INFO-VAX Fri, 25 Apr 2008 Volume 2008 : Issue 231 Contents: Re: can a rms string index be non-case sensitive Re: CPU Cooling Technologies (CoolIT Systems) Re: How to really shutdown TCPIP services ? i64 automatic reboot Re: i64 automatic reboot Re: i64 automatic reboot Re: Intel Itanium RAS Comparison with X86 Re: Intel Itanium RAS Comparison with X86 Re: Intel Itanium RAS Comparison with X86 Re: OT - good restaurants (was:Re: Intel Itanium RAS Comparison with X86) Re: OT: Need HTML Help Re: The MBM> prompt is what?? How do I exit?? Re: The Minimum You Need to Know About Service Oriented Architecture Re: The Minimum You Need to Know About Service Oriented Architecture Time synchronization Re: Time synchronization Re: Time synchronization Re: Time synchronization Re: Time synchronization Re: Time synchronization Re: Time synchronization ---------------------------------------------------------------------- Date: Fri, 25 Apr 2008 12:20:47 +0200 From: "P. Sture" Subject: Re: can a rms string index be non-case sensitive Message-ID: In article <9b616839-cb48-4d09-be63-8363940b6937@k13g2000hse.googlegroups.com>, Hein RMS van den Heuvel wrote: > On Apr 24, 11:48 pm, kczwei wrote: > > Is there a combo of RMS record options that will cause an index of > > string type to be non case sensitive? > > YES, but you would have to set up a collating sequence key. > I don't think I've seen anyone actually do that in the past 15 year. > In the XABKEY (or FDL) you would use DTP = COL or DTP = DCOL > > http://h71000.www7.hp.com/doc/731FINAL/4506/4506pro_006.html#apps_collating_ke > ys_ex > http://h71000.www7.hp.com/doc/731FINAL/4493/4493pro_contents_003.html#toc_chap > ter_16 > http://h71000.www7.hp.com/doc/731FINAL/4493/4493pro_051.html#ncs_util_routines > > The NCS manual has explicit manuals hot to fold uppercase and > lowercase as well as diacriticals > Search for: MY_CONVERSION = MCS_NODIACRITICALS * MCS_LOWER; > in: http://h71000.www7.hp.com/doc/73final/documentation/pdf/OVMS_NCS_UTIL.pdf > > Normally RMS is entirely blind to the characters > it treats each as a simple 8-bit unsigned char. > > What problem are you really trying to solve. > NCS is a somewhat big hammer. > Maybe it is easier to solve at the application level, storing a (copy > of) a key field in uppercase or lowercase. > The option of storing an all uppercase or all lowercase copy of the key is what I did many years ago, even when disk space was quite expensive. Simple to maintain as well. -- Paul Sture Sue's OpenVMS bookmarks: http://eisner.encompasserve.org/~sture/ovms-bookmarks.html ------------------------------ Date: Fri, 25 Apr 2008 09:48:28 -0500 From: David J Dachtera Subject: Re: CPU Cooling Technologies (CoolIT Systems) Message-ID: <4811EF3C.E7AD8FC0@spam.comcast.net> > Does anyone have any experience using CPU cooling technologies? > In your opinion, where do these technologies have real applications? I guess any place they're needed. That said, you'd likely be marketing to sites that self-maintain since modifying a system at that level would likely violate or void any warranty or service agreements. My $0.02... D.J.D. ------------------------------ Date: Fri, 25 Apr 2008 01:50:56 -0700 (PDT) From: etmsreec@yahoo.co.uk Subject: Re: How to really shutdown TCPIP services ? Message-ID: On 23 Apr, 16:58, JF Mezei wrote: > etmsr...@yahoo.co.uk wrote: > > Presumably NFS still had something connected (or at least thought it > > did), maintained the process/devices to keep NFS running so wouldn't > > kill INETACP off until NFS had finished, which it never did. > > NFS was dead. But when it died, it left something in the kernel that > made the kernel continue to think NFS was there. Because of that, NFS > couldn't restart (since port 2046 was already taken by a phantom NFS). > > There was no process to kill. And when shutting down TCPIP, it would > only complain about one BG device which belonged to TCPIP$INETACP ... > > On a serious system, one would expect to be able to zap NFS out of the > kernel without even having to shutdown other TCPIP services. > > On a less serious system, one would expect that just shutting down TCPIP > would work and that it would be able to get rid of all traces of itself > so it could restart cleanly. > > On a windows system, you'd expect to have to reboot to get a clean slate > because it just isn't possible to cleanly shutdown an app withot leaving > garbage around. > > Unfortunatly, VMS acted like a windows system here. I think that's inevitable with the way that TCP/IP integrates with VMS. For example, there's no way of replacing the BG device driver on a live system - the only way to replace the image is to reboot. ------------------------------ Date: Fri, 25 Apr 2008 13:49:37 +0100 From: Anton Shterenlikht Subject: i64 automatic reboot Message-ID: <20080425124937.GA78210@mech-aslap33.men.bris.ac.uk> On Alpha I use SRM variable AUTOACTION = HALT, yet I can force automatic reboot via SYSMAN> shutdown node/automatic_reboot. However, I cannot do this on I64 (rx2620). Even if I issue SYSMAN> shutdown node/automatic_reboot I'm still stopped at the boot menu and have to go via MP each time to reboot my I64. I had a look at VMS FAQ (sec 14.3.5.1 What are I64 IPB boot flag values?) and rx2620 user and install manuals in sections Booting VMS but cannot see any way of setting the boot flags for automatic reboot. Ideally I want something similar to "autoaction=halt" that can be overstepped by e.g. the above sysman command. So I'd like to have halt as a default, but to be able to force automatic reboot when required. many thanks anton -- Anton Shterenlikht Room 2.6, Queen's Building Mech Eng Dept Bristol University University Walk, Bristol BS8 1TR, UK Tel: +44 (0)117 928 8233 Fax: +44 (0)117 929 4423 ------------------------------ Date: Fri, 25 Apr 2008 06:55:17 -0700 (PDT) From: BaxterD@tessco.com Subject: Re: i64 automatic reboot Message-ID: <299f9666-d91e-4698-9a45-697cde72dfcd@p25g2000hsf.googlegroups.com> On Apr 25, 8:49=A0am, Anton Shterenlikht wrote: > On Alpha I use SRM variable AUTOACTION =3D HALT, yet > I can force automatic reboot via > =A0 =A0 =A0 =A0 SYSMAN> shutdown node/automatic_reboot. > > However, I cannot do this on I64 (rx2620). Even if I issue > =A0 =A0 =A0 =A0 SYSMAN> shutdown node/automatic_reboot > I'm still stopped at the boot menu and have to go > via MP each time to reboot my I64. > > I had a look at VMS FAQ (sec 14.3.5.1 What are I64 IPB > boot flag values?) and rx2620 user and install manuals in > sections Booting VMS but cannot see any way of setting the > boot flags for automatic reboot. > > Ideally I want something similar to "autoaction=3Dhalt" > that can be overstepped by e.g. the above sysman > command. So I'd like to have halt as a default, but > to be able to force automatic reboot when required. > > many thanks > anton > > -- > Anton Shterenlikht > Room 2.6, Queen's Building > Mech Eng Dept > Bristol University > University Walk, Bristol BS8 1TR, UK > Tel: +44 (0)117 928 8233 > Fax: +44 (0)117 929 4423 Anton, there is some information regarding this in; http://docs.hp.com/en/BA322-90077/apas05.html Dave ------------------------------ Date: Fri, 25 Apr 2008 17:13:58 +0100 From: Anton Shterenlikht Subject: Re: i64 automatic reboot Message-ID: <20080425161358.GA10108@mech-aslap33.men.bris.ac.uk> On Fri, Apr 25, 2008 at 06:55:17AM -0700, BaxterD@tessco.com wrote: > On Apr 25, 8:49 am, Anton Shterenlikht wrote: > > On Alpha I use SRM variable AUTOACTION = HALT, yet > > I can force automatic reboot via > >         SYSMAN> shutdown node/automatic_reboot. > > > > However, I cannot do this on I64 (rx2620). Even if I issue > >         SYSMAN> shutdown node/automatic_reboot > > I'm still stopped at the boot menu and have to go > > via MP each time to reboot my I64. > > > > I had a look at VMS FAQ (sec 14.3.5.1 What are I64 IPB > > boot flag values?) and rx2620 user and install manuals in > > sections Booting VMS but cannot see any way of setting the > > boot flags for automatic reboot. > > > > Ideally I want something similar to "autoaction=halt" > > that can be overstepped by e.g. the above sysman > > command. So I'd like to have halt as a default, but > > to be able to force automatic reboot when required. > > Anton, > > there is some information regarding this in; > > http://docs.hp.com/en/BA322-90077/apas05.html Dave thanks I actually have this manual in print, forgot to check there. According to the manual, it should auto boot after 10 sec. I guess I never actually gave it 10 sec, will check next time. thank you anton -- Anton Shterenlikht Room 2.6, Queen's Building Mech Eng Dept Bristol University University Walk, Bristol BS8 1TR, UK Tel: +44 (0)117 928 8233 Fax: +44 (0)117 929 4423 ------------------------------ Date: 25 Apr 2008 16:14:27 GMT From: billg999@cs.uofs.edu (Bill Gunshannon) Subject: Re: Intel Itanium RAS Comparison with X86 Message-ID: <67eeb3F2os5aqU2@mid.individual.net> In article , koehler@eisner.nospam.encompasserve.org (Bob Koehler) writes: > In article <4810a6c1$0$12290$c3e8da3@news.astraweb.com>, JF Mezei writes: >> Bob Koehler wrote: >> >>> I believe the only VAX flown on the shuttle used RAM drives. It was >>> derived from a commercially available ruggedized VAX. And yes, its >>> publically known that it was running VMS. >> >> >> It wasn't publically known... "Publically" implies marketing :-) > > It wasn't publicized, which would imply marketing, but it is > publically available information. > >> The laptops on the station are from IBM. The ones used for station >> control run Solaris. The rest run Windows. Vibration/G forces are less >> of an issue on the station, but cosmic radiation is an issue. > > I think the consoles in Houston are still Alphas running Tru64. What, not VMS? What were they thinking? bill -- Bill Gunshannon | de-moc-ra-cy (di mok' ra see) n. Three wolves billg999@cs.scranton.edu | and a sheep voting on what's for dinner. University of Scranton | Scranton, Pennsylvania | #include ------------------------------ Date: Fri, 25 Apr 2008 12:37:47 -0400 From: "FredK" Subject: Re: Intel Itanium RAS Comparison with X86 Message-ID: "Tom Linden" wrote in message news:op.t94zhqmfhv4qyg@murphus.hsd1.ca.comcast.net... > On Wed, 23 Apr 2008 08:02:36 -0700, FredK wrote: > >> >> "John Reagan" wrote in message >> news:fukt68$jic$1@usenet01.boi.hp.com... >>> Bill Gunshannon wrote: >>> >>> >>>> A few years ago is right. Fielded in 1991, approved for production in >>>> 1996, >>>> last system delivered in 2005. A total of 8 systems. There is much >>>> more >>>> information about the airframe than anything else and >>>> no mention of VMS at all. I wonder if it has been ported to Windows >>>> yet? >>>> >>>> >>> >>> A NG employee gave a keynote address at last year's OpenVMS Technical >>> Bootcamp on JSTARS. The non-disclosure prevents me from giving details >>> from the talk (which itself was very high level) plus I'd probably >>> remember incorrectly anyway. >>> >> >> JSTARS has been running on VMS since 1991 and the prototype plane was >> called >> into service during Gulf War I - if you remember the prop behind Gen >> Swartzcropf in the "mother of all retreats" - that was from JSTARS. >> >> Originally on VAX and ruggedized DEC3000 (Flamingo) systems, it migrated >> to >> modified AlphaServer ES40s (form factor modification and power supply). >> Each aircraft (Boeing 707s) contains 20 ES40s. There are 17 production >> aircraft plus backups and test. Plus training and ground stations. >> >> http://www.airforce-technology.com/projects/jstars/ >> >> Provides a high level overview of the E8. As will a Google search on >> "Northrop Grumman JSTARS" >> >> JSTARS remains an active and evolving program. >> >> All of the above can be gleaned from public materials. >> >> > Fred, do you what language they used for their code? One would, > absent any other data, assume Ada, but Northrup Grumman was a > PL/I user. > I do know. I can't tell you. But it isn't PL/1, and your assumption would not be correct. Does that help? ------------------------------ Date: Fri, 25 Apr 2008 19:23:03 +0200 From: "P. Sture" Subject: Re: Intel Itanium RAS Comparison with X86 Message-ID: In article , Sebastian Hanigk wrote: > "P. Sture" writes: > > > Yes, and if you understand it, the German version of "Das Boot" is > > better than the English one. > > Which -- with some notable exceptions (X-Files, anyone?) -- ties quite > nicely into the notion that consuming original material is usually the > better choice :-) > Indeed :-) -- Paul Sture Sue's OpenVMS bookmarks: http://eisner.encompasserve.org/~sture/ovms-bookmarks.html ------------------------------ Date: 25 Apr 2008 13:13:41 -0500 From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) Subject: Re: OT - good restaurants (was:Re: Intel Itanium RAS Comparison with X86) Message-ID: In article <4811150B.8020200@comcast.net>, bradhamilton writes: > > OK - what makes it a "Boston" pecan pie? Made by two unrelated women > who live together? :-) I don't know. Perhaps it had too much bourbon in it for Bourbon Street. ------------------------------ Date: Fri, 25 Apr 2008 09:35:38 -0500 From: David J Dachtera Subject: Re: OT: Need HTML Help Message-ID: <4811EC3A.480D26A1@spam.comcast.net> "Jeffrey H. Coffield" wrote: > > David J Dachtera wrote: > > > I'm looking for a way to do something like this: > > +----------+-------------------------------+ > > | Button 1 | | > > +----------+ | > > | Button 2 | | > > +----------+ | > > | Button 3 | | > > +----------+ | > > | Button 4 | | > > +----------+ | > > | Button 5 | | > > +----------+-------------------------------+ > > > See attached: > > Jeff C. > > --------------------------------------------------------------- > > Thanx, Jeff! Really pointed me in the right direction. For what I did with it, see my wife's up and coming website: http://www.fourpawsandfur.com/ and http://www.fourpawsandfur.com/index_live.html for the full site (not yet complete - the "home" links will take you back to the "under construction" page which has no links; so, use the browser's BACK button as needed). David J Dachtera (formerly dba) DJE Systems ------------------------------ Date: Fri, 25 Apr 2008 09:45:58 -0500 From: David J Dachtera Subject: Re: The MBM> prompt is what?? How do I exit?? Message-ID: <4811EEA6.9E341302@spam.comcast.net> Jeff Goodwin wrote: > > "George Cornelius" wrote in message > news:yMhGb59j5LCF@eisner.encompasserve.org... > > In article <1otP1vNyPmKg@eisner.encompasserve.org>, > > BEGINcornelius@decuserve.orgEND (George Cornelius) writes: > >> I guess I know that the MBM> prompt is some hardware level management > >> tool, > >> and I know that you about have to stand on your head to get into it. > >> > >> The problem is, I have a serious need to get into the system console and > >> another system administrator left this blasted thing running. How do I > >> get out of it?? > > > > > > To answer my own question: > > > > I don't _want_ to get out of it. It apparently _is_ the console > > subsystem on my ES47. > > > > I typed connect and wound up on the VMS console. > > > > Of course, I lost two days' of console messages, but what the heck... > > > > Nobody should even need console messages these days, right? > > > > -- > > George Cornelius cornelius A T eisner D O T decus D O T org > > Setting SRM_AUTO_CONNECT to ON should automatically connect you to the VMS > console on boot once the diagnostics are finished. Not sure I'd want to recommend that, though. On my primary site (GS1280s), Oracle spits out so many spurious messages, that I've found it useful to leave the MBM on the serial port, and then Telnet into the thing again and CONNECT to the SRM console from there. Before I did that, we were getting hundreds of device timeouts a day on the serial console port (OPA0:) because 9600 baud just isn't fast enough to keep up with all the stuff that Oracle spits out via OPCOM, on top of all the other messages. David J Dachtera (formerly dba) DJE Systems ------------------------------ Date: Fri, 25 Apr 2008 09:21:46 -0500 From: David J Dachtera Subject: Re: The Minimum You Need to Know About Service Oriented Architecture Message-ID: <4811E8FA.6490BF5F@spam.comcast.net> JF Mezei wrote: > > David J Dachtera wrote: > > > Seems a shame to "hide the lamp under a bushel" due to some "religious" > > thing. VMS needs all the exposure it can get, even from the "sinners". > > You don't own VMS. HP doesn't even own VMS. It owns something called > OpenVMS, and it is HP's job to give its own products all the exposure HO > wants this product to get. Well, while you underscored my point in one way, you missed it in another. "yyyc186" has some, I can only call it, "religious" aversion to Amazon, even though it's -THE- biggest, most popular book buying site on the web. (Grudges are fine as far as they go if you're into that sort of thing, but they're not very profitable.) It's rather like boycotting television or "Hollywood" movies due to the corruption, drugs, sex, rock-and-roll, etc. which characterize such things. For example, I cannot condone rap or a lot of other genres; however, I would never consider instructing my marketing people to not buy ad time/space in those media markets if I had a product that needed maximum exposure. For another example, I don't particularly appreciate PayPal's commission structures or eBay's insertion fees and commissions. Given the alternatives, however, I don't rule them out. David J Dachtera (formerly dba) DJE Systems ------------------------------ Date: 25 Apr 2008 16:10:03 GMT From: billg999@cs.uofs.edu (Bill Gunshannon) Subject: Re: The Minimum You Need to Know About Service Oriented Architecture Message-ID: <67ee2rF2os5aqU1@mid.individual.net> In article <4811E8FA.6490BF5F@spam.comcast.net>, David J Dachtera writes: > JF Mezei wrote: >> >> David J Dachtera wrote: >> >> > Seems a shame to "hide the lamp under a bushel" due to some "religious" >> > thing. VMS needs all the exposure it can get, even from the "sinners". >> >> You don't own VMS. HP doesn't even own VMS. It owns something called >> OpenVMS, and it is HP's job to give its own products all the exposure HO >> wants this product to get. > > Well, while you underscored my point in one way, you missed it in > another. > > "yyyc186" has some, I can only call it, "religious" aversion to Amazon, > even though it's -THE- biggest, most popular book buying site on the > web. (Grudges are fine as far as they go if you're into that sort of > thing, but they're not very profitable.) > > It's rather like boycotting television or "Hollywood" movies due to the > corruption, drugs, sex, rock-and-roll, etc. which characterize such > things. > > For example, I cannot condone rap or a lot of other genres; however, I > would never consider instructing my marketing people to not buy ad > time/space in those media markets if I had a product that needed maximum > exposure. > > For another example, I don't particularly appreciate PayPal's commission > structures or eBay's insertion fees and commissions. Given the > alternatives, however, I don't rule them out. > Well, having been openly cheated by Amazon and having to spend a good deal of my time and effort getting my money back for a product billed but never delivered (and not deliverable as they had no copies and it was out of print, something I would expect a book seller to understand) I wouldn't bet that his aversion to Amazon is necessarily "religious". I no longer deal with Amazon at all. Sadly, this seems to be the nature of INTERNET business. Just another play on the old "bait and switch". they establish themselves living on the INTERNET hype and then turn into just another rip-off operation. Which is why I also don't do E-bay. bill -- Bill Gunshannon | de-moc-ra-cy (di mok' ra see) n. Three wolves billg999@cs.scranton.edu | and a sheep voting on what's for dinner. University of Scranton | Scranton, Pennsylvania | #include ------------------------------ Date: Fri, 25 Apr 2008 15:46:57 +0200 From: Marc Van Dyck Subject: Time synchronization Message-ID: What is the most sensible way to synchronize the time in a mid-size OpenVMS farm today ? - Use NTP everywhere and forget the rest ? - Use DTSS with an NTP time provider ? - Synchronize time on internet time server ? - Use a radio time source ? - ... The installation I speak about is made of about 20 systems, some clustered, some not. Mix Alpha/Itanium. And surrounded by a much larger population of Unix and Windows systems. What are other people over here doing ? -- Marc Van Dyck ------------------------------ Date: Fri, 25 Apr 2008 10:20:37 -0400 From: "Richard B. Gilbert" Subject: Re: Time synchronization Message-ID: Marc Van Dyck wrote: > What is the most sensible way to synchronize the time in a mid-size > OpenVMS farm today ? > > - Use NTP everywhere and forget the rest ? > - Use DTSS with an NTP time provider ? > - Synchronize time on internet time server ? > - Use a radio time source ? > - ... > > The installation I speak about is made of about 20 systems, some > clustered, some not. Mix Alpha/Itanium. And surrounded by a much > larger population of Unix and Windows systems. > > What are other people over here doing ? > A GREAT DEAL depends on your requirements! Do you want all you systems to keep UTC, correct within a millisecond or two? Or are you happy if they all have the SAME time even if the time is in error by several minutes? I use NTP on both my VMS and my Solaris systems. It seems to work well enough. On VMS it WILL bury you in log files that usually tell you nothing worth the disk space they occupy. You will probably want/need to have a weekly or monthly cleanup job. ------------------------------ Date: Fri, 25 Apr 2008 09:59:06 -0500 From: David J Dachtera Subject: Re: Time synchronization Message-ID: <4811F1BA.61FA0F3C@spam.comcast.net> Marc Van Dyck wrote: > > What is the most sensible way to synchronize the time in a mid-size > OpenVMS farm today ? > > - Use NTP everywhere and forget the rest ? > - Use DTSS with an NTP time provider ? > - Synchronize time on internet time server ? > - Use a radio time source ? > - ... > > The installation I speak about is made of about 20 systems, some > clustered, some not. Mix Alpha/Itanium. And surrounded by a much > larger population of Unix and Windows systems. > > What are other people over here doing ? We use UCX's NTP and sync up to an external source. Typically, one node of a cluster is a stratum 2 time source, and the others sync to it at stratum 3 or higher. Stand-alones are usually stratum 3. There's been talk internally of limiting the number of stratum 2 systems inside the firewall and having all the others be stratum 3 or higher. No positive action is imminent, AFAICT. David J Dachtera (formerly dba) DJE Systems ------------------------------ Date: Fri, 25 Apr 2008 10:38:53 -0500 (CDT) From: sms@antinode.org (Steven M. Schweda) Subject: Re: Time synchronization Message-ID: <08042510385310_2020CE0A@antinode.org> From: Marc Van Dyck > What is the most sensible way to synchronize the time in a mid-size > OpenVMS farm today ? > > - Use NTP everywhere and forget the rest ? > - Use DTSS with an NTP time provider ? > - Synchronize time on internet time server ? > - Use a radio time source ? > - ... Probably "Use NTP everywhere and forget the rest". I "Use DTSS with an NTP time provider", mostly because of inertia. I have a bunch of old VAX systems (with multiple system disks) which have been using DTSS for many years, and I've been too lazy to reconfigure all of them to use NTP. ------------------------------------------------------------------------ Steven M. Schweda sms@antinode-org 382 South Warwick Street (+1) 651-699-9818 Saint Paul MN 55105-2547 ------------------------------ Date: Fri, 25 Apr 2008 16:03:39 +0000 (UTC) From: gartmann@nonsense.immunbio.mpg.de (Christoph Gartmann) Subject: Re: Time synchronization Message-ID: In article , Marc Van Dyck writes: >What is the most sensible way to synchronize the time in a mid-size >OpenVMS farm today ? > >- Use NTP everywhere and forget the rest ? >- Use DTSS with an NTP time provider ? >- Synchronize time on internet time server ? >- Use a radio time source ? >- ... > >The installation I speak about is made of about 20 systems, some >clustered, some not. Mix Alpha/Itanium. And surrounded by a much >larger population of Unix and Windows systems. > >What are other people over here doing ? Our central router uses NTP from a primary source. All other hosts use NTP to get the time from this router. Regards, Christoph Gartmann -- Max-Planck-Institut fuer Phone : +49-761-5108-464 Fax: -452 Immunbiologie Postfach 1169 Internet: gartmann@immunbio dot mpg dot de D-79011 Freiburg, Germany http://www.immunbio.mpg.de/home/menue.html ------------------------------ Date: Fri, 25 Apr 2008 16:35:38 +0000 (UTC) From: moroney@world.std.spaamtrap.com (Michael Moroney) Subject: Re: Time synchronization Message-ID: Marc Van Dyck writes: >What is the most sensible way to synchronize the time in a mid-size >OpenVMS farm today ? >- Use NTP everywhere and forget the rest ? >- Use DTSS with an NTP time provider ? >- Synchronize time on internet time server ? >- Use a radio time source ? >- ... >The installation I speak about is made of about 20 systems, some >clustered, some not. Mix Alpha/Itanium. And surrounded by a much >larger population of Unix and Windows systems. >What are other people over here doing ? I recently set up a cluster to use a router as an NTP time source. The router gets it from a radio time source. I also configured NTP for each node to treat the others as peers so if the time source is lost, they'll at least track each other. ------------------------------ Date: 25 Apr 2008 13:17:31 -0500 From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) Subject: Re: Time synchronization Message-ID: <2jNZ$lNiYzMf@eisner.encompasserve.org> In article , Marc Van Dyck writes: > > The installation I speak about is made of about 20 systems, some > clustered, some not. Mix Alpha/Itanium. And surrounded by a much > larger population of Unix and Windows systems. > > What are other people over here doing ? I find NTP is good enough for everything I need. Back when I didn't have it I did do a cluster set time via SYSMAN once a week because some of the nodes were slipping apart far enough to confuse MMS. Then I just adjusted the time when I felt like it, VAXen are pretty good at keeping time, especially when you get a lot of them together. ------------------------------ End of INFO-VAX 2008.231 ************************