but L > > the longest uptime for any member is currently just over 30 days, due to& > > monthly reboots, patches and such. > >, > > -- > > David J. Dachtera  > > dba DJE Systemsr > > http://www.djesys.com/ >  > What is a monthly reboot? % > This is a VMS forum, not windows...  > ;-)   A Yeah, well, even in the VMS world, pesky app.'s can have resource  leaks...   *SIGH*   -- a David J. Dachterae dba DJE Systemsc http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/r   ------------------------------  % Date: Thu, 13 May 2004 20:35:28 -0500'@ From: "David J. Dachtera" <djesys.nospam@NeOaSrPtAhMlNiOnWk.net>0 Subject: Re: Parity errors when HSG80 restarted?6 Message-ID: <40A42260.9277788D@NeOaSrPtAhMlNiOnWk.net>   Malcolm Dunnett wrote: > 8 > In article <40A2D791.B1F99EF4@NeOaSrPtAhMlNiOnWk.net>,G >    "David J. Dachtera" <djesys.nospam@NeOaSrPtAhMlNiOnWk.net> writes:0 > O > >> > Also, contact your support rep. about an update to the HSG firmware, but0) > >> > be ready for some unexpected news.r > >>5 > >>    Can you elaborate on what that news might be?  > >-K > > Hp no longer provides HSG firmware upgrades in the form of PCMCIA cardsrB > > - you must execute a series of commands (can be scripted usingJ > > Hyperterm(!), KEA, SmartTerm, Reflection, etc.) to patch your existing% > > cards up to the next rev. levels.a > >o > E >    Ah, way ahead of you on that one. The 8.6-4 card seems to be the0D > highest available, when I had the arrays upgraded to 8.6 last yearE > HP ( or was it still Compaq then? ) brought me some 8.6-4 cards andtD > then the tech unpacked his laptop and proceeded to load a bunch ofD > patches via the console port ( I think they used SmartTerm ). I'veF > since installed a few more recent patches that I downloaded from theG > net, using C-Kermit as the scripting tool. Not too big a problem onceiE > I worked the process out. If the HP website is to be believed ( not H > necessarily a safe assumption ) then I'm right up-to-date with patches > at 8.6-13. > I > > Be advised also that V8.7-7 takes away some functionality in HSZTERM.1I > > Not sure quite what, because my partner stopped upgrading our HSGs at F > > the -6 level. So, V8.7-6 may well be the last usable rev. of HSG80G > > firmware for sites not using PCM, ConsoleWorks, etc. to manage your 	 > > HSGs.  > M >   I'm still on 8.6 and I do the programming either via a terminal connectedoI > to the console port or via Storageworks command console. I've not tried0D > HSZTERM, but I presume it would have the same functionality as theF > "command line window" in SWCC? Being an old-timer I prefer the VT510E > connected to the console port route - less cruft to get in the way.e  D HSZTERM$SCSIPAD - the program behind SET HOST/SCSI - is more or lessE equivalent to SET HOST/HSC or SET HOST/DUP. For HSG pairs, target thec! GGA devices to minimize problems.a  G Unlike the others, however, HSZTERM accepts a single controller commanduF as $rest_of_line, or allows SYS$INPUT to be redirected to a disk file.  F Closest DEC & Co. ever came to getting it right for storage management1 automation, because at one time it was supported.a   -- 1 David J. Dachteram dba DJE Systemsm http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/1   ------------------------------  % Date: Thu, 13 May 2004 20:39:39 -0500t@ From: "David J. Dachtera" <djesys.nospam@NeOaSrPtAhMlNiOnWk.net>0 Subject: Re: Parity errors when HSG80 restarted?6 Message-ID: <40A4235B.C6E0A7E9@NeOaSrPtAhMlNiOnWk.net>   norm.raphael@metso.com wrote:x > E > "David J. Dachtera" <djesys.nospam@NeOaSrPtAhMlNiOnWk.net> wrote on  > 05/11/2004 10:53:50 PM:  >  > [snip] > >\C > > *** WARNING *** WARNING *** WARNING *** WARNING *** WARNING ***e > > K > > There is a problem with the FC-SCSI patch V5 for V7.3-1. Do not install H > > this patch, and if you have installed it, back it out IMMEDIATELY!!! > >/J > > Do not attempt to apply the replacement patch (V6) as it has also been
 > > recalled.d > >nC > > *** WARNING *** WARNING *** WARNING *** WARNING *** WARNING ***u > >  > J > I guess I was away when this happened.  Can you explain the problem withJ > the FC_SCSI V5 patch or point me to the documentation about the problem,	 > please.j > K > (I am still at V4 of this ECO, but I have some other Level 1 ECO's I wantuC > to schedule, and I certainly do not want FC problems afterwards.)n  F The scoop on this one is that at extremely high I/O rates (higher thanH the lab apparently knows how to attain), the driver does something wrongH and results in writing data with bad parity - yet get "%SYSTEM-F-PARITY,+ parity error" when you try to read it back.t  C V6 was supposed to fix this, but was recalled within hours of beingr
 announced.  ; Our primary ISV duplicated the problem in their labs, also.n   --   David J. Dachtera  dba DJE Systemsr http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/    ------------------------------  % Date: Thu, 13 May 2004 17:48:46 -04004  From: John Santos <JOHN@egh.com>G Subject: Re: Poor VMS file create performance compared to Linux/FreeBSD:6 Message-ID: <1040513172802.23626A-100000@Ives.egh.com>  $ On 13 May 2004, Simon Clubley wrote:  J > I am seeing very poor VMS I/O performance compared to Linux/FreeBSD whenJ > trying to perform the same task - unpacking a GCC tar archive - on these1 > operating systems. A table of results is below.t > N > I've gone into a lot of detail about my test environments, so that hopefullyM > if I've missed something it will be noticed. I am also interested in seeing0: > the results on other platforms if anyone is so inclined. > N > The tar archive in question is GCC 3.4.0 which contains 22735 files/dirs andO > is 191,467,520 bytes long uncompressed; as VMSTAR does not support compressedfR > archives I uncompressed all archives first[1]. The times below are for the untarN > operation only. The archive is available from the usual GCC mirrors. I used: > M > ftp://ftp.mirror.ac.uk/sites/sources.redhat.com/pub/gcc/releases/gcc-3.4.0/nI > The archive is: gcc-3.4.0.tar.bz2 (25.9MB) or gcc-3.4.0.tar.gz (34.3MB)u > E > I turned off Linux deferred writes on the filesystem in question by.G > remounting the filesystem with "-o sync" on the mount command; I havec? > included the Linux deferred write performance for comparison.s > O > In all cases, the input tar file was on the same disk as the extracted files.o >  > Machines:oH >                                                           Avg      AvgI > Machine/OS	Filesystem  Disk/Interface        RPM   Seek(ms) Latency(ms) # > AlphaStation 200 4/166, VMS 7.3-1VH >                 ODS-5       IBM DORS-32160, SCSI  5400   8.5      5.56) > Testdrive Alpha DS20, 500MHz, VMS 7.3-2EH >                 ODS-5       RZ2DD-LS, SCSI        10000  5.4      2.99" > Pentium 233Mhz, Redhat Linux 6.2H >                 ext2        ST320410A, IDE        5400   8.9      5.55 > Pentium 233Mhz, FreeBSD 4.5lG >                 FreeBSD     ST34321A, IDE         5400   11.5     5.6 
 > 		native > H > I choose the PC above so that the CPU performance would be as close as > possible to my AlphaStation. > K > In all cases, the command line was:	[tar_program] -vxf [archive_name].tar, > 
 > Results: >   > Machine		OS		Unpack time	NotesE > ===================================================================eC > AS 200 4/166	VMS 7.3-1	1 hour 46 mins	Extract didn't complete [2]i: > Pentium 233	Redhat 6.2	2 mins 10 secs	Deferred writes ON; > Pentium 233	Redhat 6.2	4 mins 22 secs	Deferred writes OFFn- > Pentium 233	FreeBSD 4.5	15 mins 55 secs	[3] C > Testdrive	VMS 7.3-2	55 mins 48 secs	Extract didn't complete [2,4]h  H Do you know the total # of I/Os (Direct and buffered), the compute time,I and the system overhead (% CPU usage in kernel, exec, super and interrupteH stack?)  (I think SUPER mode time should be charged back to the process, but inner modes may not be.)  ? The reason I ask is because everyone's immediate assumption is SB probably that it is I/O-bound, and the slowness is due to VMS fileD system overhead (extending directories, flushing attributes to disk,A etc.)  But what if something horrible is going on in some librarysA routine, and it is actually compute bound?  Maybe using differente? compiler options or a different compiler would help enormously.r  ? Someone in another followup mentioned using the DEC/Compaq/HP CaB compiler as something that might help.  I don't know what compilerA was used for the tar on Hunter's site, but I would guess that thee> GNV stuff all uses the latest and greatest DEC C, and you said that's even slower...3  F Assuming it is I/O bound, maybe VMS needs a "$ MOUNT/CACHE=DANGEROUS" I (with a synonymous "/CACHE=UNIX_STYLE") where *nothing* is flushed exceptnI when it runs out of buffers or when the volume is dismounted (or maybe oniI an explicit "set volume/cache=safe" (aka "/cache=vms_style"), which would D write everything out.) You could use this on scratch volumes or whenJ loading software, when you are willing to delete and/or re-init everythingF and start over in the event of an untimely system crash.  (If at mountF time, a volume is found to have the "dangerous" and "mounted" bits setE (indicating it wasn't cleanly dismounted), it tells you so and eitherdB write-locks it or forces you to use a /OVERRIDE=possibly_messed_up qualifier.)n  @ (Sorry for all the nested parens... too much LISP in college...)    < Maybe this flag should be on a directory basis (with default? inherited when the directory is created from its parent) rather-? than on a volume basis.  You would have to delete the directoryw> and ANALYZE/DISK/REPAIR to recover after a crash, but the rest of the disk should be fine.t   > N > For VMS, I used VMSTAR V3.4-1, obtained from Hunter's site and linked on theH > target machine. All VMS measurements were done using this version. [5] > N > Does anyone have any ideas on how I can increase VMS file create performance6 > to bring it up to Linux/FreeBSD performance levels ? >  > Simon. > P > [1] Interestingly, the uncompress operation completed in roughly the same timeK > (allowing for hardware differences) on VMS as it did on Linux. This would K > seem to suggest that raw I/O write performance on VMS is ok, and it's thecF > actual creating of files that appears to be the performance problem. > N > [2] The untar failed towards the end, because of a full path that was longerO > than 100 bytes. VMSTAR does not support LongLink tar blocks, which means thataL > VMSTAR will not fully unpack any tar archive with a full path of more thanP > 100 bytes. (I've already notified Hunter; I encountered the same problem doingH > some early testing with a GCC preview kit.) The last file successfully > extracted was: > R > Jan  4 23:18:59 2004     2316 [C3.pvg.libstdc^+^+-v3.testsuite.27_io.basic_istre& > am.extractors_arithmetic.char]13.cc;' >                               [C3.^.]u" > tar: unexpected EOF on tar file. > Q > This is sufficiently close to the end of the archive that VMSTAR shouldn't haveb? > taken more than a few minutes to finish extracting the files.i > M > [3] The FreeBSD filesystem metadata is written using sync I/O; file data is O > written using async I/O. Even allowing for the slower disk that FreeBSD is on H > (you can't place FreeBSD on an extended partition), the native FreeBSDN > filesystem appears to be slower than ext2 at writing large numbers of files.' > Any FreeBSD experts care to comment ?u > L > [4] I made sure that VMSTAR was the only process doing any significant I/O( > by using MONITOR at regular intervals. > K > [5] The VMSTAR binaries supplied as part of the GNV kits were slower thanf. > the VMSTAR that I linked. I do not know why. >  > -- rD > Simon Clubley, clubley@remove_me.eisner.decus.org-Earth.UFP       R > SCO: Proudly pushing Microsoft down to #2 on the list of most disliked companies >  >    --   John Santosh Evans Griffiths & Hart, Inc. 781-861-0670 ext 539   ------------------------------  % Date: Thu, 13 May 2004 20:47:17 -0500s@ From: "David J. Dachtera" <djesys.nospam@NeOaSrPtAhMlNiOnWk.net>G Subject: Re: Poor VMS file create performance compared to Linux/FreeBSD 6 Message-ID: <40A42525.16A56FD5@NeOaSrPtAhMlNiOnWk.net>   Simon Clubley wrote: > J > I am seeing very poor VMS I/O performance compared to Linux/FreeBSD whenJ > trying to perform the same task - unpacking a GCC tar archive - on these1 > operating systems. A table of results is below.a >  > [snip] > K > [5] The VMSTAR binaries supplied as part of the GNV kits were slower thane. > the VMSTAR that I linked. I do not know why.  @ I think there's a primary clue. tar, being a UN*X "baby", is not optimized for VMS.  A In general, I find the performance of most UN*X-ware to be ratheraH disappointing on VMS, and I don't consider it for comparison to "native"	 VMS code.e   --   David J. Dachteraa dba DJE Systemsf http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/s   ------------------------------  % Date: Thu, 13 May 2004 21:27:40 -0700 + From: "Barry Treahy, Jr." <Treahy@MMaz.com>iG Subject: Re: Poor VMS file create performance compared to Linux/FreeBSDe' Message-ID: <40A44ABC.2000702@MMaz.com>    David J. Dachtera wrote:   >Simon Clubley wrote:c >  r >sJ >>I am seeing very poor VMS I/O performance compared to Linux/FreeBSD whenJ >>trying to perform the same task - unpacking a GCC tar archive - on these1 >>operating systems. A table of results is below.t >> >>[snip] >>K >>[5] The VMSTAR binaries supplied as part of the GNV kits were slower thann. >>the VMSTAR that I linked. I do not know why. >>     >> >mA >I think there's a primary clue. tar, being a UN*X "baby", is note >optimized for VMS.l >eB >In general, I find the performance of most UN*X-ware to be ratherI >disappointing on VMS, and I don't consider it for comparison to "native".
 >VMS code. >  d >uD "Native" probably isn't the best word as anything that is compiled, F linked, and runs on VMS is "native" to the VMS OS.  The real issue is E that the API libraries that allow Un*x code to run on VMS, basically hF emulating a Un*x environment, fumbles the proverbial performance ball H therefore I would say that has less to do with the apps, otherwise they 6 would also run poorly on Un*x, which isn't the case...    F IMNHO, VMS needs to do a better job of handling the Un*x code because B Un*x predates VMS, and though many of us thrashed Un*x during the H VMS/Un*x wars, I suspect that Un*x will outlive VMS and it is something H we need to learned to live with... Frankly, I'd take a good Un*x system  over Microcrap anytime...c     Barrys     --    > Barry Treahy, Jr                       E-mail: Treahy@MMaz.com> Midwest Microwave                          Phone: 480/314-1320> Vice President & CIO                         FAX: 480/661-7028                        e   ------------------------------  % Date: Fri, 14 May 2004 07:21:03 +0200K* From: Paul Sture <nospam@sture.homeip.net>G Subject: Re: Poor VMS file create performance compared to Linux/FreeBSD1* Message-ID: <2gj3a1F3cp4iU1@uni-berlin.de>   Simon Clubley wrote:Y > In article <urMoc.1392$q53.296@news.cpqcorp.net>, hoff@hp.nospam (Hoff Hoffman) writes:7 > L >>  Search for "mybenchmark"; a set of tools made available by David Mathog.J >>  Around the discussions of the tools, you'll also find various previous' >>  discussions of the I/O performance.p >> >  > - > Thanks for the pointer; I will take a look.i >   I IIRC, one of the things which came out of David Mathog's research or one hE of the related discussions was that of default RMS extend size. IIRC bG (again), Hunter Goatley incorporated a larger value for this in HGFTP, a+ resulting in significant performance gains.f   ------------------------------  % Date: Fri, 14 May 2004 00:45:12 +0200n* From: Paul Sture <nospam@sture.homeip.net>E Subject: Re: Terry Shannon's "HP: Two Years Post-Merger Presentation"t* Message-ID: <2gic3qF3aefqU1@uni-berlin.de>   Paul Sture wrote:oJ > As reported by The Inquirer at http://www.theinquirer.net/?article=158798 > Terry Shannon's has a presentation up in PDF format at > C > http://www.shannonknowshpc.com/stories.php?story=04/05/12/7726404g > J > Terry claims "It took me 150 hours out of a 168-hour week to render the J > slideware reasonably consistent with HP standards; you can download the H > results in just a few minutes.", but I must assume that his server is I > either getting hammered or is somewhat sick, as it felt like 150 hours r > to download it.  > H > The PDF pages containing graphics took ages to render too, so to save @ > others the frustration I've knocked up some quick web pages at > 0 > http://www.sture.homeip.net/itanium/index.html >   I Well, aargh here. I have inspected my web logs, and I see that many have nF looked at the thumbnails, but not the full-screen shots. Apologies if  anyone was disappointed there.  , Sorry to all I didn't explain how it worked.  H Click on the first thumbnail and you get a full screenshot of the first H slide, click the Next link for the rest until it gets to slide 30, then  UP, Page 2 and follow the rest.a  + Simple, but then I assumed... Blegh...Sigh.  ...iH However, I can report that Apache on Alpha VMS 7.3-1 has just delivered E a load of pages to the outside world, and continues to do so at this aG moment; less than 5% CPU usage over several hours, even when I've been 1 hammering it on other things.i   ------------------------------  # Date: Thu, 13 May 2004 18:40:57 GMTn# From: hoff@hp.nospam (Hoff Hoffman) 1 Subject: Re: Timestamp format stored in RMS file?i1 Message-ID: <ZiPoc.1423$Oi3.172@news.cpqcorp.net>h  ] In article <opr7x0i8q6xckwek@diablo.uninet.ee>, Jaan Kronberg <jaan.kronberg@mail.ee> writes:n  	 :8 bytes:s :0000 36BD 04FB A200 :oL :Could anybody "translate" it to "readable" date without using any built-in  :OpenVMS function?  E   See the FAQ for an explanation of the OpenVMS quadword time format.e  H :Or, alternatively, using Perl standard functions (let's assume I don't : :have any modules and I don't know C - and I really don't)  C   Y'all might want to include "Perl" in the title, if y'all want toa,   attract the attention of the Perl folks...  C   I would tend to expect that there is some Perl module for OpenVMSiG   somewhere that handles this conversion.  Or you can call through intoeA   C code, such as something derived from the code attached below.c  J :I've read bunch of documents regarding Julian calendar, tried to compare L :some Julian calendar information with what I've got in file - nothing :) I 8 :think I'm going crazy with that (too hard case for me).  D   Details on the various tinme constructs known as Julian are listedE   in the OpenvMS FAQ, as are details around the 100ns units since the A   system base date (17-Nov-1858), as stored in the time quadword.j  2 :If question sounds very stupid - just ignore it..  E   The real question is "what are you up to, and why no system service8	   calls?"4  K :p.s. Actually those bytes _should_ represent April 1st, 2004 00:00:00[.00 o :<-- not sure about that]t  #   It is.  See example C code below.     N  ---------------------------- #include <rtfaq.h> -----------------------------K     For additional, please see the OpenVMS FAQ -- www.hp.com/go/openvms/faqgN  --------------------------- pure personal opinion ---------------------------E         Hoff (Stephen) Hoffman   OpenVMS Engineering   hoff[at]hp.comr     --   #include <descrip.h> #include <lib$routines.h>s #include <ssdef.h> #include <starlet.h> #include <stdio.h> #include <stsdef.h>  #define MAXASCTIMLEN 23  main()     {r     int RetStat;:     $DESCRIPTOR( TimeSourceAscii, "01-JAN-1995 00:00:01");X     char TimeSourceBinary[] = { 0x00, 0x000, 0x036, 0x0BD, 0x004, 0x0FB, 0x0A2, 0x000 };,     struct dsc$descriptor FormattedTimeDesc;+     char FormattedTimeBuffer[MAXASCTIMLEN];      int TimeBinaryQW[2];      unsigned char *TimeBinaryUB;       /*     //	example 1:I)     //	Convert time from binary to ASCII.y     */2     FormattedTimeDesc.dsc$w_length = MAXASCTIMLEN;2     FormattedTimeDesc.dsc$b_dtype = DSC$K_DTYPE_T;2     FormattedTimeDesc.dsc$b_class = DSC$K_CLASS_S;:     FormattedTimeDesc.dsc$a_pointer = FormattedTimeBuffer;  G     RetStat = sys$asctim( 0, &FormattedTimeDesc, TimeSourceBinary, 0 );i8     if (!$VMS_STATUS_SUCCESS( RetStat )) return RetStat;3     RetStat = lib$put_output( &FormattedTimeDesc );V8     if (!$VMS_STATUS_SUCCESS( RetStat )) return RetStat;       /*     //	example 2:y;     //	Convert time from ASCII to binary and back to ASCII.h     */;     RetStat = sys$bintim( &TimeSourceAscii, TimeBinaryQW );o8     if (!$VMS_STATUS_SUCCESS( RetStat )) return RetStat;C     RetStat = sys$asctim( 0, &FormattedTimeDesc, TimeBinaryQW, 0 );a8     if (!$VMS_STATUS_SUCCESS( RetStat )) return RetStat;)     TimeBinaryUB = (void *) TimeBinaryQW;"W     printf("0x0%2.2x 0x0%2.2x 0x0%2.2x 0x0%2.2x 0x0%2.2x 0x0%2.2x 0x0%2.2x 0x0%2.2x\n",oB 	(unsigned char) TimeBinaryUB[0], (unsigned char) TimeBinaryUB[1],B 	(unsigned char) TimeBinaryUB[2], (unsigned char) TimeBinaryUB[3],B 	(unsigned char) TimeBinaryUB[4], (unsigned char) TimeBinaryUB[5],D 	(unsigned char) TimeBinaryUB[6], (unsigned char) TimeBinaryUB[7] );3     RetStat = lib$put_output( &FormattedTimeDesc );o8     if (!$VMS_STATUS_SUCCESS( RetStat )) return RetStat;          return SS$_NORMAL;     }e   ------------------------------  % Date: Thu, 13 May 2004 17:52:59 -0400   From: John Santos <JOHN@egh.com>& Subject: Re: VMS, tftp & ftp questions6 Message-ID: <1040513175023.23626B-100000@Ives.egh.com>  ( On Thu, 13 May 2004, Hoff Hoffman wrote:  X > In article <slrnca6va1.mm4.njc@wolfgang.uucp>, Neil Cherry <njc@wolfgang.uucp> writes:F > :I need to use my Alpha (VMS 7.3) as a tftp server for my LAN. Hen IG > :ftp the files to the Alpha were should I put them so tftp can accesspE > :them. Also how do I 'touch' a file (create an empty file) so I cantB > :save my equipment binaries and config files to the tftp server? > G >   For configuring an FTP or TFTP server, please install and configureDE >   the TCP/IP Services for OpenVMS package, or one of the available 0 >   third-party TCP/IP Stacks. > ; >   For "touch", please search the OpenVMS FAQ for "touch".h  = The FAQ doesn't seem to mention that "touch" is also includedo@ in GNV, and seems to work (very brief, non-expert testing), just like the Unix version.   > @ >   To create an empty file, you can use COPY NLA0: filename, or) >   a text editor, or the CREATE command.6 > H >   Available documentation resources include the OpenVMS user's manual,G >   the OpenVMS system manager's manual, the TCP/IP services managementoG >   documentation (FTP and TFTP have chapters in this manual), and the t >   OpenVMS FAQ.   > I >   For programming, there exists the OpenVMS Programming Concepts manual.K >   and various language-specific manuals -- since you are probably working K >   in C or C++, there is a section on C and C++ programming in the OpenVMShH >   FAQ, there are C programming manuals, and (somewhat confusingly) theC >   most recent C library documentation now ships with the OpenVMS EI >   documentation set (as the run-time library itself ships with OpenVMS,eI >   and not with the compiler) and not with the C language documentation.d > K >   You'll want to upgrade to OpenVMS Alpha V7.3-2 and (whether you upgradeuI >   OpenVMS or not) you'll want to install the mandatory ECO kits for thel >   OpenVMS release in use.s > P >  ---------------------------- #include <rtfaq.h> -----------------------------M >     For additional, please see the OpenVMS FAQ -- www.hp.com/go/openvms/faqiP >  --------------------------- pure personal opinion ---------------------------G >         Hoff (Stephen) Hoffman   OpenVMS Engineering   hoff[at]hp.coma >  >  >    -- n John Santos  Evans Griffiths & Hart, Inc. 781-861-0670 ext 539   ------------------------------  % Date: Thu, 13 May 2004 23:20:39 -0500 % From: Neil Cherry <njc@wolfgang.uucp>x& Subject: Re: VMS, tftp & ftp questions. Message-ID: <slrnca8i33.skg.njc@wolfgang.uucp>  5 On Thu, 13 May 2004 15:16:16 GMT, Hoff Hoffman wrote:-X > In article <slrnca6va1.mm4.njc@wolfgang.uucp>, Neil Cherry <njc@wolfgang.uucp> writes:E >:I need to use my Alpha (VMS 7.3) as a tftp server for my LAN. Hen IsF >:ftp the files to the Alpha were should I put them so tftp can accessD >:them. Also how do I 'touch' a file (create an empty file) so I canA >:save my equipment binaries and config files to the tftp server?r > G >   For configuring an FTP or TFTP server, please install and configurenE >   the TCP/IP Services for OpenVMS package, or one of the available s >   third-party TCP/IP Stacks. > ; >   For "touch", please search the OpenVMS FAQ for "touch".  > @ >   To create an empty file, you can use COPY NLA0: filename, or) >   a text editor, or the CREATE command.a  F Thanks, do you know the path to where tftp file will land? I need that  so I can create the files there.   -- sD Linux Home Automation         Neil Cherry        ncherry@comcast.net; http://home.comcast.net/~ncherry/               (Text only) = http://linuxha.sourceforge.net/                 (SourceForge)y8 http://hcs.sourceforge.net/                     (HCS II)   ------------------------------  % Date: Thu, 13 May 2004 19:33:26 +0100s) From: Tony Arnold <tony.arnold@man.ac.uk>u Subject: Re: XDMP and TCPIP 5.4e* Message-ID: <2ghtbnF3536jU1@uni-berlin.de>   Martin,t  D Thanks for your very useful responses, however, I'm not making much 	 progress!i  I I enabled the login log file as you suggested. It was empty after trying rI to login! Note that the login hangs after entering usernmae and password wF with just the hour glass cursor sitting there. I kill the X server to 
 terminate.  G I disabled the new desk top as you suggested, but I'm not sure if this 0H has any effect on XDM logins, as I still get the new desktop style hour . glass cursor! The login still hangs as before.  E I tried adding my PC's X server as an entry in XSERVERS.TXT, and the oF login screen popped up when I restarted DECwindows and I could log in H successfully! I had trouble logging out as it seemed to kill the window # manager but left my DECterms there!S  F I can also run the command procedure generated by XDM that starts the % dtsession manually and that works OK.t  1 BTW, logging in to the workstation console is OK.    Tony.l   Martin Kirby wrote: B > I had another look at the code. If my theory is correct then theG > Traditional desktop login should work as it does not include the sames > coding error.  >  > Martin Kirby   -- 7F Tony Arnold, Deputy to the Head of COS Division, Manchester Computing,: University of Manchester, Oxford Road, Manchester M13 9PL.F T: +44 (0)161 275 6093, F: +44 (0)870 136 1004, M: +44 (0)773 330 0039E E-mail: tony.arnold@man.ac.uk, Home: http://www.man.ac.uk/Tony.Arnold    ------------------------------  + Date: Thu, 13 May 2004 17:15:55 +0000 (UTC)n From: david20@alpha2.mdx.ac.uk2 Subject: Re: You'll never guess what HP advertised) Message-ID: <c80aga$13r$1@news.mdx.ac.uk>o  Z In article <c806vr$1m1a$1@news.cybercity.dk>, "Karsten Nyblad" <nospam@nospam.com> writes: >t, ><david20@alpha2.mdx.ac.uk> wrote in message$ >news:c7vndn$ob6$1@news.mdx.ac.uk... >>I >> No what I wrote was that clustering in VMS is much better than that int >Tru64.bI >> Hence you are positing a future in which HP-UX both gets everything inl >Tru64H >> (which in itself will not be a simple job) and then improves on that.7 >> Nothing is impossible but I wouldn't hold my breath.  >>H >> You might as well posit that HP-UX will add in Fault tolerance Tandem
 >style and* >> hence HP should also not advertise NSK.E >> (after all one of the VAX systems produced in the past was a faultS	 >tolerantM >> system).3 >>L >> There are many other reasons apart from clustering or fault tolerance whyL >> certain VMS or NSK customers would probably not be happy at that prospect >not >> least of which is security. >B >Sorry, my fault.n >eM >So you would like to tell the customers that the security of HP-UX is not aslL >good as that in VMS.  The customers will hear that the security in HP-UX isM >not good enough.  Rule number A1 while selling is never ever tell a customeryL >that there is something that your product is not good at something, even ifM >it is obvious.  Been there.  Tried it.  The customer looked funny, and I gota= >a lecture from the sales person I was supporting afterwards.  >   N It depends how you do it. You just tell them that VMS security is better than 8 Unix security (not specifically HP-UX) and back that up.G Anybody who uses Unix knows how often security patches are released for? whichever variant they use.mH When doing any specific comparisons you do them against competitors UnixI offerings you don't do them specifically against your own Unix offerings.nK If someone asks specifically about HP-UX you then compare it against those r other Unix offerings. M To promote one product doesn't mean you have to disparage your other products E - you promote all of your products against your competitors products.o  I If HP followed your advice they would be reduced to only selling one OS - K probably windows - since to promote anything else would mean that they were < telling the customer that that OS was not good at something.  J No OS in existence is the best at everything and it's doubtful any OS will ever be.  J >That is also one of the reasons why it will be difficult to actively sell- >VMS clustering currently with selling HP-UX.a >    HP-UX has it's own strengths.eN When it gets the equivalent of TRU64's clustering then I don't see any problemI with promoting VMS clustering and at the same time promoting HP-UX's new w- clustering as being the best Unix clustering.h    . >I think it is time we agree that we disagree. >e  H Yes I don't see you convincing me and I somehow don't think I'm going toM convince you but I hope the discussion has at least made you consider some ofn! the issues a bit more critically.   
 David Webb VMS and Unix team leader CCSS Middlesex University     >Karsten Nyblad-  >ibpit1202 at sneakemail dot com >( >p   ------------------------------   Date: 13 May 04 15:24:08 EDT) From: cook@wvnvms.wvnet.edu (George Cook)a2 Subject: Re: You'll never guess what HP advertised! Message-ID: <TKeKOfWyGVx5@wvnvms>t  q In article <tRkaW$1d3J+M@eisner.encompasserve.org>, koehler@eisner.nospam.encompasserve.org (Bob Koehler) writes:vr > In article <283e357c10ee63b4ffa97fb790d13ebb@news.teranews.com>, JF Mezei <jfmezei.spamnot@teksavvy.com> writes: >> Bob Koehler wrote:s >> iP >> My 370 assembler card says the opposite. It has 360 instruction set with someH >> added instructions/opcode present only on 370s. Just like Intel added< >> instructions on Pentiums (such as stuff for games/video). > J >    Although the 370 instruction set is a superset of the 360 instructionF >    set, the 370 had and used a compatability mode when executing 360 >    images. > H I pulled out both my 360 and 370 System Summary books last night.  ThereJ is no mention of any type of 360 compatibility (nor simulator or emulator)K mode in the hardware or software.  The 360/25 and at least one 370 offerredtB 360/20 hardware compatibility mode (the 360/20 was a limited 360).  F Yes, I do recall the extra bits of addressing issue being discussed inF my 360/370 assembler class in college, but no longer recall much aboutI it (my assembler book/notes are packed too deeply away for quick access).nJ However both the 360 and 370 System Summary books describe an architecture with 24 bits of addressing.p     George Cookn WVNET  s   ------------------------------  % Date: Thu, 13 May 2004 19:10:59 -0400a' From: "Main, Kerry" <kerry.main@hp.com> 2 Subject: RE: You'll never guess what HP advertisedR Message-ID: <FD827B33AB0D9C4E92EACEEFEE2BA2FB3146F9@tayexc19.americas.cpqcorp.net>   > -----Original Message-----E > From: david20@alpha2.mdx.ac.uk [mailto:david20@alpha2.mdx.ac.uk]=20r > Sent: May 13, 2004 1:16 PM > To: Info-VAX@Mvb.Saic.Com-4 > Subject: Re: You'll never guess what HP advertised >=20  	 [snip...]m  : > It depends how you do it. You just tell them that VMS=20 > security is better than=20: > Unix security (not specifically HP-UX) and back that up.  E As a follow-on to David's point, many people are jumping on the LinuxeH bandwagon without fully understanding what it is they are really getting into.3  G As an example, here is a pointer on the Red Hat website that identifiesaG the Red Hat Linux security patches that are available right now (and it  only goes back to March 2003):  6 https://rhn.redhat.com/errata/rh9-errata-security.html  F Hint - 11 security patches released in April 2004 alone. These are not0 patches in general - they are security specific.   Regards,  
 Kerry Main Senior Consultant  HP Services Canada Voice: 613-592-4660/ Fax: 613-591-4477- Email: kerryDOTmainAThpDOTcome. (remove the DOT's and AT for email address)=20   ------------------------------   End of INFO-VAX 2004.266 ************************0 restarted?6 Message-ID: <40A42260.9277788D@NeOaSrPtAhMlNiOnWk.net>   Malcolm Dunnett wrote: > 8 > In article <40A2D791.B1F99EF4@NeOaSrPtAhMlNiOnWk.net>,G >    "David J. Dachtera" <djesys.nospam@NeOaSrPtAhMlNiOnWk.net> writes:0 > O > >> > Also, contact your  comando
Date: Thu, 30 Jan 2003 20:18:02 +0100
Lines: 7
Message-ID: <b1btp5$11ni2s$1@ID-175145.news.dfncis.de>
References: <HBc_9.143795$AA2.5884622@news2.tin.it> <slrnb3imgk.jb9.mlemmo-nospam@slack.deepspace> <Kpd_9.144203$AA2.5899957@news2.tin.it> <Otd_9.144236$AA2.5901305@news2.tin.it> <slrnb3iqo1.jkh.mlemmo-nospam@slack.deepspace>
Reply-To: gugmar@despammed.com
NNTP-Posting-Host: host33-34.pool62211.interbusiness.it (62.211.34.33)
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Con