1 INFO-VAX	Mon, 13 Oct 2003	Volume 2003 : Issue 567       Contents:( Re: "Old" patch site no longer updated !. Re: Another reason to loathe Micro$haft OSs...< Re: cpu upgradable in HP x4000 and Dell Precision dual xeon?< Re: cpu upgradable in HP x4000 and Dell Precision dual xeon?< Re: cpu upgradable in HP x4000 and Dell Precision dual xeon?1 Re: It's now or perhaps never for a VMS comeback! # Re: Laptop, Reflections, delete key # Re: Laptop, Reflections, delete key  Re: Latest VMS Critical Upgrade & Mapping drives between Winders and VMS* Re: Mapping drives between Winders and VMS* Re: Mapping drives between Winders and VMS PMGC-B card on DEC3000/M600  Re: Sun takes a hit 4 Re: what should SET VOLUME/LABEL do to a shadow set?4 Re: what should SET VOLUME/LABEL do to a shadow set?4 Re: what should SET VOLUME/LABEL do to a shadow set?4 Re: what should SET VOLUME/LABEL do to a shadow set?4 Re: what should SET VOLUME/LABEL do to a shadow set?( Re: where to buy memory for DEC3000/M300( RE: where to buy memory for DEC3000/M3003 Re: where to buy memory for |d|i|g|i|t|a|l| systems 8 Re: White paper: CHARON-VAX cluster w/ shared SCSI disks8 Re: White paper: CHARON-VAX cluster w/ shared SCSI disks8 Re: White paper: CHARON-VAX cluster w/ shared SCSI disks8 Re: White paper: CHARON-VAX cluster w/ shared SCSI disks8 Re: White paper: CHARON-VAX cluster w/ shared SCSI disks  Re: WTH is documentation for SLS  F ----------------------------------------------------------------------  % Date: Mon, 13 Oct 2003 00:11:36 +0000 - From: David B Sneddon <dbsneddon@bigpond.com> 1 Subject: Re: "Old" patch site no longer updated ! * Message-ID: <3F89EDB8.2070605@bigpond.com>   Martin P.J. Zinser espoused: > Dirk Munk wrote: > 2 >> Phillip Helbig---remove CLOTHES to reply wrote: >>@ >>> In article <bmb7rq$ehd$1@news1.tilbu1.nb.home.nl>, Dirk Munk >>> <munk@home.nl> writes: >>> 8 >>>> In case you still check the "old" compaq patch site >>>>: >>>> http://ftp.support.compaq.com/patches/public/vms/axp/ >>>>K >>>> for new patches, then please be advised that new patches are no longer  >>>> added to this site! >>>  >>>  > [...]  > J >>> What about this legendary Australian patch site?  Does it still exist? >> >> >> >> Haven't checked it. >> >>>  >>? > The Australian site still exists, but is not updated anymore.   C I found the Sydney site to be rather well designed and very usable. C I sent a request to the itrc people asking that they reorganize the C itrc site along the lines of the Sydney site since the structure of + the Sydney site was infinitely more usable. E Maybe more requests along those lines would be productive... at least ( some areas of hp appear to be listening.   Regards, Dave.  --  I David B Sneddon (dbs)    VMS Systems Programmer     dbsneddon@bigpond.com I Sneddo's quick guide ...          http://www.users.bigpond.com/dbsneddon/ I DBS freeware at ...   http://www.users.bigpond.com/dbsneddon/software.htm I "Life is what happens to you while you're busy making other plans" Lennon    ------------------------------  % Date: Sun, 12 Oct 2003 20:22:41 +0200 ( From: "H Vlems" <hvlems.nieuw@zonnet.nl>7 Subject: Re: Another reason to loathe Micro$haft OSs... 9 Message-ID: <bmc69b$l1b1c$1@ID-143435.news.uni-berlin.de>   . <VAXman- @SendSpamHere.ORG> schreef in bericht* news:00A2744C.4F760F91@SendSpamHere.ORG...D > Wife's Laptop with Weendoze-2000 "Professional" (tongue in cheek). > K > I was able to get the LNCO2 "driver" installed but I cannot get the setup K > window to accept a "$" when I specify the queue it is to connect to (LPD)  > on my Alpha. > I > Any tips?  Is there some secret <ALT><CTRL><SHIFT> finger knotting that  > will make it accept the "$"? >  > --  2 > VAXman- A Bored Certified VMS Kernel Mode Hacker VAXman(at)TMESIS(dot)COM > 6 >   "Well my son, life is like a beanstalk, isn't it?"  I Err, use a logical name on the VMS system without a "$"? I know, it feels L like giving in to the evil empire here but life is just too short to try and make a bent tool useful...   ------------------------------  % Date: Sun, 12 Oct 2003 20:14:55 +0200 ( From: "H Vlems" <hvlems.nieuw@zonnet.nl>E Subject: Re: cpu upgradable in HP x4000 and Dell Precision dual xeon? 9 Message-ID: <bmc5qp$l6j6l$1@ID-143435.news.uni-berlin.de>   3 Now why didn't I think of that answer :-) (or simh)   A "Wilm Boerhout" <w.boerhoutOLD@PAINTplanet.nl> schreef in bericht $ news:bmbdt8$91d$1@reader11.wxs.nl...H > VMS 7.3 running on CHARON-VAX running on Xeon is certainly no joke. Or* > if it is, it's a fast and cheap joke :-) >  > H Vlems wrote: > 7 > > "Michael Unger" <unger@decus.de> schreef in bericht 7 > > news:bm9f0i$k5o9n$1@ID-152801.news.uni-berlin.de...  > > ) > >>On 2003-10-11 19:18, "H Vlems" wrote:  > >> > >>; > >>>"Lia" <ballbustingbabe@hotmail.com> schreef in bericht < > >>>news:f98a9e28.0310110725.6e78bd74@posting.google.com... > >>>  > >>>>Hey there, > >>>>H > >>>>I'm looking at two used systems to buy right now...a Dell 620 withJ > >>>>dual xeon 1ghz processors, and an HP x4000 with dual xeon p4 1.7 ghz > >>>>processors.  > >>>> > >>>>[...]  > >>> J > >>>Glad to help Lia, but one of the rules in this group is that you have toI > >>>specify the version of the operating system you're running. Now what  > >  > > version  > > 2 > >>>of VMS do you (want to) run on those systems? > >> > >>VMS running on XEONs???  > >> > >>Michael  > >> > > H > > You wouldn't recognize a joke unless it hits you square in the face, would 	 > > you ?  > >  > >  >  > --   > Wilm Boerhout  >  > w.boerhoutOLD@PAINTplanet.nl > + > remove OLD PAINT from reply address first ) > *** my opinions are strictly my own ***  >    ------------------------------  % Date: Sun, 12 Oct 2003 20:15:49 +0200 ( From: "H Vlems" <hvlems.nieuw@zonnet.nl>E Subject: Re: cpu upgradable in HP x4000 and Dell Precision dual xeon? 9 Message-ID: <bmc5sf$kl000$1@ID-143435.news.uni-berlin.de>   9 "Paul Sture" <nospam@sture.homeip.net> schreef in bericht * news:3F898C03.744E25B9@sture.homeip.net... > H Vlems wrote: > > : > > "Lia" <ballbustingbabe@hotmail.com> schreef in bericht; > > news:f98a9e28.0310110725.6e78bd74@posting.google.com...  > > > Hey there, > > > H > > > I'm looking at two used systems to buy right now...a Dell 620 withJ > > > dual xeon 1ghz processors, and an HP x4000 with dual xeon p4 1.7 ghz > > > processors.  > > > F > > > Two questions: First, are these cpu chips upgradable and to what= > > > speed? Second, I'm not sure if the Dell is a P3 or a P4 8 > > > processor...what's the difference between the two? > > > ) > > > I look forward to hearing from you!  > > > 	 > > > Lia  > > L > > Glad to help Lia, but one of the rules in this group is that you have toH > > specify the version of the operating system you're running. Now what version 1 > > of VMS do you (want to) run on those systems?  > >  > > Hans >  > LOL Hans :-)
 > A nice one!  >   3 Thanks Paul, Michael's reply left me wondering ....    ------------------------------  % Date: Sun, 12 Oct 2003 20:33:41 +0200 $ From: Michael Unger <unger@decus.de>E Subject: Re: cpu upgradable in HP x4000 and Dell Precision dual xeon? 9 Message-ID: <bmc82g$koqnk$1@ID-152801.news.uni-berlin.de>   % On 2003-10-12 20:15, "H Vlems" wrote:   ; > "Paul Sture" <nospam@sture.homeip.net> schreef in bericht , > news:3F898C03.744E25B9@sture.homeip.net... >> H Vlems wrote:  >> >; >> > "Lia" <ballbustingbabe@hotmail.com> schreef in bericht < >> > news:f98a9e28.0310110725.6e78bd74@posting.google.com... >> > > Hey there,  >> > > >> > > [...] >> >M >> > Glad to help Lia, but one of the rules in this group is that you have to I >> > specify the version of the operating system you're running. Now what 	 > version 2 >> > of VMS do you (want to) run on those systems? >> >	 >> > Hans  >> >> LOL Hans :-)  >> A nice one! >> > 5 > Thanks Paul, Michael's reply left me wondering ....   H Apparently the OP isn't a "regular" so starting with a joke might not beF appropriate. (Assuming of course the posting made its way to the groupF it was intended to be sent to. And yes, I'm aware of the platforms VMS is running on.)    Michael    --  ; Real names enhance the probability of getting real answers. @ Please do *not* send "Security Patch Notifications" or "SecurityA Updates"; this system isn't running a Micro$oft operating system. = And don't annoy me <mailto:postmaster@[127.0.0.1]> please ;-)    ------------------------------  % Date: Sun, 12 Oct 2003 18:35:19 -0500 1 From: "David J. Dachtera" <djesys.nospam@fsi.net> : Subject: Re: It's now or perhaps never for a VMS comeback!' Message-ID: <3F89E537.2961DDD1@fsi.net>    Didier Morandi wrote:  >  > Bob Ceculski wrote:  > 4 > > is Carly smart enough to take advantage of this?1 > > It is now or perhaps never for a final chance 2 > > for the best os in the world to reclaim number > > one ...  > H > Please excuse me, Bob. I always tried to avoid to comments your posts,< > as I am not sure what you try to do/demonstrate with them. > I > But VMS *is* back. Why are you insisting so much? Did you have recently F > a chance to have a beer/glass of wine/milk with some VMS specialistsC > from former DEC (people with badge numbers less than 200'000) who F > crossed the COMPAQ era? They say: "today, we really feel that we are > back to DEC".   H It's possible that those customers who disagree about "back to DEC" will) follow-up to your post in regard to that.    --   David J. Dachtera  dba DJE Systems  http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/    ------------------------------  % Date: Sun, 12 Oct 2003 18:31:08 -0500 1 From: "David J. Dachtera" <djesys.nospam@fsi.net> , Subject: Re: Laptop, Reflections, delete key' Message-ID: <3F89E43C.580D0D99@fsi.net>    Paul Sture wrote:  >  > David J. Dachtera wrote: > > % > > VAXman-, @SendSpamHere.ORG wrote:  > > > J > > > Argh!  I have to use a CRAPTOP and it has Reflections.  I can TELNETI > > > to my systems but I can't seem to figure out how to make the delete / > > > key send a delete instead of a backspace.  > > > J > > > I've looked at the keymap stuff and I think I have it setup to map aH > > > PeeCee delete to the VT keyboard delete but it still doesn't work. > > >  > > > Any pointers?  > > > K > > > FWIW, this exercise only goes to enforce my disdain for these loathe- K > > > some piece of shit toys.  I have diddled about with this cybertoy and K > > > its playwarez for hours and still can't do a simple character delete.  > > K > > Depending what version of WRQ Reflection (singular), Setup -> Terminal, K > > click the Keyboard tab, select the radio button that says "VT Backspace L > > sends" (*)Delete. Then, click <OK> and save your settings (File -> Save) > > at some appropriate point. > >  > I > I am sligtly suprised here, because although it's been something like a I > decade since I used Reflections, it appeared to be one of the better Vt  > emulators.  A Reflection (singular, see the title bar of the product window for 9 verification) is still the Cadillac of terminal programs.   E Reflection/2 is now known as "Reflection for UNIX and Digital". (Note / that UNIX is listed first in the product name.)   = Reflection/4 is now known as "Reflection for ReGIS Graphics".   G I suspect that the out-of-the-box-default (<DEL> key sends <BS>) has to H do with UN*X's dominance vs. VMS. WRQ, like the rest of the market, goes with what is most prominent.   --   David J. Dachtera  dba DJE Systems  http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/    ------------------------------  # Date: Mon, 13 Oct 2003 00:14:49 GMT 4 From: brad@.gateway.2wire.net (Bradford J. Hamilton), Subject: Re: Laptop, Reflections, delete key/ Message-ID: <Z7mib.747381$Ho3.188675@sccrnsc03>   [ In article <3F89E43C.580D0D99@fsi.net>, "David J. Dachtera" <djesys.nospam@fsi.net> writes:  !snip! ! B !Reflection (singular, see the title bar of the product window for: !verification) is still the Cadillac of terminal programs.  K I must humbly disagree.  Kermit has more features, at lower cost.  The only M area in which WRQ Reflection surpasses Kermit, is that it is easier to set up  (and clone) terminal sessions.  K WRQ Reflection only recently (V10) acquired SSH capability, and the onerous I license costs do not automatically allow for "free" upgrades from earlier 8 products, unless one buys the "top-of-the-line" license.    !snip! !--  !David J. Dachtera !dba DJE Systems !http://www.djesys.com/  ! ) !Unofficial Affordable OpenVMS Home Page:   !http://www.djesys.com/vms/soho/  J __________________________________________________________________________A Bradford J. Hamilton                    "All opinions are my own" K bMradAhamiPltSon-at-coMmcAast.nPeSt     "Lose the MAPS, and replace '-at-'  0                                          with @"   ------------------------------  + Date: Sun, 12 Oct 2003 22:09:17 -0500 (CDT)  From: sms@antinode.org( Subject: Re: Latest VMS Critical Upgrade) Message-ID: <03101222091782@antinode.org>   H    As the current junk-e-mail tide ebbs and flows, I wonder about simple@ enhancements to the TCPIP SMTP system to help cope.  Rather thanF attempting to build every useful anti-junk feature into the main code,G how about adding a big hook which would let the local system manager or  user participate?   G    For example, how difficult would it be to accommodate something like F running a DCL procedure (TCPIP$SMTP_COMMON:SMTP_MESSAGE_CHECK.COM) forC each incoming message?  It would be handy if DCL symbols or logical D names (or even P1-P8) could be provided to hold useful data like theD recipient, the source IP address, popular header items, and the fileG spec where the body of the message could be found.  The procedure could D do its analysis as it pleased, act accordingly, and then exit with aF status which would determine whether the message would be delivered or
 discarded.  '    Easy?  Hard?  Troublesome?  Useless?   H ------------------------------------------------------------------------  4    Steven M. Schweda               (+1) 651-699-98183    382 South Warwick Street        sms@antinode,org     Saint Paul  MN  55105-2547    ------------------------------    Date: 12 Oct 2003 17:32:55 -0700* From: denny.rich@swagelok.com (Denny Rich)/ Subject: Mapping drives between Winders and VMS < Message-ID: <d28306e.0310121632.5a427982@posting.google.com>  : I need to map a drive on an XP Winders (r) box to my Alpha' workstations, or vice versa would work.   D The important thing is to be able to transfer data by simply puttingF it into a folder on one machine and then reading it out on the other. : Data is pretty much text, or possibly zip files (VMS ZIP).  C I have a vague recollection that this is possible. So I looked into E NFS, but there doesn't seem to be a complement to NFS on the XP side. A (I have done this just fine between a Silicon Graphics box and an $ Alpha server a couple of years ago.)  
 Any hints?  E If I had to pick only one, I would like to be able to move files from > XP to a disk on the VMS box. Although both ways would be good.   Thanks   Denny    ------------------------------  # Date: Mon, 13 Oct 2003 01:23:26 GMT 4 From: brad@.gateway.2wire.net (Bradford J. Hamilton)3 Subject: Re: Mapping drives between Winders and VMS > Message-ID: <i8nib.461378$2x.180542@rwcrnsc52.ops.asp.att.net>  i In article <d28306e.0310121632.5a427982@posting.google.com>, denny.rich@swagelok.com (Denny Rich) writes: ; !I need to map a drive on an XP Winders (r) box to my Alpha ( !workstations, or vice versa would work. !snip! !  !Any hints?    How about SAMBA?   ! F !If I had to pick only one, I would like to be able to move files from? !XP to a disk on the VMS box. Although both ways would be good.  !  !Thanks  !  !Denny  J __________________________________________________________________________A Bradford J. Hamilton                    "All opinions are my own" K bMradAhamiPltSon-at-coMmcAast.nPeSt     "Lose the MAPS, and replace '-at-'  0                                          with @"   ------------------------------  # Date: Mon, 13 Oct 2003 01:43:59 GMT + From: Ken Robinson <sendspamhere@rbnsn.com> 3 Subject: Re: Mapping drives between Winders and VMS @ Message-ID: <0d727c4470e3070b46783436a37fb8cc@news.teranews.com>  . denny.rich@swagelok.com (Denny Rich) wrote in 4 news:d28306e.0310121632.5a427982@posting.google.com:  < > I need to map a drive on an XP Winders (r) box to my Alpha) > workstations, or vice versa would work.  > F > The important thing is to be able to transfer data by simply puttingH > it into a folder on one machine and then reading it out on the other. < > Data is pretty much text, or possibly zip files (VMS ZIP). > L looks like you're running a hobbyist version of VMS, so your best bet would = be Samba. This was recently announced on the Samba VMS list:  - (<http://lists.samba.org/listinfo/samba-vms>)   . Subject: Samba/VMS 2.2.8 New version availableI A new version is available on http://www.pi-net.dyndns.org/anonymous/jyc/    Ken Robinson kenrbnsn (at) rbnsn (dot) com    ------------------------------    Date: 12 Oct 2003 17:21:46 -0700* From: denny.rich@swagelok.com (Denny Rich)$ Subject: PMGC-B card on DEC3000/M600< Message-ID: <d28306e.0310121621.3226d511@posting.google.com>   Hello,A I have a Sandpiper (DEC3000/M600) with a PMAGC-B Turbochannel I/O E card, and a VRT19-HA monitor.  The card is running firmware V1.2, and " the Sandpiper is running VMS7.3-1.  > When I boot VMS, VMs complains about DECW$DEVICE-I-NODEVICE No graphics device found.  F The SANDPIPER will operate on the VRT19 as a console device (characterF cell display) or it will send serial console traffic to the modem portC on the back panel if switch S3 is in the proper position. My laptop 3 will act as a console with the proper serial cable.   D The VRT19-HA monitor displays like a character cell terminal on thisA machine, but does a fine DecWindows job on its brother machine, a B DEC3000/M300.  So I conclude the monitor is fine. (The M300 has an= embedded graphics device, not on a turbochannel option card.)   B VMS on the Sandpiper reports a GVA0: device, but it is offline.  IC edited a number of DECW$*.COM files to add a device "GV", but to no  avail.  E Any idea how I can get this Sandpiper to recognize a graphics device?    Thanks.    Denny    ------------------------------  % Date: Mon, 13 Oct 2003 01:01:12 +0200 * From: Paul Sture <nospam@sture.homeip.net> Subject: Re: Sun takes a hit0 Message-ID: <3F89F958.4B3E9C0A@sture.homeip.net>   Paul Sture wrote:  >  Whatever I did write.   F That was a mistake posting from the address I did, beacuse it resulted@ in loads of spam, mainly, but not quite all, of the virus type.   F My conclusion is that TCP/IP V5.3 with RBL and reverse address lookup,C plus NOSPAM as my username (yes, a real one - only one spam so far)  seems quite effective :-)    --  
 Paul Sture   ------------------------------  # Date: Sun, 12 Oct 2003 18:17:25 GMT " From: Lee <lytmah@telusplanet.net>= Subject: Re: what should SET VOLUME/LABEL do to a shadow set? * Message-ID: <VUgib.8355$Pe5.5830@edtnps84>  I The volume label is changed only on the node you executed the command on. B You will have to execute the SET command on all the cluster nodes.  / Phillip Helbig---remove CLOTHES to reply wrote:   G >If one executes the command SET VOLUME/LABLE with a shadow set as the  @ >argument, should it change the volume label on all the members? >  >    >    ------------------------------  + Date: Sun, 12 Oct 2003 22:11:18 +0000 (UTC) P From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply)= Subject: Re: what should SET VOLUME/LABEL do to a shadow set? # Message-ID: <bmcji6$h7$1@online.de>   G In article <VUgib.8355$Pe5.5830@edtnps84>, Lee <lytmah@telusplanet.net>  writes:   K > The volume label is changed only on the node you executed the command on. D > You will have to execute the SET command on all the cluster nodes.  I What does that mean?  Since the HELP says that it does NOT change DISK$*  H logicals, locks etc which use the old name, it seems to me---also based C on the name of the command---that it must change something on disk.   A Suppose I execute the command differently on several nodes, shut  G everything down and remount the disk on a completely different cluster.  What label will it have?   ------------------------------  # Date: Sun, 12 Oct 2003 22:56:06 GMT " From: Lee <lytmah@telusplanet.net>= Subject: Re: what should SET VOLUME/LABEL do to a shadow set? + Message-ID: <a_kib.11617$Pe5.4903@edtnps84>   , This is a multi-part message in MIME format.& --------------0907030001090907060208039 Content-Type: text/plain; charset=us-ascii; format=flowed  Content-Transfer-Encoding: 7bit    Good question.D I posted after I actually went in and changed the volume label of a  shadow on one node of B the cluster.  The other nodes were perfectly happy with their own  different volume labels.0 I was able to execute DIR commands on two nodes.F In the past, I have had to change volume labels of clusterwide shadow  sets.  However, I I made sure I implemented the change on all the nodes immediately and also   revised the A shadow mount procedure which is identical on all the system disk.   H I performed a test (using a non-shadow disk) which shows that the change0 must be done on the disk.  Here are the results.  
 On node A:  ' $ MOUNT /OVER=IDEN /NOUNLOAD $2$DUA668: 2 %MOUNT-I-MOUNTED, UHSJ8 mounted on _$2$DUA668: (L)   $ SHOW DEV $2$DUA668:   G Device                  Device           Error    Volume         Free   	 Trans Mnt G  Name                   Status           Count     Label        Blocks  	 Count Cnt ? $2$DUA668:         (L)  Mounted alloc        0  UHSJ8            9488988     1   1    ! $ SET VOL/LAB=NEWLABEL $2$DUA668:     $ SHOW DEV $2$DUA668:   G Device                  Device           Error    Volume         Free   	 Trans Mnt G  Name                   Status           Count     Label        Blocks  	 Count Cnt ? $2$DUA668:         (L)  Mounted alloc        0  NEWLABEL         9488988     1   1     $ DISM $2$DUA668:    On node  B:   ' $ MOUNT /OVER=IDEN /NOUNLOAD $2$DUA668: : %MOUNT-I-MOUNTED, NEWLABEL mounted on _$2$DUA668: (HSJC06)    $ SHO DEV $2$DUA668:  G Device                  Device           Error    Volume         Free   	 Trans Mnt G  Name                   Status           Count     Label        Blocks  	 Count Cnt ? $2$DUA668:    (HSJC06)  Mounted alloc        0  NEWLABEL         9488988     1   1   G  From this, I would suspect that the change is written out to the disk   when it is dismounted.) Or of course, when the node is shut down.     / Phillip Helbig---remove CLOTHES to reply wrote:   H >In article <VUgib.8355$Pe5.5830@edtnps84>, Lee <lytmah@telusplanet.net>	 >writes:   >  >    > K >>The volume label is changed only on the node you executed the command on. D >>You will have to execute the SET command on all the cluster nodes. >>     >> > J >What does that mean?  Since the HELP says that it does NOT change DISK$* I >logicals, locks etc which use the old name, it seems to me---also based  D >on the name of the command---that it must change something on disk. > B >Suppose I execute the command differently on several nodes, shut H >everything down and remount the disk on a completely different cluster. >What label will it have?  >    >   & --------------090703000109090706020803) Content-Type: text/html; charset=us-ascii  Content-Transfer-Encoding: 7bit   ? <!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">  <html> <head>I   <meta http-equiv="Content-Type" content="text/html;charset=ISO-8859-1">    <title></title>  </head> ' <body text="#000000" bgcolor="#ffffff">  Good question.<br>C I posted after I actually went in and changed the volume label of a  shadow on one node of<br> F the cluster.&nbsp; The other nodes were perfectly happy with their own different volume labels.<br>4 I was able to execute DIR commands on two nodes.<br>E In the past, I have had to change volume labels of clusterwide shadow  sets.&nbsp; However, I<br>C made sure I implemented the change on all the nodes immediately and  also revised the<br>E shadow mount procedure which is identical on all the system disk.<br>  <br>L I performed a test (using a non-shadow disk) which shows that the change<br>9 must be done on the disk.&nbsp; Here are the results.<br>  <br> On node A:<br> <br>+ $ MOUNT /OVER=IDEN /NOUNLOAD $2$DUA668:<br> 6 %MOUNT-I-MOUNTED, UHSJ8 mounted on _$2$DUA668: (L)<br> <br> $ SHOW DEV $2$DUA668:<br>  <br>	Device&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Device&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Error&nbsp;&nbsp;&nbsp; Volume&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Free&nbsp; 
 Trans Mnt<br> &nbsp;Name&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Status&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Count&nbsp;&nbsp;&nbsp;&nbsp; Label&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Blocks
 Count Cnt<br>  $2$DUA668:&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; (L)&nbsp; Mounted alloc&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; 0&nbsp; <b>UHSJ8</b>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; 3 9488988&nbsp;&nbsp;&nbsp;&nbsp; 1&nbsp;&nbsp; 1<br> 
 &nbsp;<br>% $ SET VOL/LAB=NEWLABEL $2$DUA668:<br> 
 &nbsp;<br> $ SHOW DEV $2$DUA668:<br>  <br>	Device&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Device&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Error&nbsp;&nbsp;&nbsp; Volume&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Free&nbsp; 
 Trans Mnt<br> &nbsp;Name&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Status&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Count&nbsp;&nbsp;&nbsp;&nbsp; Label&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Blocks
 Count Cnt<br>  $2$DUA668:&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; (L)&nbsp; Mounted alloc&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; 0&nbsp; <b>NEWLABEL</b>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;3 9488988&nbsp;&nbsp;&nbsp;&nbsp; 1&nbsp;&nbsp; 1<br> 
 &nbsp;<br> $ DISM $2$DUA668:<br>  <br> On node&nbsp; B:<br> <br>+ $ MOUNT /OVER=IDEN /NOUNLOAD $2$DUA668:<br> E %MOUNT-I-MOUNTED, <b>NEWLABEL</b> mounted on _$2$DUA668: (HSJC06)<br> 
 &nbsp;<br> $ SHO DEV $2$DUA668:<br> <br>	Device&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Device&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Error&nbsp;&nbsp;&nbsp; Volume&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Free&nbsp; 
 Trans Mnt<br> &nbsp;Name&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Status&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Count&nbsp;&nbsp;&nbsp;&nbsp; Label&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Blocks
 Count Cnt<br>  $2$DUA668:&nbsp;&nbsp;&nbsp; (HSJC06)&nbsp; Mounted alloc&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; 0&nbsp; <b>NEWLABEL</b>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; 3 9488988&nbsp;&nbsp;&nbsp;&nbsp; 1&nbsp;&nbsp; 1<br>  <br>E From this, I would suspect that the change is written out to the disk  when it is dismounted.<br>- Or of course, when the node is shut down.<br>  <br> <br>3 Phillip Helbig---remove CLOTHES to reply wrote:<br> 8 <blockquote type="cite" cite="midbmcji6$h7$1@online.de">   <pre wrap="">In article &lt;VUgib.8355$Pe5.5830@edtnps84&gt;, Lee <a class="moz-txt-link-rfc2396E" href="mailto:lytmah@telusplanet.net">&lt;lytmah@telusplanet.net&gt;</a> writes:      </pre>   <blockquote type="cite">Z     <pre wrap="">The volume label is changed only on the node you executed the command on.B You will have to execute the SET command on all the cluster nodes.
     </pre>   </blockquote>    <pre wrap=""><!---->I What does that mean?  Since the HELP says that it does NOT change DISK$*  H logicals, locks etc which use the old name, it seems to me---also based C on the name of the command---that it must change something on disk.   A Suppose I execute the command differently on several nodes, shut  G everything down and remount the disk on a completely different cluster.  What label will it have?   </pre>
 </blockquote>  </body>  </html>   ( --------------090703000109090706020803--   ------------------------------  # Date: Sun, 12 Oct 2003 23:31:50 GMT " From:   VAXman-  @SendSpamHere.ORG= Subject: Re: what should SET VOLUME/LABEL do to a shadow set? 0 Message-ID: <00A2747F.F91E0910@SendSpamHere.ORG>  P In article <a_kib.11617$Pe5.4903@edtnps84>, Lee <lytmah@telusplanet.net> writes:- >This is a multi-part message in MIME format. ' >--------------090703000109090706020803 : >Content-Type: text/plain; charset=us-ascii; format=flowed  >Content-Transfer-Encoding: 7bit >  >Good question. E >I posted after I actually went in and changed the volume label of a   >shadow on one node ofC >the cluster.  The other nodes were perfectly happy with their own   >different volume labels. 1 >I was able to execute DIR commands on two nodes. G >In the past, I have had to change volume labels of clusterwide shadow   >sets.  However, IJ >made sure I implemented the change on all the nodes immediately and also  >revised theB >shadow mount procedure which is identical on all the system disk.  H The change is done to the shadow virtual disk and thus to the underlyingH member disks.  The difference you see in the cluster is to the mount useG of the volume lock.  If you were to issue a SET VOLUME/LABEL and $ DIS- G MOUNT/CLUSTER the DSAn device and then reMOUNT/CLUSTER the DSAn device, F you will see that the label is most definately written to the shadow's
 member disks.    --  L VAXman- A Bored Certified VMS Kernel Mode Hacker    VAXman(at)TMESIS(dot)COM             5   "Well my son, life is like a beanstalk, isn't it?"     ------------------------------  + Date: Mon, 13 Oct 2003 03:03:50 +0000 (UTC) 7 From: moroney@world.std.spaamtrap.com (Michael Moroney) = Subject: Re: what should SET VOLUME/LABEL do to a shadow set? ( Message-ID: <bmd4mm$s8c$1@pcls4.std.com>  R helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) writes:  H >In article <VUgib.8355$Pe5.5830@edtnps84>, Lee <lytmah@telusplanet.net>	 >writes:    L >> The volume label is changed only on the node you executed the command on.E >> You will have to execute the SET command on all the cluster nodes.   J >What does that mean?  Since the HELP says that it does NOT change DISK$* I >logicals, locks etc which use the old name, it seems to me---also based  D >on the name of the command---that it must change something on disk.  I Specifically written to the home block (LBN 1) and the backup homeblocks.   B >Suppose I execute the command differently on several nodes, shut H >everything down and remount the disk on a completely different cluster. >What label will it have?   E The value set by the last node, of course.  However, I'm not sure all D nodes are 'notified' when another node executes a $ SET VOLUME/LABEL- command so $ SHOW DEV may show the old label.  --   -Mike    ------------------------------    Date: 12 Oct 2003 18:26:05 -0700* From: denny.rich@swagelok.com (Denny Rich)1 Subject: Re: where to buy memory for DEC3000/M300 < Message-ID: <d28306e.0310121726.1f6bdd02@posting.google.com>  E Like the original message but my system is: DEC 3000/M300. Looks like D 72pin PC stuff. What specs? is it really 34 bit?  and I know if have- to add it in pairs. But what sort of memory ?    thanks   denny     j Chris Scheers <chris@applied-synergy.com> wrote in message news:<3F89AE76.C8369396@applied-synergy.com>...1 > Phillip Helbig---remove CLOTHES to reply wrote:  > > B > > I'm looking to buy some additional memory, particularly for anK > > ALPHAstation 255/233.  Can anyone recommend somewhere where I can order K > > the stuff online, pay by credit card, be 100% sure that what I get will L > > work, have it shipped to Germany with no hassle or exhorbitant cost, and" > > get it for a reasonable price? > > F > > What IS a reasonable price for DEC(-compatible) memory these days? > 2 > I just added 256MB (4x64MB SIMMs) to my 255/233. > 4 > You need 72pin, parity, non-registered 70ns SIMMs. > F > I think I paid about $45 (including shipping) for the 256MB.  It wasH > really memory meant for a HP server, but it works fine in the 255/233. > 2 > I have also used IBM parity memory successfully. >  > Good luck! >  > I > -----------------------------------------------------------------------l& > Chris Scheers, Applied Synergy, Inc. > E > Voice: 817-237-3360            Internet: chris@applied-synergy.com c >   Fax: 817-237-3074    ------------------------------  % Date: Sun, 12 Oct 2003 18:21:52 -0700a# From: "Tom Linden" <tom@kednos.com> 1 Subject: RE: where to buy memory for DEC3000/M300c9 Message-ID: <CIEJLCMNHNNDLLOOGNJIAEMKIDAA.tom@kednos.com>   C It is _almost_ pc memory.  I tried some 4 years ago and it only sawd5 the first 8 MB, so I ended up buying Kingston memory.    >-----Original Message-----l2 >From: Denny Rich [mailto:denny.rich@swagelok.com]' >Sent: Sunday, October 12, 2003 6:26 PM  >To: Info-VAX@Mvb.Saic.Com2 >Subject: Re: where to buy memory for DEC3000/M300 >  > F >Like the original message but my system is: DEC 3000/M300. Looks likeE >72pin PC stuff. What specs? is it really 34 bit?  and I know if have . >to add it in pairs. But what sort of memory ? >l >thankss >e >denny >s >e; >Chris Scheers <chris@applied-synergy.com> wrote in messageh0 >news:<3F89AE76.C8369396@applied-synergy.com>...2 >> Phillip Helbig---remove CLOTHES to reply wrote: >> >C >> > I'm looking to buy some additional memory, particularly for ancL >> > ALPHAstation 255/233.  Can anyone recommend somewhere where I can orderL >> > the stuff online, pay by credit card, be 100% sure that what I get willC >> > work, have it shipped to Germany with no hassle or exhorbitant>
 >cost, and# >> > get it for a reasonable price?  >> >G >> > What IS a reasonable price for DEC(-compatible) memory these days?@ >>3 >> I just added 256MB (4x64MB SIMMs) to my 255/233." >>5 >> You need 72pin, parity, non-registered 70ns SIMMs.q >>G >> I think I paid about $45 (including shipping) for the 256MB.  It wassI >> really memory meant for a HP server, but it works fine in the 255/233.  >>3 >> I have also used IBM parity memory successfully.h >>
 >> Good luck!l >> >>J >> -----------------------------------------------------------------------' >> Chris Scheers, Applied Synergy, Inc.  >>E >> Voice: 817-237-3360            Internet: chris@applied-synergy.come >>   Fax: 817-237-3074 >o >---' >Incoming mail is certified Virus Free. ; >Checked by AVG anti-virus system (http://www.grisoft.com).hA >Version: 6.0.518 / Virus Database: 316 - Release Date: 9/11/2003n >h ---i& Outgoing mail is certified Virus Free.: Checked by AVG anti-virus system (http://www.grisoft.com).@ Version: 6.0.518 / Virus Database: 316 - Release Date: 9/11/2003   ------------------------------  % Date: Sun, 12 Oct 2003 14:41:42 -0500c/ From: Chris Scheers <chris@applied-synergy.com>u< Subject: Re: where to buy memory for |d|i|g|i|t|a|l| systems3 Message-ID: <3F89AE76.C8369396@applied-synergy.com>   / Phillip Helbig---remove CLOTHES to reply wrote:  > @ > I'm looking to buy some additional memory, particularly for anI > ALPHAstation 255/233.  Can anyone recommend somewhere where I can order$I > the stuff online, pay by credit card, be 100% sure that what I get willeJ > work, have it shipped to Germany with no hassle or exhorbitant cost, and  > get it for a reasonable price? > D > What IS a reasonable price for DEC(-compatible) memory these days?  0 I just added 256MB (4x64MB SIMMs) to my 255/233.  2 You need 72pin, parity, non-registered 70ns SIMMs.  D I think I paid about $45 (including shipping) for the 256MB.  It wasF really memory meant for a HP server, but it works fine in the 255/233.  0 I have also used IBM parity memory successfully.  
 Good luck!    G -----------------------------------------------------------------------a$ Chris Scheers, Applied Synergy, Inc.  C Voice: 817-237-3360            Internet: chris@applied-synergy.com l   Fax: 817-237-3074    ------------------------------  % Date: Sun, 12 Oct 2003 20:20:37 +0200 ( From: "H Vlems" <hvlems.nieuw@zonnet.nl>A Subject: Re: White paper: CHARON-VAX cluster w/ shared SCSI disks 9 Message-ID: <bmc65f$kq8ef$1@ID-143435.news.uni-berlin.de>c   Wilm  K if you describe your set up as a "SCSI cluster" (with or without the prefix J shared) most people in this newsgroup will translate that as a VMS clusterJ where data travels a shared SCSI bus. Whatever way you may want to put it,J for VMS system managers the Charon setup is a straight forward NI cluster.8 Which incidentally may also be achieved if you run simh.   Hans  A "Wilm Boerhout" <w.boerhoutOLD@PAINTplanet.nl> schreef in berichti$ news:bmbn9m$lml$1@reader11.wxs.nl...J > Thanks Rob, I will keep this channel open for a few weeks and then write > a version 1.1. >hH > As for the naming, the "Guidelines for OpenVMS Cluster Systems" manualF > speaks of a "SCSI multihost OpenVMS Cluster" where Alpha systems areD > concernd. So, maybe a "SCSI multihost OpenVMS VAXcluster" would beF > appropriate? I like your suggestion of "Mixed Interconnect" as well. >  > Rob Lyons wrote:J > > Thanks for getting the word out to the rest of us who were not able to' attend the Dutch Technical Update Days.y > > K > > I suggest the name for your configuration is LAVC with shared SCSI or an6 Mixed Interconnect Cluster since the interconnect thatK > > carries the cluster traffic names the cluster type, by convention.  Thei< term "SCSI-cluster" incorrectly implies that the SCSI bus is! > > carrying SCA cluster traffic.. > >oE > > Be wary of using the 100Mb Ethernet setting for the NI speed.  In A clusters we've built on the other side of the pond, activities as-K > > simple as disk to disk copy between clusters can swamp the NI and causem8 the cluster to become unstable.  The 10Mb setting avoids > > that pitfall.n > > K > > Shared SCSI has a very limited number of configurations possible due tot< the cable length restrictions of SCSI.  The best way we haveJ > > found to build scalable clusters is to use Fibre Channel.  The same PK6 port driver is used for SCSI and Fibre Channel and theF > > performance profile with an HSG80 or MSA1000 storage controller is impressive.o > > I > > As you find additional hints and kinks, keep the information flowing.z5 The community at large all benefit from your efforts.  > > 
 > > Rob Lyons  > > Resilient Systems Inc. > > w-w-w.resilientsys.com > >b > >d >  > --   > Wilm Boerhout, >m > w.boerhoutOLD@PAINTplanet.nl >P+ > remove OLD PAINT from reply address firste) > *** my opinions are strictly my own ***d >    ------------------------------  % Date: Sun, 12 Oct 2003 20:55:32 +0200 5 From: Wilm Boerhout <wboerhoutOLD@PAINTvxcompany.com>cA Subject: Re: White paper: CHARON-VAX cluster w/ shared SCSI diskse* Message-ID: <bmc8ae$e0t$1@reader11.wxs.nl>  E But Hans, in this setup, the data *does* travel on the SCSI bus. The  3 lock manager traffic user the network interconnect.   E Robert Lyons naming it "Mixed Interconnect"is right in that respect.  E That is exactly why we demonstrated it. Please read the setup in hte H$ white paper, and judge for yourself.  F I may not have been clear enough in my description, please comment so  that it can be made clear.   Wilm   H Vlems wrote: > Wilm > M > if you describe your set up as a "SCSI cluster" (with or without the prefixVL > shared) most people in this newsgroup will translate that as a VMS clusterL > where data travels a shared SCSI bus. Whatever way you may want to put it,L > for VMS system managers the Charon setup is a straight forward NI cluster.: > Which incidentally may also be achieved if you run simh. >  > Hans > C > "Wilm Boerhout" <w.boerhoutOLD@PAINTplanet.nl> schreef in bericht(& > news:bmbn9m$lml$1@reader11.wxs.nl... > J >>Thanks Rob, I will keep this channel open for a few weeks and then write >>a version 1.1. >>H >>As for the naming, the "Guidelines for OpenVMS Cluster Systems" manualF >>speaks of a "SCSI multihost OpenVMS Cluster" where Alpha systems areD >>concernd. So, maybe a "SCSI multihost OpenVMS VAXcluster" would beF >>appropriate? I like your suggestion of "Mixed Interconnect" as well. >> >>Rob Lyons wrote: >>I >>>Thanks for getting the word out to the rest of us who were not able toi > ) > attend the Dutch Technical Update Days.$ > J >>>I suggest the name for your configuration is LAVC with shared SCSI or a > 8 > Mixed Interconnect Cluster since the interconnect that > J >>>carries the cluster traffic names the cluster type, by convention.  The > > > term "SCSI-cluster" incorrectly implies that the SCSI bus is >   >>>carrying SCA cluster traffic. >>>iD >>>Be wary of using the 100Mb Ethernet setting for the NI speed.  In > C > clusters we've built on the other side of the pond, activities asc > J >>>simple as disk to disk copy between clusters can swamp the NI and cause > : > the cluster to become unstable.  The 10Mb setting avoids >  >>>that pitfall. >>> J >>>Shared SCSI has a very limited number of configurations possible due to > > > the cable length restrictions of SCSI.  The best way we have > I >>>found to build scalable clusters is to use Fibre Channel.  The same PKd > 8 > port driver is used for SCSI and Fibre Channel and the > E >>>performance profile with an HSG80 or MSA1000 storage controller isi > 
 > impressive.o > H >>>As you find additional hints and kinks, keep the information flowing. > 7 > The community at large all benefit from your efforts./ >  >>>Rob Lyons >>>Resilient Systems Inc.n >>>w-w-w.resilientsys.comd >>>. >>>m >> >>--   >>Wilm Boerhout- >> >>w.boerhoutOLD@PAINTplanet.nl >>+ >>remove OLD PAINT from reply address firste) >>*** my opinions are strictly my own ***R >> >  >  >    --  
 Wilm Boerhout8 VX Company, The Netherlandsa   wboerhoutOLD@PAINTvxcompany.comc'   (remove OLD PAINT from reply address)m  ' *** my opinions are strictly my own ***v   ------------------------------  % Date: Sun, 12 Oct 2003 21:27:24 +0200m( From: "H Vlems" <hvlems.nieuw@zonnet.nl>A Subject: Re: White paper: CHARON-VAX cluster w/ shared SCSI disks 9 Message-ID: <bmca2m$ldp2h$1@ID-143435.news.uni-berlin.de>a   Wilm,u  L you've got me thoroughly confused by now. A shared SCSI cluster has at leastK one SCSI bus with two VMS hosts attached to the bus (those buses). The dataeF travels the SCSI bus(es) the lockmanager uses, by default, the fastest" common NI it can lay its hands on.K AFAIK Charon is an emulator that runs on a Windows platform and shared SCSIcJ (with both windows hosts at work) just does not exist. IMO there's no such) thing as a simulated shwared SCSI bus :-)d  6 So obviously I missed something, somewhere, but what ?   hansD "Wilm Boerhout" <wboerhoutOLD@PAINTvxcompany.com> schreef in bericht$ news:bmc8ae$e0t$1@reader11.wxs.nl...F > But Hans, in this setup, the data *does* travel on the SCSI bus. The5 > lock manager traffic user the network interconnect.  >oF > Robert Lyons naming it "Mixed Interconnect"is right in that respect.F > That is exactly why we demonstrated it. Please read the setup in hte& > white paper, and judge for yourself. > G > I may not have been clear enough in my description, please comment soi > that it can be made clear. >l > Wilm >  > H Vlems wrote: > > Wilm > >SH > > if you describe your set up as a "SCSI cluster" (with or without the prefixF > > shared) most people in this newsgroup will translate that as a VMS cluster J > > where data travels a shared SCSI bus. Whatever way you may want to put it,/E > > for VMS system managers the Charon setup is a straight forward NIw cluster.< > > Which incidentally may also be achieved if you run simh. > >  > > Hans > >ME > > "Wilm Boerhout" <w.boerhoutOLD@PAINTplanet.nl> schreef in berichtL( > > news:bmbn9m$lml$1@reader11.wxs.nl... > >mL > >>Thanks Rob, I will keep this channel open for a few weeks and then write > >>a version 1.1. > >>J > >>As for the naming, the "Guidelines for OpenVMS Cluster Systems" manualH > >>speaks of a "SCSI multihost OpenVMS Cluster" where Alpha systems areF > >>concernd. So, maybe a "SCSI multihost OpenVMS VAXcluster" would beH > >>appropriate? I like your suggestion of "Mixed Interconnect" as well. > >> > >>Rob Lyons wrote: > >>K > >>>Thanks for getting the word out to the rest of us who were not able to  > >i+ > > attend the Dutch Technical Update Days.a > >tL > >>>I suggest the name for your configuration is LAVC with shared SCSI or a > >o: > > Mixed Interconnect Cluster since the interconnect that > >DL > >>>carries the cluster traffic names the cluster type, by convention.  The > > @ > > term "SCSI-cluster" incorrectly implies that the SCSI bus is > >_" > >>>carrying SCA cluster traffic. > >>>rF > >>>Be wary of using the 100Mb Ethernet setting for the NI speed.  In > >lE > > clusters we've built on the other side of the pond, activities as  > > L > >>>simple as disk to disk copy between clusters can swamp the NI and cause > >,< > > the cluster to become unstable.  The 10Mb setting avoids > >: > >>>that pitfall. > >>>gL > >>>Shared SCSI has a very limited number of configurations possible due to > >i@ > > the cable length restrictions of SCSI.  The best way we have > >PK > >>>found to build scalable clusters is to use Fibre Channel.  The same PKk > >u: > > port driver is used for SCSI and Fibre Channel and the > >eG > >>>performance profile with an HSG80 or MSA1000 storage controller is  > >c > > impressive.o > >mJ > >>>As you find additional hints and kinks, keep the information flowing. > >H9 > > The community at large all benefit from your efforts.D > >m > >>>Rob Lyons > >>>Resilient Systems Inc.  > >>>w-w-w.resilientsys.comt > >>>t > >>>i > >> > >>-- c > >>Wilm Boerhouta > >>  > >>w.boerhoutOLD@PAINTplanet.nl > >>- > >>remove OLD PAINT from reply address firstr+ > >>*** my opinions are strictly my own ***a > >> > >  > >h > >i >i > -- t > Wilm Boerhoutd > VX Company, The Netherlandsu >b! > wboerhoutOLD@PAINTvxcompany.coms) >   (remove OLD PAINT from reply address)- >-) > *** my opinions are strictly my own ***- >-   ------------------------------  % Date: Sun, 12 Oct 2003 21:44:18 +020035 From: Wilm Boerhout <wboerhoutOLD@PAINTvxcompany.com>iA Subject: Re: White paper: CHARON-VAX cluster w/ shared SCSI disks * Message-ID: <bmcb5s$imm$1@reader11.wxs.nl>  F Look at the picture in the white paper. The SCSI box has a multipoint E controller -and some internal redundancy as well-. I connect two VMS eF hosts (well, CHARON-VAX emulated hosts) using CHARON-VAX's ability to F connect to physical SCSI disks, bypassing Windoze disk device drivers  altogether.p  H More graphical: from the >>> boot prompt on one (emulated) VAX, you see G all SCSI disks (but not the other VAX host). From the other VAX host's aH boot prompt you see those same SCSI disks (but not the first host). So, F for all practical purposes, the VAXen share the SCSI disks. They look I the same, they have the same names, so let's assume they are the same :-)c  G You need the VMS lock manager of course to ensure the integrity of the kH I/O to / from the disks. Windoze (sort of in between, but not very much H so with CHARON-VAX) knows absolutely nothing about synchronizing I/O on 3 these same disks. You need VMS clusters to do that.d  H In fact, I like to think it's more like DSSI than shared SCSI on Alpha, D but it's not the same. It works -has been running for days with all ? kinds of loads- not only by haphazard operational luck, but by .H VAXcluster architectural design as well. Of course, the VAX cluster SPD & will not be rewritten because of this.  H Is it a bird? Is it a plane? It's... it's... it's a VAXcluster! We hope G to have one on line before long so the community can take a look at it a from the inside.   H Vlems wrote:   > Wilm,S > N > you've got me thoroughly confused by now. A shared SCSI cluster has at leastM > one SCSI bus with two VMS hosts attached to the bus (those buses). The datatH > travels the SCSI bus(es) the lockmanager uses, by default, the fastest$ > common NI it can lay its hands on.M > AFAIK Charon is an emulator that runs on a Windows platform and shared SCSI-L > (with both windows hosts at work) just does not exist. IMO there's no such+ > thing as a simulated shwared SCSI bus :-)i > 8 > So obviously I missed something, somewhere, but what ? >  > hansF > "Wilm Boerhout" <wboerhoutOLD@PAINTvxcompany.com> schreef in bericht& > news:bmc8ae$e0t$1@reader11.wxs.nl... > F >>But Hans, in this setup, the data *does* travel on the SCSI bus. The5 >>lock manager traffic user the network interconnect.  >>F >>Robert Lyons naming it "Mixed Interconnect"is right in that respect.F >>That is exactly why we demonstrated it. Please read the setup in hte& >>white paper, and judge for yourself. >>G >>I may not have been clear enough in my description, please comment sos >>that it can be made clear. >> >>Wilm >> >>H Vlems wrote: >> >>>WilmS >>>:G >>>if you describe your set up as a "SCSI cluster" (with or without the2 >  > prefix > E >>>shared) most people in this newsgroup will translate that as a VMSo > 	 > cluster  > I >>>where data travels a shared SCSI bus. Whatever way you may want to put  >  > it,O > D >>>for VMS system managers the Charon setup is a straight forward NI > 
 > cluster. > ; >>>Which incidentally may also be achieved if you run simh.g >>>. >>>Hanse >>> D >>>"Wilm Boerhout" <w.boerhoutOLD@PAINTplanet.nl> schreef in bericht' >>>news:bmbn9m$lml$1@reader11.wxs.nl...r >>>g >>>wL >>>>Thanks Rob, I will keep this channel open for a few weeks and then write >>>>a version 1.1. >>>>J >>>>As for the naming, the "Guidelines for OpenVMS Cluster Systems" manualH >>>>speaks of a "SCSI multihost OpenVMS Cluster" where Alpha systems areF >>>>concernd. So, maybe a "SCSI multihost OpenVMS VAXcluster" would beH >>>>appropriate? I like your suggestion of "Mixed Interconnect" as well. >>>> >>>>Rob Lyons wrote: >>>> >>>>K >>>>>Thanks for getting the word out to the rest of us who were not able tow >>> * >>>attend the Dutch Technical Update Days. >>>o >>>eL >>>>>I suggest the name for your configuration is LAVC with shared SCSI or a >>>n9 >>>Mixed Interconnect Cluster since the interconnect thatd >>>, >>>iL >>>>>carries the cluster traffic names the cluster type, by convention.  The >>>n? >>>term "SCSI-cluster" incorrectly implies that the SCSI bus isa >>>h >>>e" >>>>>carrying SCA cluster traffic. >>>>>-F >>>>>Be wary of using the 100Mb Ethernet setting for the NI speed.  In >>>eD >>>clusters we've built on the other side of the pond, activities as >>>F >>>BL >>>>>simple as disk to disk copy between clusters can swamp the NI and cause >>>w; >>>the cluster to become unstable.  The 10Mb setting avoidst >>>n >>>s >>>>>that pitfall. >>>>>uL >>>>>Shared SCSI has a very limited number of configurations possible due to >>>r? >>>the cable length restrictions of SCSI.  The best way we have  >>>o >>>aK >>>>>found to build scalable clusters is to use Fibre Channel.  The same PK- >>>-9 >>>port driver is used for SCSI and Fibre Channel and the  >>>a >>>uG >>>>>performance profile with an HSG80 or MSA1000 storage controller ist >>>e >>>impressive. >>>$ >>>8J >>>>>As you find additional hints and kinks, keep the information flowing. >>>m8 >>>The community at large all benefit from your efforts. >>>l >>>  >>>>>Rob Lyons >>>>>Resilient Systems Inc.r >>>>>w-w-w.resilientsys.comt >>>>>c >>>>>S >>>> >>>>--   >>>>Wilm Boerhouts >>>>  >>>>w.boerhoutOLD@PAINTplanet.nl >>>>- >>>>remove OLD PAINT from reply address first-+ >>>>*** my opinions are strictly my own ***c >>>> >>>1 >>>0 >>>U >>-- r >>Wilm Boerhoutm >>VX Company, The Netherlandsi >>! >>wboerhoutOLD@PAINTvxcompany.comR) >>  (remove OLD PAINT from reply address)d >>) >>*** my opinions are strictly my own ***  >> >  >  >    -- 5
 Wilm Boerhouts VX Company, The Netherlandsn   wboerhoutOLD@PAINTvxcompany.coml'   (remove OLD PAINT from reply address)   ' *** my opinions are strictly my own ***e   ------------------------------  + Date: Sun, 12 Oct 2003 22:30:44 +0000 (UTC)P7 From: moroney@world.std.spaamtrap.com (Michael Moroney)tA Subject: Re: White paper: CHARON-VAX cluster w/ shared SCSI disks ( Message-ID: <bmckmk$ru2$1@pcls4.std.com>  0 "Rob Lyons" <rob.lyons@resilientsys.com> writes:  o >Thanks for getting the word out to the rest of us who were not able to attend the Dutch Technical Update Days.b   >I suggest the name for your configuration is LAVC with shared SCSI or a Mixed Interconnect Cluster since the interconnect thatj >carries the cluster traffic names the cluster type, by convention.  The term "SCSI-cluster" incorrectly implies that the SCSI bus ise >carrying SCA cluster traffic.  I Well, no, to be picky the term LAVC was obsolete with VMS V5.  In the oldfB days (VMS V4.x) 'big vax' clusters (CI) were different beasts fromD microvax clusters (LAVC) running MicroVMS.  Nowadays 'a cluster is a cluster is a cluster'.  B re base note: Since Charon-VAX emulates a VAX not Alpha and sharedB SCSI was never on VAX, the VAX SCSI drivers may assume they're theC only thing accessing the bus and there may some 'strangeness' seen.  But interesting nonetheless. -- l -Mike    ------------------------------  # Date: Sun, 12 Oct 2003 21:16:24 GMTm6 From: peter@langstoeger.at (Peter 'EPLAN' LANGSTOEGER)) Subject: Re: WTH is documentation for SLSw4 Message-ID: <Iwjib.84974$KP4.1013109@news.chello.at>  h In article <PUgfb.10601$LI.5774@twister.nyroc.rr.com>, "Jeff Goodwin" <jgoodwin@maine.rrr-r.com> writes: >It's downloadable from here:) > ' >http://www.support.compaq.com/sms/sls/   F Please note also, that such old CPQ URLs seem now meaningless for HPQ.= SLS _is_ currently in fieldtest but no longer from this site.c5 (same or similar statements can be made for DFG, ...)    -- 0 Peter "EPLAN" LANGSTOEGER % Network and OpenVMS system specialist$ E-mail  peter@langstoeger.atF A-1030 VIENNA  AUSTRIA              I'm not a pessimist, I'm a realist   ------------------------------   End of INFO-VAX 2003.567 ************************