1 INFO-VAX	Fri, 12 Sep 2003	Volume 2003 : Issue 506       Contents:+ Re: "HP breakup on the way" - Merrill Lynch + Re: "HP breakup on the way" - Merrill Lynch + Re: "HP breakup on the way" - Merrill Lynch + Re: "HP breakup on the way" - Merrill Lynch + Re: "HP breakup on the way" - Merrill Lynch + Re: "HP breakup on the way" - Merrill Lynch + Re: "HP breakup on the way" - Merrill Lynch + Re: "HP breakup on the way" - Merrill Lynch + Re: "HP breakup on the way" - Merrill Lynch + Re: "HP breakup on the way" - Merrill Lynch + Re: "HP breakup on the way" - Merrill Lynch  Anonymous FTP configuration @ RE: Anyone else having problems with VMS and HSG80-based arrays?@ Re: Anyone else having problems with VMS and HSG80-based arrays?@ Re: Anyone else having problems with VMS and HSG80-based arrays?@ Re: Anyone else having problems with VMS and HSG80-based arrays?@ Re: Anyone else having problems with VMS and HSG80-based arrays?@ Re: Anyone else having problems with VMS and HSG80-based arrays?@ Re: Anyone else having problems with VMS and HSG80-based arrays?0 Re: First VMS IPF difference... not very nice...0 Re: First VMS IPF difference... not very nice...0 Re: First VMS IPF difference... not very nice...0 Re: First VMS IPF difference... not very nice...P I have a fix!!!  Re: Anyone else having problems with VMS and HSG80-based arraysP re: I have a fix!!!  Re: Anyone else having problems with VMS and HSG80-based ar( Re: IDL on VMS. Was: Image tools for VMS( Re: IDL on VMS. Was: Image tools for VMS8 Re: Image tools for VMS - follow up - image file formats& Re: Mac OS X -> Pathworks/Mac problems Re: Mailboxes and SYS$OUTPUT Re: MESSAGE problem  Re: MESSAGE problem  Re: MESSAGE problem  RE: MESSAGE problem  RE: MESSAGE problem  RE: MESSAGE problem  RE: MESSAGE problem  Newbie question about NTP  Re: Newbie question about NTP  RE: Newbie question about NTP  Re: Newbie question about NTP  Re: Newbie question about NTP  Re: Newbie question about NTP  Re: Newbie question about NTP  Re: Newbie question about NTP  RE: Newbie question about NTP  RE: Newbie question about NTP  Re: Newbie question about NTP  Re: Newbie question about NTP  Re: Newbie question about NTP  RE: Newbie question about NTP  RE: Newbie question about NTP  Re: Newbie question about NTP  Re: Newbie question about NTP C Re: OpenVMS clusters (Alpha) - standalone workstation joined to the # Re: Question on running GnuPG 1.2.0 B Reminder, the OpenVMS Advanced Technical Boot Camp is now 50% full= Re: Runtime linking of Java to AWT graphics libraries problem 3 TCPIP 5.1 SMTP alternate gateway on different port?  Re: VMS website updated. Re: VMS website updated. Re: VMS website updated. Re: VMS website updated.2 Re: Why does MAIL DIR go slow during mail receive?2 Re: Why does MAIL DIR go slow during mail receive?  F ----------------------------------------------------------------------  % Date: Fri, 12 Sep 2003 09:46:07 +0100 O From: Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com> 4 Subject: Re: "HP breakup on the way" - Merrill Lynch0 Message-ID: <bjs14g$9pj$1@new-usenet.uk.sun.com>   Ken Farmer wrote: - > Re: "HP breakup on the way" - Merrill Lynch  > : > Yea, they know how to invest peoples money also...wrong. > M > Any comments on VeriSign booting Sun out of the Internic for Red Hat Linux?  > That has to hurt.  > 5 > Red Hat Providing VeriSign With Linux-Based Servers C > http://www.enterpriselinux.org/stories.php?story=03/09/10/1221561  >  >  > -- > Kenneth Farmer  <><  > SpyderByte.com >   : If you can find a reference to Sun in the article then you win a prize.  ( As it is you get a resounding raspberry.  ; Verisign use Sun, HP and IBM, the article doesn't say which : vendors platforms are being replace by Linux it could well
 be all three.   = I also cannot quite see why this would be a cause for crowing  on your part anyway.  @ Any migration from RISC-UNIX to Linux-IA32 only serves to remove> a potential customer from the division of HP that can possiblyD post merger say 2006 make money (Enterprise Systems) to the one that demonstrably can't (PCs).   B I just hope for Verisigns sake that they have managed to negociate? a very very substantial discount off the RetHat AS subscription C service ~90% would be good otherwise they are looking at a solution 5 thats radically more expensive than the alternatives.   D The economics of Linux are interesting and complicated but one thingB you can guarantee is that paid for Linux is very very difficult to justify from a cost standpoint.    Regards  Andrew Harrison    ------------------------------  % Date: Fri, 12 Sep 2003 10:43:20 +0100 O From: Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com> 4 Subject: Re: "HP breakup on the way" - Merrill Lynch0 Message-ID: <bjs4fp$at7$1@new-usenet.uk.sun.com>   Rick Jones wrote: R > Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com> wrote: > ? >>The bug is a problem for non IP network traffic that does not  >>do checksumming on blocks. >  > H > I take it then the data corruption happens after the GbE chip does theD > Internet checksum offload?  Or is Sun not taking advantage of that$ > feature of the Broadcom GbE chips? >  > 7 >>On an affected V240 (not all the servers manufactured < >>use the batch of NICs with the manufacturing defect) using7 >>IP this results in ~0.3% retransmission rate which is  >>pretty much background noise.  >  > E > Yes and no.  For bulk data transfer with a "large enough" window to E > still get fast RTX and such, 0.3% is not going to bring things to a F > screaching halt.  However, for stuff where the windows are not largeH > enough to get fast RTX, or the connection latency is large (high-speed@ > WAN where you need to keep cwnd large) then 0.3% can indeed be
 > plusungood.  >   ? 0.3% is the theoretical maximum. Sun shipped a resonable number B of V210/V240's before the NIC manufacturing defect was discovered.  A We didn't to my knowledge have any reports of network performance C issues, the issue with the defect was triggered by a customer using D a non IP protocol. I am working with an ex OpenVMS exchange customer@ who ran a series of network performance tests on V240's with theC defect for their WAN connected gateway servers they didn't find any B anomilies and were rather disgruntled when we refused to ship themB any more systems after we discovered the defect because it clearly didn't impact them.   C Their application is a market data type gateway, by nature it isn't  doing bulk transfer.  A When the issue was discovered Sun stopped all V210/V240 deliverys ? and we are replacing all the affected V210/240 motherboards out 
 in the field.   ? Ironically your estimates of the theoretical effects 0.3% error @ rate in Sun's NIC are rather similar to the actual effect of the7 remedy offered by Intel to Itanium 2 customers with the : most recent Itanium bug, clock back from 1 Ghz to 800 Mhz.  / http://www.computerweekly.com/Article121736.htm   2 IBM like Sun with the V210/V240 felt that this was1 unnaceptable and stopped shipping the x450 server ( other vendors didn't take the same view.   Regards  Andrew Harrison E > Some old data showing the effect on bulk transfer of having to wait 8 > for RTX timeouts at various RTX rates can be found at: > ? > ftp://ftp.cup.hp.com/dist/networking/briefs/SE_372_2_99.ps.gz  > > > Which showed that for a 0.01% loss rate the timeout-only TCPG > connection lost 20% performance, going to a loss of about 65% at 0.1%  > and 95% by 1%. > B > When the connection was able to fast RTX, a 0.1% throughput held< > steady (with a 32KB window), and was down about 55% at 1%. > F > Also, while 0.3% might be background noise for an Internet connectedF > server, for something in the department or intranet that is going to& > be quite large comapred with normal. > E > Lets assume we have a synchronous request/reply application going - F > say very much like single instance of netperf TCP_RR :).  A UP J5000G > (440 MHz PA-8500) can do roughly 13100 single-byte TCP_RR's through a H > J6825A GbE NIC (Broadcom like the V2[14]0, but 5701 rather than 5704).8 > That suggests an RTT of 1/13000 or ~0.08 milliseconds. > G > If there were a 0.3% retransmission rate, then that RTT would become:  > ! >   (.997 * 0.08) + (.003 * 500)   > E > or ~1.57 milliseconds per transaction, which would be a transaction   > rate of ~635 or a loss of 95%. > H > (If one is using the typical tcp_rexmit_interval_min. I'm ass-u-me-ingC > that Solaris doesn't use the 1 second default the RFC's suggest - # > HP-UX uses 500 ms as the default)  > E > Now, it is perhaps unreasonable to expect real applications to have E > 0.08 ms response time, and I've no idea if a V240 can achieve 13000 G > TCP_RR transactions per second through one of its GbE interfaces with D > a single instance of netperf, packet losses or no, so lets insteadB > assume that the application RTT is 1 millisecond when the NIC is> > functioning correctly.  That suggests it would be doing 1000D > transactions per second without loss.  With that 0.3% loss rate we > instead have >  >   (.997 * 1) + (.003 * 500)  > G > or 2.497 ms response time, which means a transaction rate of ~400 per & > second or a 60% loss of performance. >  > 9 >>Of course if you are using a non IP based protocol with , >>no checksumming then 0.3% is unacceptable. >  > E > Not all IP-based protocols do retransmissions - UDP leaves it up to F > the application.  NFS, DNS etc would have to do their retransmission7 > things, and both are rather request/reply in nature.   >  > rick jones   ------------------------------   Date: 12 Sep 2003 12:53:22 GMT( From: bill@cs.uofs.edu (Bill Gunshannon)4 Subject: Re: "HP breakup on the way" - Merrill Lynch9 Message-ID: <bjsfk1$mo8ea$1@ID-135708.news.uni-berlin.de>   3 In article <adj8QgBMN5pi@eisner.encompasserve.org>, > 	koehler@eisner.nospam.encompasserve.org (Bob Koehler) writes:j > In article <bjqhni$l7pq2$1@ID-135708.news.uni-berlin.de>, bill@gw5.cs.uofs.edu (Bill Gunshannon) writes: >>  A >> Apology accepted.  :-)  Some of us still take our PDP-11s very D >> seriously.  And we commiserate with the former TOPS-20 users. :-( > A >    Some of us still have out PDP-11s, unfortunately RSX doesn't  >    run on a Pro350.  >   1 Hmmmm.....   I thought POS was an RSX derivative.    bill   --  J Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>       ------------------------------    Date: 12 Sep 2003 07:37:20 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) 4 Subject: Re: "HP breakup on the way" - Merrill Lynch3 Message-ID: <adj8QgBMN5pi@eisner.encompasserve.org>   h In article <bjqhni$l7pq2$1@ID-135708.news.uni-berlin.de>, bill@gw5.cs.uofs.edu (Bill Gunshannon) writes: >   @ > Apology accepted.  :-)  Some of us still take our PDP-11s veryC > seriously.  And we commiserate with the former TOPS-20 users. :-(   ?    Some of us still have out PDP-11s, unfortunately RSX doesn't     run on a Pro350.    ------------------------------    Date: 12 Sep 2003 07:38:39 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) 4 Subject: Re: "HP breakup on the way" - Merrill Lynch3 Message-ID: <kR$VOvzUosg+@eisner.encompasserve.org>    In article <bjs14g$9pj$1@new-usenet.uk.sun.com>, Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com> writes:  B > Any migration from RISC-UNIX to Linux-IA32 only serves to remove@ > a potential customer from the division of HP that can possiblyF > post merger say 2006 make money (Enterprise Systems) to the one that > demonstrably can't (PCs).   F    Nope.  IBM has shown quite clearly that there is a market for Linux    on enterprise class servers.    ------------------------------  # Date: Fri, 12 Sep 2003 13:55:43 GMT 2 From: "Ken Farmer" <KFarmer@NOSPAM.SpyderByte.com>4 Subject: Re: "HP breakup on the way" - Merrill Lynch6 Message-ID: <zfk8b.58$XA5.28@twister.southeast.rr.com>  K "Andrew Harrison SUNUK Consultancy" <Andrew_No.Harrison_No@nospamn.sun.com> ; wrote in message news:bjs14g$9pj$1@new-usenet.uk.sun.com...  > Ken Farmer wrote: H > > Any comments on VeriSign booting Sun out of the Internic for Red Hat Linux? > > That has to hurt.  > > 7 > > Red Hat Providing VeriSign With Linux-Based Servers E > > http://www.enterpriselinux.org/stories.php?story=03/09/10/1221561  >> > < > If you can find a reference to Sun in the article then you > win a prize.  L There is no reference to Sun.  I do know that when I first started an ISP inL this local area, Internic's DNS was all Sun.  I even spoke with them back inH 93 about the best servers to run to get started.  Sun, Sun, Sun is all IL heard.  That was my first venture into Unix.  Bought a couple of Sparc5s andH we were off to the races.  I actually still have that first Sparc5 and aL dual Sparc20.  I loved those machines.  Worked several Solaris sysadmin jobs3 for years after that using many larger Sun servers.   ? Of course that all changed when I meet my first Alpha.  Smokin!   * > As it is you get a resounding raspberry.   :)  = > Verisign use Sun, HP and IBM, the article doesn't say which < > vendors platforms are being replace by Linux it could well > be all three.   I An article elsewhere did mention there was a mixture.  It didn't speak of " the root DNS servers specifically.  ? > I also cannot quite see why this would be a cause for crowing  > on your part anyway.  C Crowing?  I'm not a one trick pony dude, I'm a fan of many OSes and : platforms.  I wanted to see if you knew anything about it.  7 Poor thing.  Kinda feel like everyone's out to get you?      -- Kenneth Farmer  <><  SpyderByte.com    EnterpriseUnix.org  |  Tru64.org OpenVMS.org  |  dcl.OpenVMS.org $ EnterpriseLinux.org  |  LinuxHPC.org   ------------------------------  % Date: Fri, 12 Sep 2003 15:59:14 +0100 O From: Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com> 4 Subject: Re: "HP breakup on the way" - Merrill Lynch0 Message-ID: <bjsn02$71j$1@new-usenet.uk.sun.com>   Bob Koehler wrote: > In article <bjs14g$9pj$1@new-usenet.uk.sun.com>, Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com> writes: >  > B >>Any migration from RISC-UNIX to Linux-IA32 only serves to remove@ >>a potential customer from the division of HP that can possiblyF >>post merger say 2006 make money (Enterprise Systems) to the one that >>demonstrably can't (PCs).  >  > H >    Nope.  IBM has shown quite clearly that there is a market for Linux! >    on enterprise class servers.  >     6 Ahh the de-commoditise something whose major advantage8 if any is that is based on a commodity platform aproach.  3 The economics of Linux on x86 are reasonably dodgy, 4 putting it on Power or something even more daft like, zSeries only serves to compound the problem.  5 Plus there is an extreme absence of any ISV's (except . IBM) who are prepared to support the platform.   Regards  Andrew Harrison    ------------------------------    Date: 12 Sep 2003 09:30:05 -07001 From: keithparris_NOSPAM@yahoo.com (Keith Parris) 4 Subject: Re: "HP breakup on the way" - Merrill Lynch= Message-ID: <cf15391e.0309120830.620c7182@posting.google.com>    Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com> wrote in message news:<bjni1b$ipu$1@new-usenet.uk.sun.com>... : > Losing market share in all their markets as HP have done  D Given that you recklessly used the word "all" in your hyperbole, all3 it takes is one counter-example to prove you wrong:   E "IDC reports that HP increased its market leadership in notebooks and = accounted for 17.3 percent of the unit market share in Q2. In ; addition, HP increased its notebook unit sales 48.1 percent E year-over-year while its notebook unit growth more than doubled (22.4 ) percent) the overall market unit growth."   9 http://www.hp.com/hpinfo/newsroom/press/2003/030902a.html    ------------------------------    Date: 12 Sep 2003 12:08:35 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) 4 Subject: Re: "HP breakup on the way" - Merrill Lynch3 Message-ID: <TNTDAjb1xytP@eisner.encompasserve.org>   d In article <bjsfk1$mo8ea$1@ID-135708.news.uni-berlin.de>, bill@cs.uofs.edu (Bill Gunshannon) writes:5 > In article <adj8QgBMN5pi@eisner.encompasserve.org>, @ > 	koehler@eisner.nospam.encompasserve.org (Bob Koehler) writes:k >> In article <bjqhni$l7pq2$1@ID-135708.news.uni-berlin.de>, bill@gw5.cs.uofs.edu (Bill Gunshannon) writes:  >>>   B >>> Apology accepted.  :-)  Some of us still take our PDP-11s veryE >>> seriously.  And we commiserate with the former TOPS-20 users. :-(  >>  B >>    Some of us still have out PDP-11s, unfortunately RSX doesn't >>    run on a Pro350. >>   > 3 > Hmmmm.....   I thought POS was an RSX derivative.   $    Yep, but it's not the Real Thing.   ------------------------------    Date: 12 Sep 2003 12:11:24 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) 4 Subject: Re: "HP breakup on the way" - Merrill Lynch3 Message-ID: <2HHqHSuT5+MT@eisner.encompasserve.org>    In article <bjsn02$71j$1@new-usenet.uk.sun.com>, Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com> writes: > 5 > The economics of Linux on x86 are reasonably dodgy, 6 > putting it on Power or something even more daft like. > zSeries only serves to compound the problem.  F    IBM disagrees.  Who's business judgement do you think I value more?:    IBM knows how to make money.  You know how to make FUD.   ------------------------------  % Date: Fri, 12 Sep 2003 18:16:23 +0100 U From: Andrew Harrison SUNUK Consultancy <andrew_nospam.harrison_remove_this@sun#.com> 4 Subject: Re: "HP breakup on the way" - Merrill Lynch0 Message-ID: <bjsv17$9ke$1@new-usenet.uk.sun.com>   Keith Parris wrote:  > Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com> wrote in message news:<bjni1b$ipu$1@new-usenet.uk.sun.com>...  > : >>Losing market share in all their markets as HP have done >  > F > Given that you recklessly used the word "all" in your hyperbole, all5 > it takes is one counter-example to prove you wrong:  > G > "IDC reports that HP increased its market leadership in notebooks and ? > accounted for 17.3 percent of the unit market share in Q2. In = > addition, HP increased its notebook unit sales 48.1 percent G > year-over-year while its notebook unit growth more than doubled (22.4 + > percent) the overall market unit growth."  > ; > http://www.hp.com/hpinfo/newsroom/press/2003/030902a.html     4 Ohh goody notebooks I guess the board can go back to0 the shareholders and say that it has in fact all0 turned out OK because although HP's market share6 in every other market has declined they have increased# their share of the notebook market.   # And do HP make money on notebooks ?    Answers on a post card   Regards  Andrew Harrison    ------------------------------   Date: 12 Sep 2003 16:23:54 GMT6 From: DAVISM@er6.eng.ohio-state.edu (Michael T. Davis)$ Subject: Anonymous FTP configuration: Message-ID: <bjsruq$9e9$1@charm.magnus.acs.ohio-state.edu>   	We're running...   ?   Compaq TCP/IP Services for OpenVMS Alpha Version V5.1 - ECO 3 3   on a AlphaServer 800 5/500 running OpenVMS V7.2-1   D 	Should TCPIP$FTP_ANONYMOUS_DIRECTORY be defined as a search list orM a comma-separated string?  Can this logical name be defined to provide access L to locations on the server _in_addition_to_ a login dev./dir. configured forG the ANONYMOUS account (in SYSUAF), or are these two mechanisms mutually H exclusive?  I believe that if the ANONYMOUS account is configured with aG bogus login dev./dir., the FTP server will look ONLY to the equivalence I name(s) for TCPIP$FTP_ANONYMOUS_DIRECTORY as viable navigation locations, L right?  If that's the case, and an anonymous FTP user connects to the "root"M of our system via anonymous FTP, will (s)he be presented with the equivalence N name(s) of the TCPIP$FTP_ANONYMOUS_DIRECTORY logical name to which (s)he could then navigate?  A 	For example, let's say ANONYMOUS has a login location defined as E "NL:".  With TCPIP$FTP_ANONYMOUS_DIRECTORY defined as "DIR1,DIR2" (or F should that be "DIR1","DIR2"?), what will the anonymous FTP client seeI when (s)he navigates to "ftp://<our-server's-FQDN>/" with a commonly-used  Web browser (e.g. IE)?   Thanks,  Mike --K              Michael T. Davis              |    Systems Specialist: ChE,MSE N   E-mail: davism@er6.eng.ohio-state.edu    | Departmental Networking/ComputingJ            -or- DAVISM+@osu.edu            |     The Ohio State UniversityJ http://www.er6.eng.ohio-state.edu/~davism/ |     197 Watts, (614) 292-6928   ------------------------------  $ Date: Fri, 12 Sep 2003 8:51:36 -0400 From: koskaj@bender.com I Subject: RE: Anyone else having problems with VMS and HSG80-based arrays? ; Message-ID: <03091208513602.139.10330689@alaxp3.bender.com>    comments in-line  6 >From:	SMTP%"svieth@wi.rr.com" 11-SEP-2003 14:51:40.72 >To:	Info-VAX@Mvb.Saic.Com >CC:	 C >Subj:	Anyone else having problems with VMS and HSG80-based arrays?  >  >Hi: > E >I've got an ES40 running VMS 7.3-1 which is connected to an ESA12000  >(running ACS 8.7S-2).   I'm running 8.7F.    <snip> .  .  .  > D >environment from tape.  The odd thing that I noticed is that when IB >tried a "restart this" on one of the HSG80s, the CLI hung on bothF >controllers.  I had to have an operator hit the buttons on the OCP on) >the front of the HSG80s to restart them.   I I've seen this problem when cache module is flaky. When the cache module  H (not the DIMMs) finally failed, the redundant controller handled things L fine.  Until till then, it behave like you say or very close to it.  Volume P shadowing between ESA12000 cabinets saved us during this flaky hardware problem.   <snip> .  .  . G >The only hunch we have about this bug in the ESA12000 is that it seems = >to be related to periods of high I/O activity.  On all three F >occassions, we were hit during the time that our backups run.  We are) >also using controller-based snapshots...  >   I If you don't I would make sure you are using mirror cache option between  K dual redundant HSG-80 controllers.  Also, have you change default settings  K of the HSG-80's to retain cache for longer periods?  If so, you might want  1 to return to defaults to see if problem persists.    :) jck Koskaj@hatespam.bender.com   >-Sleepless Scott in Milwaukee.  >svieth@wi.rr.nospam.com   ------------------------------    Date: 12 Sep 2003 07:58:09 -0700$ From: svieth@wi.rr.com (Scott Vieth)I Subject: Re: Anyone else having problems with VMS and HSG80-based arrays? = Message-ID: <5a85bce2.0309120658.34955dfe@posting.google.com>   f young_r@encompasserve.org (Rob Young) wrote in message news:<PEbWgoyWsM7q@eisner.encompasserve.org>...  C > > AND THIS MORNING, we got hit again.  Same symptoms.  Two DGAnn: G > > devices went mntverifytimeout.  Had to reboot without shutting down I > > Cache.  Got lucky once more and the Cache data files were not corrupt J > > after reboot.  Tried a "restart other" on the HSG80s.  The CLI on bothJ > > controllers hung.  Had to walk an operator through hitting the buttonsJ > > on the front of the HSG80s to restart them.  Does anyone see a pattern	 > > here?  > >  >  > 	Yes I see a pattern.    > C > 	Why July 2nd?  Did you upgrade firmware July 1st?  What changed?   B No firmware upgrade or any other change on July 1st.  We make it aE policy to NOT make systems changes within five days of the end of the C month.  July 1st was when we started the month-end batch processing @ for June.  On the night of July 1st, we had three Backup streamsB running as well as our month-end processing.  The HSG80s locked-up2 during the month-end and made quite a mess for us.   > Q >         Seems like the heavy IO from concurrent operations, controller copies,  G > 	backups and night jobs is stretching the HSG80s to a breaking point.  > 	 > 				Rob   D That is my suspicion as well: we are overwhelming the array with I/O! at night while backup is running.    -Scott   ------------------------------    Date: 12 Sep 2003 08:01:06 -0700$ From: svieth@wi.rr.com (Scott Vieth)I Subject: Re: Anyone else having problems with VMS and HSG80-based arrays? = Message-ID: <5a85bce2.0309120701.5eb12f06@posting.google.com>   s al5vf03p02@sneakemail.com (William Webb) wrote in message news:<d5ce4b06.0309111322.4a54edcc@posting.google.com>...  >  > Are you using XFC or VIOC?   > @ > We decided to stay with VIOC 'cuz we suspect it's more stable. >   D Hoff told me that it was okay to go ahead and use the XFC.  It kicksE ass when compared to the VIOC.  If you are running 7.3-1 and have all D your patches in place, then turn on the XFC.  I have the cache limitE set to 12GB.  I am going to increase it this weekend.  We are getting D pretty respectible cache hit numbers with 12 gigs but I think we can
 do better.  D The fastest disk I/O is the one that doesn't have to go to disk. :^)   -Scott   ------------------------------    Date: 12 Sep 2003 08:03:47 -0700$ From: svieth@wi.rr.com (Scott Vieth)I Subject: Re: Anyone else having problems with VMS and HSG80-based arrays? = Message-ID: <5a85bce2.0309120703.15fb6cf4@posting.google.com>   v keithparris_NOSPAM@yahoo.com (Keith Parris) wrote in message news:<cf15391e.0309111325.141d63e6@posting.google.com>...  $ > > Does anyone have advice on how I9 > > escalate this problem higher-up in the HP food chain?  > D > Start by calling the CSC back, asking for the Manager On Duty, and/ > expressing your concerns about the situation.   B I did that yesterday.  "Pulled the fire alarm" at HP (in a virtualC manner).  Quite a few HP people are talking today to find out why I > have an open IPMT case that is 8 weeks old and I haven't heardD anything from Storage Engineering.  I suspect that someone's butt is& going to be in a sling shortly...  :^)   ------------------------------    Date: 12 Sep 2003 08:09:08 -0700$ From: svieth@wi.rr.com (Scott Vieth)I Subject: Re: Anyone else having problems with VMS and HSG80-based arrays? < Message-ID: <5a85bce2.0309120709.c4eed15@posting.google.com>  p spamdump@mccready.com (Gary McCready) wrote in message news:<ffd79a6c.0309111355.7313ffe7@posting.google.com>...  E >  With the problems that we have had to deal with, you just learn to F > lean on them to get the problem addressed to your satisfaction. FeelE > free to email me at Gary _at_ McCready -dot- com, and perhaps I can = > give you some hints based upon what you have not tried yet.  >  > Good luck,   Gary:     ? I think that my emails yesterday light a fire under quite a few F people.  I received a couple of phone calls from "higher-up" people atF HP who promised to bring the resources together to address this issue.E  One person I spoke to said he was talking with another HP person and D said "This is bad.  Scott asked to speak to a V.P. on August 4th and< no one got back to him."  The other person replied "You meanF *September* 4th, right?"  and the first person said "No, I mean AUGUST 4th"  C I have had enough of working with "storage engineering" and playing ? the game where the engineer working on the IPMT emails a senior C support person and then that support person acts an interpreter and A asks me for data.  i mail the log files back to the middleman who F forwards the data back to the engineer.  That is a load of crap.  FromA now on, I only talk with "VP" or higher on this problem.  No more  storage techies.    -Scott "Takin names, kickin ass"   ------------------------------    Date: 12 Sep 2003 10:10:52 -0500+ From: young_r@encompasserve.org (Rob Young) I Subject: Re: Anyone else having problems with VMS and HSG80-based arrays? 3 Message-ID: <6TuFyCkqXoIm@eisner.encompasserve.org>   d In article <5a85bce2.0309120658.34955dfe@posting.google.com>, svieth@wi.rr.com (Scott Vieth) writes:h > young_r@encompasserve.org (Rob Young) wrote in message news:<PEbWgoyWsM7q@eisner.encompasserve.org>... > D >> > AND THIS MORNING, we got hit again.  Same symptoms.  Two DGAnn:H >> > devices went mntverifytimeout.  Had to reboot without shutting downJ >> > Cache.  Got lucky once more and the Cache data files were not corruptK >> > after reboot.  Tried a "restart other" on the HSG80s.  The CLI on both K >> > controllers hung.  Had to walk an operator through hitting the buttons K >> > on the front of the HSG80s to restart them.  Does anyone see a pattern 
 >> > here? >> >   >>   >> 	Yes I see a pattern.     >>  D >> 	Why July 2nd?  Did you upgrade firmware July 1st?  What changed? > D > No firmware upgrade or any other change on July 1st.  We make it aG > policy to NOT make systems changes within five days of the end of the E > month.  July 1st was when we started the month-end batch processing B > for June.  On the night of July 1st, we had three Backup streamsD > running as well as our month-end processing.  The HSG80s locked-up4 > during the month-end and made quite a mess for us. >   B 	Okay.  So why didn't it occur June 1st, May 1st, April 1st, etc.?> 	Were backup schedules modified such that 3 are running at oneD 	time, prior to that 1 ran?  In other words, for a short term fix go? 	back to prior processes/processing and cut down on concurrent  * 	IO and see if it mysteriously disappears.   				Rob    ------------------------------    Date: 12 Sep 2003 10:27:50 -0500+ From: young_r@encompasserve.org (Rob Young) I Subject: Re: Anyone else having problems with VMS and HSG80-based arrays? 3 Message-ID: <fj48oEeLfZ4O@eisner.encompasserve.org>   d In article <5a85bce2.0309120701.5eb12f06@posting.google.com>, svieth@wi.rr.com (Scott Vieth) writes:u > al5vf03p02@sneakemail.com (William Webb) wrote in message news:<d5ce4b06.0309111322.4a54edcc@posting.google.com>...  >>   >> Are you using XFC or VIOC?    >>  A >> We decided to stay with VIOC 'cuz we suspect it's more stable.  >>   > F > Hoff told me that it was okay to go ahead and use the XFC.  It kicksG > ass when compared to the VIOC.  If you are running 7.3-1 and have all F > your patches in place, then turn on the XFC.  I have the cache limitG > set to 12GB.  I am going to increase it this weekend.  We are getting F > pretty respectible cache hit numbers with 12 gigs but I think we can > do better. > F > The fastest disk I/O is the one that doesn't have to go to disk. :^) >   ; 	This is easy to check and more doesn't always mean better.   = 	Firstly, Cache will be caching hot global regions and as the < 	installation guide says you can run GLOSTAT to determine if 	you need to bump up buffers:   6 http://platinum.intersystems.com/GCI/GCI_vmsparms.html  K "You can use the statistics produced by the GLOSTAT utility to determine if F adding more global buffers will reduce disk access and thereby improve
 performance."   ; 	Give it as many as it needs, etc.  But XFC comes into play ? 	as only so much will be utilized and a utility like GLOSTAT is ? 	conservative (as is MEMREQ ;-).  I'd be intersested in seeing  & 	your cache hit rates and Free MBytes:   	$ show memory/cache  F 	During high IO times during the day, is all your memory dedicated to E 	cache (not to be confused with Cache) in use?  If not, why dedicate   	more?   				Rob    ------------------------------    Date: 12 Sep 2003 07:44:45 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) 9 Subject: Re: First VMS IPF difference... not very nice... 3 Message-ID: <rDoJJO$Aa2Pm@eisner.encompasserve.org>   U In article <00A25C0A.C6B4D74B@SendSpamHere.ORG>,   VAXman-  @SendSpamHere.ORG writes:   E > Now if they'd just put the manuals in IPF$DOCS in a readable format F > like PostScript, I'd be on my way.  Of specific interest this momentD > is the Calling Standard docs.  If VMS manuals are still done using0 > DOCUMENT, producing PS should be a no brainer.  F    I got a copy of an IA64 calling standard doc from Cmopaq's OpenVMS G    web site prior to the merger.  Is it out of date, insuffient, or you     just haven't got it?    ------------------------------  # Date: Fri, 12 Sep 2003 13:48:20 GMT & From: Joshua Cope <Joshua.Cope@hp.com>9 Subject: Re: First VMS IPF difference... not very nice... & Message-ID: <3F61CEA3.D139E9E8@hp.com>   > >> DELTA! < > >> %SYSTEM-F-NOCMEXEC, operation requires CMEXEC privilege > C > I don't know if it's a bug or feature.  Just reporting the facts.   G I think Larry's question was, "why is this a VMS IPF difference?" On my0J V7.3-1 system, trying to run DELTA as a non-privileged user gives the same error message.   ------------------------------  # Date: Fri, 12 Sep 2003 17:14:18 GMT " From:   VAXman-  @SendSpamHere.ORG9 Subject: Re: First VMS IPF difference... not very nice...a0 Message-ID: <00A25CB8.3D049EC1@SendSpamHere.ORG>  O In article <3F61CEA3.D139E9E8@hp.com>, Joshua Cope <Joshua.Cope@hp.com> writes:  >> >> DELTA!= >> >> %SYSTEM-F-NOCMEXEC, operation requires CMEXEC privilegey >>D >> I don't know if it's a bug or feature.  Just reporting the facts. >oH >I think Larry's question was, "why is this a VMS IPF difference?" On myK >V7.3-1 system, trying to run DELTA as a non-privileged user gives the same  >error message.i  % Doesn't or didn't prior to V7.3(-1?).e  J Just tried to define LIB$DEBUG to DELTA and run a simple program and I get
 a stack dump.V  ? Really hell bent on keeping me out of elevated access modes eh?S  d --L VAXman- A Bored Certified VMS Kernel Mode Hacker    VAXman(at)TMESIS(dot)COM            s5   "Well my son, life is like a beanstalk, isn't it?"     ------------------------------    Date: 12 Sep 2003 12:56:54 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler).9 Subject: Re: First VMS IPF difference... not very nice...M3 Message-ID: <YsvNuQYryzHO@eisner.encompasserve.org>y  G    Hm.  XF90 doesn't seem to have a forsysdef.tlb.  I wonder if someoneID    in VMS engineering would actualy like to keep track of the littleH    nits we find.  I suspect things like forsysdef.tlb simply reflect theH    early state of the compiler, but eventually someone will want to knowF    whether they've addressed such things.  Certainly they're still too+    busy to jump into all of them right off.r  D    Maybe I should post this to vmsnet.sdk.openvms.fieldtest?  Is 8.0    considered an SDK?T  I    I must say, I really do like having access to these systems already.  e8    I was suspecting not to see it for a few more months.   ------------------------------    Date: 12 Sep 2003 08:34:40 -0700$ From: svieth@wi.rr.com (Scott Vieth)Y Subject: I have a fix!!!  Re: Anyone else having problems with VMS and HSG80-based arraysq= Message-ID: <5a85bce2.0309120734.7b37cbdf@posting.google.com>q  i svieth@wi.rr.com (Scott Vieth) wrote in message news:<5a85bce2.0309111054.6f6fdbac@posting.google.com>...n > Hi:h > F > I've got an ES40 running VMS 7.3-1 which is connected to an ESA12000 > (running ACS 8.7S-2).0  A ...whichs falls down periodically in the middle of the night. :^)   E I am here this morning to post a fix (or two) for the storage issue Ip described yesterday.  F After "lighting a fire" under HP yesterday via email and phone call, I: received an email from one of the gods in VMS Engineering.  E He said that the problem with our VMS system and the ESA12000 sounded-; very similar to a problem experienced at another site.  VMS:F Engineering had also reproduced this problem in their test lab (prolly' where that funny white elephant lives).b  C Anywho, the root problem that we are seeing is that our mighty ES40tB running OpenVMS 7.3-1 is simply OVERWHELMING the ESA12000 with I/O> requests.  The VMS Engineering person told me that some of theF performance tweaks in 7.3-1 really make VMS fly when it comes to I/O. F Now our ES40 is demanding data so fast from the HSG80s that eventually the HSG80s tip over.  F This makes perfect sense.  The three times that we have been bitten byB this problem, we had *extremely* heavy I/O on the ES40.  The firstD time, we were running three concurrent backup streams from snapshotsF AND running our month-end batch processing.  The poor HSG80s could notE handle the load and gave up (which corrupted our Cache data files andn made us restore from tape).e  E The VMS Engineering person told me that the HSG80s have a total queueoF depth (if that is the correct term) of *240* outstanding I/O requests.E  After that, the controllers try to tell the host system to slow down B a little.  But the ES40 and VMS 7.3-1 are hungry for more data and finally the HSG80 faints.o  D First Fix: The guru from VMS Engineering asked me to check the DIOLMC setting on the account that we use to run out backup jobs.  Knowing C that the HSG80s have a maximum queue depth of 240, we don't want tonC bury the HSG80s any more.  In Authorize, I found that DIOLM for oursB backup account was set to 32767.  Three backup jobs running at the> same time under that account were issuing TONS and TONs of I/O  requests and burying the HSG80s.  D So, per his advice, I set the DIOLM for our backup account to "32". B This will give very good backup performance and still not bury the HSG80s.s  7 Second fix: The VMS Engineering guru told me to installRE DEC-AXPVMS-VMS731_MSA1000-V0100 as soon as I can.  It fixes a timeouteF value for fibre channel read/writes.  The value got set to "4 seconds"A in VMS 7.3-1 and this patch changes the timeout value back to "24uE seconds".  This will help the OS be more tolerant when the HSG80s arem* being pokey with returning requested data.  < How do you know if you are burying your HSG80s?  You can run
 "anal/sys" and then do a "fc stdt/all"a  E look at the far-righthand column.  It is labelled "Seq Tmo".  If thatfE number is huge, that is a bad sign.  Also look at the column labelled 5 "QF Seen".  We don't want that value to large either.   D So, that appears to be the problem I ran into and those are the waysA to fix it.  The Mighty VMS Operating system (7.3-1) is simply toohE powerful for the HSG80 controllers.  By throttling the I/Os a little, E we can get the OS to "behave" until we get rid of the HSG80s and movea
 to our....   [E]xtra-high-performance [V]MS-readyt [A]rrayc   :^) :^) :^)h  D You may have noticed that I did not reveal the name of the person inB VMS Engineering who helped me.  He asked that I not post his name,2 phone number or email address (or yearbook photo).  < You will need to open *your own* case and get *your own* VMS; Engineering god to help you out.  Those are the breaks. :^)   E Now that I know what is causing our 'mntverifytimeouts' in the middleaC of the night and how to fix it, I feel like a large weight has beene< lifted from my shoulders.  I can finally get some sleep. :^)  F Many, many thanks to the person from VMS Engineering who took the timeE to email me last night and talk with me on the phone this morning.  I F learned more in 30 minutes talking with this person than I did workingA with the storage group for eight weeks.  VMS Engineering rocks!! n+ Storage Engineering...well, you kinda suck.e   ------------------------------  % Date: Fri, 12 Sep 2003 13:53:03 -0400c From: koskaj@bender.com Y Subject: re: I have a fix!!!  Re: Anyone else having problems with VMS and HSG80-based ara; Message-ID: <03091213530381.139.10369833@alaxp3.bender.com>e  6 >From:	SMTP%"svieth@wi.rr.com" 12-SEP-2003 11:31:24.43 >To:	Info-VAX@Mvb.Saic.Com >CC:	cX >Subj:	I have a fix!!!  Re: Anyone else having problems with VMS and HSG80-based arrays? >wj >svieth@wi.rr.com (Scott Vieth) wrote in message news:<5a85bce2.0309111054.6f6fdbac@posting.google.com>... >> Hi: >> aG >> I've got an ES40 running VMS 7.3-1 which is connected to an ESA12000t >> (running ACS 8.7S-2). >oB >...whichs falls down periodically in the middle of the night. :^) > F >I am here this morning to post a fix (or two) for the storage issue I >described yesterday.. > G >After "lighting a fire" under HP yesterday via email and phone call, It; >received an email from one of the gods in VMS Engineering.t >cF >He said that the problem with our VMS system and the ESA12000 sounded< >very similar to a problem experienced at another site.  VMSG >Engineering had also reproduced this problem in their test lab (prolly ( >where that funny white elephant lives). >iD >Anywho, the root problem that we are seeing is that our mighty ES40C >running OpenVMS 7.3-1 is simply OVERWHELMING the ESA12000 with I/Om? >requests.  The VMS Engineering person told me that some of the0G >performance tweaks in 7.3-1 really make VMS fly when it comes to I/O. 2G >Now our ES40 is demanding data so fast from the HSG80s that eventuallyt >the HSG80s tip over.3  L Hmmm... I have a GS160 Model 16 with 7.3-1 and all but last couple of weeks N of patches installed with 6 dual redundant pairs of HSG-80's that are 256 meg N mirrored caches, 12 fibre HBAs, and 3 pair redundant switches, and 14 gigs of J XFC.  To me, my GS160 might be able to present similar I/O load like your 1 ES40 to HSG-80's, if my application drives it to.s  F Granted, your I/O load may be higher than mine.  But my MONITOR stats J indicate I have what I consider fairly high I/O load over extended time.  L My I/O request queue length are in the thousands (2K to 5K) on DGA devices, M as are I/O operation rates (15K to 36K) on all my DSA devices, and (1K - 2K) iH on my DGA devices.  My ECP stats also seem to match up. In general, I/O I bound system with spread I/O across the disk farm we could afford (about .N 140 of 15Krpm drives).  I'm not doing cloning with HSG-80's like you, so load L on HSG-80 may not be as high.  But I do think I have high sustained (hours) 9 I/O load to my HSG-80's, with about 50/50 read/write mix.i  K I guess I will have to measure doing ANAL/SYS sometime and see how big the sK numbers are. "Huge" and "large" are quite relative terms.  Can you be more  L quantitative?  If not, I understand.  I don't have the problem you see/had, M but I am curious to compare, and to take action to avoid the problem if need r be.s   :) jck" Koskaj@hatespam.bender.nospam.com    ------------------------------  # Date: Fri, 12 Sep 2003 12:47:01 GMTa# From: "John Smith" <a@nonymous.com>i1 Subject: Re: IDL on VMS. Was: Image tools for VMSbF Message-ID: <9fj8b.4815$Rm1.2582@news02.bloor.is.net.cable.rogers.com>  H The impressions fostered by Palmer, Capellas, and Fiorina about VMS have come home to roost.o       Jonathan Boswell wrote:h > Keith Cayemberg wrote:> >> Let it not be said that there are no graphics imaging tools >> for OpenVMS...o >w > [snip] >  >> RSI >> http://www.rsinc.com/ > @ > This doubtlessly refers to the excellent IDL product, numerousE > licenses for which I have purchased over the years starting in 1985PE > when it ran quite nicely on a VAX 11/750.  The only trouble is that.@ > Kodak acquired RSI and (if memory serves) in less than a monthG > dropped support for IDL on Alpha/VMS.  Support for IDL on VAX/VMS waslE > dropped years before that.  Support continues for "Compaq" Tru64 tor > this day.t > F > The following note was penned late last night by Jim Kelley, the RSIG > product manager for IDL and OK'd by him for posting here.  Please sitm > down before reading. >o >  - JB.9 >                 _______________________________________M >iA > RSI dropped support for VMS several years after DEC had droppednD > support for that operating system.  The decision by RSI was drivenE > primarily by a number of various factors that essentially convergedo > at about the same time. A > First, when DEC announced they were dropping support for VMS itsG > triggered customers initiated their own plans to migrate to different  > operating F > systems.  Also as a result of DEC dropping VMS support,  a number ofB > third party vendors that has libraries or code used by IDL beganF > phasing out their support as well.  RSI realized that our ability toF > continue releasing IDL on VMS would end as soon as these third party > dependencies in IDL were nonC > longer available for VMS.  We announced our intention to drop VMS G > about 12 months prior to the last release, in hopes of giving as mucho) > lead-time as possible to our customers.a >aD > Having dropped the VMS platform many releases ago, it would a huge	 > task tot= > try to resurrect IDL on VMS today.  But, in addition to the G > engineering challenge that would certainly not be cost effective,  we E > really could not replicate IDL 6.0 today on VMS because of the manyMF > 3rd party components and libraries that also no longer offer support > for the VMS operating system.n > E > RSI does not drop support for hardware all that often, we typicallysC > retain support for certain systems long past the point that it is F > economically advantageous for the company.  But at some point we are@ > faced with choosing between moving the software forward to theG > benefit of many users at the expense of having to drop older systems.  >iG > The  fact of the matter is that DEC built really great hardware,  andsD > while many of their systems in the field are really old by today'sG > standards, some of these systems will like be running years from now.S   ------------------------------    Date: 12 Sep 2003 12:19:30 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler)31 Subject: Re: IDL on VMS. Was: Image tools for VMS,3 Message-ID: <4sAKc8qF9NZV@eisner.encompasserve.org>   l In article <9fj8b.4815$Rm1.2582@news02.bloor.is.net.cable.rogers.com>, "John Smith" <a@nonymous.com> writes:J > The impressions fostered by Palmer, Capellas, and Fiorina about VMS have > come home to roost.  >   B    Despite RSI's mistatements about Digital dropping VMS, they didE    react to actual business fact.  They dropped IDL on VMS just aftereI    one of thier major customers moved from VMS to Solaris.   Competitive  J    products, such as MATLAB, made similar decisions for similar reasons.    G    Since these products market don't attract the same attention as the eK    niches DEC, Compaq, and HP have been willing to suggest VMS is good at, h&    those facts will likely not change.  ;    But like old VAXen, old copies of IDL just keep running.m   ------------------------------    Date: 12 Sep 2003 10:25:34 +0100K From: pmoreau@ath.cena.fr (Patrick MOREAU, CENA Athis, Tel: 01.69.57.68.40) A Subject: Re: Image tools for VMS - follow up - image file formats ! Message-ID: <FA6AjDSSoK8b@sinead>n   Imagemagick is OK with PNG.   I I also use Imagemagick to create photo indexes to serve them (indexes and:L photos) via OSU server on my home LAN (mosty PCs plus a few VMS boxes). IdemK with music files made from my own CDs. XV is also used for color editing, agK great tool and I rarely use the PC photo editors I have (Paint Shop Pro ande Picture Publisher).    Patrick   . In article <bjrc9p$avo0$1@news3.infoave.net>, 2 "Jeff Morgan" <vmswiz@geonospamcities.com> writes:H > Last time I looked, IMAGEMAGIK and XV didn't support PNG. Of course, II > haven't felt a need to upgrade these tools for years since they've been6 > working just fine. > H > I use IMAGEMAGIK to auto-create thumbnails in batch and XV to do colorN > editing and touchups for all my family photo albums. XV has an awesome colorN > editor. It will take the yellow out of scanned antique photos and can turn aK > dark picture into full daylight. You can increase the saturation on fadedfM > photos and actually bring the colors back out. You can even use it to eraseeI > scratches but it is a crude pixel editor. Considering its age, it is ann > astounding tool. >  --O ===============================================================================rN pmoreau@ath.cena.fr  (CENA)      ______      ___   _          (Patrick MOREAU)4 moreau_p@decus.fr (DECUS)       / /   /     / /|  /|J CENA/Athis-Mons FRANCE         / /___/     / / | / |   __   __   __   __  N BP 205                        / /         / /  |/  |  |  | |__| |__  |__| |  |N 94542 ORLY AEROGARE CEDEX    / /   ::    / /       |  |__| | \  |__  |  | |__|N http://www.ath.cena.fr/~pmoreau/            http://www.multimania.com/pmoreau/O ===============================================================================    ------------------------------    Date: 12 Sep 2003 07:36:31 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler)7/ Subject: Re: Mac OS X -> Pathworks/Mac problemsd3 Message-ID: <C7n5Fay6Dlyu@eisner.encompasserve.org>c  M In article <bjr0i70ofp@enews1.newsguy.com>, healyzh@NOaracnetSPAM.com writes:c > J > I actually happened to be playing with just this feat this morning afterN > reading a Macworld article.  It worked OK to a Unix system, however, I endedM > up with the same results with OpenVMS.  I'm using OpenVMS 7.2-1H1 and TCPIPc > 5.3 with Mac OS X 10.2.6."  C    Does TCP/IP 5.3 have an emulate-UNIX setting, like Multinet has?    ------------------------------    Date: 12 Sep 2003 07:18:12 -0500 From: briggs@encompasserve.org% Subject: Re: Mailboxes and SYS$OUTPUTo3 Message-ID: <H+xc6gECIDQp@eisner.encompasserve.org>r  i In article <ae38b.41$K_3.22054@news.uswest.net>, "Michael D. Ober" <obermd-@-alum-mit-edu-nospam> writes:dH > Does anyone have sample code that creates a temporary mailbox and thenC > redirects the SYS$OUTPUT of a spawned subprogram to that mailbox?t $ type backup_to_mbx.for&         options         /extend_source           implicit        none  >         integer         lib$spawn, sys$crembx, lib$get_foreign           integer *2      chan           integer         status           character *63   disk         integer *2      disk_l  @         status = lib$get_foreign ( disk, 'What disk? ', disk_l )          if ( .not. status ) then7             type *, 'Failed to retrieve disk selection'r*             call sys$exit ( %val(status) )         end if  M         status = sys$crembx ( %val(1), %ref(chan), %val(8192), %val(32768),,,..         2                     'backup_mbx',, )          if ( .not. status ) then.             type *, 'Failed to create mailbox'*             call sys$exit ( %val(status) )         end if  S         status = lib$spawn ( 'backup /log /block=8192 /image /ignore=interlock ' //mG         1                    disk(1:disk_l) // ': backup_mbx:. /save' )e          if ( .not. status ) then"             type *, 'Spawn failed'*             call sys$exit ( %val(status) )         end if  @ 	! Separate external program that reads from the mailbox omitted< 	! I think I was piping it elsewhere using rcp or some such.           end    ------------------------------  % Date: Fri, 12 Sep 2003 09:58:33 +0100b& From: Edward Brocklesby <ejb@goth.net> Subject: Re: MESSAGE problem= Message-ID: <4Uf8b.6443$YL.4016@news-binary.blueyonder.co.uk>l   Wayne Sewell wrote:lC > You have to access the message code by value, not by reference.  y  F Thanks Wayne (and Keith Lewis who provided the PL/I equivalent)--this  appears to be the solution.o  H Now, is this not documented in the MESSAGE manual, or did I simply miss J it?  (The only example code I could see was FORTRAN, which I don't speak).   Regards, Edward.    ------------------------------    Date: 12 Sep 2003 07:33:30 -0500- From: Kilgallen@SpamCop.net (Larry Kilgallen)0 Subject: Re: MESSAGE problem3 Message-ID: <DDJwq6Ur5OdH@eisner.encompasserve.org>s  f In article <4Uf8b.6443$YL.4016@news-binary.blueyonder.co.uk>, Edward Brocklesby <ejb@goth.net> writes: > Wayne Sewell wrote:iD >> You have to access the message code by value, not by reference.   > H > Thanks Wayne (and Keith Lewis who provided the PL/I equivalent)--this  > appears to be the solution.  > J > Now, is this not documented in the MESSAGE manual, or did I simply miss L > it?  (The only example code I could see was FORTRAN, which I don't speak).  E I would expect to find it in the PL/I documentation, which I have notc read in depth.   ------------------------------    Date: 12 Sep 2003 07:36:32 -0500 From: briggs@encompasserve.org Subject: Re: MESSAGE problem3 Message-ID: <qS3Wgs8uLvu5@eisner.encompasserve.org>t  f In article <4Uf8b.6443$YL.4016@news-binary.blueyonder.co.uk>, Edward Brocklesby <ejb@goth.net> writes: > Wayne Sewell wrote: D >> You have to access the message code by value, not by reference.   > H > Thanks Wayne (and Keith Lewis who provided the PL/I equivalent)--this  > appears to be the solution.e > J > Now, is this not documented in the MESSAGE manual, or did I simply miss L > it?  (The only example code I could see was FORTRAN, which I don't speak).  C It's not an issue with the MESSAGE utility.  It's an issue with the E language you use to write the code that is linked against the messageY file.R  C When you resolve external references at link time, you need to telle? your compiler whether you're importing addresses or values.  Ini= general and by default, your compiler will assume that you'reh importing addresses.  ? So you can either use a compiler directive to override that ands> have the compiler treat the external references as values.  Or: can write code that uses the imported "addresses" as such.  > Adding a single ampersand will fix your C code, allowing it toC pass the resolved address by value rather than passing the contents , of the memory cell at that address by value.   	John Briggs   ------------------------------  % Date: Fri, 12 Sep 2003 06:25:18 -07006# From: "Tom Linden" <tom@kednos.com>- Subject: RE: MESSAGE problem9 Message-ID: <CIEJLCMNHNNDLLOOGNJIEEGHHPAA.tom@kednos.com>A   >-----Original Message----- 5 >From: Larry Kilgallen [mailto:Kilgallen@SpamCop.net]0) >Sent: Friday, September 12, 2003 5:34 AMn >To: Info-VAX@Mvb.Saic.Com >Subject: Re: MESSAGE problem: >4 >2? >In article <4Uf8b.6443$YL.4016@news-binary.blueyonder.co.uk>, r) >Edward Brocklesby <ejb@goth.net> writes:  >> Wayne Sewell wrote:E >>> You have to access the message code by value, not by reference.  C >> SI >> Thanks Wayne (and Keith Lewis who provided the PL/I equivalent)--this e >> appears to be the solution. >> aK >> Now, is this not documented in the MESSAGE manual, or did I simply miss  @ >> it?  (The only example code I could see was FORTRAN, which I  >don't speak). > F >I would expect to find it in the PL/I documentation, which I have not >read in depth.h  H Why would you expect to find it there?  It seems to me the interfaces toI MESSAGE should be provided by the MESSAGE documentation, which is clearlyeK deficient, which I can confirm after having perused the manual.  Instead ofaI Fortran or Cobol example they should have provided a rigourous definitionh of the interface with SDL.  $ It is not in the PL/I documentation.   >g >---' >Incoming mail is certified Virus Free. ; >Checked by AVG anti-virus system (http://www.grisoft.com). A >Version: 6.0.512 / Virus Database: 309 - Release Date: 8/19/2003k >u ---u& Outgoing mail is certified Virus Free.: Checked by AVG anti-virus system (http://www.grisoft.com).@ Version: 6.0.512 / Virus Database: 309 - Release Date: 8/19/2003   ------------------------------  % Date: Fri, 12 Sep 2003 06:31:33 -0700s# From: "Tom Linden" <tom@kednos.com>d Subject: RE: MESSAGE problem9 Message-ID: <CIEJLCMNHNNDLLOOGNJIMEGHHPAA.tom@kednos.com>-   >-----Original Message-----0A >From: briggs@encompasserve.org [mailto:briggs@encompasserve.org]:) >Sent: Friday, September 12, 2003 5:37 AMS >To: Info-VAX@Mvb.Saic.Com >Subject: Re: MESSAGE problemf >o >n> >In article <4Uf8b.6443$YL.4016@news-binary.blueyonder.co.uk>,) >Edward Brocklesby <ejb@goth.net> writes:s >> Wayne Sewell wrote:C >>> You have to access the message code by value, not by reference.i >>H >> Thanks Wayne (and Keith Lewis who provided the PL/I equivalent)--this >> appears to be the solution. >>J >> Now, is this not documented in the MESSAGE manual, or did I simply miss? >> it?  (The only example code I could see was FORTRAN, which I  >don't speak). >tD >It's not an issue with the MESSAGE utility.  It's an issue with theF >language you use to write the code that is linked against the message >file.  K Yes and no.  It is really a documantation deficiency pertaining to MESSAGE.rL If you provide a utility, it seems to me that you need to provide a complete interface spec.t   >iD >When you resolve external references at link time, you need to tell@ >your compiler whether you're importing addresses or values.  In> >general and by default, your compiler will assume that you're >importing addresses.c >-@ >So you can either use a compiler directive to override that and? >have the compiler treat the external references as values.  Orl; >can write code that uses the imported "addresses" as such.8 >t? >Adding a single ampersand will fix your C code, allowing it toeD >pass the resolved address by value rather than passing the contents- >of the memory cell at that address by value.- >-
 >	John Briggsp >0 >---' >Incoming mail is certified Virus Free. ; >Checked by AVG anti-virus system (http://www.grisoft.com). A >Version: 6.0.512 / Virus Database: 309 - Release Date: 8/19/2003t >. ---e& Outgoing mail is certified Virus Free.: Checked by AVG anti-virus system (http://www.grisoft.com).@ Version: 6.0.512 / Virus Database: 309 - Release Date: 8/19/2003   ------------------------------    Date: 12 Sep 2003 11:58:57 -0500- From: Kilgallen@SpamCop.net (Larry Kilgallen)n Subject: RE: MESSAGE problem3 Message-ID: <x4ctMgOSWCrG@eisner.encompasserve.org>a  _ In article <CIEJLCMNHNNDLLOOGNJIEEGHHPAA.tom@kednos.com>, "Tom Linden" <tom@kednos.com> writes:  >  >  >>-----Original Message-----6 >>From: Larry Kilgallen [mailto:Kilgallen@SpamCop.net]* >>Sent: Friday, September 12, 2003 5:34 AM >>To: Info-VAX@Mvb.Saic.Com  >>Subject: Re: MESSAGE problem >> >>@ >>In article <4Uf8b.6443$YL.4016@news-binary.blueyonder.co.uk>, * >>Edward Brocklesby <ejb@goth.net> writes: >>> Wayne Sewell wrote:yF >>>> You have to access the message code by value, not by reference.   >>> J >>> Thanks Wayne (and Keith Lewis who provided the PL/I equivalent)--this  >>> appears to be the solution.n >>> L >>> Now, is this not documented in the MESSAGE manual, or did I simply miss A >>> it?  (The only example code I could see was FORTRAN, which I a >>don't speak).  >>G >>I would expect to find it in the PL/I documentation, which I have notu >>read in depth. > ( > Why would you expect to find it there?  > I would expect to find explanation of how to import values vs.< the addresses of values.  Other languages have that in their< documentation, and I imagine it is in the PL/I documentation also.n  " > It seems to me the interfaces toK > MESSAGE should be provided by the MESSAGE documentation, which is clearly-M > deficient, which I can confirm after having perused the manual.  Instead ofuK > Fortran or Cobol example they should have provided a rigourous definitions > of the interface with SDL.  A Well SDL has never been a supported product, so it is unlikely tor@ get documentation support.  Using MESSAGE/SDL does save time for@ those who do a lot of linking (compared to compiling) but it canE lead to programmers forgetting to explicitly link the message module.i  & > It is not in the PL/I documentation.  C It looks to me like Section 12.2.2 of the PL/I User Manual explainseD how to do it.  The Message value usage might be added to the list in# Section 12.2.3 to be more complete.t   ------------------------------  % Date: Fri, 12 Sep 2003 10:03:56 -0700m# From: "Tom Linden" <tom@kednos.com>e Subject: RE: MESSAGE problem9 Message-ID: <CIEJLCMNHNNDLLOOGNJIOEHAHPAA.tom@kednos.com>w   >-----Original Message-----o5 >From: Larry Kilgallen [mailto:Kilgallen@SpamCop.net].) >Sent: Friday, September 12, 2003 9:59 AMm >To: Info-VAX@Mvb.Saic.Com >Subject: RE: MESSAGE problemg >" >n? >In article <CIEJLCMNHNNDLLOOGNJIEEGHHPAA.tom@kednos.com>, "Tomo! >Linden" <tom@kednos.com> writes:e >> >> >>>-----Original Message-----e7 >>>From: Larry Kilgallen [mailto:Kilgallen@SpamCop.net]s+ >>>Sent: Friday, September 12, 2003 5:34 AMp >>>To: Info-VAX@Mvb.Saic.Com >>>Subject: Re: MESSAGE problem  >>>e >>>h@ >>>In article <4Uf8b.6443$YL.4016@news-binary.blueyonder.co.uk>,+ >>>Edward Brocklesby <ejb@goth.net> writes:g >>>> Wayne Sewell wrote:E >>>>> You have to access the message code by value, not by reference.- >>>>J >>>> Thanks Wayne (and Keith Lewis who provided the PL/I equivalent)--this  >>>> appears to be the solution. >>>>@ >>>> Now, is this not documented in the MESSAGE manual, or did I >simply missA >>>> it?  (The only example code I could see was FORTRAN, which I4 >>>don't speak). >>>oH >>>I would expect to find it in the PL/I documentation, which I have not >>>read in depth.m >>) >> Why would you expect to find it there?o >.? >I would expect to find explanation of how to import values vs. = >the addresses of values.  Other languages have that in theirw= >documentation, and I imagine it is in the PL/I documentationC >also.  I Ah, yes.  Misunderstood the point you were making.  It is inded the VALUEH attribute on the declaration >n# >> It seems to me the interfaces tooL >> MESSAGE should be provided by the MESSAGE documentation, which is clearlyB >> deficient, which I can confirm after having perused the manual. >Instead of.L >> Fortran or Cobol example they should have provided a rigourous definition >> of the interface with SDL.r >yB >Well SDL has never been a supported product, so it is unlikely toA >get documentation support.  Using MESSAGE/SDL does save time forwA >those who do a lot of linking (compared to compiling) but it can F >lead to programmers forgetting to explicitly link the message module.  H Pity,  it is a great facility.  After all these years IBM is pushing LE,1 but they still don't have an SDL equivalent (yet)d > ' >> It is not in the PL/I documentation.c > D >It looks to me like Section 12.2.2 of the PL/I User Manual explainsE >how to do it.  The Message value usage might be added to the list ino$ >Section 12.2.3 to be more complete.  G Possibly, but in order to be complete then, I suppose one ought to list A all global symbols and where they are used.  Is that what you are0 suggesting?_   >o >---' >Incoming mail is certified Virus Free. ; >Checked by AVG anti-virus system (http://www.grisoft.com).gA >Version: 6.0.512 / Virus Database: 309 - Release Date: 8/19/2003c >0 ---0& Outgoing mail is certified Virus Free.: Checked by AVG anti-virus system (http://www.grisoft.com).@ Version: 6.0.512 / Virus Database: 309 - Release Date: 8/19/2003   ------------------------------  # Date: Fri, 12 Sep 2003 13:31:32 GMT ) From: "Stefano Borselli" <ibbo@libero.it>p" Subject: Newbie question about NTP9 Message-ID: <UUj8b.313625$Ny5.9772186@twister2.libero.it>    Hi, i'm Stefano Borselli. F In Unix exist command "ntpdate". This command print the difference (inJ seconds) between this machine and a server that running ntpd. In VMS exist "ntpdate" or similar?s  
 Thank you.   Stefano.   ------------------------------  % Date: Fri, 12 Sep 2003 09:09:24 -0500 ( From: brandon@dalsemi.com (John Brandon)& Subject: Re: Newbie question about NTP1 Message-ID: <03091209092453@dscis6-0.dalsemi.com>    > Hi, i'm Stefano Borselli.yH > In Unix exist command "ntpdate". This command print the difference (inL > seconds) between this machine and a server that running ntpd. In VMS exist > "ntpdate" or similar?e >  > Thank you. > 
 > Stefano.   $ tcpip show ver  A   DIGITAL TCP/IP Services for OpenVMS Alpha Version V5.0A - ECO 3l>   on a Compaq AlphaServer GS160 6/731 running OpenVMS V7.2-1H1   $ show sym ntp*t,   NTPDATE == "$SYS$SYSTEM:TCPIP$NTPDATE.EXE"(   NTPDC == "$SYS$SYSTEM:TCPIP$NTPDC.EXE"&   NTPQ == "$SYS$SYSTEM:TCPIP$NTPQ.EXE".   NTPTRACE == "$SYS$SYSTEM:TCPIP$NTPTRACE.EXE"	 $ ntpdatey  C Usage:  ntpdate [-d] [-o version] [-p packets] [-q] host [host ...]   *            -d            turn on debugging3            -o version    NTP version (default is 3)hC            -p packets    number of packets to sample (default is 4)-<            -q            simple query, don't set system time         J*o*h*n B*r*a*n*d*o*ne VMS Systems Administratorj* firstname.lastname.spam.me.not@dalsemi.com   ------------------------------  % Date: Fri, 12 Sep 2003 07:39:19 -0700c# From: "Tom Linden" <tom@kednos.com>G& Subject: RE: Newbie question about NTP9 Message-ID: <CIEJLCMNHNNDLLOOGNJIIEGKHPAA.tom@kednos.com>e  C tcpip needs some works.  By extension and symmetry you would expecte to find something underC tcpip> helpr   This prompted my curiosity  
 on Tru64 5.1a   . bash-2.05b# /usr/sbin/ntpdate -q 207.126.97.57C server 207.126.97.57, stratum 2, offset 10830.964511, delay 0.05774 D 12 Sep 07:34:42 ntpdate[6587]: step time server 207.126.97.57 offset 10830.964511 sec  
 and on VMS7.3-   FREJA> ntpdate -q 207.126.97.57e Socket bind error.C Verify that the NTP server (TCPIP$NTP) is not running on this host.r! %SYSTEM-F-DUPLNAM, duplicate namei% FREJA> pipe sho sys|sear sys$pipe ntphF 39A0021F TCPIP$NTP_1     LEF     10  1234883   0 00:00:16.03       564 204  N1 FREJA> pipe tcpip sho service | sear sys$pipe ntp ; NTP                  123  UDP      TCPIP$NTP        0.0.0.0w Enableds   What am I missing?   >-----Original Message----- 0 >From: John Brandon [mailto:brandon@dalsemi.com]) >Sent: Friday, September 12, 2003 7:09 AM- >To: Info-VAX@Mvb.Saic.Com' >Subject: Re: Newbie question about NTPe >p >e >> Hi, i'm Stefano Borselli.I >> In Unix exist command "ntpdate". This command print the difference (inaC >> seconds) between this machine and a server that running ntpd. Inh
 >VMS exist >> "ntpdate" or similar? >>
 >> Thank you.O >> >> Stefano.  >u >$ tcpip show verg > B >  DIGITAL TCP/IP Services for OpenVMS Alpha Version V5.0A - ECO 3? >  on a Compaq AlphaServer GS160 6/731 running OpenVMS V7.2-1H1i >r >$ show sym ntp*- >  NTPDATE == "$SYS$SYSTEM:TCPIP$NTPDATE.EXE"e) >  NTPDC == "$SYS$SYSTEM:TCPIP$NTPDC.EXE" ' >  NTPQ == "$SYS$SYSTEM:TCPIP$NTPQ.EXE" / >  NTPTRACE == "$SYS$SYSTEM:TCPIP$NTPTRACE.EXE" 
 >$ ntpdate >mD >Usage:  ntpdate [-d] [-o version] [-p packets] [-q] host [host ...] >t+ >           -d            turn on debugging 4 >           -o version    NTP version (default is 3)D >           -p packets    number of packets to sample (default is 4)= >           -q            simple query, don't set system time2 >: >- >0 >r >J*o*h*n B*r*a*n*d*o*n >VMS Systems Administrator+ >firstname.lastname.spam.me.not@dalsemi.com. >n >---' >Incoming mail is certified Virus Free.n; >Checked by AVG anti-virus system (http://www.grisoft.com). A >Version: 6.0.512 / Virus Database: 309 - Release Date: 8/19/2003c >l ---e& Outgoing mail is certified Virus Free.: Checked by AVG anti-virus system (http://www.grisoft.com).@ Version: 6.0.512 / Virus Database: 309 - Release Date: 8/19/2003   ------------------------------  % Date: Fri, 12 Sep 2003 09:55:41 -0500i( From: brandon@dalsemi.com (John Brandon)& Subject: Re: Newbie question about NTP1 Message-ID: <03091209554173@dscis6-0.dalsemi.com>i  E > tcpip needs some works.  By extension and symmetry you would expects > to find something unders
 > tcpip> help    $ help TCPIP_Services ntph   TCPIP_SERVICES     NTP-  F        The Network Time Protocol (NTP) provides a means to synchronize@        time and coordinate time distribution throughout a TCP/IPG        network. NTP aims to provide accurate and dependable timekeepingc$        for hosts on TCP/IP networks.  H        NTP provides synchronization traceable to clocks of high absoluteF        accuracy and avoids synchronization to clocks keeping incorrect        time.  E        Configuring and managing NTP on your OpenVMS host involves theo        following tasks:t  D        o  Determining which network hosts to use for synchronization  E        o  Creating and populating the NTP configuration file with theP%           list of participating hosts   A        o  Defining a time-zone differential (offset) logical name   B        o  Using the NTPDATE program to set the local date and time  G        o  Using the NTPTRACE utility to determine the source from whicho(           an NTP server obtains its time  F        o  Using the NTPDC query program to make runtime changes to NTP     >  > This prompted my curiosity >  > on Tru64 5.1ap > 0 > bash-2.05b# /usr/sbin/ntpdate -q 207.126.97.57E > server 207.126.97.57, stratum 2, offset 10830.964511, delay 0.05774fF > 12 Sep 07:34:42 ntpdate[6587]: step time server 207.126.97.57 offset > 10830.964511 sec >  > and on VMS7.3h > ! > FREJA> ntpdate -q 207.126.97.57p > Socket bind error.E > Verify that the NTP server (TCPIP$NTP) is not running on this host.m# > %SYSTEM-F-DUPLNAM, duplicate name ' > FREJA> pipe sho sys|sear sys$pipe ntphH > 39A0021F TCPIP$NTP_1     LEF     10  1234883   0 00:00:16.03       564 > 204  N3 > FREJA> pipe tcpip sho service | sear sys$pipe ntp = > NTP                  123  UDP      TCPIP$NTP        0.0.0.0t	 > Enabledp >  > What am I missing?  I Not sure... me too... similiar problems... more investigation required...o   $ ntpdate -q 207.126.97.57 Socket bind error.! %SYSTEM-F-DUPLNAM, duplicate namee    $ pipe sho sys|sear sys$pipe ntpN 20A82BA4 TCPIP$NTP       LEF      9 22411342   0 03:21:07.91       364    277  N   , $ pipe tcpip sho service | sear sys$pipe ntpO NTP                  123  UDP      TCPIP$NTP        0.0.0.0             Enableda  L Do note the lack of error on ntpdate and the Enabled message on sho service.     J*o*h*n B*r*a*n*d*o*n  VMS Systems Administratorf* firstname.lastname.spam.me.not@dalsemi.com   ------------------------------  % Date: Fri, 12 Sep 2003 09:59:00 -0500u( From: brandon@dalsemi.com (John Brandon)& Subject: Re: Newbie question about NTP1 Message-ID: <03091209585999@dscis6-0.dalsemi.com>t  5 > > FREJA> pipe tcpip sho service | sear sys$pipe ntp ? > > NTP                  123  UDP      TCPIP$NTP        0.0.0.0u < > Enabled N > Do note the lack of error on ntpdate and the Enabled message on sho service.  ; Never mind on the Enabled - you wrapped and I did not look.          J*o*h*n B*r*a*n*d*o*n  VMS Systems Administratora* firstname.lastname.spam.me.not@dalsemi.com   ------------------------------    Date: 12 Sep 2003 10:04:46 -0500+ From: young_r@encompasserve.org (Rob Young)a& Subject: Re: Newbie question about NTP3 Message-ID: <WLUnzj467x4x@eisner.encompasserve.org>   \ In article <03091209554173@dscis6-0.dalsemi.com>, brandon@dalsemi.com (John Brandon) writes:  s >> What am I missing?b > K > Not sure... me too... similiar problems... more investigation required...s >  > $ ntpdate -q 207.126.97.57 > Socket bind error.# > %SYSTEM-F-DUPLNAM, duplicate names >    	Must be your stack.     $ multinet show /versionM Process Software MultiNet V4.4 Rev A-X, Digital AlphaStation 500/266, OpenVMSe
 AXP V7.3-1 $  $ ntpdate -q 207.126.97.57? server 207.126.97.57, stratum 2, offset 0.001197, delay 0.11340^D 12 Sep 11:03:24 adjust time server 207.126.97.57 offset 0.001197 sec   				Robn   ------------------------------  % Date: Fri, 12 Sep 2003 18:31:58 +0300o* From: Mike Rechtman <rechtman@tzora.co.il>& Subject: Re: Newbie question about NTP* Message-ID: <3F61E6EE.1040902@tzora.co.il>   Tom Linden wrote: E > tcpip needs some works.  By extension and symmetry you would expect  > to find something undert
 > tcpip> help  >  > This prompted my curiosity >  > on Tru64 5.1ay > 0 > bash-2.05b# /usr/sbin/ntpdate -q 207.126.97.57E > server 207.126.97.57, stratum 2, offset 10830.964511, delay 0.05774eF > 12 Sep 07:34:42 ntpdate[6587]: step time server 207.126.97.57 offset > 10830.964511 sec >  > and on VMS7.3p > ! > FREJA> ntpdate -q 207.126.97.57a > Socket bind error.E > Verify that the NTP server (TCPIP$NTP) is not running on this host.u# > %SYSTEM-F-DUPLNAM, duplicate name-' > FREJA> pipe sho sys|sear sys$pipe ntpcH > 39A0021F TCPIP$NTP_1     LEF     10  1234883   0 00:00:16.03       564 > 204  N3 > FREJA> pipe tcpip sho service | sear sys$pipe ntp1= > NTP                  123  UDP      TCPIP$NTP        0.0.0.0t	 > Enabled" >  > What am I missing? > C IIRC (and this is documented...) you cannot run NTPDATE if the NTP s service is running.hG Try @SYS$STARTUP:TCPIP$NTP_SHUTDOWN (or similar name, sorry not near a nC VMS machine right now) and then MC NTPDATE -d <time-server-address>>   Mike   >  >>-----Original Message-----1 >>From: John Brandon [mailto:brandon@dalsemi.com]n* >>Sent: Friday, September 12, 2003 7:09 AM >>To: Info-VAX@Mvb.Saic.Comh( >>Subject: Re: Newbie question about NTP >> >> >> >>>Hi, i'm Stefano Borselli.I >>>In Unix exist command "ntpdate". This command print the difference (inaC >>>seconds) between this machine and a server that running ntpd. Ind >> >>VMS existi >> >>>"ntpdate" or similar? >>> 
 >>>Thank you.s >>>n >>>Stefano.0 >> >>$ tcpip show ver >>B >> DIGITAL TCP/IP Services for OpenVMS Alpha Version V5.0A - ECO 3? >> on a Compaq AlphaServer GS160 6/731 running OpenVMS V7.2-1H1n >> >>$ show sym ntp*.- >> NTPDATE == "$SYS$SYSTEM:TCPIP$NTPDATE.EXE"i) >> NTPDC == "$SYS$SYSTEM:TCPIP$NTPDC.EXE"S' >> NTPQ == "$SYS$SYSTEM:TCPIP$NTPQ.EXE"0/ >> NTPTRACE == "$SYS$SYSTEM:TCPIP$NTPTRACE.EXE"n >>$ ntpdate  >>E >>Usage:  ntpdate [-d] [-o version] [-p packets] [-q] host [host ...]s >>+ >>          -d            turn on debuggings4 >>          -o version    NTP version (default is 3)D >>          -p packets    number of packets to sample (default is 4)= >>          -q            simple query, don't set system times >> >> >> >> >>J*o*h*n B*r*a*n*d*o*nc >>VMS Systems Administratora, >>firstname.lastname.spam.me.not@dalsemi.com >> >>---S( >>Incoming mail is certified Virus Free.< >>Checked by AVG anti-virus system (http://www.grisoft.com).B >>Version: 6.0.512 / Virus Database: 309 - Release Date: 8/19/2003 >> >  > ---i( > Outgoing mail is certified Virus Free.< > Checked by AVG anti-virus system (http://www.grisoft.com).B > Version: 6.0.512 / Virus Database: 309 - Release Date: 8/19/2003 >        -- K  & New to c.o.vms? allow me to recommend:6 http://eisner.encompasserve.org/~rechtman/post_hlp.htm@ Other useful links at http://eisner.encompasserve.org/~rechtman/E ---------------------------------------------------------------------nE Usual disclaimer: All opinions are mine alone, perhaps not even that.I? Mike Rechtman                            *rechtman@tzora.co.il*,E Kibbutz Tzor'a.                          Voice (home):(972)-2-9908337iC    "20% of a job takes 80% of the time, the rest takes another 80%"lE ---------------------------------------------------------------------o   ------------------------------  % Date: Fri, 12 Sep 2003 10:41:48 -0500t( From: brandon@dalsemi.com (John Brandon)& Subject: Re: Newbie question about NTP1 Message-ID: <03091210414870@dscis6-0.dalsemi.com>-  E > IIRC (and this is documented...) you cannot run NTPDATE if the NTP < > service is running.cI > Try @SYS$STARTUP:TCPIP$NTP_SHUTDOWN (or similar name, sorry not near a 5E > VMS machine right now) and then MC NTPDATE -d <time-server-address>  >  > Mike   Verified - thanks Mike.   ! $ @SYS$STARTUP:TCPIP$NTP_SHUTDOWNa! TCPIP$NTP server 20E01034 stoppedb   $ ntpdate -q 207.126.97.57/ Server 207.126.97.57 (207.126.97.57), stratum 0f# offset +0.0000000, delay +0.0000000RA No server suitable for synchronization found from those provided.e       J*o*h*n B*r*a*n*d*o*ns VMS Systems Administratorv* firstname.lastname.spam.me.not@dalsemi.com   ------------------------------  % Date: Fri, 12 Sep 2003 08:41:07 -0700 # From: "Tom Linden" <tom@kednos.com>v& Subject: RE: Newbie question about NTP9 Message-ID: <CIEJLCMNHNNDLLOOGNJIOEGNHPAA.tom@kednos.com>   & FREJA> @SYS$STARTUP:TCPIP$NTP_SHUTDOWN %TCPIP-I-INFO, service disabledh9 %TCPIP-I-INFO, process TCPIP$NTP_1 (pid=39A0021F) stoppedh$ %TCPIP-I-INFO, logical names deleted9 %TCPIP-I-INFO, image SYS$SYSTEM:TCPIP$NTP.EXE deinstalledm/ %TCPIP-S-SHUTDONE, TCPIP$NTP shutdown completed  FREJA> ntpdate -q 207.126.97.57_/ Server 207.126.97.57 (207.126.97.57), stratum 2 # offset -0.0075878, delay +0.0519035 B Selected server is 207.126.97.57 (207.126.97.57) offset -0.0075878  D I am sure there is a good explanation for this, but it is definitely goofy.   >-----Original Message-----w2 >From: Mike Rechtman [mailto:rechtman@tzora.co.il]) >Sent: Friday, September 12, 2003 8:32 AM  >To: Info-VAX@Mvb.Saic.Com' >Subject: Re: Newbie question about NTP  >e >t >Tom Linden wrote:F >> tcpip needs some works.  By extension and symmetry you would expect >> to find something under >> tcpip> help >> s >> This prompted my curiosityb >>   >> on Tru64 5.1a >> g1 >> bash-2.05b# /usr/sbin/ntpdate -q 207.126.97.57 F >> server 207.126.97.57, stratum 2, offset 10830.964511, delay 0.05774G >> 12 Sep 07:34:42 ntpdate[6587]: step time server 207.126.97.57 offset, >> 10830.964511 secs >> o >> and on VMS7.3 >> l" >> FREJA> ntpdate -q 207.126.97.57 >> Socket bind error. F >> Verify that the NTP server (TCPIP$NTP) is not running on this host.$ >> %SYSTEM-F-DUPLNAM, duplicate name( >> FREJA> pipe sho sys|sear sys$pipe ntpI >> 39A0021F TCPIP$NTP_1     LEF     10  1234883   0 00:00:16.03       564y	 >> 204  Na4 >> FREJA> pipe tcpip sho service | sear sys$pipe ntp> >> NTP                  123  UDP      TCPIP$NTP        0.0.0.0
 >> Enabled >> o >> What am I missing?  >> aD >IIRC (and this is documented...) you cannot run NTPDATE if the NTP  >service is running.H >Try @SYS$STARTUP:TCPIP$NTP_SHUTDOWN (or similar name, sorry not near a D >VMS machine right now) and then MC NTPDATE -d <time-server-address> >e >Mikee >  >> n >>>-----Original Message-----g2 >>>From: John Brandon [mailto:brandon@dalsemi.com]+ >>>Sent: Friday, September 12, 2003 7:09 AM  >>>To: Info-VAX@Mvb.Saic.Com) >>>Subject: Re: Newbie question about NTPs >>>t >>>d >>>m >>>>Hi, i'm Stefano Borselli. J >>>>In Unix exist command "ntpdate". This command print the difference (inD >>>>seconds) between this machine and a server that running ntpd. In >>>l >>>VMS exist >>>I >>>>"ntpdate" or similar?  >>>> >>>>Thank you. >>>> >>>>Stefano. >>>m >>>$ tcpip show vera >>>eC >>> DIGITAL TCP/IP Services for OpenVMS Alpha Version V5.0A - ECO 3i@ >>> on a Compaq AlphaServer GS160 6/731 running OpenVMS V7.2-1H1 >>>e >>>$ show sym ntp*. >>> NTPDATE == "$SYS$SYSTEM:TCPIP$NTPDATE.EXE"* >>> NTPDC == "$SYS$SYSTEM:TCPIP$NTPDC.EXE"( >>> NTPQ == "$SYS$SYSTEM:TCPIP$NTPQ.EXE"0 >>> NTPTRACE == "$SYS$SYSTEM:TCPIP$NTPTRACE.EXE" >>>$ ntpdate >>>oF >>>Usage:  ntpdate [-d] [-o version] [-p packets] [-q] host [host ...] >>>M, >>>          -d            turn on debugging5 >>>          -o version    NTP version (default is 3)tE >>>          -p packets    number of packets to sample (default is 4)l> >>>          -q            simple query, don't set system time >>>g >>>  >>>t >>>  >>>J*o*h*n B*r*a*n*d*o*n >>>VMS Systems Administrator- >>>firstname.lastname.spam.me.not@dalsemi.com  >>>: >>>---) >>>Incoming mail is certified Virus Free.,= >>>Checked by AVG anti-virus system (http://www.grisoft.com).rC >>>Version: 6.0.512 / Virus Database: 309 - Release Date: 8/19/2003e >>>  >> g >> ---) >> Outgoing mail is certified Virus Free.t= >> Checked by AVG anti-virus system (http://www.grisoft.com). C >> Version: 6.0.512 / Virus Database: 309 - Release Date: 8/19/2003h >> c >l >m >  >--  >w' >New to c.o.vms? allow me to recommend: 7 >http://eisner.encompasserve.org/~rechtman/post_hlp.htmvA >Other useful links at http://eisner.encompasserve.org/~rechtman/cF >---------------------------------------------------------------------F >Usual disclaimer: All opinions are mine alone, perhaps not even that.@ >Mike Rechtman                            *rechtman@tzora.co.il*F >Kibbutz Tzor'a.                          Voice (home):(972)-2-9908337D >   "20% of a job takes 80% of the time, the rest takes another 80%"F >--------------------------------------------------------------------- >t >---' >Incoming mail is certified Virus Free.q; >Checked by AVG anti-virus system (http://www.grisoft.com).YA >Version: 6.0.512 / Virus Database: 309 - Release Date: 8/19/2003n >h ---o& Outgoing mail is certified Virus Free.: Checked by AVG anti-virus system (http://www.grisoft.com).@ Version: 6.0.512 / Virus Database: 309 - Release Date: 8/19/2003   ------------------------------  % Date: Fri, 12 Sep 2003 08:44:11 -0700.# From: "Tom Linden" <tom@kednos.com> & Subject: RE: Newbie question about NTP9 Message-ID: <CIEJLCMNHNNDLLOOGNJIEEGOHPAA.tom@kednos.com>    >-----Original Message-----|0 >From: John Brandon [mailto:brandon@dalsemi.com]) >Sent: Friday, September 12, 2003 8:42 AM_ >To: Info-VAX@Mvb.Saic.Com' >Subject: Re: Newbie question about NTP_ >| > F >> IIRC (and this is documented...) you cannot run NTPDATE if the NTP  >> service is running.J >> Try @SYS$STARTUP:TCPIP$NTP_SHUTDOWN (or similar name, sorry not near a F >> VMS machine right now) and then MC NTPDATE -d <time-server-address> >> r >> Mikee >i >Verified - thanks Mike. > " >$ @SYS$STARTUP:TCPIP$NTP_SHUTDOWN" >TCPIP$NTP server 20E01034 stopped >D >$ ntpdate -q 207.126.97.57a0 >Server 207.126.97.57 (207.126.97.57), stratum 0$ >offset +0.0000000, delay +0.0000000B >No server suitable for synchronization found from those provided.   Why did this not work for you?   FREJA> ntpdate -q 207.126.97.57U/ Server 207.126.97.57 (207.126.97.57), stratum 2r# offset -0.0075878, delay +0.0519035MB Selected server is 207.126.97.57 (207.126.97.57) offset -0.0075878 >3 >v >n >J*o*h*n B*r*a*n*d*o*n >VMS Systems Administrator+ >firstname.lastname.spam.me.not@dalsemi.com  >1 >---' >Incoming mail is certified Virus Free.j; >Checked by AVG anti-virus system (http://www.grisoft.com).@A >Version: 6.0.512 / Virus Database: 309 - Release Date: 8/19/2003s >t ---,& Outgoing mail is certified Virus Free.: Checked by AVG anti-virus system (http://www.grisoft.com).@ Version: 6.0.512 / Virus Database: 309 - Release Date: 8/19/2003   ------------------------------  % Date: Fri, 12 Sep 2003 10:56:03 -0500i( From: brandon@dalsemi.com (John Brandon)& Subject: Re: Newbie question about NTP1 Message-ID: <03091210560317@dscis6-0.dalsemi.com>   $ > >$ @SYS$STARTUP:TCPIP$NTP_SHUTDOWN$ > >TCPIP$NTP server 20E01034 stopped > >  > >$ ntpdate -q 207.126.97.57i2 > >Server 207.126.97.57 (207.126.97.57), stratum 0& > >offset +0.0000000, delay +0.0000000D > >No server suitable for synchronization found from those provided. >   > Why did this not work for you?  5 Probably because 207.126.97.57 is not my time server.  Whereas this one is:   $ ntpdate -q 180.0.34.2 ) Server 180.0.34.2 (180.0.34.2), stratum 3,# offset -0.0013439, delay +0.0264688 < Selected server is 180.0.34.2 (180.0.34.2) offset -0.0013439       J*o*h*n B*r*a*n*d*o*n  VMS Systems Administratort* firstname.lastname.spam.me.not@dalsemi.com   ------------------------------  # Date: Fri, 12 Sep 2003 15:40:14 GMTl4 From: brad@.gateway.2wire.net (Bradford J. Hamilton)& Subject: Re: Newbie question about NTP. Message-ID: <yNl8b.427506$uu5.76887@sccrnsc04>  a In article <WLUnzj467x4x@eisner.encompasserve.org>, young_r@encompasserve.org (Rob Young) writes:w] !In article <03091209554173@dscis6-0.dalsemi.com>, brandon@dalsemi.com (John Brandon) writes:- !  !>> What am I missing? !> 3L !> Not sure... me too... similiar problems... more investigation required... !> s !> $ ntpdate -q 207.126.97.57s !> Socket bind error.k$ !> %SYSTEM-F-DUPLNAM, duplicate name !> c !s !	Must be your stack.  e !o !$ multinet show /versionsN !Process Software MultiNet V4.4 Rev A-X, Digital AlphaStation 500/266, OpenVMS !AXP V7.3-1o !$ !$ ntpdate -q 207.126.97.57n@ !server 207.126.97.57, stratum 2, offset 0.001197, delay 0.11340E !12 Sep 11:03:24 adjust time server 207.126.97.57 offset 0.001197 secd  O This also works fine with TCPWare.  I see the following on VMS 7.3, version 5.1g of TCP/IP Services:r    XXXXXX::BRADH$ tcpip sho version  7   Compaq TCP/IP Services for OpenVMS Alpha Version V5.1g0   on a AlphaStation 255/233 running OpenVMS V7.3  ' XXXXXX::BRADH$ ntpdate -q dev.tummy.comt Socket bind error.C Verify that the NTP server (TCPIP$NTP) is not running on this host.n! %SYSTEM-F-DUPLNAM, duplicate namey   !n !				Rob !   J __________________________________________________________________________A Bradford J. Hamilton                    "All opinions are my own"yK bMradAhamiPltSon-at-coMmcAast.nPeSt     "Lose the MAPS, and replace '-at-' s0                                          with @"   ------------------------------  + Date: Fri, 12 Sep 2003 15:40:49 +0000 (UTC)  From: david20@alpha2.mdx.ac.uk& Subject: Re: Newbie question about NTP) Message-ID: <bjspe1$sf8$1@news.mdx.ac.uk>g  e In article <UUj8b.313625$Ny5.9772186@twister2.libero.it>, "Stefano Borselli" <ibbo@libero.it> writes:d >Hi, i'm Stefano Borselli.G >In Unix exist command "ntpdate". This command print the difference (in K >seconds) between this machine and a server that running ntpd. In VMS existt >"ntpdate" or similar? >  Yes.   Alpha2:ucx sh vern  ?   Compaq TCP/IP Services for OpenVMS Alpha Version V5.3 - ECO 2 4   on a AlphaServer 2100 5/300 running OpenVMS V7.3-1    Alpha2:ntpdate  C Usage:  ntpdate [-d] [-o version] [-p packets] [-q] host [host ...]   *            -d            turn on debugging3            -o version    NTP version (default is 4) C            -p packets    number of packets to sample (default is 4)e<            -q            simple query, don't set system time    1 and from another system not running an NTP server      $ ntpdate -q alpha2d& Server alpha2 (158.94.0.14), stratum 3# offset -0.0040905, delay +0.0270902 9 Selected server is alpha2 (158.94.0.14) offset -0.0040905     
 David Webb VMS and Unix team leader CCSS Middlesex University   >Thank you.e >n	 >Stefano.y >  >o >  >o >g   ------------------------------  + Date: Fri, 12 Sep 2003 15:50:37 +0000 (UTC)t From: david20@alpha2.mdx.ac.uk& Subject: RE: Newbie question about NTP) Message-ID: <bjsq0d$sr7$1@news.mdx.ac.uk>3  _ In article <CIEJLCMNHNNDLLOOGNJIIEGKHPAA.tom@kednos.com>, "Tom Linden" <tom@kednos.com> writes:CD >tcpip needs some works.  By extension and symmetry you would expect >to find something under >tcpip> help >a >This prompted my curiosityi >, >on Tru64 5.1a > / >bash-2.05b# /usr/sbin/ntpdate -q 207.126.97.57eD >server 207.126.97.57, stratum 2, offset 10830.964511, delay 0.05774E >12 Sep 07:34:42 ntpdate[6587]: step time server 207.126.97.57 offsetW >10830.964511 sec  >  >and on VMS7.3 >t  >FREJA> ntpdate -q 207.126.97.57 >Socket bind error.>D >Verify that the NTP server (TCPIP$NTP) is not running on this host." >%SYSTEM-F-DUPLNAM, duplicate name  D The VMS 7.3 system is already an NTP server and is getting it's time" synchronised by other NTP servers.  2 To see what servers it is synchronised against use  	 ntptrace u   eg   Alpha2:ntptrace alpha2; alpha2: stratum 3, offset -0.001953, synch distance 0.05070l: alpha1: stratum 2, offset 0.005999, synch distance 0.02756L ntp2.ja.net: stratum 1, offset 0.005758, synch distance 0.00075, refid 'GPS'  K The implementation of ntpdate on VMS TCPIP Services does not allow it to beaO used when the server is an NTP server. I don't know whether this limitation is tH restricted to VMS and TCPIP Services or is a general feature of ntpdate.M The usual use of ntpdate is to set the time from a remote NTP server when youhI are not running an NTP server. Hence if you could run it on an NTP server : system it would only make sense to run it in query mode ie
 ntpdate -q        
 David Webb VMS and Unix team leader CCSS Middlesex University          & >FREJA> pipe sho sys|sear sys$pipe ntpG >39A0021F TCPIP$NTP_1     LEF     10  1234883   0 00:00:16.03       564  >204  Nm2 >FREJA> pipe tcpip sho service | sear sys$pipe ntp< >NTP                  123  UDP      TCPIP$NTP        0.0.0.0 >Enabled >r >What am I missing?  >o >>-----Original Message-----1 >>From: John Brandon [mailto:brandon@dalsemi.com]p* >>Sent: Friday, September 12, 2003 7:09 AM >>To: Info-VAX@Mvb.Saic.ComG( >>Subject: Re: Newbie question about NTP >> >> >>> Hi, i'm Stefano Borselli.,J >>> In Unix exist command "ntpdate". This command print the difference (inD >>> seconds) between this machine and a server that running ntpd. In >>VMS exist  >>> "ntpdate" or similar?e >>>  >>> Thank you. >>>s >>> Stefano. >> >>$ tcpip show ver >>C >>  DIGITAL TCP/IP Services for OpenVMS Alpha Version V5.0A - ECO 3s@ >>  on a Compaq AlphaServer GS160 6/731 running OpenVMS V7.2-1H1 >> >>$ show sym ntp*I. >>  NTPDATE == "$SYS$SYSTEM:TCPIP$NTPDATE.EXE"* >>  NTPDC == "$SYS$SYSTEM:TCPIP$NTPDC.EXE"( >>  NTPQ == "$SYS$SYSTEM:TCPIP$NTPQ.EXE"0 >>  NTPTRACE == "$SYS$SYSTEM:TCPIP$NTPTRACE.EXE" >>$ ntpdatei >>E >>Usage:  ntpdate [-d] [-o version] [-p packets] [-q] host [host ...]  >>, >>           -d            turn on debugging5 >>           -o version    NTP version (default is 3)eE >>           -p packets    number of packets to sample (default is 4)r> >>           -q            simple query, don't set system time >> >> >> >> >>J*o*h*n B*r*a*n*d*o*n  >>VMS Systems Administratori, >>firstname.lastname.spam.me.not@dalsemi.com >> >>---e( >>Incoming mail is certified Virus Free.< >>Checked by AVG anti-virus system (http://www.grisoft.com).B >>Version: 6.0.512 / Virus Database: 309 - Release Date: 8/19/2003 >> >---' >Outgoing mail is certified Virus Free. ; >Checked by AVG anti-virus system (http://www.grisoft.com).vA >Version: 6.0.512 / Virus Database: 309 - Release Date: 8/19/2003i >    ------------------------------    Date: 12 Sep 2003 11:11:51 -0500+ From: young_r@encompasserve.org (Rob Young)e& Subject: RE: Newbie question about NTP3 Message-ID: <GsblZnmwDwIo@eisner.encompasserve.org>i  _ In article <CIEJLCMNHNNDLLOOGNJIOEGNHPAA.tom@kednos.com>, "Tom Linden" <tom@kednos.com> writes:0( > FREJA> @SYS$STARTUP:TCPIP$NTP_SHUTDOWN! > %TCPIP-I-INFO, service disabledm; > %TCPIP-I-INFO, process TCPIP$NTP_1 (pid=39A0021F) stoppedA& > %TCPIP-I-INFO, logical names deleted; > %TCPIP-I-INFO, image SYS$SYSTEM:TCPIP$NTP.EXE deinstalledE1 > %TCPIP-S-SHUTDONE, TCPIP$NTP shutdown completedm! > FREJA> ntpdate -q 207.126.97.57 1 > Server 207.126.97.57 (207.126.97.57), stratum 2l% > offset -0.0075878, delay +0.0519035mD > Selected server is 207.126.97.57 (207.126.97.57) offset -0.0075878 > F > I am sure there is a good explanation for this, but it is definitely > goofy. >   > 	Yes, as Multinet does not require ntp to be shutdown to query
 	via ntpdate:    $ multinet netc xntp show - Connected to NETCONTROL server on "LOCALHOST"yF < node.site.com Network Control V4.4(10) at Fri 12-Sep-2003 12:05PM-ED ToH <      remote           local      st poll reach  delay   offset    dispI < =======================================================================aH < +137.236.57.250  199.99.99.99    2  512  333 0.05753 -0.013734 0.01459H < *122.125.23.11   199.99.99.99    2  512  377 0.04189 -0.000907 0.00194I < +122.22.135.71    199.99.99.99    2  512  377 0.01532 -0.000363 0.01048oH < +112.142.55.100  199.99.99.99    2  512  337 0.01167  0.001729 0.00089    7 	I rewrote those IPs.  Not giving away my good sources.c   $ ntpdate -q 122.125.23.11@ server 122.125.23.11, stratum 2, offset -0.000279, delay 0.06752E 12 Sep 12:08:54 adjust time server 122.125.23.11 offset -0.000279 secm   				Robn   ------------------------------   Date: 12 Sep 03 18:33:16 +0200) From: p_sture@elias.decus.ch (Paul Sture)s& Subject: Re: Newbie question about NTP) Message-ID: <63Zvudk$Y6$f@elias.decus.ch>t  \ In article <03091209092453@dscis6-0.dalsemi.com>, brandon@dalsemi.com (John Brandon) writes: >> Hi, i'm Stefano Borselli.I >> In Unix exist command "ntpdate". This command print the difference (inhM >> seconds) between this machine and a server that running ntpd. In VMS existl >> "ntpdate" or similar? >> U
 >> Thank you.a >>   >> Stefano.  >  > $ tcpip show ver > C >   DIGITAL TCP/IP Services for OpenVMS Alpha Version V5.0A - ECO 3-@ >   on a Compaq AlphaServer GS160 6/731 running OpenVMS V7.2-1H1 >  > $ show sym ntp*". >   NTPDATE == "$SYS$SYSTEM:TCPIP$NTPDATE.EXE"* >   NTPDC == "$SYS$SYSTEM:TCPIP$NTPDC.EXE"( >   NTPQ == "$SYS$SYSTEM:TCPIP$NTPQ.EXE"0 >   NTPTRACE == "$SYS$SYSTEM:TCPIP$NTPTRACE.EXE" > $ ntpdate  > E > Usage:  ntpdate [-d] [-o version] [-p packets] [-q] host [host ...]b > , >            -d            turn on debugging5 >            -o version    NTP version (default is 3)nE >            -p packets    number of packets to sample (default is 4) > >            -q            simple query, don't set system time >   D It's probably worth mentioning that those (and other TCP/IP related); commands are not there by default. They are set up with the6A following command, which I normally put in either my login or thej system wide one.  $ $ @sys$startup:tcpip$define_commands   ------------------------------    Date: 12 Sep 2003 12:21:00 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler)l& Subject: Re: Newbie question about NTP3 Message-ID: <O+ar5Iz1TqgT@eisner.encompasserve.org>E  e In article <UUj8b.313625$Ny5.9772186@twister2.libero.it>, "Stefano Borselli" <ibbo@libero.it> writes:  > Hi, i'm Stefano Borselli.wH > In Unix exist command "ntpdate". This command print the difference (inL > seconds) between this machine and a server that running ntpd. In VMS exist > "ntpdate" or similar?f  C    Like all things TCP/IP, it depends on what stack you're running.hH    Multinet includes the rdate command, which has similar functionality.   ------------------------------   Date: 12 Sep 2003 00:35 CDTk' From: carl@gerg.tamu.edu (Carl Perkins)nL Subject: Re: OpenVMS clusters (Alpha) - standalone workstation joined to the- Message-ID: <12SEP200300355076@gerg.tamu.edu>   7 Jeffrey Green <pleasereply@thisnewsgroup.com> writes...p@ }The message I saw when I booted up said that it could not load H }Net-app-sup-200 (during boot but I don't remember the exact wording on G }the message) and when I went into single user mode it said that I did o }not have a cluster license. } H }Unfortunately the room these machines are in (basically a storage room I }without a door) is nowhwere near a PC with internet access.  I will try  F }to grab the errors out of operator.log and post them.  I was unaware I }that the satellites needed a cluster license, and I did try a mod_units yI }on the net-app-sup-200 on the cluster master from 1050 to 4200 just for e }grins and no change.P }  }Jeffrey Green wrote:h  5 You should have a separate NAS license for each node.h9 Each of them should be assigned to the specific node thaty@ it is for via LIC MOD/INCLUDE (or in SYS$UPDATE:VMSLICENSE.COM).= Ditto for the VMS license. (Actually, if each system is usingr> its own separate license database I'm not certain that this is necessary, but it can't hurt.)  A If I recall correctly, you have multiple systems that you want toaA boot from their own individual system disks local to that system,iB and they were all running as standalone systems before adding them to the cluster.:  @ Were licenses loaded onto each system when running as standaloneB systems prior to adding them to the cluster? If not, they wouldn'tD have allowed much to be done on them then either. If so, each systemE should have its own license database on its own system disk. What you ? say above sounds like each system may be using the same licensem& database instead of its own local one.  E Are you certain each system is still booting from its own system diskeG and not from the disk of a cluster boot server now? If they are booting-B as satellite nodes using a system disk on a boot server that could7 explain why they don't see the right license databases.   D An alternative explanation is that you were premature or overzealous@ in defining logical names in SYLOGICALS.COM on each node. If youA defined LMF$LICENSE in there to point to a license database othera? than the local one but did not add each node's licenses to thatn: database, then you could also have this licensing problem.   --- Carl   ------------------------------  + Date: Fri, 12 Sep 2003 16:13:16 +0000 (UTC)u, From: lewis@mazda.mitre.org (Keith A. Lewis), Subject: Re: Question on running GnuPG 1.2.0. Message-ID: <bjsras$9l9$1@newslocal.mitre.org>  l healyzh@NOaracnetSPAM.com writes in article <bjrgpc02agj@enews4.newsguy.com> dated 12 Sep 2003 04:07:08 GMT:K >I got the GnuPG 1.2.0 kit from HP and have finally gotten it running after N >rebuilding it from the source.  That is it will run from the SYSTEM account. D >What privileges are needed for a user account to be able to run it?  L The PGP I built back in the day did not require any special privileges.  AllI it does is read and write files and some heavy number crunching, right?  c  G Are you getting an error when you run it under a normal user account?  0 If so, post it.n  + --Keith Lewis              klewis$mitre.orgs> The above may not (yet) represent the opinions of my employer.   ------------------------------    Date: 12 Sep 2003 06:21:43 -07001 From: susan_skonetski@hotmail.com (Sue Skonetski)rK Subject: Reminder, the OpenVMS Advanced Technical Boot Camp is now 50% fulld= Message-ID: <857e9e41.0309120521.7e81324a@posting.google.com>0  ? If you are planning on attending you may want to register soon.e  & http://www.hp.com/go/openvms/bootcamp   
 Warm Regards,S sue    ------------------------------    Date: 12 Sep 2003 02:04:46 -0700& From: theo.platt@ftid.com (Theo Platt)F Subject: Re: Runtime linking of Java to AWT graphics libraries problem= Message-ID: <7c9ad507.0309120104.33b3bee0@posting.google.com>3   > F > This problem was fixed a short time back, and the fix will be in the@ > next 1.4.1 kit.  The new kit is expected to be available soon.    # Thanks Jeff - I'll look out for it.-   Theo   ------------------------------  % Date: Fri, 12 Sep 2003 10:55:12 +0200:3 From: "Thomas Kratz" <ThomasKratz@REMOVEwebCAPS.de>v< Subject: TCPIP 5.1 SMTP alternate gateway on different port?2 Message-ID: <3f618a4f.0@juno.wiesbaden.netsurf.de>  L SGksDQoNCmlzIGl0IHBvc3NpYmxlIHRvIHNwZWNpZnkgYSBwb3J0ICE9IDI1IGZvciB0aGUgYWx0L ZXJuYXRpdmUgZ2F0ZXdheSBlbnRyeSBpbiBTTVRQIGNvbmZpZ3VyYXRpb24/IElmIHllcywgaG93L Pw0KDQpJIHRyeSB0byB1c2UgYSBub24gdHJhbnNwYXJlbnQgU01UUC1wcm94eS4NCg0KVGhhbmtz DQpUaG9tYXM=   ------------------------------  % Date: Fri, 12 Sep 2003 09:18:28 +01001( From: "John Travell" <john@jomatech.com>! Subject: Re: VMS website updated.$9 Message-ID: <bjrvgp$mfqkt$1@ID-120847.news.uni-berlin.de>N  : "Jonathan Boswell" <jsb@ost.cdrh.fda.gov> wrote in message* news:3F60F4D1.53B8969F@ost.cdrh.fda.gov... > John Travell wrote:nE > > I am well aware how this group are not representative of the vast 
 ma[jo]rity of G > > internet users, but I had not fully appreciated just how much a few) vocal 9 > > members seem to have theirs heads buried in the sand.  > J > OK, that is such an outrageous statement you have brought me out of lurk mode.oL > For the record, I might run some up-to-date browser on my VMS systems if IE > really felt like it.  But most of the time I browse on Windows WITHe
 JAVASCRIPTG > TURNED OFF.  I rarely use NS or IE because it is so difficult to turna	 scripting L > back on again when some braying jackass forces me into it just to navigate their I > obnoxious website.  So usually I use Opera, where the Quick Preferencess pulldown4 > menu allows me easily to turn scripting on or off.  H if you are running on billyware you might take a look at Firebird. It isD easy to disable javascript, blocks unrequested windows, and with theD 'adblock' add-in is quite effective at cutting out much of the crap.  I > Here are the facts which you have not acknowledged.  In addition to thec security > riskL There really is NO security risk, as long as you have not been stupid enough& to use billyware's 'profile assistant'  K > and risk to privacy, some of us find the typical uses of javascript to be H > so annoying, we turn it off for that reason alone.  What do you use it for?J > Maybe a mouse-over is not so bad, though I have seen people try to mouse overL > the text made visible by this event, only to find the text disappear.  Not good.r >tL > Also, pop-up animated-GIF banner ads, or worse still pop-under banner ads, are F > widely loathed by all.  You tell even clueless newbies that they can	 eliminatefK > these obnoxious things by turning off javascript, and off it goes without 9 > further ado and will probably never get turned back on.   : Firebird blocks all these, without turning off javascript.  J > And then there was the form popup window that I couldn't dismiss without fillingsK > in a valid email address...  The morons who programmed that behavior werePE > permanently blacklisted, nor did they get any email address from mem	 because I H > used the task manager to kill my brain-dead browser (which was IE as I recall).  ? Ah, you are using billyware...  (so do I some of the time :-( )7  H > Or how about taking away my status/URL text in favor of some scrollingK > nonsense?  Ha!  Does anybody really care to see something scrolling theren wherehH > their status line should be?  I don't think so.  Or the drop-down menu thatK > launches something (you know not what) because scripting selects whateverd youaK > were on without the use of a submit button.  Then you try the back button  onlyK > to find that you can't go back due to some damnable meta refresh tag withe zero2 > wait.  These things make steam come out my ears.   Yes, they annoy me too..  L > But my "favorite" scripting blunder is the website that requires scriptingJ > enabled to navigate beyond the front page.  But when you enable it, your browserrH > crashes!  Don't you love it when that happens?  (It has happened to me+ > repeatedly with various versions of NS4.)3  K But all variants of NS4 are a bag of shite, the later flavors are just less$ so.nI NS6 and later, browsers based on Mozilla, are a different kettle of fish. L I find Firebird to be particularly good (as far as anything on billyware can	 be good!)   $ > Can you recover the customers lostL > when their browser crashed because of your crufty js?  Can you *guarantee* thatK > your command of scripting is so immaculate that it won't crash somebody's H > browser?  Of course not; you already admitted that you don't use older	 browsers,s > so how could you know?  5 Ah but I do, just not every variant of every browser.   : > Just because YOU want to go with the bleeding edge, waitH > impatiently for 6.0.1.01.005C and download it instantly at 3AM when it becomes L > available, along with the patch du jour for gross security flaws triggeredJ > sometimes simply by reading your email...  Just because you are that way doesK > NOT mean that anybody else feels the same intensity about the technology.: TheyH > don't.  To many of us, it is just a tool.  And millions stick with the browsertE > that they first got with their Internet service, never bothering toe upgrade even > once.-  G You misread me, I tend to be quite cautious about what software I allowI loose on any of my machines.A If millions stay put, there are hundreds of millions that do not.   K > Bottom line: You should not script nor mark up your web documents for anyaJ > particular user agent, configured in any specific way, because that willL > alienate some fraction of your customers who will go elsewhere rather thanJ > struggle to give you their business.  What fraction of your prospects do you0K > want to alienate?  5%?  10%?  How many will find that your script doesn't  workG > for them, or does something they didn't expect or don't want?  Do you  reallyF > have any idea how many customers you are alienating?  No you do not.  G I hope not to alienate anyone, I just got a bit ratty with a few people K apparently saying that all use of javascript is bad and giving me grief fordK even using it at all. I DO try to accomodate those people who prefer not tocF use it, and intend to impove still further that support for browsers I cannot test it on.D I have had some useful feedback, and have fixed a few problems sinceI starting this thread. There are more still to fix. Thank you to all those.* who have contributed, whatever the manner.   -- John Travell" Independent VMS crashdump analyst. john- at - jomatech - dot - comT +44-(0)23-92552229 http://www.jomatech.com/           ---a& Outgoing mail is certified Virus Free.: Checked by AVG anti-virus system (http://www.grisoft.com).A Version: 6.0.515 / Virus Database: 313 - Release Date: 01/09/2003    ------------------------------  % Date: Fri, 12 Sep 2003 09:59:38 +0100 O From: Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com>.! Subject: Re: VMS website updated.P0 Message-ID: <bjs1tr$9ve$1@new-usenet.uk.sun.com>   Larry Kilgallen wrote:f > In article <bjo2kv$kr94l$1@ID-120847.news.uni-berlin.de>, "John Travell" <john@jomatech.com> writes: > ; >>"Paul Repacholi" <prep@prep.synonet.com> wrote in messagea) >>news:87ekyojypo.fsf@prep.synonet.com...e >>- >>>"John Travell" <john@jomatech.com> writes:> >>>c >>>:J >>>>If anyone who has any of those other browsers can bother to look at my >>>e >>site >>+ >>>>with them, I would appreciate feedback.  >>> C >>>Works fine with Opera 7.11 with js enabled. Why not use ordinary A >>>links? Touting the security of VMS then using Javascript seemss) >>>somewhat contarian to say the least :)r >>>  >>/ >>Why? MY JavaScript is not a security problem!V >  > J > But _you_ are not running that JavaScript.  You are asking others to run > it on _their_ machines.A > I > The big security problem with Java, JavaScript, ActiveX etc. is not theEH > nature of the programming language but the notion of mobile code.  TheG > alleged sandbox environments provided by browsers are _not_ adequate.n    @ Provide references for Java that show your assertion to be true.  A Almost all the Java vunerabilites have been problems not with the)@ way the sandbox is designed but the way it has been implimented.    > This is CERTS advice probably a better source than some of the" opinions circulated in this group.   "Should I disable Java applets?a  F The risk associated with Java applets is significantly different from E some of the other technologies. Java has a robust security mechanism wD designed to deal with situations like these that prevents sensitive J information from being disclosed or client information from being damaged.  G However, Java applets written by an attacker can still be loaded while 5H your are viewing a legitimate web page. The problems that can arise are H similar to those involving the <FORM> and other HTML tags. For example, B an attacker could develop a "Trojan Horse" program that presented I misleading information and prompted you for a password. If you failed to -G recognize the malicious applet for what it was, you could accidentally e  disclose sensitive information."  F This is CERTS advice, the second paragraph in also applies to anything! that uses <FORM> in an HTML page.3       Regardsa Andrew Harrison    ------------------------------    Date: 12 Sep 2003 07:32:28 -0500- From: Kilgallen@SpamCop.net (Larry Kilgallen)-! Subject: Re: VMS website updated. 3 Message-ID: <rBTUqc$f4qun@eisner.encompasserve.org>m   In article <bjs1tr$9ve$1@new-usenet.uk.sun.com>, Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com> writes: > Larry Kilgallen wrote:  0 >>>Why? MY JavaScript is not a security problem! >>   >> SK >> But _you_ are not running that JavaScript.  You are asking others to run  >> it on _their_ machines. >>  J >> The big security problem with Java, JavaScript, ActiveX etc. is not theI >> nature of the programming language but the notion of mobile code.  TheqH >> alleged sandbox environments provided by browsers are _not_ adequate. >  > B > Provide references for Java that show your assertion to be true. > C > Almost all the Java vunerabilites have been problems not with theaB > way the sandbox is designed but the way it has been implimented.  C And that does not matter at all.  _Theoretically_ it is possible ton: write perfect software.  In _practice_ it does not happen.   ------------------------------  # Date: Fri, 12 Sep 2003 14:08:54 GMT@2 From: "Ken Farmer" <KFarmer@NOSPAM.SpyderByte.com>! Subject: Re: VMS website updated.s5 Message-ID: <Wrk8b.63$XA5.5@twister.southeast.rr.com>g  L Starting to feel like you walked into a PETA meeting wearing a fur coat?  :)   kf   -- Kenneth Farmer  <><d SpyderByte.com    EnterpriseUnix.org  |  Tru64.org OpenVMS.org  |  dcl.OpenVMS.org)$ EnterpriseLinux.org  |  LinuxHPC.org   ------------------------------   Date: 12 Sep 2003 01:02 CDTf' From: carl@gerg.tamu.edu (Carl Perkins)o; Subject: Re: Why does MAIL DIR go slow during mail receive?-- Message-ID: <12SEP200301025336@gerg.tamu.edu>a   bleau@umtof.umd.edu writes...:K }I checked this other user's mail file and noticed a message in the NEWMAILeL }folder that corresponds to the time of the freeze.  A DIR/FULL showed me itI }had 45 records and was stored in an external file.  This user's MAIL.DIRd  }file size is about 1800 blocks. } L }Fact: For "large" mail messages, VMS MAIL stores the message in an externalI }file and stores a pointer to that file.  Also, when creating a file in a4L }directory that has many files, sometimes VMS needs to reshuffle the entries }in the .DIR file.  I }What the theory does not explain is why users other than the user who is9L }receiving the message and whose MAIL.DIR file is being reshuffled should be
 }affected.  E }So, does it sound that the theory is correct?  If it is, what is thesG }missing part: why does it affect other users?  Is there some lock MAIL0I }takes out?  Or is this just a cpu or i/o intensive operation that causes$H }the other user to be scheduled much later? (that's stretching it, imho)  ( }And,is there a way to fix this problem?   }Lawrence Bleau   M It can be worse than just reshuffling. All .DIR files are contiguous. Always.@I So if the new file created to hold the mail message requires the MAIL.DIRcK file to need to grow and that growth pushes it over the currently allocated J size, then there are two things that can happen. The first one is that theF cluster of blocks (or possibly the default extension number of blocks,F rounded up to full clusters) on the disk immediately following the endG of the MAIL.DIR file is not allocated, in which case it should (I hope)eD allocate them and then do the writing of the new record and probablyE shuffling the contents of the file as well. If, however, those blocks>E are already allocated, a new chunk of disk big enough to hold the new J larger MAIL.DIR file is located and then allocated and the entire MAIL.DIRM file is copied to this location (it may be efficient enough to simultaneously7E add the new record and shuffle the rest back into the file at the newDH positions, or it may not be which requires an additional add and shuffleG after the copy) - this is all done at rather high priority and can keep>E the disk pretty busy seeking back and forth until the file is copied.]  D The best solution is probably a faster disk, or collection of disks.E You might check to see what disk caching you are using as well, none,aC VIOC, or XFC - VIOC is better than none and XFC (if you have it - I B cut the part of the message that gives your VMS version) is betterA than that. If you are not at the latest version of VMS, upgradingtE could also help as then have made improvements in the I/O processing.iF If you have a multiprocessor system, and are running V7.3-1 setting upE "fastpath" might help a little (it may offload some of the processinga- for this sort of thing from the primary CPU).   G Still, a delay of more than a few seconds does seem excessive. The disk/E wouldn't happen to be an MSCP served disk on another node served over  the network, would it?   --- Carl   ------------------------------  % Date: Fri, 12 Sep 2003 10:15:06 +0100s* From: Nic Clews <sendspamhere@[127.0.0.1]>; Subject: Re: Why does MAIL DIR go slow during mail receive?-' Message-ID: <bjs2nm$145$1@lore.csc.com>-   Lawrence Bleau wrote:  > $ [cut problem description and theory]  2 After a little pondering, a few things to look at.  H Have you checked a MONITOR FILE to see the effectiveness or otherwise of the file caches?> Is systemwide locking exhibiting normal or a slowed behaviour?@ Are the nonpaged pool and paged pool significantly expanded (and fragmented)?? What is the level of file fragmentation, particularly directorye fragmentation?  D You could consider ACP settings in SYSGEN and disk mount qualifiers.  H Justifying this would only result in a long woffly and theoretical post.E I believe DFU can be used to compress directories that may also help.t  G Have you any opportunity to create a new mail file in an "archive" area > with the old (and majority) of the messages in that 1800 block
 directory?   -- m? Regards, Nic Clews a.k.a. Mr. CP Charges, CSC Computer SciencesP nclews at csc dot comH   ------------------------------   End of INFO-VAX 2003.506 ************************[mailto:brandon@dalsemi.com]) >Sent: Friday, September 12, 2003 8:42 AM_ >To: Info-VAX@Mvb.Saic.Com' >Subject: Re: Newbie question about NTP_ >| > F >> IIRC (and this is documented...) you cannot run NTPDATE if the NTP  >> service is running.J >> Try @SYS$STARTUP:TCPIP$NTP_SHUTDOWN (or similar name, sorry not near a F >> VMS machine right now) and then MC NTPDATE -d <time-server-address> >> r >> MikPORT 147,162,156,46,200,346 >>> 200 Port 200.34 at Host 147.162.156.46 accepted.  <<< PORT 147,162,156,46,200,346 >>> 200 Port 200.34 at Host 147.162.156.46 accepted.  <<< PORT 147,162,156,46,200,346 >>> 200 Port 200.34 at Host 147.162.156.46 accepted.  <<< PORT 147,162,156,46,200,346 >>> 200 Port 200.34 at Host 147.162.156.46 accepted.  <<< PORT 147,162,156,46,200,346 >>> 200 Port 200.34 at Host 147.162.156.46 accepted.  <<< PORT 147,162,156,46,200,346 >>> 200 Port 200.34 at Host 147.