INFO-VAX Thu, 24 May 2007 Volume 2007 : Issue 283 Contents: C Formatted output. Re: C Formatted output. Re: C Formatted output. Re: CDC software (formerly known as Ross Systems) to drop Gembase VMS support Re: DECC, printf and sys$output Re: DECC, printf and sys$output DFG Config Question Re: DFG Config Question Re: Does NASA/JPL etc still use VMS? Re: Does NASA/JPL etc still use VMS? Re: Does NASA/JPL etc still use VMS? Re: Does NASA/JPL etc still use VMS? Re: Is VMS losing the Financial Sector, also? RE: Is VMS losing the Financial Sector, also? Migrating ODS-2 to ODS-5 Re: Migrating ODS-2 to ODS-5 Re: Migrating ODS-2 to ODS-5 Re: Migrating ODS-2 to ODS-5 Re: Migrating ODS-2 to ODS-5 Re: Migrating ODS-2 to ODS-5 Re: Migrating ODS-2 to ODS-5 Re: Migrating ODS-2 to ODS-5 Re: Migrating ODS-2 to ODS-5 Re: Migrating ODS-2 to ODS-5 Re: Migrating ODS-2 to ODS-5 Re: Migrating ODS-2 to ODS-5 Re: New Zealnders now laughing at global warming Re: New Zealnders now laughing at global warming Re: New Zealnders now laughing at global warming Re: New Zealnders now laughing at global warming Re: New Zealnders now laughing at global warming Re: Noahs ark found! Re: Noahs ark found! Re: Noahs ark found! Re: Noahs ark found! OpenVMS 2007 Bootcamp Re: OpenVMS 2007 Bootcamp Re: OpenVMS 2007 Bootcamp Re: OpenVMS 2007 Bootcamp Re: recognizing newly created device on HSG80 Re: recognizing newly created device on HSG80 Re: recognizing newly created device on HSG80 Re: recognizing newly created device on HSG80 Re: recognizing newly created device on HSG80 Re: recognizing newly created device on HSG80 Re: recognizing newly created device on HSG80 Re: recognizing newly created device on HSG80 Re: recognizing newly created device on HSG80 Re: recognizing newly created device on HSG80 Re: recognizing newly created device on HSG80 Re: Rename Volume Labels Re: TCPIP programming (sockaddr_in question) Re: TCPIP programming (sockaddr_in question) Re: TCPIP programming (sockaddr_in question) Re: TCPIP Services SMTP, RBLs blocking all inbound email Re: TCPIP Services SMTP, RBLs blocking all inbound email Re: TCPIP Services SMTP, RBLs blocking all inbound email Re: TCPIP Services SMTP, RBLs blocking all inbound email Re: World's Fastest Processor? For Now, Anyway RE: World's Fastest Processor? For Now, Anyway Re: World's Fastest Processor? For Now, Anyway Re: World's Fastest Processor? For Now, Anyway Re: World's Fastest Processor? For Now, Anyway Re: World's Fastest Processor? For Now, Anyway Re: World's Fastest Processor? For Now, Anyway Re: World's Fastest Processor? For Now, Anyway Re: Xpdf 3.02 for OpenVMS/Itanium Re: Xpdf 3.02 for OpenVMS/Itanium Re: Xpdf 3.02 for OpenVMS/Itanium ---------------------------------------------------------------------- Date: Wed, 23 May 2007 15:18:58 -0700 From: "Tom Linden" Subject: C Formatted output. Message-ID: I looked in the CRTL Ref Lib for OpenVMS to understand the following statement sprintf (buf, "%#*.*E", width, frac, value); Suppose width is say 23 and frac is 7, does that mean that 23 chars are written to buf and what would it look like? -- Using Opera's revolutionary e-mail client: http://www.opera.com/mail/ ------------------------------ Date: Thu, 24 May 2007 09:32:52 +1000 From: Jim Duff Subject: Re: C Formatted output. Message-ID: <4654cf24$1@dnews.tpgi.com.au> Tom Linden wrote: > > I looked in the CRTL Ref Lib for OpenVMS to understand the following > statement > > sprintf (buf, "%#*.*E", width, frac, value); > > Suppose width is say 23 and frac is 7, does that mean that 23 chars > are written to buf and what would it look like? > > Yes, format a double precision number 23 characters wide with 7 significant places after the decimal point (in scientific notation with an explicit decimal point): $ ty t.c #include #include int main (void) { char buf[23+1]; (void)sprintf (buf, "%#*.*E", 23, 7, 123.456789); (void)printf ("\"%s\"", buf); } $ cc t $ link t $ r t " 1.2345679E+02" $ HTH, Jim. -- www.eight-cubed.com ------------------------------ Date: Wed, 23 May 2007 16:41:23 -0700 From: "Tom Linden" Subject: Re: C Formatted output. Message-ID: On Wed, 23 May 2007 16:32:52 -0700, Jim Duff wrote: > Tom Linden wrote: >> >> I looked in the CRTL Ref Lib for OpenVMS to understand the following >> statement >> >> sprintf (buf, "%#*.*E", width, frac, value); >> >> Suppose width is say 23 and frac is 7, does that mean that 23 chars >> are written to buf and what would it look like? >> >> > > Yes, format a double precision number 23 characters wide with 7 > significant places after the decimal point (in scientific notation with > an explicit decimal point): > > $ ty t.c > #include > #include > > int main (void) { > > char buf[23+1]; > > (void)sprintf (buf, "%#*.*E", 23, 7, 123.456789); > (void)printf ("\"%s\"", buf); > } > $ cc t > $ link t > $ r t > " 1.2345679E+02" > $ > > HTH, > Jim. Thanks -- Using Opera's revolutionary e-mail client: http://www.opera.com/mail/ ------------------------------ Date: Thu, 24 May 2007 00:20:22 +0200 From: "Dr. Dweeb" Subject: Re: CDC software (formerly known as Ross Systems) to drop Gembase VMS support Message-ID: <4654be7f$0$7607$157c6196@dreader2.cybercity.dk> Syltrem wrote: > Hello group > > For those of you running Gembase on their Alpha / Itanium servers, > know that CDC software's roadmap shows that the next version of > Gembase (8.x) will be the last one supported on VMS. > So by 2011 at most that will be the end for us. > > I don't think they realize how much that will cost their customers. > We have many applications that are linked to Gembase -- mainly > through Oracle and RMS files but we also have complex, external > coding using Basic subroutines to do some suff not otherwise possible > with Gembase. And of course a lot of DCL around everything. > We've been with VMS for 25+ years... and I don`t see what other OS > could do some of the things we're used to... > Let's just hope hey change their mind, or that HP does something to > persuade them not to drop VMS. > > Porting this to another platform (and ensuring proper disaster > recovery capabilities) will be very time consuming and costly, but > switching to another ERP system would be much worse, as changing > platform should not affect end users where the latter would (of > course). > So that's another software vendor that will drive VMS down the drain > a bit further. > > Those of you concerned by this should complain to CDC and HP. Right > now they (CDC) apparently only talked to a few customers to conclude > they will all happily drop VMS in a couple of years from now, so why > should they keep VMS in their supported platforms portfolio? It > appears like their VMS customer base is on a slow downhill slope. Is > this going to continue until they are not in sufficient number to > justify the expense? Will all VMS customers drop it in favor of > something else ? Are we (my company) so different ? Pls speak ! > > We have very costly journey in front of us... Bite the bullet now. VMS is at best in palliative care and no matter what you and I think of it's merits (and they are legion), the owner of VMS wants it dead, and has wanted it dead for a long time. This will come to pass. Accept it and start planning an d executing. And do us all a favour by never spending another dime with HP :) Dr. Dweeb (who has embraced M$ - warts and all - because that is the only game in town) ------------------------------ Date: Wed, 23 May 2007 14:06:08 -0400 From: JF Mezei Subject: Re: DECC, printf and sys$output Message-ID: Bob Koehler wrote: > What > version of the compiler are you using? $ cc/version HP C V7.1-015 on OpenVMS Alpha V8.3 OK, I tried it again this afternoon, and this time define/user sys$output works !!!!! And this is from the same account, same process, same decterm as I was using this morning. Must be one more of those instances where I am temporarily moved to a different dimension where VMS works differently. ------------------------------ Date: Wed, 23 May 2007 14:26:07 -0400 From: JF Mezei Subject: Re: DECC, printf and sys$output Message-ID: <8e629$46548757$cef8887a$12904@TEKSAVVY.COM> Chris Sharman wrote: > The other thing which would cause that is a ctrl/Y immediately before > the def/user - I've been got by that several times - image rundown (& > deletion of user logicals) then happens immediately before chocolate. Dring ! Dring ! Bingo ! Jackpot ! Dring ! Dring ! Yep, this is is exactly as described, and in hindsight, it makes sense. If I run y image to make sure it works, ctrl-y it , define the logical and run it again to generate the full output to file, the output still goes to terminal. As a coorrolary, I tried: run chocolate define/user sys$output temp.txt continue but alas, the output still went to the terminal (and it makes sense since the output channel to the terminal is already opened). ------------------------------ Date: Wed, 23 May 2007 14:48:31 -0400 From: Chuck Aaron Subject: DFG Config Question Message-ID: I am using DFG disk defragmenter on vms 8.2, alpha server ds25. How do I configure the scheduler to start the defragmentation of a particular disk say web$disk to run at 10pm instead of midnight? Thanks, Chuck ------------------------------ Date: 23 May 2007 11:57:05 -0700 From: Peter Weaver Subject: Re: DFG Config Question Message-ID: <1179946625.564776.324730@a26g2000pre.googlegroups.com> On May 23, 2:48 pm, Chuck Aaron wrote: > I am using DFG disk defragmenter on vms 8.2, alpha server ds25. > How do I configure the scheduler to start the defragmentation of > a particular disk say web$disk to run at 10pm instead of midnight? > > Thanks, > Chuck DEFRAG MODIFY scriptname /AFTER=22:00/INTERVAL=1-: Should run the job every day at 22:00. ------------------------------ Date: Wed, 23 May 2007 14:57:40 -0400 From: JF Mezei Subject: Re: Does NASA/JPL etc still use VMS? Message-ID: <15ef0$46548ebc$cef8887a$2846@TEKSAVVY.COM> This just in: (of course, as expected, VMS is not mentioned) UPDATE: H-P Wins Massive Seven-year NASA Contract MainSMDS By Rex Crum SAN FRANCISCO (Dow Jones) -- Hewlett-Packard Co. won a seven-year technologyand services contract worth as much as $5.6 billion from the NationalAeronautics and Space Administration, the company said Wednesday. The contract calls for Palo Alto, Calif.-based H-P (HPQ) to provide a range ofcomputing products, such as desktop and workstation computers, blade PCs thatrun on the Linux and Unix operating systems, enterprise servers and printersunder what's called the NASA Solutions for Enterprise-Wide Procurement IVcontract. Federal agencies and their authorized contractors will be able to purchaseequipment from H-P under the contract. H-P said that it has delivered more than$620 million worth of products and services between 1992 and 2007 under previousNASA contracts. H-P's shares, part of the Dow Jones Industrial Average, rose 20 cents to $45.80 in morning action. (END) Dow Jones Newswires ------------------------------ Date: 23 May 2007 19:06:49 GMT From: bill@cs.uofs.edu (Bill Gunshannon) Subject: Re: Does NASA/JPL etc still use VMS? Message-ID: <5bjhm9F2sujtoU1@mid.individual.net> In article <15ef0$46548ebc$cef8887a$2846@teksavvy.com>, JF Mezei writes: > > This just in: (of course, as expected, VMS is not mentioned) Probably because it is not included!! bill -- Bill Gunshannon | de-moc-ra-cy (di mok' ra see) n. Three wolves bill@cs.scranton.edu | and a sheep voting on what's for dinner. University of Scranton | Scranton, Pennsylvania | #include ------------------------------ Date: 23 May 2007 14:40:33 -0500 From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) Subject: Re: Does NASA/JPL etc still use VMS? Message-ID: In article <15ef0$46548ebc$cef8887a$2846@TEKSAVVY.COM>, JF Mezei writes: > > This just in: (of course, as expected, VMS is not mentioned) [...] > By Rex Crum > Would Crum know or recognise the difference between UNIX and VMS, or is he one of those folks who thinks if it's not desktop it must be UNIX? ------------------------------ Date: Wed, 23 May 2007 23:39:48 -0700 From: Vance Haemmerle Subject: Re: Does NASA/JPL etc still use VMS? Message-ID: <46553334.9060104@toyvax.Glendale.CA.US> Bob Koehler wrote: > In article <5be4vcF2sv7e3U1@mid.individual.net>, bill@cs.uofs.edu (Bill Gunshannon) writes: > >>Well, we were talking about NASA and JPL. I meant the Pioneer 10 Team. :-) >> > > > I don't know about the data, but Pioneer 10 finally died several > months ago. > Actually it was several years ago. Last contact was January 2003. No one knows exactly when the power source failed. http://sse.jpl.nasa.gov/news/display.cfm?News_ID=4618 Vance ------------------------------ Date: Wed, 23 May 2007 21:22:08 -0500 From: David J Dachtera Subject: Re: Is VMS losing the Financial Sector, also? Message-ID: <4654F6D0.DE5C246C@spam.comcast.net> "Main, Kerry" wrote: > > > -----Original Message----- > > From: JF Mezei [mailto:jfmezei.spamnot@vaxination.ca] > > Sent: May 20, 2007 12:58 PM > > To: Info-VAX@Mvb.Saic.Com > > Subject: Re: Is VMS losing the Financial Sector, also? > > > > re: mainframes > > > > [snip ...] > > > > > The big question is whether Oracle will succeed in making Linux a > > serious platform, acceptable to banks for serious applications. > > mmm... with 5-20 security patches released each and every month? > > Good luck to them .. > > Yeah, I can see the banks jumping all over this as they are not > concerned to much about security are they? > > And in terms of cost savings, I know the billions of $'s in profit banks > make each year is not much, so I can see them really being under > pressure to get rid of those nice, safe back end systems. All I can say is, read the paged linked to the URL in the OP. Pressure or not, it's happening. -- David J Dachtera dba DJE Systems http://www.djesys.com/ Unofficial OpenVMS Marketing Home Page http://www.djesys.com/vms/market/ Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/ Unofficial OpenVMS-IA32 Home Page: http://www.djesys.com/vms/ia32/ Unofficial OpenVMS Hobbyist Support Page: http://www.djesys.com/vms/support/ ------------------------------ Date: Thu, 24 May 2007 00:10:24 -0400 From: "Main, Kerry" Subject: RE: Is VMS losing the Financial Sector, also? Message-ID: > -----Original Message----- > From: John Smith [mailto:a@nonymous.com] > Sent: May 23, 2007 10:01 AM > To: Info-VAX@Mvb.Saic.Com > Subject: Re: Is VMS losing the Financial Sector, also? >=20 > Main, Kerry wrote: > >> -----Original Message----- > >> From: JF Mezei [mailto:jfmezei.spamnot@vaxination.ca] > >> Sent: May 20, 2007 12:58 PM > >> To: Info-VAX@Mvb.Saic.Com > >> Subject: Re: Is VMS losing the Financial Sector, also? > >> > >> re: mainframes > >> > > [snip...] >=20 > Being *very* active in the financial services indusrty, I can assure > you > that Oracle, notwithstanding its mult-patches per month, and Linux > with its > patch du jour approach are in fact supplanting most things. >=20 > There are very few customers still using VMS, and of those, some are > fully > committed to VMS whilst most are looking to migrate to unix...usually > to > Solaris or AIX. I can't recall the last time I saw a shop have any > significant PHUX in-house. >=20 >=20 Yeah, have seen a number of cases where the OS religion over-rides sound business judgement. The techies go down a certain road and their managers do not have the will or the smarts to reign them in.... kind of like sailing down the rapids in a canoe with no paddle. Perhaps this is not politically correct, but where have all the IT managers with sound business sense disappeared to these days? I guess the banks will have to experience a major hack and major audit before someone asks the question "Excuse me - are you saying you put our mission critical XYZ application on an OS platform that you *knew* had 5-20 security patches released each and every month?" Yep, auditors will love this when someone finally does get around to asking those tough questions. Reminds me of movie called Swordfish where the hacker spends a great deal of time using a super computer to hack into a secure bank. With Linux on that platform, how long do you think that would hold up?=20 And please do not say banks do not need to worry because they have good firewalls - 50-60% of significant security events are internal issues. Gotta love the IT hype these days.=20 Regards Kerry Main Senior Consultant HP Services Canada Voice: 613-592-4660 Fax: 613-591-4477 kerryDOTmainAThpDOTcom (remove the DOT's and AT)=20 OpenVMS - the secure, multi-site OS that just works. ------------------------------ Date: Wed, 23 May 2007 12:32:44 -0700 From: "Tom Linden" Subject: Migrating ODS-2 to ODS-5 Message-ID: Currently have an ODS-2 file system DSA0: Mounted 0 COMMON 79911522 49 6 $1$DGA7: (HAFNER) ShadowSetMember 0 (member of DSA0:) $1$DGA8: (HAFNER) ShadowSetMember 0 (member of DSA0:) This is in on HSG80 and was migrated from a BA356 that had host-based shadowing as it does here. Now I have a new device which is a striped mirror set(raid0+1) $1$DGA11: (HAFNER) Online Which I want to be an ODS-5 and to replace DSA0, what steps must i go through? ------------------------------ Date: Wed, 23 May 2007 20:25:05 GMT From: VAXman- @SendSpamHere.ORG Subject: Re: Migrating ODS-2 to ODS-5 Message-ID: <00A680DF.629F3DE3@SendSpamHere.ORG> In article , "Tom Linden" writes: > > > >Currently have an ODS-2 file system >DSA0: Mounted 0 COMMON 79911522 >49 6 >$1$DGA7: (HAFNER) ShadowSetMember 0 (member of DSA0:) >$1$DGA8: (HAFNER) ShadowSetMember 0 (member of DSA0:) > >This is in on HSG80 and was migrated from a BA356 that had host-based >shadowing >as it does here. > >Now I have a new device which is a striped mirror set(raid0+1) >$1$DGA11: (HAFNER) Online > >Which I want to be an ODS-5 and to replace DSA0, what steps must i go >through? Backup DSA0 to $1$DGA11. Then $ SET VOLUME/STRUCTURE_LEVEL=5 $1$DGA11: -- VAXman- A Bored Certified VMS Kernel Mode Hacker VAXman(at)TMESIS(dot)COM "Well my son, life is like a beanstalk, isn't it?" ------------------------------ Date: Wed, 23 May 2007 14:25:54 -0700 From: "Tom Linden" Subject: Re: Migrating ODS-2 to ODS-5 Message-ID: On Wed, 23 May 2007 13:25:05 -0700, VAXman- <@SendSpamHere.ORG> wrote: > In article , "Tom Linden" = > writes: >> >> >> >> Currently have an ODS-2 file system >> DSA0: Mounted 0 COMMON 7991152= 2 >> 49 6 >> $1$DGA7: (HAFNER) ShadowSetMember 0 (member of DSA0:) >> $1$DGA8: (HAFNER) ShadowSetMember 0 (member of DSA0:) >> >> This is in on HSG80 and was migrated from a BA356 that had host-based= >> shadowing >> as it does here. >> >> Now I have a new device which is a striped mirror set(raid0+1) >> $1$DGA11: (HAFNER) Online >> >> Which I want to be an ODS-5 and to replace DSA0, what steps must i go= >> through? > > Backup DSA0 to $1$DGA11. Then > > $ SET VOLUME/STRUCTURE_LEVEL=3D5 $1$DGA11: Don't I need to INIT first? What about mounting? > -- = Using Opera's revolutionary e-mail client: http://www.opera.com/mail/ ------------------------------ Date: Thu, 24 May 2007 02:32:21 +0300 From: =?ISO-8859-15?Q?Uusim=E4ki?= Subject: Re: Migrating ODS-2 to ODS-5 Message-ID: <4654cec7$0$15296$9b536df3@news.fv.fi> There are several ways to do it, but one simple way is: $ mount/fore $1$DGA11: $ back/imag dsa0: $1$dga11: It would of course be better to do it with "mini-VMS". The resulting backup would then be fully consistent. This way there might be lots of open files, which aren't backed up consistently. When the backup has finished, you dismount the disk $1$dga11: and remount it e.g. $ mount/clus $1$dga11: common Now you can access the disk from the mounting node and the other cluster nodes too. If it is possible - technically seen - you could also mount the disk directly from the other cluster nodes, too. (provided that the other nodes have direct access to the disk also via the SAN). Otherwise the other nodes access the disk via the MSCP server. Tom Linden wrote: > On Wed, 23 May 2007 13:25:05 -0700, VAXman- <@SendSpamHere.ORG> wrote: > >> In article , "Tom Linden" >> writes: >>> >>> >>> >>> Currently have an ODS-2 file system >>> DSA0: Mounted 0 COMMON 79911522 >>> 49 6 >>> $1$DGA7: (HAFNER) ShadowSetMember 0 (member of DSA0:) >>> $1$DGA8: (HAFNER) ShadowSetMember 0 (member of DSA0:) >>> >>> This is in on HSG80 and was migrated from a BA356 that had host-based >>> shadowing >>> as it does here. >>> >>> Now I have a new device which is a striped mirror set(raid0+1) >>> $1$DGA11: (HAFNER) Online >>> >>> Which I want to be an ODS-5 and to replace DSA0, what steps must i go >>> through? >> >> Backup DSA0 to $1$DGA11. Then >> >> $ SET VOLUME/STRUCTURE_LEVEL=5 $1$DGA11: > > Don't I need to INIT first? What about mounting? >> > > > ------------------------------ Date: Wed, 23 May 2007 16:48:26 -0700 From: "Tom Linden" Subject: Re: Migrating ODS-2 to ODS-5 Message-ID: On Wed, 23 May 2007 16:32:21 -0700, Uusimäki wrote: > > There are several ways to do it, but one simple way is: > > $ mount/fore $1$DGA11: > $ back/imag dsa0: $1$dga11: > > It would of course be better to do it with "mini-VMS". The resulting > backup would then be fully consistent. This way there might be lots of > open files, which aren't backed up consistently. Well that is hard to do in a cluster unless it is dismount from all but one node > > When the backup has finished, you dismount the disk $1$dga11: and > remount it e.g. > > $ mount/clus $1$dga11: common well, yes, but I also obviously need dismount the originsl disk from the cluster, and modify the startup mount commands. > > Now you can access the disk from the mounting node and the other cluster > nodes too. If it is possible - technically seen - you could also mount > the disk directly from the other cluster nodes, too. (provided that the > other nodes have direct access to the disk also via the SAN). > Otherwise the other nodes access the disk via the MSCP server. > > > > > > > > > Tom Linden wrote: >> On Wed, 23 May 2007 13:25:05 -0700, VAXman- <@SendSpamHere.ORG> wrote: >> >>> In article , "Tom Linden" >>> writes: >>>> >>>> >>>> >>>> Currently have an ODS-2 file system >>>> DSA0: Mounted 0 COMMON 79911522 >>>> 49 6 >>>> $1$DGA7: (HAFNER) ShadowSetMember 0 (member of DSA0:) >>>> $1$DGA8: (HAFNER) ShadowSetMember 0 (member of DSA0:) >>>> >>>> This is in on HSG80 and was migrated from a BA356 that had host-based >>>> shadowing >>>> as it does here. >>>> >>>> Now I have a new device which is a striped mirror set(raid0+1) >>>> $1$DGA11: (HAFNER) Online >>>> >>>> Which I want to be an ODS-5 and to replace DSA0, what steps must i go >>>> through? >>> >>> Backup DSA0 to $1$DGA11. Then >>> >>> $ SET VOLUME/STRUCTURE_LEVEL=5 $1$DGA11: >> Don't I need to INIT first? What about mounting? >>> >> -- Using Opera's revolutionary e-mail client: http://www.opera.com/mail/ ------------------------------ Date: Wed, 23 May 2007 20:17:47 -0400 From: JF Mezei Subject: Re: Migrating ODS-2 to ODS-5 Message-ID: <85a34$4654d9c3$cef8887a$14810@TEKSAVVY.COM> Tom Linden wrote: >> $ SET VOLUME/STRUCTURE_LEVEL=5 $1$DGA11: > > Don't I need to INIT first? What about mounting? The whole point of SET VOLUME/STRUCTURE is to allow an in-situ conversion of an existing drive. And SET VOLUME wants to play on a mounted drive. You probably want to quiesce the drive first, deinstall images etc. (not sure if it is required though). ------------------------------ Date: Wed, 23 May 2007 17:20:27 -0700 From: "Tom Linden" Subject: Re: Migrating ODS-2 to ODS-5 Message-ID: On Wed, 23 May 2007 16:32:21 -0700, Uusimäki wrote: > > There are several ways to do it, but one simple way is: > > $ mount/fore $1$DGA11: > $ back/imag dsa0: $1$dga11: > > It would of course be better to do it with "mini-VMS". The resulting > backup would then be fully consistent. This way there might be lots of > open files, which aren't backed up consistently. > > When the backup has finished, you dismount the disk $1$dga11: and > remount it e.g. > > $ mount/clus $1$dga11: common > > Now you can access the disk from the mounting node and the other cluster > nodes too. If it is possible - technically seen - you could also mount > the disk directly from the other cluster nodes, too. (provided that the > other nodes have direct access to the disk also via the SAN). > Otherwise the other nodes access the disk via the MSCP server. > > Now, I have some processes that run off of this disk like, MX, HGFTP, WASD and a few others, When I dismount the old and mount the new, I presume I need to restart these processes? > > > > > > > Tom Linden wrote: >> On Wed, 23 May 2007 13:25:05 -0700, VAXman- <@SendSpamHere.ORG> wrote: >> >>> In article , "Tom Linden" >>> writes: >>>> >>>> >>>> >>>> Currently have an ODS-2 file system >>>> DSA0: Mounted 0 COMMON 79911522 >>>> 49 6 >>>> $1$DGA7: (HAFNER) ShadowSetMember 0 (member of DSA0:) >>>> $1$DGA8: (HAFNER) ShadowSetMember 0 (member of DSA0:) >>>> >>>> This is in on HSG80 and was migrated from a BA356 that had host-based >>>> shadowing >>>> as it does here. >>>> >>>> Now I have a new device which is a striped mirror set(raid0+1) >>>> $1$DGA11: (HAFNER) Online >>>> >>>> Which I want to be an ODS-5 and to replace DSA0, what steps must i go >>>> through? >>> >>> Backup DSA0 to $1$DGA11. Then >>> >>> $ SET VOLUME/STRUCTURE_LEVEL=5 $1$DGA11: >> Don't I need to INIT first? What about mounting? >>> >> -- Using Opera's revolutionary e-mail client: http://www.opera.com/mail/ ------------------------------ Date: Thu, 24 May 2007 01:41:30 GMT From: VAXman- @SendSpamHere.ORG Subject: Re: Migrating ODS-2 to ODS-5 Message-ID: <00A6810B.9651DFA6@SendSpamHere.ORG> In article , "Tom Linden" writes: {...snip...} >> Backup DSA0 to $1$DGA11. Then >> >> $ SET VOLUME/STRUCTURE_LEVEL=3D5 $1$DGA11: > >Don't I need to INIT first? What about mounting? I thought your initial question was about replacing one system device (DSA0, a shadow set) with a RAIDed device ($1$DGA11). One way or another, I would assume you'd have to reboot the system to boot up off of the new $1$DGA11 system device. 1. Boot your system off of the CD. 2. Backup the present system drive to the new device. (No need to init it unless you have some desires to change things like allocation size, etc. Consult the OpenVMS guide to backup if you need help there.) 3. Once you have the system on $1$DGA11 mount it and issue the $ SET VOLUME/STRUCTURE_LEVEL=5 $1$DGA11 command. 4. Boot up off of the new device. Obviously, I'm not intimately knowledgeable of your system's configuration, so I can't offer up anything more detailed. Generally, going from an ODS-2 to an ODS-5 volume structure is really as simple as: $ SET VOLUME/STRUCTURE_LEVEL=5 volume. -- VAXman- A Bored Certified VMS Kernel Mode Hacker VAXman(at)TMESIS(dot)COM "Well my son, life is like a beanstalk, isn't it?" ------------------------------ Date: Wed, 23 May 2007 18:54:23 -0700 From: "Tom Linden" Subject: Re: Migrating ODS-2 to ODS-5 Message-ID: On Wed, 23 May 2007 18:41:30 -0700, VAXman- <@SendSpamHere.ORG> wrote: > In article , "Tom Linden" = > writes: > {...snip...} >>> Backup DSA0 to $1$DGA11. Then >>> >>> $ SET VOLUME/STRUCTURE_LEVEL=3D3D5 $1$DGA11: >> >> Don't I need to INIT first? What about mounting? > > I thought your initial question was about replacing one system device = = > (DSA0, > a shadow set) with a RAIDed device ($1$DGA11). > > One way or another, I would assume you'd have to reboot the system to = = > boot > up off of the new $1$DGA11 system device. This not a system disk > > 1. Boot your system off of the CD. > > 2. Backup the present system drive to the new device. > (No need to init it unless you have some desires to change > things like allocation size, etc. Consult the OpenVMS guide > to backup if you need help there.) > > 3. Once you have the system on $1$DGA11 mount it and issue the > $ SET VOLUME/STRUCTURE_LEVEL=3D5 $1$DGA11 command. > > 4. Boot up off of the new device. > > Obviously, I'm not intimately knowledgeable of your system's = > configuration, > so I can't offer up anything more detailed. Generally, going from an = = > ODS-2 > to an ODS-5 volume structure is really as simple as: > > $ SET VOLUME/STRUCTURE_LEVEL=3D5 volume. > -- = Using Opera's revolutionary e-mail client: http://www.opera.com/mail/ ------------------------------ Date: Thu, 24 May 2007 02:18:13 GMT From: "John E. Malmberg" Subject: Re: Migrating ODS-2 to ODS-5 Message-ID: JF Mezei wrote: > Tom Linden wrote: > >>> $ SET VOLUME/STRUCTURE_LEVEL=5 $1$DGA11: >> >> >> Don't I need to INIT first? What about mounting? > > > The whole point of SET VOLUME/STRUCTURE is to allow an in-situ > conversion of an existing drive. And SET VOLUME wants to play on a > mounted drive. > > You probably want to quiesce the drive first, deinstall images etc. (not > sure if it is required though). A volume to be converted must be mounted private to the process that is doing the conversion. For a system disk, you need to be booted off of a different disk, like the installation CDROM/DVD. -John wb8tyw@qsl.network Personal Opinion Only ------------------------------ Date: Wed, 23 May 2007 21:34:12 -0500 From: David J Dachtera Subject: Re: Migrating ODS-2 to ODS-5 Message-ID: <4654F9A4.59D2C089@spam.comcast.net> Tom Linden wrote: > > Currently have an ODS-2 file system > DSA0: Mounted 0 COMMON 79911522 > 49 6 > $1$DGA7: (HAFNER) ShadowSetMember 0 (member of DSA0:) > $1$DGA8: (HAFNER) ShadowSetMember 0 (member of DSA0:) > > This is in on HSG80 and was migrated from a BA356 that had host-based > shadowing > as it does here. > > Now I have a new device which is a striped mirror set(raid0+1) > $1$DGA11: (HAFNER) Online > > Which I want to be an ODS-5 and to replace DSA0, what steps must i go > through? Why not just add DGA11 to DSA0:, wait for the shadow-copy to finish, then drop DGA7 and DGA8, MOUNT DSA0 privately, SET VOL/STRU=5 DSA0:, then MOUNT DSA0 (single member: DGA11) back to the /SYSTEM and copy in a nother member? Doesn't provide the defrag. of the BACKUP technique, but could work. -- David J Dachtera dba DJE Systems http://www.djesys.com/ Unofficial OpenVMS Marketing Home Page http://www.djesys.com/vms/market/ Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/ Unofficial OpenVMS-IA32 Home Page: http://www.djesys.com/vms/ia32/ Unofficial OpenVMS Hobbyist Support Page: http://www.djesys.com/vms/support/ ------------------------------ Date: Thu, 24 May 2007 05:00:15 +0200 From: "P. Sture" Subject: Re: Migrating ODS-2 to ODS-5 Message-ID: In article , "Tom Linden" wrote: > On Wed, 23 May 2007 16:32:21 -0700, Uusimäki > wrote: > > > > > There are several ways to do it, but one simple way is: > > > > $ mount/fore $1$DGA11: > > $ back/imag dsa0: $1$dga11: > > > > It would of course be better to do it with "mini-VMS". The resulting > > backup would then be fully consistent. This way there might be lots of > > open files, which aren't backed up consistently. > > > > When the backup has finished, you dismount the disk $1$dga11: and > > remount it e.g. > > > > $ mount/clus $1$dga11: common > > > > Now you can access the disk from the mounting node and the other cluster > > nodes too. If it is possible - technically seen - you could also mount > > the disk directly from the other cluster nodes, too. (provided that the > > other nodes have direct access to the disk also via the SAN). > > Otherwise the other nodes access the disk via the MSCP server. > > > > > Now, I have some processes that run off of this disk like, MX, HGFTP, WASD > and a few others, When I dismount the old and mount the new, I presume > I need to restart these processes? > Yes. And obviously you need to stop those processes before dismounting the original disk. -- Paul Sture ------------------------------ Date: Wed, 23 May 2007 18:42:57 +0000 (UTC) From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) Subject: Re: New Zealnders now laughing at global warming Message-ID: In article <1179890412.224639.196720@u30g2000hsc.googlegroups.com>, Neil Rieck writes: > > Greenland's ice cap alone is capable of significant ocean level rise on > > the planet. Remember that it is only water added to the top of oceans, > > it has no relation to the water volume below current sea level (current > > volume of water in oceans). > > While I believe in man-made global warming, I disagree with this > statement. I think it was Carl Sagan who said that if the earth was > the size of a basket ball then our atmosphere would be no thicker than > a layer of paint on that ball. Taken from this perspective we notice > that Greenland's ice doesn't rise too far from the ocean's surface. > This means the volume is low and, for the purpose of this discussion, > we can get away with comparing the surface are of Greenland to the > surface area of all the oceans; the resulting fraction is quite small. This is just wrong. You have to multiply the fraction by the thickness of the ice to get the sea-level rise. Suppose the fraction of surface areas is 1/100. If the ice is 1 km thick, then you have a rise in sea level of 10m! These figures are just examples; plug in the numbers. As JF mentioned, the VOLUME of the oceans is essentially irrelevant; we're concerned with how high the ADDITIONAL water will raise the sea level. I haven't worked out the fraction; thickness is of the order km. Of course, much less than 10m would still be catastrophic for people living in coastal regions. > Don't take my conclusion the wrong way. Global warming is real and > mankind must act now. BTW, the Antartic ice does rise quite a bit > higher than Greenland although still not as high as any mountain made > from rock. True, but the weight of the ice also pushes the land beneath downward, so there is more ice than one might assume. Also, warming will increase the volume of the oceans due to thermal expansion. ------------------------------ Date: Wed, 23 May 2007 23:11:28 +0200 From: Dirk Munk Subject: Re: New Zealnders now laughing at global warming Message-ID: Bill Gunshannon wrote: > In article , > JF Mezei writes: >> Consider a city like New Orleans. Raise ocean level by half a metre and >> the next hurricane to hit will definitely push a surge of water that >> will go over the levies every year. >> > > And yet, after loosing everything and really having nothing to go back > there for people are flocking back and being strongly encouraged to do > so by the local government knowing full well that they are very likely > to get wiped out yet again. Go figure!! People here do realize that > the US is so big there is plenty of land available to relocate all those > people permanently at a tremendous saving over the cost of continuously > rebuilding New Orleans. > > And someone in an earlier post mentioned how much time JFK would spend > underwater if the oceans rose even a little bit. Can you say Schipol? Actually it is Schiphol, and it means Ship's Hell. The old Schiphol airport (now Schiphol East) was build on the bottom of a former big lake, the Haarlemmermeer. Storms from the West (the most common direction) drove ships to that part of the lake, and many ships sank there. Hence the name Ship's Hell for that part of the lake. In the mid 19th a big storm almost caused the lake to flood Amsterdam. So it was decided to pump the lake dry, and since it was to big to pump it dry with just windmills, they used steam driven pumps for the first time. One of the pumps is called Qruquius, it has the biggest steamengine ever build (in pure size that is), and it is now a museum. > I knew you could!! > > bill > ------------------------------ Date: Thu, 24 May 2007 00:07:37 +0200 From: "Dr. Dweeb" Subject: Re: New Zealnders now laughing at global warming Message-ID: <4654bb82$0$7610$157c6196@dreader2.cybercity.dk> JF Mezei wrote: > Neil Rieck wrote: >> that Greenland's ice doesn't rise too far from the ocean's surface. > > It does, unless you consider 6km to be thin. This represents a HUGE > volume of stored water. And it also represents a huge area of land > which, as it starts to get clear of snow, will capture much more > sunlight and accelerate the warming of the island and accelerate the > melting of the ice. > Yes. But .. there is strong evidence to suggest that the polar caps will not melt - ever. Just oscillate in size as they have done so for millenia. As a matter of passing interest, the inland ice weighs so much that Greenland is severely depressed and will "pop-up" should all the ice suddenly and mysteriously melt (which is extremely unlikely btw). I do not recall the exact figues but they are to be found. Dweeb > Remember that it won't take much of a rise in ocean levels to cause a > lot of problems. > > Consider a city like New Orleans. Raise ocean level by half a metre > and the next hurricane to hit will definitely push a surge of water > that will go over the levies every year. > > Consider the main new york airports. They are not exactly high about > sea level, and a big wind storm with slightly higher water levels > will just push water onto the runways. Osaka's airport will become > under sea level. Much of the netherland may not be able to cope with > higher ocean levels. And Venice, already too low, will suffer greatly. > > Areas such as Maldives and florida keys will get washed away with > every typhoon/hurricane/cyclone. > > Consider the economic impacts on insurance companies when the weather > causes more and more destruction year after year. What happens when a > big storm hits new york and water enters subway tunnels ? You may not > be flooded where you are in Ontario, but in the end, the insurance > rates you pay for your home will rise because the big insurance > wholesales like Lloyds of London will charge a lot more money for > insurance on insurance companies. > > > Nobody will cry when homes of the rich and famous along the california > coast are destroyed by some storm with waves crashing into their > living rooms. But insurance companies won't like this happening > every year. > > > > > > > >> This means the volume is low and, for the purpose of this discussion, >> we can get away with comparing the surface are of Greenland to the >> surface area of all the oceans; the resulting fraction is quite >> small. Don't take my conclusion the wrong way. Global warming is real and >> mankind must act now. BTW, the Antartic ice does rise quite a bit >> higher than Greenland although still not as high as any mountain made >> from rock. >> >> p.s. Today the BBC put a reporter on that 10 mile long 3 mile wide >> piece of ice that just broke away from Greenland. This thing is huge >> and the scientists being interviewed stated that we will all be >> screwed if this thing drifts into the oil rigs just off the coast of >> New Foundland. >> >> Neil Rieck >> Kitchener/Waterloo/Cambridge, >> Ontario, Canada. >> http://www3.sympatico.ca/n.rieck/ ------------------------------ Date: Thu, 24 May 2007 00:10:34 +0200 From: "Dr. Dweeb" Subject: Re: New Zealnders now laughing at global warming Message-ID: <4654bc33$0$7608$157c6196@dreader2.cybercity.dk> JF Mezei wrote: > Neil Rieck wrote: >> p.s. Today the BBC put a reporter on that 10 mile long 3 mile wide >> piece of ice that just broke away from Greenland. This thing is huge >> and the scientists being interviewed stated that we will all be >> screwed if this thing drifts into the oil rigs just off the coast of >> New Foundland. > Ice, like water "flows", albeit very slowly, bits break off all the time - no news there. I admit it looks impressive when a huge chunk breaks off, but it is not unexpected and not unnatural in decreasing as well as increasing environments. Dweeb. > It actually broke away a year or two ago. But it has been noticed last > year and this is the first attempt to land on it. > > And no, ice that far north doesn't flow down to newfoundland, it > rotates around the pole and will end up in russia and then northern > alaska. (or perhaps the other way around). > > There isn't much off-shore oil drilling (yet) in the arctic due to the > ice. When I was there in 1989, there were test rigs next the Hershell > Island and some north of Tuktoyaktuk. Not sure how they dealt with > winter. ------------------------------ Date: Wed, 23 May 2007 20:19:58 -0400 From: JF Mezei Subject: Re: New Zealnders now laughing at global warming Message-ID: <2e6c3$4654da47$cef8887a$14810@TEKSAVVY.COM> Dr. Dweeb wrote: > Yes. But .. Did you type this in the same tone as "Little Britain" ? :-) :-) :-) > there is strong evidence to suggest that the polar caps will > not melt - ever. Just oscillate in size as they have done so for millenia. Current trends measure massive changes in a 100 year scale, whereas the normal cycles are measured in 10s of thousands of years. ------------------------------ Date: Wed, 23 May 2007 13:52:28 -0400 From: JF Mezei Subject: Re: Noahs ark found! Message-ID: Bill Gunshannon wrote: > "Better to remain silent and be thought a fool than to speak out > and remove all doubt." Abraham Lincoln The other side of the coin: Better to be yourself and express your opinions than to pretend to be someone else by staying quiet. If the fear of being different prevents people from speaking out, it essentially breaks how democracy works and a majority of people may decide to remain silent and just go with "the flow". This is exactly how the USA population re-elected an idiot lying son of a bitch war criminal as their president. Those few who spoke out against him were ridiculed and made to shut up. Phil Donahue lost his job at NBC because he was against the war and NBC made a decision to go with the flow and not criticise Bush. (who decided what the "flow" was ?) Susan Sarandon was prevented from attending her niece's school play because that school decided that because she had questioned the war, she was to be considered a terrorist sympathizer and could not enter the school grounds. And then there were the Dixie chicks whose music was banned from many radio stations because they were against the war. The list is very long. ------------------------------ Date: Wed, 23 May 2007 23:31:32 +0200 From: Dirk Munk Subject: Re: Noahs ark found! Message-ID: Andrew wrote: > On 30 Apr, 15:25, Dirk Munk wrote: >> gen...@marblecliff.com wrote: >>> the same researcher who helped bring about the discovery >>> of Mount Sinai now believes he has found Noahs ark on a >>> mountain in Iran ... scroll to the bottom and look at the >>> pictures and video ... >>> http://www.arkfever.com/ >> This time Boob is right. What you see on the pictures is the stern of >> the ark. On it they found the inscription "Noahs Ark - Monrovia" in >> Hebrew. That proofs Boob is not mistaken this time. > > Thats what you get if you register your ark under a flag of > convenience, wildly off course and stuck up a mountain with no > navigable water in site. > > Regards > Andrew Harrison > Nice that someone got the Monrovia joke :) But now to get On Topic again on this Off Topic subject, two nights ago I saw a National Geographic TV program on the subject of the big flood. It showed Boob's ark, as well as three other arks in that area. Boob's ark seems to be made out of stone. Somehow I think stone doesn't float that well, but who knows. At the end they gave two possible explanations for the Big Flood stories in the Middle East. The first one was a meteorite plunging into the Indian Ocean near Madagaskar. That would have caused 180 meter (600 feet) tsunamis hitting the coast of the Indian Ocean. That would have happened some 5000 years ago if I remember correctly. The second explanation was the flooding of the Black Sea. They found beaches of a sweet water lake 150 meters below the present surface of the sea. It seems about 7600 years ago the landbridge at the Bosperus was flooded and gave way. Within 60 to 90 days the Black Sea was flooded, and the sound and tremors of the waterfall must have be enormous. ------------------------------ Date: 23 May 2007 16:19:48 -0700 From: AEF Subject: Re: Noahs ark found! Message-ID: <1179962387.949225.81960@h2g2000hsg.googlegroups.com> On May 22, 8:50 pm, s...@antinode.org (Steven M. Schweda) wrote: > From: AEF > > > But newsgroups for topics of great general interest are often filled > > with severely obnoxious posters. > > Sadly, this is hardly a distinguishing characteristic. One might ask Oh, please. Check out some other groups. Try sci.physics. Here's a good recent post: [beg quote] Physicists don't have a clue, which should not surprise anybody. For all their insufferable pomposity, physicists don't really know shit. ahahaha... Anyway, if you know the right answer to the question, you deserve at least a hall pass. You should get a Nobel prize but the Nobel is only reserved for clueless physicists. ahahaha... I happen to know the answer. So I am the judge here. ahahaha... Let's see who the real smart ones are. Louis Savain (ad removed) [end quote] Another thread has a title that denise the Holocaust. I only had to look at the first page of sci.physics to find these "gems". Take a look for yourself. And try some of the politics groups! One I stumbled upon once was even worse!!! Please, COV is paradise compared to some other groups. > why you waste space here to complain about junk elsewhere, instead of > striving to reduce junk here (which can often be done simply by > remaining silent). But I won't. Why are you wasting space telling me about questions you're not asking? :-) I miss the "Which news group is this? Am I lost?" jokes. > > ------------------------------------------------------------------------ > > Steven M. Schweda sms@antinode-org > 382 South Warwick Street (+1) 651-699-9818 > Saint Paul MN 55105-2547 AEF ------------------------------ Date: 23 May 2007 16:20:49 -0700 From: AEF Subject: Re: Noahs ark found! Message-ID: <1179962449.533140.307310@p47g2000hsd.googlegroups.com> On May 22, 9:02 pm, "Richard B. Gilbert" wrote: > AEF wrote: > > On May 21, 11:56 am, "FredK" wrote: > > >>"Andrew" wrote in message > > >>news:1179760297.803505.65120@36g2000prm.googlegroups.com... > > >>>On 30 Apr, 15:25, Dirk Munk wrote: > > >>>>gen...@marblecliff.com wrote: > > >>>>>the same researcher who helped bring about the discovery > >>>>>of Mount Sinai now believes he has found Noahs ark on a > > > > > > > But newsgroups for topics of great general interest are often filled > > with severely obnoxious posters. > > > AEF > > The same could be said of a newsgroup devoted an unmarketed O/S. You haven't looked very hard. See my reply to Schweda's post. AEF ------------------------------ Date: 23 May 2007 11:46:24 -0700 From: DaveG Subject: OpenVMS 2007 Bootcamp Message-ID: <1179945984.859314.205100@z28g2000prd.googlegroups.com> For those in attendance: Care to share what is going on in ZKO land this week. Dave... ------------------------------ Date: Wed, 23 May 2007 14:49:21 -0400 From: JF Mezei Subject: Re: OpenVMS 2007 Bootcamp Message-ID: <38257$46548cc9$cef8887a$1729@TEKSAVVY.COM> DaveG wrote: > Care to share what is going on in ZKO land this week. The response you will get is: What goes in at bootcamp stays at bootcamp. So even if they confirm that VMS is being ported to the 8086, the attendees are under a strict non-disclosure and cannot convey the good news and will have to make statements such as "I am not aware of any plans to port VMS to that architecture". ------------------------------ Date: Wed, 23 May 2007 21:30:10 -0500 From: David J Dachtera Subject: Re: OpenVMS 2007 Bootcamp Message-ID: <4654F8B2.B4747E5B@spam.comcast.net> JF Mezei wrote: > > DaveG wrote: > > Care to share what is going on in ZKO land this week. > > The response you will get is: > > What goes in at bootcamp stays at bootcamp. > > So even if they confirm that VMS is being ported to the 8086, the > attendees are under a strict non-disclosure and cannot convey the good > news and will have to make statements such as "I am not aware of any > plans to port VMS to that architecture". I would hope the port is to 80x86 and/or x86-64 rather than a 16-bit processor like 8086. -- David J Dachtera dba DJE Systems http://www.djesys.com/ Unofficial OpenVMS Marketing Home Page http://www.djesys.com/vms/market/ Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/ Unofficial OpenVMS-IA32 Home Page: http://www.djesys.com/vms/ia32/ Unofficial OpenVMS Hobbyist Support Page: http://www.djesys.com/vms/support/ ------------------------------ Date: 23 May 2007 20:17:59 -0700 From: sean@obanion.us Subject: Re: OpenVMS 2007 Bootcamp Message-ID: <1179976679.501723.319800@u30g2000hsc.googlegroups.com> Clearly, none of you are drinking enough. An "Out of order" sign on a urinal at the hotel has been clarified: "This must be a Microsoft product, because even if you piss on it, it doen't work." Pictures to be posted - I could not possibly make that up, even with more Tequila... Sean On May 23, 7:30 pm, David J Dachtera wrote: > JF Mezei wrote: > > > DaveG wrote: > > > Care to share what is going on in ZKO land this week. > > > The response you will get is: > > > What goes in at bootcamp stays at bootcamp. > > > So even if they confirm that VMS is being ported to the 8086, the > > attendees are under a strict non-disclosure and cannot convey the good > > news and will have to make statements such as "I am not aware of any > > plans to port VMS to that architecture". > > I would hope the port is to 80x86 and/or x86-64 rather than a 16-bit processor > like 8086. > > -- > David J Dachtera > dba DJE Systemshttp://www.djesys.com/ > > Unofficial OpenVMS Marketing Home Pagehttp://www.djesys.com/vms/market/ > > Unofficial Affordable OpenVMS Home Page:http://www.djesys.com/vms/soho/ > > Unofficial OpenVMS-IA32 Home Page:http://www.djesys.com/vms/ia32/ > > Unofficial OpenVMS Hobbyist Support Page:http://www.djesys.com/vms/support/ ------------------------------ Date: 23 May 2007 10:55:06 -0700 From: BaxterD@tessco.com Subject: Re: recognizing newly created device on HSG80 Message-ID: <1179942905.975140.17320@u30g2000hsc.googlegroups.com> On May 23, 1:20 pm, "Tom Linden" wrote: > On Wed, 23 May 2007 10:31:40 -0700, Uusim=E4ki =20 > > wrote: > > > Yes, exactly, just for the nodes with FC-HBA's. > > When I try that > > HSG80-BOT>set d11 enable=3Dhafner > Error 1240: Connection hafner not found > > > > > > > > > > > You have to do it only on one controller (you are running them in =20 > > failover mode of course). > Yes. > > > Tom Linden wrote: > >> On Wed, 23 May 2007 10:03:00 -0700, Tom Linden =20 > >> wrote: > > >>> On Wed, 23 May 2007 09:49:55 -0700, Uusim=E4ki =20 > >>> wrote: > > >>>> Just some checks: > > >>>> Did you do > >>>> HSG80> SET D11 ID=3D11 ! This sets the identifier needed for VMS > >>>> HSG80> SET D11 ENA=3Dnode1,node2,... ! This allows the access to th= e =20 > >>>> disk for the cluster nodes > > >>> I presume you only enter those nodes which have an HBA? > >> Oh, and for each controller? > > >>>> Then you should find them by doing: > > >>>> $ mc sysman io auto > >>>> $ show devi dg > > >>>> norm.raph...@metso.com wrote: > >>>>> SYSMAN> help io > >>>>> IO > >>>>> There are several IO subcommands. Select the one you want from > >>>>> the choices below. > >>>>> Additional information available: > >>>>> AUTOCONFIGURE CONNECT FIND_WWID LOAD REBUILD > >>>>> REPLACE_WWID SCSI_PATH_VERIFY SET SHOW > >>>>> IO Subtopic? > >>>>> Topic? io find > >>>>> IO > >>>>> FIND_WWID > >>>>> The IO FIND_WWID command probes all Fibre Channel ports, =20 > >>>>> detects > >>>>> all previously undiscovered tapes and medium changers behind= a > >>>>> Modular Data Router (MDR), and assigns a worldwide identifier > >>>>> (WWID) to each one. The command also displays a list of the > >>>>> devices and their assigned device names and automatically =20 > >>>>> records > >>>>> this information in the SYS$SYSTEM:SYS$DEVICES.DAT file. =20 > >>>>> Finally, > >>>>> the command updates relevant local and clusterwide memory > >>>>> structures. > >>>>> To configure newly attached Fibre Channel tapes, use this =20 > >>>>> command > >>>>> prior to running the SYSMAN command IO AUTOCONFIGURE. > >>>>> You must have CMKRNL privilege to use the IO FIND_WWID =20 > >>>>> command. > >>>>> For more information about Fibre Channel, see the Guideline= s =20 > >>>>> for > >>>>> OpenVMS Cluster Configurations. > >>>>> Format > >>>>> IO FIND_WWID > >>>>> Additional information available: > >>>>> Parameters Qualifiers Description Example > >>>>> "Tom Linden" wrote on 05/23/2007 11:37:14= =20 > >>>>> AM: > > >>>>>> On Wed, 23 May 2007 08:39:07 -0700, Peter 'EPLAN' LANGSTOeGER > >>>>>> wrote: > > >>>>>>> In article , "Tom Linden" > >>>>>>> writes: > >>>>>>>> Using the wizard on a W2K box talking to SWCC I created a new =20 > >>>>>>>> device > >>>>>>>> Striped + raid, now what do I need to do to get the cluster node= s =20 > >>>>>>>> to > >>>>>>>> recognize it? I tried sysgen> io autoconfigure but that did not > >>>>>>>> work > >>>>>>> Normally the CONFIGURE process does this for, that means they do =20 > >>>>>>> appear > >>>>>>> automatically if properly configured. (Did you do a SET Dxx =20 > >>>>>>> ID=3Dxxyy as > > >>>>>>> well?) > > >>>>>> Do you mean > >>>>>> sysman> config set ? > > >>>>>> The device I created, should have shown up as > >>>>>> $1$DGA11 > > >>>>>> Haven't done this before. > > >>>>>> -- Using Opera's revolutionary e-mail client: =20 > >>>>>>http://www.opera.com/mail/ > > -- > Using Opera's revolutionary e-mail client:http://www.opera.com/mail/- Hid= e quoted text - > > - Show quoted text - On the HSG80 controller, type SHOW CONNECTIONS This will show what the controller is considering acceptable "connections". It has been a while, however I don't think that the HSG80 controllers know about 'nodes', they only know about connections (the names of which are user defined and can obviously include the Node name.) The connection is really the "path", so there will (normally) be multiple connections for each node, depending on number of hba's and fabrics. Dave. ------------------------------ Date: 23 May 2007 10:55:28 -0700 From: BaxterD@tessco.com Subject: Re: recognizing newly created device on HSG80 Message-ID: <1179942928.834884.264300@w5g2000hsg.googlegroups.com> On May 23, 1:20 pm, "Tom Linden" wrote: > On Wed, 23 May 2007 10:31:40 -0700, Uusim=E4ki =20 > > wrote: > > > Yes, exactly, just for the nodes with FC-HBA's. > > When I try that > > HSG80-BOT>set d11 enable=3Dhafner > Error 1240: Connection hafner not found > > > > > > > > > > > You have to do it only on one controller (you are running them in =20 > > failover mode of course). > Yes. > > > Tom Linden wrote: > >> On Wed, 23 May 2007 10:03:00 -0700, Tom Linden =20 > >> wrote: > > >>> On Wed, 23 May 2007 09:49:55 -0700, Uusim=E4ki =20 > >>> wrote: > > >>>> Just some checks: > > >>>> Did you do > >>>> HSG80> SET D11 ID=3D11 ! This sets the identifier needed for VMS > >>>> HSG80> SET D11 ENA=3Dnode1,node2,... ! This allows the access to th= e =20 > >>>> disk for the cluster nodes > > >>> I presume you only enter those nodes which have an HBA? > >> Oh, and for each controller? > > >>>> Then you should find them by doing: > > >>>> $ mc sysman io auto > >>>> $ show devi dg > > >>>> norm.raph...@metso.com wrote: > >>>>> SYSMAN> help io > >>>>> IO > >>>>> There are several IO subcommands. Select the one you want from > >>>>> the choices below. > >>>>> Additional information available: > >>>>> AUTOCONFIGURE CONNECT FIND_WWID LOAD REBUILD > >>>>> REPLACE_WWID SCSI_PATH_VERIFY SET SHOW > >>>>> IO Subtopic? > >>>>> Topic? io find > >>>>> IO > >>>>> FIND_WWID > >>>>> The IO FIND_WWID command probes all Fibre Channel ports, =20 > >>>>> detects > >>>>> all previously undiscovered tapes and medium changers behind= a > >>>>> Modular Data Router (MDR), and assigns a worldwide identifier > >>>>> (WWID) to each one. The command also displays a list of the > >>>>> devices and their assigned device names and automatically =20 > >>>>> records > >>>>> this information in the SYS$SYSTEM:SYS$DEVICES.DAT file. =20 > >>>>> Finally, > >>>>> the command updates relevant local and clusterwide memory > >>>>> structures. > >>>>> To configure newly attached Fibre Channel tapes, use this =20 > >>>>> command > >>>>> prior to running the SYSMAN command IO AUTOCONFIGURE. > >>>>> You must have CMKRNL privilege to use the IO FIND_WWID =20 > >>>>> command. > >>>>> For more information about Fibre Channel, see the Guideline= s =20 > >>>>> for > >>>>> OpenVMS Cluster Configurations. > >>>>> Format > >>>>> IO FIND_WWID > >>>>> Additional information available: > >>>>> Parameters Qualifiers Description Example > >>>>> "Tom Linden" wrote on 05/23/2007 11:37:14= =20 > >>>>> AM: > > >>>>>> On Wed, 23 May 2007 08:39:07 -0700, Peter 'EPLAN' LANGSTOeGER > >>>>>> wrote: > > >>>>>>> In article , "Tom Linden" > >>>>>>> writes: > >>>>>>>> Using the wizard on a W2K box talking to SWCC I created a new =20 > >>>>>>>> device > >>>>>>>> Striped + raid, now what do I need to do to get the cluster node= s =20 > >>>>>>>> to > >>>>>>>> recognize it? I tried sysgen> io autoconfigure but that did not > >>>>>>>> work > >>>>>>> Normally the CONFIGURE process does this for, that means they do =20 > >>>>>>> appear > >>>>>>> automatically if properly configured. (Did you do a SET Dxx =20 > >>>>>>> ID=3Dxxyy as > > >>>>>>> well?) > > >>>>>> Do you mean > >>>>>> sysman> config set ? > > >>>>>> The device I created, should have shown up as > >>>>>> $1$DGA11 > > >>>>>> Haven't done this before. > > >>>>>> -- Using Opera's revolutionary e-mail client: =20 > >>>>>>http://www.opera.com/mail/ > > -- > Using Opera's revolutionary e-mail client:http://www.opera.com/mail/- Hid= e quoted text - > > - Show quoted text - On the HSG80 controller, type SHOW CONNECTIONS This will show what the controller is considering acceptable "connections". It has been a while, however I don't think that the HSG80 controllers know about 'nodes', they only know about connections (the names of which are user defined and can obviously include the Node name.) The connection is really the "path", so there will (normally) be multiple connections for each node, depending on number of hba's and fabrics. Dave. ------------------------------ Date: 23 May 2007 10:59:10 -0700 From: BaxterD@tessco.com Subject: Re: recognizing newly created device on HSG80 Message-ID: <1179943150.803416.267950@m36g2000hse.googlegroups.com> On May 23, 1:55 pm, Baxt...@tessco.com wrote: > On May 23, 1:20 pm, "Tom Linden" wrote: > > > > > > > On Wed, 23 May 2007 10:31:40 -0700, Uusim=E4ki =20 > > > wrote: > > > > Yes, exactly, just for the nodes with FC-HBA's. > > > When I try that > > > HSG80-BOT>set d11 enable=3Dhafner > > Error 1240: Connection hafner not found > > > > You have to do it only on one controller (you are running them in =20 > > > failover mode of course). > > Yes. > > > > Tom Linden wrote: > > >> On Wed, 23 May 2007 10:03:00 -0700, Tom Linden =20 > > >> wrote: > > > >>> On Wed, 23 May 2007 09:49:55 -0700, Uusim=E4ki =20 > > >>> wrote: > > > >>>> Just some checks: > > > >>>> Did you do > > >>>> HSG80> SET D11 ID=3D11 ! This sets the identifier needed for VMS > > >>>> HSG80> SET D11 ENA=3Dnode1,node2,... ! This allows the access to = the =20 > > >>>> disk for the cluster nodes > > > >>> I presume you only enter those nodes which have an HBA? > > >> Oh, and for each controller? > > > >>>> Then you should find them by doing: > > > >>>> $ mc sysman io auto > > >>>> $ show devi dg > > > >>>> norm.raph...@metso.com wrote: > > >>>>> SYSMAN> help io > > >>>>> IO > > >>>>> There are several IO subcommands. Select the one you want f= rom > > >>>>> the choices below. > > >>>>> Additional information available: > > >>>>> AUTOCONFIGURE CONNECT FIND_WWID LOAD REBUILD > > >>>>> REPLACE_WWID SCSI_PATH_VERIFY SET SHOW > > >>>>> IO Subtopic? > > >>>>> Topic? io find > > >>>>> IO > > >>>>> FIND_WWID > > >>>>> The IO FIND_WWID command probes all Fibre Channel ports, =20 > > >>>>> detects > > >>>>> all previously undiscovered tapes and medium changers behi= nd a > > >>>>> Modular Data Router (MDR), and assigns a worldwide identif= ier > > >>>>> (WWID) to each one. The command also displays a list of the > > >>>>> devices and their assigned device names and automatically =20 > > >>>>> records > > >>>>> this information in the SYS$SYSTEM:SYS$DEVICES.DAT file. =20 > > >>>>> Finally, > > >>>>> the command updates relevant local and clusterwide memory > > >>>>> structures. > > >>>>> To configure newly attached Fibre Channel tapes, use this= =20 > > >>>>> command > > >>>>> prior to running the SYSMAN command IO AUTOCONFIGURE. > > >>>>> You must have CMKRNL privilege to use the IO FIND_WWID =20 > > >>>>> command. > > >>>>> For more information about Fibre Channel, see the Guideli= nes =20 > > >>>>> for > > >>>>> OpenVMS Cluster Configurations. > > >>>>> Format > > >>>>> IO FIND_WWID > > >>>>> Additional information available: > > >>>>> Parameters Qualifiers Description Example > > >>>>> "Tom Linden" wrote on 05/23/2007 11:37:= 14 =20 > > >>>>> AM: > > > >>>>>> On Wed, 23 May 2007 08:39:07 -0700, Peter 'EPLAN' LANGSTOeGER > > >>>>>> wrote: > > > >>>>>>> In article , "Tom Linden" > > >>>>>>> writes: > > >>>>>>>> Using the wizard on a W2K box talking to SWCC I created a new =20 > > >>>>>>>> device > > >>>>>>>> Striped + raid, now what do I need to do to get the cluster no= des =20 > > >>>>>>>> to > > >>>>>>>> recognize it? I tried sysgen> io autoconfigure but that did = not > > >>>>>>>> work > > >>>>>>> Normally the CONFIGURE process does this for, that means they d= o =20 > > >>>>>>> appear > > >>>>>>> automatically if properly configured. (Did you do a SET Dxx =20 > > >>>>>>> ID=3Dxxyy as > > > >>>>>>> well?) > > > >>>>>> Do you mean > > >>>>>> sysman> config set ? > > > >>>>>> The device I created, should have shown up as > > >>>>>> $1$DGA11 > > > >>>>>> Haven't done this before. > > > >>>>>> -- Using Opera's revolutionary e-mail client: =20 > > >>>>>>http://www.opera.com/mail/ > > > -- > > Using Opera's revolutionary e-mail client:http://www.opera.com/mail/-Hi= de quoted text - > > > - Show quoted text - > > On the HSG80 controller, type SHOW CONNECTIONS > > This will show what the controller is considering acceptable > "connections". It has been a while, however I don't think that the > HSG80 controllers know about 'nodes', they only know about > connections (the names of which are user defined and can obviously > include the Node name.) The connection is really the "path", so > there will (normally) be multiple connections for each node, depending > on number of hba's and fabrics. > > Dave.- Hide quoted text - > > - Show quoted text - Sorry about the double post.. Just to finish off, once you have decided which connections apply to your cluster HBA.s, you must do the Set D11 enable=3D You must include at least one connection for each cluster node it this disk is to be mounted cluster-wide. Dave. ------------------------------ Date: Wed, 23 May 2007 21:03:30 +0300 From: =?ISO-8859-15?Q?Uusim=E4ki?= Subject: Re: recognizing newly created device on HSG80 Message-ID: <465481b3$0$15293$9b536df3@news.fv.fi> Well, then you need to look at what your connections (=HBA WWID's) are called: HSG80> show conn Then you'll see a list of your HBA WWID's. If you have other units (Dxxx) allocated for the same node already, you find the connections even easier by looking at the other units parameters: HSG80> sho Dyyy Where it shows (among the other info) the connections which have access to that unit. Then you can just use the same names: HSG80> set Dxxx enab=connection_name Btw, the unit initialization takes a while. The time is directly related to the storage set size (amount of disks). You can find out how it is proceeding by: HSG80> show R1 (e.g a raidset, which name is R1) Tom Linden wrote: > On Wed, 23 May 2007 10:31:40 -0700, Uusimäki > wrote: > >> >> Yes, exactly, just for the nodes with FC-HBA's. > > When I try that > > HSG80-BOT>set d11 enable=hafner > Error 1240: Connection hafner not found > >> >> You have to do it only on one controller (you are running them in >> failover mode of course). > Yes. >> >> >> >> >> >> Tom Linden wrote: >>> On Wed, 23 May 2007 10:03:00 -0700, Tom Linden >>> wrote: >>> >>>> On Wed, 23 May 2007 09:49:55 -0700, Uusimäki >>>> wrote: >>>> >>>>> >>>>> Just some checks: >>>>> >>>>> Did you do >>>>> HSG80> SET D11 ID=11 ! This sets the identifier needed for VMS >>>>> HSG80> SET D11 ENA=node1,node2,... ! This allows the access to the >>>>> disk for the cluster nodes >>>> >>>> I presume you only enter those nodes which have an HBA? >>> Oh, and for each controller? >>> >>>> >>>>> >>>>> Then you should find them by doing: >>>>> >>>>> $ mc sysman io auto >>>>> $ show devi dg >>>>> >>>>> >>>>> >>>>> >>>>> >>>>> norm.raphael@metso.com wrote: >>>>>> SYSMAN> help io >>>>>> IO >>>>>> There are several IO subcommands. Select the one you want from >>>>>> the choices below. >>>>>> Additional information available: >>>>>> AUTOCONFIGURE CONNECT FIND_WWID LOAD REBUILD >>>>>> REPLACE_WWID SCSI_PATH_VERIFY SET SHOW >>>>>> IO Subtopic? >>>>>> Topic? io find >>>>>> IO >>>>>> FIND_WWID >>>>>> The IO FIND_WWID command probes all Fibre Channel ports, >>>>>> detects >>>>>> all previously undiscovered tapes and medium changers behind a >>>>>> Modular Data Router (MDR), and assigns a worldwide identifier >>>>>> (WWID) to each one. The command also displays a list of the >>>>>> devices and their assigned device names and automatically >>>>>> records >>>>>> this information in the SYS$SYSTEM:SYS$DEVICES.DAT file. >>>>>> Finally, >>>>>> the command updates relevant local and clusterwide memory >>>>>> structures. >>>>>> To configure newly attached Fibre Channel tapes, use this >>>>>> command >>>>>> prior to running the SYSMAN command IO AUTOCONFIGURE. >>>>>> You must have CMKRNL privilege to use the IO FIND_WWID >>>>>> command. >>>>>> For more information about Fibre Channel, see the >>>>>> Guidelines for >>>>>> OpenVMS Cluster Configurations. >>>>>> Format >>>>>> IO FIND_WWID >>>>>> Additional information available: >>>>>> Parameters Qualifiers Description Example >>>>>> "Tom Linden" wrote on 05/23/2007 11:37:14 >>>>>> AM: >>>>>> >>>>>>> On Wed, 23 May 2007 08:39:07 -0700, Peter 'EPLAN' LANGSTOeGER >>>>>>> wrote: >>>>>>> >>>>>>>> In article , "Tom Linden" >>>>>>>> writes: >>>>>>>>> Using the wizard on a W2K box talking to SWCC I created a new >>>>>>>>> device >>>>>>>>> Striped + raid, now what do I need to do to get the cluster >>>>>>>>> nodes to >>>>>>>>> recognize it? I tried sysgen> io autoconfigure but that did not >>>>>>>>> work >>>>>>>> Normally the CONFIGURE process does this for, that means they do >>>>>>>> appear >>>>>>>> automatically if properly configured. (Did you do a SET Dxx >>>>>>>> ID=xxyy as >>>>>> >>>>>>>> well?) >>>>>>>> >>>>>>> Do you mean >>>>>>> sysman> config set ? >>>>>>> >>>>>>> The device I created, should have shown up as >>>>>>> $1$DGA11 >>>>>>> >>>>>>> Haven't done this before. >>>>>>> >>>>>>> -- Using Opera's revolutionary e-mail client: >>>>>>> http://www.opera.com/mail/ >>>>>> >>>> >>>> >>>> >>> > > > ------------------------------ Date: Wed, 23 May 2007 11:23:50 -0700 From: "Tom Linden" Subject: Re: recognizing newly created device on HSG80 Message-ID: On Wed, 23 May 2007 11:03:30 -0700, Uusimäki wrote: > > Well, then you need to look at what your connections (=HBA WWID's) are > called: > > HSG80> show conn > > Then you'll see a list of your HBA WWID's. If you have other units > (Dxxx) allocated for the same node already, you find the connections > even easier by looking at the other units parameters: > > HSG80> sho Dyyy > > Where it shows (among the other info) the connections which have access > to that unit. Then you can just use the same names: > > HSG80> set Dxxx enab=connection_name well that ends up being quite a list since I have 2 HBAs in each node, connected to to two switches, each with two connections to the two controllers, Oh well. It wasn't bad with cut and paste, but after i was done, I doscovered I could have said set D11 enable=all OK, so far so good. Peter indicated to run CONFIGURE, but that seems to just sit there HAFNER> mc configure %MSCPLOAD-I-CONFIGSCAN, enabled automatic disk serving %TMSCPLOAD-I-CONFIGSCAN, enabled automatic tape serving > > Btw, the unit initialization takes a while. The time is directly related > to the storage set size (amount of disks). You can find out how it is > proceeding by: > > HSG80> show R1 (e.g a raidset, which name is R1) > > > > > > Tom Linden wrote: >> On Wed, 23 May 2007 10:31:40 -0700, Uusimäki >> wrote: >> >>> >>> Yes, exactly, just for the nodes with FC-HBA's. >> When I try that >> HSG80-BOT>set d11 enable=hafner >> Error 1240: Connection hafner not found >> >>> >>> You have to do it only on one controller (you are running them in >>> failover mode of course). >> Yes. >>> >>> >>> >>> >>> >>> Tom Linden wrote: >>>> On Wed, 23 May 2007 10:03:00 -0700, Tom Linden >>>> wrote: >>>> >>>>> On Wed, 23 May 2007 09:49:55 -0700, Uusimäki >>>>> wrote: >>>>> >>>>>> >>>>>> Just some checks: >>>>>> >>>>>> Did you do >>>>>> HSG80> SET D11 ID=11 ! This sets the identifier needed for VMS >>>>>> HSG80> SET D11 ENA=node1,node2,... ! This allows the access to the >>>>>> disk for the cluster nodes >>>>> >>>>> I presume you only enter those nodes which have an HBA? >>>> Oh, and for each controller? >>>> >>>>> >>>>>> >>>>>> Then you should find them by doing: >>>>>> >>>>>> $ mc sysman io auto >>>>>> $ show devi dg >>>>>> >>>>>> >>>>>> >>>>>> >>>>>> >>>>>> norm.raphael@metso.com wrote: >>>>>>> SYSMAN> help io >>>>>>> IO >>>>>>> There are several IO subcommands. Select the one you want >>>>>>> from >>>>>>> the choices below. >>>>>>> Additional information available: >>>>>>> AUTOCONFIGURE CONNECT FIND_WWID LOAD REBUILD >>>>>>> REPLACE_WWID SCSI_PATH_VERIFY SET SHOW >>>>>>> IO Subtopic? >>>>>>> Topic? io find >>>>>>> IO >>>>>>> FIND_WWID >>>>>>> The IO FIND_WWID command probes all Fibre Channel ports, >>>>>>> detects >>>>>>> all previously undiscovered tapes and medium changers >>>>>>> behind a >>>>>>> Modular Data Router (MDR), and assigns a worldwide >>>>>>> identifier >>>>>>> (WWID) to each one. The command also displays a list of the >>>>>>> devices and their assigned device names and automatically >>>>>>> records >>>>>>> this information in the SYS$SYSTEM:SYS$DEVICES.DAT file. >>>>>>> Finally, >>>>>>> the command updates relevant local and clusterwide memory >>>>>>> structures. >>>>>>> To configure newly attached Fibre Channel tapes, use this >>>>>>> command >>>>>>> prior to running the SYSMAN command IO AUTOCONFIGURE. >>>>>>> You must have CMKRNL privilege to use the IO FIND_WWID >>>>>>> command. >>>>>>> For more information about Fibre Channel, see the >>>>>>> Guidelines for >>>>>>> OpenVMS Cluster Configurations. >>>>>>> Format >>>>>>> IO FIND_WWID >>>>>>> Additional information available: >>>>>>> Parameters Qualifiers Description Example >>>>>>> "Tom Linden" wrote on 05/23/2007 11:37:14 >>>>>>> AM: >>>>>>> >>>>>>>> On Wed, 23 May 2007 08:39:07 -0700, Peter 'EPLAN' LANGSTOeGER >>>>>>>> wrote: >>>>>>>> >>>>>>>>> In article , "Tom Linden" >>>>>>>>> writes: >>>>>>>>>> Using the wizard on a W2K box talking to SWCC I created a new >>>>>>>>>> device >>>>>>>>>> Striped + raid, now what do I need to do to get the cluster >>>>>>>>>> nodes to >>>>>>>>>> recognize it? I tried sysgen> io autoconfigure but that did >>>>>>>>>> not >>>>>>>>>> work >>>>>>>>> Normally the CONFIGURE process does this for, that means they do >>>>>>>>> appear >>>>>>>>> automatically if properly configured. (Did you do a SET Dxx >>>>>>>>> ID=xxyy as >>>>>>> >>>>>>>>> well?) >>>>>>>>> >>>>>>>> Do you mean >>>>>>>> sysman> config set ? >>>>>>>> >>>>>>>> The device I created, should have shown up as >>>>>>>> $1$DGA11 >>>>>>>> >>>>>>>> Haven't done this before. >>>>>>>> >>>>>>>> -- Using Opera's revolutionary e-mail client: >>>>>>>> http://www.opera.com/mail/ >>>>>>> >>>>> >>>>> >>>>> >>>> >> -- Using Opera's revolutionary e-mail client: http://www.opera.com/mail/ ------------------------------ Date: Wed, 23 May 2007 15:15:54 -0400 From: Jur van der Burg <"vdburg at hotmail dot com"> Subject: Re: recognizing newly created device on HSG80 Message-ID: <465492e5$0$330$e4fe514c@news.xs4all.nl> > HAFNER> mc configure Don't do that. The configure process is started automatically when it's needed and that's not for fibrechannel devices. mc sysman io auto/log should be able to get them visible on vms provided that the hsg is setup the proper way. Jur. Tom Linden wrote: > On Wed, 23 May 2007 11:03:30 -0700, Uusimäki > wrote: > >> >> Well, then you need to look at what your connections (=HBA WWID's) are >> called: >> >> HSG80> show conn >> >> Then you'll see a list of your HBA WWID's. If you have other units >> (Dxxx) allocated for the same node already, you find the connections >> even easier by looking at the other units parameters: >> >> HSG80> sho Dyyy >> >> Where it shows (among the other info) the connections which have >> access to that unit. Then you can just use the same names: >> >> HSG80> set Dxxx enab=connection_name > well that ends up being quite a list since I have 2 HBAs in each node, > connected to > to two switches, each with two connections to the two controllers, Oh > well. It wasn't > bad with cut and paste, but after i was done, I doscovered I could have > said > set D11 enable=all > OK, so far so good. Peter indicated to run CONFIGURE, but that seems > to just sit there > HAFNER> mc configure > %MSCPLOAD-I-CONFIGSCAN, enabled automatic disk serving > %TMSCPLOAD-I-CONFIGSCAN, enabled automatic tape serving > > >> >> Btw, the unit initialization takes a while. The time is directly >> related to the storage set size (amount of disks). You can find out >> how it is proceeding by: >> >> HSG80> show R1 (e.g a raidset, which name is R1) >> >> >> >> >> >> Tom Linden wrote: >>> On Wed, 23 May 2007 10:31:40 -0700, Uusimäki >>> wrote: >>> >>>> >>>> Yes, exactly, just for the nodes with FC-HBA's. >>> When I try that >>> HSG80-BOT>set d11 enable=hafner >>> Error 1240: Connection hafner not found >>> >>>> >>>> You have to do it only on one controller (you are running them in >>>> failover mode of course). >>> Yes. >>>> >>>> >>>> >>>> >>>> >>>> Tom Linden wrote: >>>>> On Wed, 23 May 2007 10:03:00 -0700, Tom Linden >>>>> wrote: >>>>> >>>>>> On Wed, 23 May 2007 09:49:55 -0700, Uusimäki >>>>>> wrote: >>>>>> >>>>>>> >>>>>>> Just some checks: >>>>>>> >>>>>>> Did you do >>>>>>> HSG80> SET D11 ID=11 ! This sets the identifier needed for VMS >>>>>>> HSG80> SET D11 ENA=node1,node2,... ! This allows the access to >>>>>>> the disk for the cluster nodes >>>>>> >>>>>> I presume you only enter those nodes which have an HBA? >>>>> Oh, and for each controller? >>>>> >>>>>> >>>>>>> >>>>>>> Then you should find them by doing: >>>>>>> >>>>>>> $ mc sysman io auto >>>>>>> $ show devi dg >>>>>>> >>>>>>> >>>>>>> >>>>>>> >>>>>>> >>>>>>> norm.raphael@metso.com wrote: >>>>>>>> SYSMAN> help io >>>>>>>> IO >>>>>>>> There are several IO subcommands. Select the one you want >>>>>>>> from >>>>>>>> the choices below. >>>>>>>> Additional information available: >>>>>>>> AUTOCONFIGURE CONNECT FIND_WWID LOAD REBUILD >>>>>>>> REPLACE_WWID SCSI_PATH_VERIFY SET SHOW >>>>>>>> IO Subtopic? >>>>>>>> Topic? io find >>>>>>>> IO >>>>>>>> FIND_WWID >>>>>>>> The IO FIND_WWID command probes all Fibre Channel ports, >>>>>>>> detects >>>>>>>> all previously undiscovered tapes and medium changers >>>>>>>> behind a >>>>>>>> Modular Data Router (MDR), and assigns a worldwide >>>>>>>> identifier >>>>>>>> (WWID) to each one. The command also displays a list of the >>>>>>>> devices and their assigned device names and automatically >>>>>>>> records >>>>>>>> this information in the SYS$SYSTEM:SYS$DEVICES.DAT file. >>>>>>>> Finally, >>>>>>>> the command updates relevant local and clusterwide memory >>>>>>>> structures. >>>>>>>> To configure newly attached Fibre Channel tapes, use >>>>>>>> this command >>>>>>>> prior to running the SYSMAN command IO AUTOCONFIGURE. >>>>>>>> You must have CMKRNL privilege to use the IO FIND_WWID >>>>>>>> command. >>>>>>>> For more information about Fibre Channel, see the >>>>>>>> Guidelines for >>>>>>>> OpenVMS Cluster Configurations. >>>>>>>> Format >>>>>>>> IO FIND_WWID >>>>>>>> Additional information available: >>>>>>>> Parameters Qualifiers Description Example >>>>>>>> "Tom Linden" wrote on 05/23/2007 >>>>>>>> 11:37:14 AM: >>>>>>>> >>>>>>>>> On Wed, 23 May 2007 08:39:07 -0700, Peter 'EPLAN' LANGSTOeGER >>>>>>>>> wrote: >>>>>>>>> >>>>>>>>>> In article , "Tom Linden" >>>>>>>>>> writes: >>>>>>>>>>> Using the wizard on a W2K box talking to SWCC I created a new >>>>>>>>>>> device >>>>>>>>>>> Striped + raid, now what do I need to do to get the cluster >>>>>>>>>>> nodes to >>>>>>>>>>> recognize it? I tried sysgen> io autoconfigure but that did >>>>>>>>>>> not >>>>>>>>>>> work >>>>>>>>>> Normally the CONFIGURE process does this for, that means they >>>>>>>>>> do appear >>>>>>>>>> automatically if properly configured. (Did you do a SET Dxx >>>>>>>>>> ID=xxyy as >>>>>>>> >>>>>>>>>> well?) >>>>>>>>>> >>>>>>>>> Do you mean >>>>>>>>> sysman> config set ? >>>>>>>>> >>>>>>>>> The device I created, should have shown up as >>>>>>>>> $1$DGA11 >>>>>>>>> >>>>>>>>> Haven't done this before. >>>>>>>>> >>>>>>>>> -- Using Opera's revolutionary e-mail client: >>>>>>>>> http://www.opera.com/mail/ >>>>>>>> >>>>>> >>>>>> >>>>>> >>>>> >>> > > > ------------------------------ Date: 23 May 2007 21:28:56 +0200 From: peter@langstoeger.at (Peter 'EPLAN' LANGSTOeGER) Subject: Re: recognizing newly created device on HSG80 Message-ID: <4654b218$1@news.langstoeger.at> In article , "Tom Linden" writes: >OK, so far so good. Peter indicated to run CONFIGURE, but that seems to >just sit there >HAFNER> mc configure >%MSCPLOAD-I-CONFIGSCAN, enabled automatic disk serving >%TMSCPLOAD-I-CONFIGSCAN, enabled automatic tape serving No. I didn't tell to run CONFIGURE! I wrote, that a process named "CONFIGURE" - started by VMS startup - runs CONFIGURE.EXE. Why don't you do a SHOW SYS/CLU/PROC=*CONF* in you cluster? -- Peter "EPLAN" LANGSTOEGER Network and OpenVMS system specialist E-mail peter@langstoeger.at A-1030 VIENNA AUSTRIA I'm not a pessimist, I'm a realist ------------------------------ Date: Wed, 23 May 2007 12:18:35 -0700 From: "Tom Linden" Subject: Re: recognizing newly created device on HSG80 Message-ID: On Wed, 23 May 2007 12:15:54 -0700, Jur van der Burg <"vdburg at hotmail dot com"> wrote: > > HAFNER> mc configure > > Don't do that. The configure process is started automatically when it's > needed > and that's not for fibrechannel devices. > > mc sysman io auto/log should be able to get them visible on vms provided > that the hsg is setup the proper way. > Excellent, thanks, the identifier wasn't set. > Jur. > > > Tom Linden wrote: >> On Wed, 23 May 2007 11:03:30 -0700, Uusimäki >> wrote: >> >>> >>> Well, then you need to look at what your connections (=HBA WWID's) are >>> called: >>> >>> HSG80> show conn >>> >>> Then you'll see a list of your HBA WWID's. If you have other units >>> (Dxxx) allocated for the same node already, you find the connections >>> even easier by looking at the other units parameters: >>> >>> HSG80> sho Dyyy >>> >>> Where it shows (among the other info) the connections which have >>> access to that unit. Then you can just use the same names: >>> >>> HSG80> set Dxxx enab=connection_name >> well that ends up being quite a list since I have 2 HBAs in each node, >> connected to >> to two switches, each with two connections to the two controllers, Oh >> well. It wasn't >> bad with cut and paste, but after i was done, I doscovered I could have >> said >> set D11 enable=all >> OK, so far so good. Peter indicated to run CONFIGURE, but that seems >> to just sit there >> HAFNER> mc configure >> %MSCPLOAD-I-CONFIGSCAN, enabled automatic disk serving >> %TMSCPLOAD-I-CONFIGSCAN, enabled automatic tape serving >> >>> >>> Btw, the unit initialization takes a while. The time is directly >>> related to the storage set size (amount of disks). You can find out >>> how it is proceeding by: >>> >>> HSG80> show R1 (e.g a raidset, which name is R1) >>> >>> >>> >>> >>> >>> Tom Linden wrote: >>>> On Wed, 23 May 2007 10:31:40 -0700, Uusimäki >>>> wrote: >>>> >>>>> >>>>> Yes, exactly, just for the nodes with FC-HBA's. >>>> When I try that >>>> HSG80-BOT>set d11 enable=hafner >>>> Error 1240: Connection hafner not found >>>> >>>>> >>>>> You have to do it only on one controller (you are running them in >>>>> failover mode of course). >>>> Yes. >>>>> >>>>> >>>>> >>>>> >>>>> >>>>> Tom Linden wrote: >>>>>> On Wed, 23 May 2007 10:03:00 -0700, Tom Linden >>>>>> wrote: >>>>>> >>>>>>> On Wed, 23 May 2007 09:49:55 -0700, Uusimäki >>>>>>> wrote: >>>>>>> >>>>>>>> >>>>>>>> Just some checks: >>>>>>>> >>>>>>>> Did you do >>>>>>>> HSG80> SET D11 ID=11 ! This sets the identifier needed for VMS >>>>>>>> HSG80> SET D11 ENA=node1,node2,... ! This allows the access to >>>>>>>> the disk for the cluster nodes >>>>>>> >>>>>>> I presume you only enter those nodes which have an HBA? >>>>>> Oh, and for each controller? >>>>>> >>>>>>> >>>>>>>> >>>>>>>> Then you should find them by doing: >>>>>>>> >>>>>>>> $ mc sysman io auto >>>>>>>> $ show devi dg >>>>>>>> >>>>>>>> >>>>>>>> >>>>>>>> >>>>>>>> >>>>>>>> norm.raphael@metso.com wrote: >>>>>>>>> SYSMAN> help io >>>>>>>>> IO >>>>>>>>> There are several IO subcommands. Select the one you want >>>>>>>>> from >>>>>>>>> the choices below. >>>>>>>>> Additional information available: >>>>>>>>> AUTOCONFIGURE CONNECT FIND_WWID LOAD REBUILD >>>>>>>>> REPLACE_WWID SCSI_PATH_VERIFY SET SHOW >>>>>>>>> IO Subtopic? >>>>>>>>> Topic? io find >>>>>>>>> IO >>>>>>>>> FIND_WWID >>>>>>>>> The IO FIND_WWID command probes all Fibre Channel ports, >>>>>>>>> detects >>>>>>>>> all previously undiscovered tapes and medium changers >>>>>>>>> behind a >>>>>>>>> Modular Data Router (MDR), and assigns a worldwide >>>>>>>>> identifier >>>>>>>>> (WWID) to each one. The command also displays a list of >>>>>>>>> the >>>>>>>>> devices and their assigned device names and automatically >>>>>>>>> records >>>>>>>>> this information in the SYS$SYSTEM:SYS$DEVICES.DAT file. >>>>>>>>> Finally, >>>>>>>>> the command updates relevant local and clusterwide memory >>>>>>>>> structures. >>>>>>>>> To configure newly attached Fibre Channel tapes, use >>>>>>>>> this command >>>>>>>>> prior to running the SYSMAN command IO AUTOCONFIGURE. >>>>>>>>> You must have CMKRNL privilege to use the IO FIND_WWID >>>>>>>>> command. >>>>>>>>> For more information about Fibre Channel, see the >>>>>>>>> Guidelines for >>>>>>>>> OpenVMS Cluster Configurations. >>>>>>>>> Format >>>>>>>>> IO FIND_WWID >>>>>>>>> Additional information available: >>>>>>>>> Parameters Qualifiers Description Example >>>>>>>>> "Tom Linden" wrote on 05/23/2007 >>>>>>>>> 11:37:14 AM: >>>>>>>>> >>>>>>>>>> On Wed, 23 May 2007 08:39:07 -0700, Peter 'EPLAN' LANGSTOeGER >>>>>>>>>> wrote: >>>>>>>>>> >>>>>>>>>>> In article , "Tom Linden" >>>>>>>>>>> writes: >>>>>>>>>>>> Using the wizard on a W2K box talking to SWCC I created a new >>>>>>>>>>>> device >>>>>>>>>>>> Striped + raid, now what do I need to do to get the cluster >>>>>>>>>>>> nodes to >>>>>>>>>>>> recognize it? I tried sysgen> io autoconfigure but that did >>>>>>>>>>>> not >>>>>>>>>>>> work >>>>>>>>>>> Normally the CONFIGURE process does this for, that means they >>>>>>>>>>> do appear >>>>>>>>>>> automatically if properly configured. (Did you do a SET Dxx >>>>>>>>>>> ID=xxyy as >>>>>>>>> >>>>>>>>>>> well?) >>>>>>>>>>> >>>>>>>>>> Do you mean >>>>>>>>>> sysman> config set ? >>>>>>>>>> >>>>>>>>>> The device I created, should have shown up as >>>>>>>>>> $1$DGA11 >>>>>>>>>> >>>>>>>>>> Haven't done this before. >>>>>>>>>> >>>>>>>>>> -- Using Opera's revolutionary e-mail client: >>>>>>>>>> http://www.opera.com/mail/ >>>>>>>>> >>>>>>> >>>>>>> >>>>>>> >>>>>> >>>> >> -- Using Opera's revolutionary e-mail client: http://www.opera.com/mail/ ------------------------------ Date: Wed, 23 May 2007 12:20:31 -0700 From: "Tom Linden" Subject: Re: recognizing newly created device on HSG80 Message-ID: On Wed, 23 May 2007 12:28:56 -0700, Peter 'EPLAN' LANGSTOeGER = wrote: > In article , "Tom Linden" = > writes: >> OK, so far so good. Peter indicated to run CONFIGURE, but that seem= s = >> to >> just sit there >> HAFNER> mc configure >> %MSCPLOAD-I-CONFIGSCAN, enabled automatic disk serving >> %TMSCPLOAD-I-CONFIGSCAN, enabled automatic tape serving > > No. I didn't tell to run CONFIGURE! > I wrote, that a process named "CONFIGURE" - started by VMS startup - r= uns > CONFIGURE.EXE. Why don't you do a SHOW SYS/CLU/PROC=3D*CONF* in you = > cluster? > Thanks, Peter, I have bungled my through it now, more to come. -- = Using Opera's revolutionary e-mail client: http://www.opera.com/mail/ ------------------------------ Date: Wed, 23 May 2007 16:13:19 -0700 From: "Malcolm Dunnett" Subject: Re: recognizing newly created device on HSG80 Message-ID: <4654ca90@flight> wrote in message news:1179943150.803416.267950@m36g2000hse.googlegroups.com... >Just to finish off, once you have decided which connections apply to >your cluster HBA.s, you must do the >Set D11 enable= The enable= should not be necessary, by default the HSG80 will talk to all connections. The "set D11 ident=11" (on the HSG80 console) and a "$SYSMAN IO AUTO" from a suitably privileged account on VMS should do the trick. With regard to identifiers, IIRC the HSG80 won't allow an identifier to be greater than 255. ------------------------------ Date: Wed, 23 May 2007 21:28:42 -0500 From: David J Dachtera Subject: Re: recognizing newly created device on HSG80 Message-ID: <4654F85A.CBF9C8A6@spam.comcast.net> Uusimäki wrote: > > Just some checks: > > Did you do > HSG80> SET D11 ID=11 ! This sets the identifier needed for VMS > HSG80> SET D11 ENA=node1,node2,... ! This allows the access to the disk > for the cluster nodes Are they not enabled for all connections by default? -- David J Dachtera dba DJE Systems http://www.djesys.com/ Unofficial OpenVMS Marketing Home Page http://www.djesys.com/vms/market/ Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/ Unofficial OpenVMS-IA32 Home Page: http://www.djesys.com/vms/ia32/ Unofficial OpenVMS Hobbyist Support Page: http://www.djesys.com/vms/support/ ------------------------------ Date: 23 May 2007 14:30:58 -0500 From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) Subject: Re: Rename Volume Labels Message-ID: In article , koehler@eisner.nospam.encompasserve.org (Bob Koehler) writes: > In article , "P. Sture" writes: >> >> But presumably _your_ logical names do reference DISK$label-goes-here? >> > > No. When you put a logical name at the end of the MONUT command then > the logical name translates to the device name. Of course, it may not work if you misspell MOUNT. (Did I get it right that time?) ------------------------------ Date: Thu, 24 May 2007 08:24:42 +0800 From: "Richard Maher" Subject: Re: TCPIP programming (sockaddr_in question) Message-ID: Hi Brian, My best guess is that some software on your box is creating a socket using the address family AF_INET6. Why someone uses IPV6 (apart from running out of addresses) I don't know. New Multicast stuff? Something else wiz-bang? Why not try this: - Use the normal 16 byte IPv4 sockaddr structure and if the first word ends up "26" the do a with the 28 byte structure. (At that stage can you dump out the owner of the BG device so we can see who it is and what software they're running?) If you manage to retrieve the Port Number then please let me know. Actually, if your observations and experiences vary *at all* from what I described in the previous reply then please let us all know. From what I think I've seen, I still say it's a bug. John G is too quiet at the moment considering he gets up hours before me and, once Bootcamp winds-up, I think Matt goes on the after-dinner circuit promoting his new book "Deadlines - Smedlines" so I don't expect to hear much from that direction :-) Cheers Richard Maher wrote in message news:00A680B2.B7839670@SendSpamHere.ORG... > In article , "Richard Maher" writes: > > > > > >What software created the BG device for which you are now doing a > >io$_sensemode? > > > >Cheers Richard Maher > > Web browsers, remote ssh, remote sftp and ftp, mail servers, etc. > Like I said before, I'm trying to "programmatically" perform the > equivalent of TCPIP SHOW DEVICE bgxxxx/FULL > > -- > VAXman- A Bored Certified VMS Kernel Mode Hacker VAXman(at)TMESIS(dot)COM > > "Well my son, life is like a beanstalk, isn't it?" ------------------------------ Date: Thu, 24 May 2007 01:46:01 GMT From: VAXman- @SendSpamHere.ORG Subject: Re: TCPIP programming (sockaddr_in question) Message-ID: <00A6810C.37D24202@SendSpamHere.ORG> In article , "Richard Maher" writes: > > >Hi Brian, > >My best guess is that some software on your box is creating a socket using >the address family AF_INET6. Why someone uses IPV6 (apart from running out >of addresses) I don't know. New Multicast stuff? Something else wiz-bang? Well, the "some software" would be TCP/IP services. When testing this, the bg device was associated with a telnet session. >Why not try this: - > >Use the normal 16 byte IPv4 sockaddr structure and if the first word ends up >"26" the do a with the 28 byte structure. (At >that stage can you dump out the owner of the BG device so we can see who it >is and what software they're running?) If you manage to retrieve the Port >Number then please let me know. Actually, if your observations and >experiences vary *at all* from what I described in the previous reply then >please let us all know. Why would I do 2 $QIOs when I need only one. I've been calling the $QIO and passing a descriptor pointing to ample space to contain the IPv6 sockaddr_in structure. There are several ways available to determine what was returned and to parse out the data I desire. >From what I think I've seen, I still say it's a bug. John G is too quiet at >the moment considering he gets up hours before me and, once Bootcamp >winds-up, I think Matt goes on the after-dinner circuit promoting his new >book "Deadlines - Smedlines" so I don't expect to hear much from that >direction :-) LOL! :) -- VAXman- A Bored Certified VMS Kernel Mode Hacker VAXman(at)TMESIS(dot)COM "Well my son, life is like a beanstalk, isn't it?" ------------------------------ Date: Thu, 24 May 2007 12:03:42 +0800 From: "Richard Maher" Subject: Re: TCPIP programming (sockaddr_in question) Message-ID: Hi Brian, > Well, the "some software" would be TCP/IP services. When testing this, the > bg device was associated with a telnet session. Hey, me too (Specifically Telnet); I wonder why? This is what I get with a show service on telnet: - Flags: Listen Rtty IPv6 I don't seem to have a description in the ucx help for /FLAGS=IPv6 or /PROTOCOL=Family=V6 on my version so I assume that you can somehow communicate the fact that you want the Auxillary Server to listen on an INET6 socket so that all the progeny get IPv6 capabilities. > Why would I do 2 $QIOs when I need only one. I've been calling the $QIO and > passing a descriptor pointing to ample space to contain the IPv6 sockaddr_in > structure. There are several ways available to determine what was returned > and to parse out the data I desire. I was about to say "So you can at least get the bloody port number for the v4 devices" but then I checked, and my Telnet BG devices *are* returning the correct port number, so you're probably wondering what I'm babbling on about. Then I realized that I'd been looking at the client-side BG devices (I'm testing with client and server on the same box) and 'cos they're all connecting to port 1024 or "0004" in network byte order that's why I was constantly getting a perfectly reasonable port_number, Doh! I still think the fact that you can get a IPv6 Socket back without the io$m_extend is a buglet, but in the abscence of differing behaviour in Multinet or TCPWare, who cares? I'll stop taking up your time now :-) > LOL! :) Yeah, but you just know that there are those who never laugh at such things. (Makes it all worthwhile really :-) Cheers Richard Maher wrote in message news:00A6810C.37D24202@SendSpamHere.ORG... > In article , "Richard Maher" writes: > > > > > >Hi Brian, > > > >My best guess is that some software on your box is creating a socket using > >the address family AF_INET6. Why someone uses IPV6 (apart from running out > >of addresses) I don't know. New Multicast stuff? Something else wiz-bang? > > Well, the "some software" would be TCP/IP services. When testing this, the > bg device was associated with a telnet session. > > > > >Why not try this: - > > > >Use the normal 16 byte IPv4 sockaddr structure and if the first word ends up > >"26" the do a with the 28 byte structure. (At > >that stage can you dump out the owner of the BG device so we can see who it > >is and what software they're running?) If you manage to retrieve the Port > >Number then please let me know. Actually, if your observations and > >experiences vary *at all* from what I described in the previous reply then > >please let us all know. > > Why would I do 2 $QIOs when I need only one. I've been calling the $QIO and > passing a descriptor pointing to ample space to contain the IPv6 sockaddr_in > structure. There are several ways available to determine what was returned > and to parse out the data I desire. > > > > >From what I think I've seen, I still say it's a bug. John G is too quiet at > >the moment considering he gets up hours before me and, once Bootcamp > >winds-up, I think Matt goes on the after-dinner circuit promoting his new > >book "Deadlines - Smedlines" so I don't expect to hear much from that > >direction :-) > > LOL! :) > > -- > VAXman- A Bored Certified VMS Kernel Mode Hacker VAXman(at)TMESIS(dot)COM > > "Well my son, life is like a beanstalk, isn't it?" ------------------------------ Date: 23 May 2007 11:51:36 -0700 From: Sam Hoblit Subject: Re: TCPIP Services SMTP, RBLs blocking all inbound email Message-ID: <1179946295.929742.317050@a26g2000pre.googlegroups.com> > Anyone experienced this before? Its like the RBL lookup functionality > has deadlocked itself. I wrote to Spamhaus (and did some google > searches for recent spamhaus news) in case the problem was at their > end but haven't found (or received) anything back. So far only the > three VMS sites have been affected; two PC mail servers (one using > spamhaus) have not shown the same problems. We had the same problem a while back, but I never took time to try to resolve it. We're running 7.3-2, TCPIP v. 5.4, eco6, but we might have still been on eco5 when I tried using the RBLs. We also used sbl- xbl.spamhaus.org and combined.njabl.org, as well as a few others. Once I noticed no incoming email to the whole system, I just deleted the RBL lists in smtp.config and the problem went away. We don't have a terrible spam problem here, so removing the RBL lists wasnt' that big a deal. Still, it would be nice to know the source of this problem. ------------------------------ Date: Wed, 23 May 2007 15:07:12 -0400 From: JF Mezei Subject: Re: TCPIP Services SMTP, RBLs blocking all inbound email Message-ID: Did a bit of reading on the spamhaus web site. sbl-xbl combined list still exists apparently, but it is not listed in the main lists as the "zen" is now the recommended list. Some ISPs , notably AT&T are known to block DNS requests to the well known RBLs (supposedly to reduce load on their DNS servers). (no sure is this is the old AT&T or the newly reconstituted at&t (aka southwest Bell that bought AT&T and BellSouth) ------------------------------ Date: 23 May 2007 13:21:10 -0700 From: Rich Jordan Subject: Re: TCPIP Services SMTP, RBLs blocking all inbound email Message-ID: <1179951670.110831.42200@k79g2000hse.googlegroups.com> On May 23, 1:20 pm, JF Mezei wrote: > Rich Jordan wrote: > > I've had three sites suddenly stop receiving email since Monday; they > > are Alphas using the TCPIP Services SMTP service. All have RBL > > filtering enabled, all had sbl-xbl.spamhaus.org, > > From the node that blocks email, have you tried to nslookup some IP to > see how it responds. > > Say you get a SMTP connection from 20.30.40.50 > > $ nslookup 50.40.30.20.zen.spamhaus.org > > It should return a "not found" error (or an IP address in a 127.0.0.* > range if that IP is on their list. > > You can also make tests: > Testing your SBL Setup > Once you have set up your mail server to use sbl.spamhaus.org, you can > test to see if the SBL blocking is working by sending an email (any > email) to: nelson-sbl-t...@crynwr.com (you must send the email from the > mail server which you wish to test). The Crynwr system robot will answer > you to tell you if your server is correctly blocking SBL-listed IPs or not. > > Note that their web pages don't seem to mention sbl-xbl anymore. They > talk about zen.spamhaus.org which is the combined xbl, sbl and pbl Thanks for replying; I will try the tests. No response from spamhaus yet (not surprising) and also still no indication that this might be an external problem (nobody else complaining)... but three systems all at once makes me think _something_ has changed. I'll probably try the zen list too. Thanks! Rich ------------------------------ Date: 23 May 2007 14:23:29 -0700 From: Rich Jordan Subject: Re: TCPIP Services SMTP, RBLs blocking all inbound email Message-ID: <1179955409.582761.233310@h2g2000hsg.googlegroups.com> On May 23, 2:07 pm, JF Mezei wrote: > Did a bit of reading on the spamhaus web site. > > sbl-xbl combined list still exists apparently, but it is not listed in > the main lists as the "zen" is now the recommended list. > > Some ISPs , notably AT&T are known to block DNS requests to the well > known RBLs (supposedly to reduce load on their DNS servers). > > (no sure is this is the old AT&T or the newly reconstituted at&t (aka > southwest Bell that bought AT&T and BellSouth) Problem localized. The three sites also have Dotster domain registration and DNS service in common. Dotster has apparently decided to 'monetize' all those wildcard and mis-spelled subdomains of their customer domains... 'anything'.mydomain.com that does not have its own specific record in that domain's DNS now resolves to a paid advertising site. When I do a lookup for ##.##.##.##.zen.spamhaus.org from any of the three sites I get response back ##.##.##.##.zen.spamhaus.org.mydomain.com with the two parking site IP addresses; a positive response so its blacklisted. I can now browse to any random word.mydomain.com and get to that damned search page... I AM PISSED. There was NO notification of this (the customer service rep confirmed that). She did provide a workaround... set a *.mydomain.com CNAME pointing to 'anything' (we pointed it to the website address). Have to wait to see if it works. And then decide if its time to change registrars and service providers again. ------------------------------ Date: 23 May 2007 14:31:55 -0500 From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) Subject: Re: World's Fastest Processor? For Now, Anyway Message-ID: In article <5bjc4rF2r5absU1@mid.individual.net>, bill@cs.uofs.edu (Bill Gunshannon) writes: > > Your joking, right? > > "PL/I was developed by IBM, at its Hursley Laboratories in the > United Kingdom, as part of the development of System/360. " > I though PL/I came out about 5 years later, after the 360 was well established. ------------------------------ Date: Wed, 23 May 2007 15:57:04 -0400 From: "Main, Kerry" Subject: RE: World's Fastest Processor? For Now, Anyway Message-ID: Some history on FP - http://www.cs.berkeley.edu/~wkahan/ieee754status/754story.html Interesting extract: "Intel had decided they wanted really good arithmetic. I suggested that DEC VAX's floating-point be copied because it was very good for its time. But Intel wanted the `best' arithmetic." Warning - you better love this FP topic before starting to read the article - its long. :-) Kerry Main Senior Consultant HP Services Canada Voice: 613-592-4660 Fax: 613-591-4477 kerryDOTmainAThpDOTcom (remove the DOT's and AT)=20 OpenVMS - the secure, multi-site OS that just works. ------------------------------ Date: Wed, 23 May 2007 19:03:31 -0400 From: Chip Coldwell Subject: Re: World's Fastest Processor? For Now, Anyway Message-ID: On Wed, 23 May 2007, Dan Foster wrote: > In article <66dfp8Idh60R@eisner.encompasserve.org>, Larry Kilgallen wrote: > >> > >> What kind basis do they have for making this claim that it's the first > >> one to do it in hardware? Haven't there been a number of > >> microprocessors that did this long ago? > > > > Perhaps they were not _UNIX_ microprocessors, like this one is :-) > > That's true; but the Motorola 68040 had built-in FP support in hardware > some 17 years ago. Ah, but that was *binary* floating-point. What IBM is describing is *decimal* floating-point. Decimal floating-point is standard in pocket calculators, because user's get annoyed when 0.1 (which has no exact binary representation) gets rounded to the nearest binary floating-point equivalent. IBM had decimal arithmetic in hardware in their mainframes in the '60s, but that may have been fixed-point. Chip -- Charles M. "Chip" Coldwell "Turn on, log in, tune out" ------------------------------ Date: Wed, 23 May 2007 16:40:27 -0700 From: "Tom Linden" Subject: Re: World's Fastest Processor? For Now, Anyway Message-ID: On Wed, 23 May 2007 16:03:31 -0700, Chip Coldwell wrote: > On Wed, 23 May 2007, Dan Foster wrote: > >> In article <66dfp8Idh60R@eisner.encompasserve.org>, Larry Kilgallen >> wrote: >> >> >> >> What kind basis do they have for making this claim that it's the >> first >> >> one to do it in hardware? Haven't there been a number of >> >> microprocessors that did this long ago? >> > >> > Perhaps they were not _UNIX_ microprocessors, like this one is :-) >> >> That's true; but the Motorola 68040 had built-in FP support in hardware >> some 17 years ago. > > Ah, but that was *binary* floating-point. What IBM is describing is > *decimal* floating-point. > > Decimal floating-point is standard in pocket calculators, because user's > get annoyed when 0.1 (which has no exact binary representation) gets > rounded to the nearest binary floating-point equivalent. > > IBM had decimal arithmetic in hardware in their mainframes in the '60s, > but that may have been fixed-point. That was scaled fixed decimal. so you might, for example, declare in your program something like declare foo fixed decimal(34,15); which would give you 19 digits left of the decimal point and 15 right you are free to choose p and q in fixed decimal(p,q) subject to implementation limits > > Chip > -- Using Opera's revolutionary e-mail client: http://www.opera.com/mail/ ------------------------------ Date: 24 May 2007 00:27:53 GMT From: bill@cs.uofs.edu (Bill Gunshannon) Subject: Re: World's Fastest Processor? For Now, Anyway Message-ID: <5bk4g9F2th5pnU1@mid.individual.net> In article , Chip Coldwell writes: > On Wed, 23 May 2007, Dan Foster wrote: > >> In article <66dfp8Idh60R@eisner.encompasserve.org>, Larry Kilgallen wrote: >> >> >> >> What kind basis do they have for making this claim that it's the first >> >> one to do it in hardware? Haven't there been a number of >> >> microprocessors that did this long ago? >> > >> > Perhaps they were not _UNIX_ microprocessors, like this one is :-) >> >> That's true; but the Motorola 68040 had built-in FP support in hardware >> some 17 years ago. > > Ah, but that was *binary* floating-point. What IBM is describing is > *decimal* floating-point. > > Decimal floating-point is standard in pocket calculators, because user's > get annoyed when 0.1 (which has no exact binary representation) gets > rounded to the nearest binary floating-point equivalent. I have never seen a pocket calculator that did not have the standard floating point errors. > > IBM had decimal arithmetic in hardware in their mainframes in the '60s, > but that may have been fixed-point. It was. bill -- Bill Gunshannon | de-moc-ra-cy (di mok' ra see) n. Three wolves bill@cs.scranton.edu | and a sheep voting on what's for dinner. University of Scranton | Scranton, Pennsylvania | #include ------------------------------ Date: Wed, 23 May 2007 21:26:27 -0400 From: =?ISO-8859-1?Q?Arne_Vajh=F8j?= Subject: Re: World's Fastest Processor? For Now, Anyway Message-ID: <4654e9b9$0$90276$14726298@news.sunsite.dk> Main, Kerry wrote: >>>> The POWER6 supposedly is the first UNIX microprocessor able to >>>> calculate decimal floating-point arithmetic in hardware. Until now, >>>> calculations involving decimal numbers with floating decimal points >>>> were done using software. The built-in decimal floating-point >>>> capability gives advantage to enterprises running complex tax, >>>> financial, and ERP programs, among others. > VAX 11780/11785/11750 ..all had FP HW options .. Almost all computers has binary floating point. The early VAX'es had decimal fixed point (BCD). Decimal floating point is a third thing. Arne ------------------------------ Date: Wed, 23 May 2007 21:33:27 -0400 From: =?ISO-8859-1?Q?Arne_Vajh=F8j?= Subject: Re: World's Fastest Processor? For Now, Anyway Message-ID: <4654eb5e$0$90275$14726298@news.sunsite.dk> Bill Gunshannon wrote: > they are claiming Decimal Floating > Point. That would be Floating Point without the rounding errors. Decimal FP has similar representation problems as binary FP. Decimal FP just works well with 1/10's similar to how binary FP works well with 1/2's. And 1/10's is the standard used by people. It should also be noted that decimal FP still has basic FP characteristics so that X > 0 does not imply that Y + X > Y etc.. Arne ------------------------------ Date: Thu, 24 May 2007 05:20:19 +0200 From: "P. Sture" Subject: Re: World's Fastest Processor? For Now, Anyway Message-ID: In article <5bk4g9F2th5pnU1@mid.individual.net>, bill@cs.uofs.edu (Bill Gunshannon) wrote: > In article , > Chip Coldwell writes: > > On Wed, 23 May 2007, Dan Foster wrote: > > > >> In article <66dfp8Idh60R@eisner.encompasserve.org>, Larry Kilgallen > >> wrote: > >> >> > >> >> What kind basis do they have for making this claim that it's the first > >> >> one to do it in hardware? Haven't there been a number of > >> >> microprocessors that did this long ago? > >> > > >> > Perhaps they were not _UNIX_ microprocessors, like this one is :-) > >> > >> That's true; but the Motorola 68040 had built-in FP support in hardware > >> some 17 years ago. > > > > Ah, but that was *binary* floating-point. What IBM is describing is > > *decimal* floating-point. > > > > Decimal floating-point is standard in pocket calculators, because user's > > get annoyed when 0.1 (which has no exact binary representation) gets > > rounded to the nearest binary floating-point equivalent. > > I have never seen a pocket calculator that did not have the standard > floating point errors. Which is why years ago I invested in a desktop calculator for doing accounts. I have no idea what the inner logic used, but you didn't get the type of rounding errors found with a pocket calculator. -- Paul Sture ------------------------------ Date: Wed, 23 May 2007 19:12:11 -0400 From: Chip Coldwell Subject: Re: Xpdf 3.02 for OpenVMS/Itanium Message-ID: OK, if anybody is interested I've built a package for Xpdf 3.02/Itanium (thanks to the HP test-drive cluster). A few notes: 1. This one doesn't use shared libraries for t1lib or freetype since I don't have the required privileges to install those on the HP system. 2. I have no idea what the platform name for OpenVMS/Itanium is ... AXPVMS is Alpha so I guessed IPFVMS ... but somebody should correct me. 3. You still need ghostscript (for the type 1 fonts). I dunno if the freeware cd has been ported to IPF or not. You can get the package here ftp://frank.harvard.edu/pub/coldwell/vms/xpdf/freeware-ipfvms-xpdf-v0302-0-1.pcsi let me know if it works .... Chip -- Charles M. "Chip" Coldwell "Turn on, log in, tune out" ------------------------------ Date: Thu, 24 May 2007 02:26:48 GMT From: "John E. Malmberg" Subject: Re: Xpdf 3.02 for OpenVMS/Itanium Message-ID: Chip Coldwell wrote: > OK, if anybody is interested I've built a package for Xpdf 3.02/Itanium > (thanks to the HP test-drive cluster). A few notes: You were able to display an X-11 image from them on a local X-server for testing? Interesting. > 1. This one doesn't use shared libraries for t1lib or freetype since I > don't have the required privileges to install those on the HP system. You do not need any privileges. The shared images do not need to be installed to be used for development. Nor are they required to be installed for use at runtime. It is just more efficient if they are. See the gnv directory at anonymous ftp from encompasserve.org for a number of shared images. It might be useful to someone if Itanium versions of them were built. > 2. I have no idea what the platform name for OpenVMS/Itanium is ... > AXPVMS is Alpha so I guessed IPFVMS ... but somebody should correct me. I think that is right. I created IPF versions of the PCSI$* files for the kits I put in the above directory. -John wb8tyw@qsl.network Personal Opinion Only ------------------------------ Date: Wed, 23 May 2007 22:20:39 -0400 From: Chip Coldwell Subject: Re: Xpdf 3.02 for OpenVMS/Itanium Message-ID: On Wed, 23 May 2007, Chip Coldwell wrote: > > OK, if anybody is interested I've built a package for Xpdf 3.02/Itanium cue the crickets .... Chip -- Charles M. "Chip" Coldwell "Turn on, log in, tune out" ------------------------------ End of INFO-VAX 2007.283 ************************