INFO-VAX Thu, 11 Jan 2007 Volume 2007 : Issue 22 Contents: DECW$STARTUP RESTART oddity ES45 versus ES47 Re: ES45 versus ES47 Re: Hopeless HP support infrastructure Re: Hopeless HP support infrastructure JAVA sanity check please Re: MONITOR X-axis change Re: MONITOR X-axis change Moving from Bind 8 to Bind 9 Re: Moving from Bind 8 to Bind 9 Re: Need help setting IP address on DS20 (VMS 7.3) Re: Need help setting IP address on DS20 (VMS 7.3) Re: Need help setting IP address on DS20 (VMS 7.3) Re: On-shoring of VMS support?? Re: On-shoring of VMS support?? OT: Example of software development delays/failure Re: Star Trek and $CONFIGURE GALAXY Re: Strategy for common DECNET proxy files in cluster ? Re: VMS732_UPDATE-V0900 fails Would anyone have a copy of the Availability Manager V2.5 ? ---------------------------------------------------------------------- Date: Thu, 11 Jan 2007 08:04:19 -0500 From: JF Mezei Subject: DECW$STARTUP RESTART oddity Message-ID: <45a63555$0$17837$c3e8da3@news.astraweb.com> This has happened to be a couple of times: Alpha DS10L, Radeon 7500 card, VMS 8.3 For whatever reason, I want to restart the dec windows server. So I go to a VT terminal, login to that node, type @DECW$STARTUP RESTART, answer "YES" to confirm and it does it trick. As the server restarts, it displays the "grey" patterned screen, followed by the login screen. However, the login screen does not fill the whole display, AND the mouse pointer is displayed but not functional. Keyboard is functional. Solution is to kill the server process (and any other decw processies that might still be around), then get into the console mode, login and then type the @DECW$STARTUP from the workstation keyboard. ------------------------------ Date: 11 Jan 2007 08:01:45 -0800 From: "jbigboote" Subject: ES45 versus ES47 Message-ID: <1168531305.301331.216830@p59g2000hsd.googlegroups.com> >From what I've read it looks like there is no clear winner in performance between the ES45 and ES47; specifically with a 4 CPU configuration (assming 1.25GHz and 1.15GHz respectively), and 8-16GB RAM. Of course the ES45s are cheaper than the ES47s, and cost is a consideration. Does the performance change at all when they are clustered (two or three nodes)? How about long-term supportability? Any reason to think the ES47s might have a longer supported lifespan than the ES45s? ------------------------------ Date: Thu, 11 Jan 2007 11:21:02 -0500 From: "Island Computers, D B Turner" Subject: Re: ES45 versus ES47 Message-ID: <12qcovkthg1ae79@news.supernews.com> Oops I meant $4000 for the memory not $400! -- Island Computers US Corp 2700 Gregory St Savannah GA 31404 Tel: 912 447 6622 x201 Mail: dturner-atnospam-islandco-com (You know what to do with the dashes) "Island Computers, D B Turner" wrote in message news:12qcocl9eu3j8bd@news.supernews.com... > Look at memory pricing on these as well > > a 4GB kit for the ES45 is about $400 > The ES47 is about $7200 (special pricing) > > DT > -- > Island Computers US Corp > 2700 Gregory St > Savannah GA 31404 > Tel: 912 447 6622 x201 > Mail: dturner-atnospam-islandco-com > (You know what to do with the dashes) > "jbigboote" wrote in message > news:1168531305.301331.216830@p59g2000hsd.googlegroups.com... > > >From what I've read it looks like there is no clear winner in > > performance between the ES45 and ES47; specifically with a 4 CPU > > configuration (assming 1.25GHz and 1.15GHz respectively), and 8-16GB > > RAM. Of course the ES45s are cheaper than the ES47s, and cost is a > > consideration. Does the performance change at all when they are > > clustered (two or three nodes)? > > How about long-term supportability? Any reason to think the ES47s might > > have a longer supported lifespan than the ES45s? > > > > ------------------------------ Date: 11 Jan 2007 07:57:12 -0600 From: Kilgallen@SpamCop.net (Larry Kilgallen) Subject: Re: Hopeless HP support infrastructure Message-ID: In article , clubley@remove_me.eisner.decus.org-Earth.UFP (Simon Clubley) writes: > UCX is acting as the primary Internet facing SMTP server for my network, > and spammers are periodically sending spam with MAIL FROM: lines of the > form username@open (ie, without a qualified domain). > > UCX is incorrectly adding the local domain while receiving the message, so > that the message appears to come from username@open.my_company.com. > > This means that when the message is bounced, UCX then tries to contact a > non-existant host within my local network and because of the way the > firewall is setup, UCX has to wait for the connection attempt to timeout, > which causes severe backlogs during spam floods. Responding to spam by attempting to send a new message back to the claimed originator is WRONG WRONG WRONG. An SMTP Reject is the only honorable way to do this and avoid pestering innocent victims. Spammers lie. ------------------------------ Date: 11 Jan 2007 08:16:51 -0600 From: clubley@remove_me.eisner.decus.org-Earth.UFP (Simon Clubley) Subject: Re: Hopeless HP support infrastructure Message-ID: In article , Kilgallen@SpamCop.net (Larry Kilgallen) writes: > > Responding to spam by attempting to send a new message back to the > claimed originator is WRONG WRONG WRONG. > > An SMTP Reject is the only honorable way to do this and avoid pestering > innocent victims. > > Spammers lie. I agree. Unfortunately, SMTP Reject appears to only be an option in newer versions of UCX. BTW, I have just been told by HP Canada that adding the local domain onto unqualified domains in incoming email is the expected and designed behaviour. :-( Simon. -- Simon Clubley, clubley@remove_me.eisner.decus.org-Earth.UFP If Google's motto is "don't be evil", then how did we get Google Groups 2 ? ------------------------------ Date: Thu, 11 Jan 2007 08:10:02 -0500 From: JF Mezei Subject: JAVA sanity check please Message-ID: <45a636c5$0$15365$c3e8da3@news.astraweb.com> Could someone go to http://www.dslreports.com/tools , select the speed test tool, then select the java based tool, then select any site and try to run the java based tool (javascript and java need to be enabled) ? Can anyone succeed in making it run on a VMS 8.3 with JAVA 1.5.0 and Mozilla 1.7.13 ? For me, the java app starts (the java logo appears, but then I get the X on upper left corner an an indication that the applet failed. The last remaining "hurdle" was removed when I moved the user disk over to an ODS5 structure. The java software resided on an ODS5 system disk already. Going to a site to test my java version results in a succesful Java applet execution which tells me my version etc. Also, I seem to be unable to get the java console to come up. (either right key on the java applet area in the HTML window, or using one of the menu optiosn in "TOOLS" at top of screen). If one person can run this speed test with a similar software config, then i know there is something wrong with my system. But if other people are also unable to run this app, then I know it isn't my fault. ------------------------------ Date: 11 Jan 2007 07:42:18 -0600 From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) Subject: Re: MONITOR X-axis change Message-ID: In article <1168456073.209495.15040@o58g2000hsb.googlegroups.com>, "Ian Miller" writes: > MONITOR on recent versions of VMS was re-written in another language so > it can run on I64 and they can make enhancements to it. Gee that sounds like a waste of time and money compared to what Kednos would want for a compiler license. ------------------------------ Date: 11 Jan 2007 12:32:02 -0600 From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) Subject: Re: MONITOR X-axis change Message-ID: In article , John Reagan writes: > > I think most folks overestimate the amount of PL/I in MONITOR. All I know for sure is that early VMS 3.x releases would product %PLI- error messages. Later on the same messages got the MONITOR facility name. ------------------------------ Date: Thu, 11 Jan 2007 09:11:27 -0500 From: JF Mezei Subject: Moving from Bind 8 to Bind 9 Message-ID: Going from VAX to Alpha brings one from TCPIP Services Bind8 to a newer Bind9 implementation with many new features. The manual discusses moving from UCX to Bind9 (eg: convert the config files), but I didn't see stuff from Bind8 to Bind 9. The server options file TCPIP$BIND.CONF gets many new options. Many of them are documented in the manual. Do the actual zone files remain the same with no new options ? Or does the updated VMS software now support more record types (such as the geographical location of a server, not supported on VAX (forget the record type for those, but I recall trying them and it failed). Does anyone know if different views are working on the TCPIP Services 5.6 ? (this would allow the DNS server to serve different answers depending on whether the request comes from within the lan or from the outside world for instance). Any hints ? recommendations ? any warnings ? gotchas for such a move ? ------------------------------ Date: 11 Jan 2007 08:55:03 -0800 From: davidc@montagar.com Subject: Re: Moving from Bind 8 to Bind 9 Message-ID: <1168534503.484435.312510@k58g2000hse.googlegroups.com> JF Mezei wrote: > Going from VAX to Alpha brings one from TCPIP Services Bind8 to a newer > Bind9 implementation with many new features. Here are some entries you will want to include in your tcpip$bind.conf configuration: logging { category lame-servers { null ; } ; } ; There are lots of "lame" servers out there, and if you don't turn off lame-servers logging, you'll get lots of activity on OPCOM, etc. Also, there is a DDOS that takes advantage of DNS recursive queries. There's a CERT about it at http://www.us-cert.gov/reading_room/DNS-recursion033006.pdf You will only want to allow recusion by your local network, so add lines like these (note, this is from my tcpip$bind.conf file): // list your internal and own IP address ranges acl internal { 10.0.0.0/8 ; 72.64.118.0/24 ; } ; options { directory "SYS$SPECIFIC:[TCPIP$BIND]"; allow-transfer { // List networks allows to zone transfer 72.64.118.0/24 ; 10.0.0.0/24 ; // ... } ; allow-recursion { internal ; } ; } ; If you want an "internal" DNS server to handle local names on your internal IP address range, you will either want to run a different name server for that, or set up "views". ------------------------------ Date: 11 Jan 2007 08:44:20 +0100 From: peter@langstoeger.at (Peter 'EPLAN' LANGSTOEGER) Subject: Re: Need help setting IP address on DS20 (VMS 7.3) Message-ID: <45a5f8e4@news.langstoeger.at> In article , Z writes: >I inherited a DS20 ... I was warned that I would not be able to change >the IP address of it, the warning mentioned that others before me had >tried before, but they always just gave up and let it be. Why would a IP address be so hard to change? What if you boot a Tru64 or a new VMS installation? >Anyway, I *need* to change the IP. Check your SRM variables if there is any IP address set. I don't know why one would have done this but just to be sure. >This was originally a 7.1-2 box, and was upgraded to 7.3 some time back. btw: You know that V8.3 is current? >Apparently, perhaps form the upgrade, it's set up in some sort of UCX >compatibility mode - I say that because when I remove the TCP*.DAT and >UCX*.DAT files, and run TCPIP$CONFIG, it creates the TCP*.DAT files >then, also, one UCX dat file. Also, UCX$CONFIG runs TCPIP$CONFIG (after >telling you that it will). Do you mean the (empty) UCX$SERVICE.DAT? >In TCPIP$CONFIG, there is exactly one interface: IE0. (The SRM >identifies the NIC as EIA0 and TCPIP$CONFIG calls it IE0, if that >matters at all.) EIA0 is what, a DE600? Did your old TCPIP support them already? Could well be, but I don't know. What does TCPIP>LIST COMMUNICATION_CONTROLLER tell? Maybe someone defined the EI COMMUNICATION_CONTROLLER per hand and wrong? Compare the EI definitions with the eg. EW >In TCPIP$CONFIG, I set that interface to IP address 10.210.54.67 (which >I've confirmed is not in use on the subnet), and the mask to >255.255.255.0. I plug this system into a hub that also has a PC plugged >in (PC's IP is 10.210.54.66). > >I can ping the DS20 from the DS20 (by its name, by its IP and by its >127.0.0.1 address). > >I cannot ping the PC from the DS20 or the DS20 from the PC (by IP >address). I can ping the other systems in the hub from the PC. > >Have I missed a crucial NIC configuration step? btw: Has the PC a firewall software running which prevents sending an echo response? >Also, I know that on another 7.3 system I worked on, I had to start the >INET driver before I could even see the interfaces with TCPIP SHOW INT. > >But that's not the case on this DS20. Even so, I tried starting INET >and still the pings between the PC and the DS20 all failed. > >Help! What am I doing wrong or forgetting to do? What does the ARP table tell? Any foreign addresses there? And so on What does the routing table tell? Default route there and correct? ... -- Peter "EPLAN" LANGSTOEGER Network and OpenVMS system specialist E-mail peter@langstoeger.at A-1030 VIENNA AUSTRIA I'm not a pessimist, I'm a realist ------------------------------ Date: 11 Jan 2007 07:55:50 -0600 From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) Subject: Re: Need help setting IP address on DS20 (VMS 7.3) Message-ID: <1vYsgBjvcVT7@eisner.encompasserve.org> In article , Z writes: > I inherited a DS20 ... I was warned that I would not be able to change > the IP address of it, the warning mentioned that others before me had > tried before, but they always just gave up and let it be. > > Anyway, I *need* to change the IP. > > This was originally a 7.1-2 box, and was upgraded to 7.3 some time back. > > Apparently, perhaps form the upgrade, it's set up in some sort of UCX > compatibility mode - I say that because when I remove the TCP*.DAT and > UCX*.DAT files, and run TCPIP$CONFIG, it creates the TCP*.DAT files > then, also, one UCX dat file. Also, UCX$CONFIG runs TCPIP$CONFIG (after > telling you that it will). > I'm reading between the lines here, but are you using TCPIP Services for VMS, or a real IP stack? In any case you can change the IP address. How is stack specific. The very worst thing one of the stacks might do is force you to reboot. ------------------------------ Date: Thu, 11 Jan 2007 10:20:43 -0800 From: Z Subject: Re: Need help setting IP address on DS20 (VMS 7.3) Message-ID: david20@alpha1.mdx.ac.uk wrote: > At the >>> what does > show eia*mode > display Aha! eia0_mode FastFD (Full Duplex) The hub is a Netgear EN1906TP ... specs say 10 Mbps. What's the command to change the DS20's NIC to 10MBps in the SRM? ------------------------------ Date: 11 Jan 2007 07:52:45 -0600 From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) Subject: Re: On-shoring of VMS support?? Message-ID: In article <45a566cd$0$137$157c6196@dreader2.cybercity.dk>, "Dr. Dweeb" writes: > That is the difference, the native English English speakers have little > problem with Subcontinentals, Aantipodeans, Americans or Liverpudlians ! > This is a defining characteristic, and distinguishes us from Americans who - > jusdging by the constant flak one sees in the US computer press - have great > difficulties understanding anyone other than themselves. It's not our fault if we can't understand Texans. Folks from India, the UK, and Oz at least try to be understood. ------------------------------ Date: 11 Jan 2007 09:00:09 -0800 From: davidc@montagar.com Subject: Re: On-shoring of VMS support?? Message-ID: <1168534809.591995.273960@i39g2000hsf.googlegroups.com> Richard B. Gilbert wrote: > It has been many years since I was last forced to speak with a Texan but > they can, with a great deal of effort, be understood. The effort is not > usually justified! You just need the Redneck to English Translation Book! If a Texan asks you for some "All", he's not asking for laundy detergent. He wants some "oil". I'm here oil week, try the Veal. ------------------------------ Date: Thu, 11 Jan 2007 07:04:48 -0500 From: JF Mezei Subject: OT: Example of software development delays/failure Message-ID: <45a62763$0$17795$c3e8da3@news.astraweb.com> Westjet started as a low cost carrier in Canada which became the defacto competitor to Air Canada when Canadian Airlines disapeared. They have since grown tremendously and have been coveted by some international carriers to act as a feeder to their flights. To do so, Westjet needs to upgrade their "no frills" low cost reservation system to one which has the ability to cpnnect to global reservation systems and interline with other major airlines. So, it sought out the latest and greatest: http://www.go-aires.com/questions.htm through a company called Travelport which operates some .com name brands such as Orbitz and inherited some "legacy" reservation systems such as Gallileo. The iaRES is touted as being new generation using the latest and greatest tools such as Java etc. Well, guess what ? A couple months before scheduled implementation into production, the promised product isn't ready ! Surprise, surprise. In December, Westjet announced that it had negotiated an extension with its old provider so that they can continue to operate their older no-frill reservation system until the software issues are sorted out. Yesterday, they announced that work on the new system was halted, pending new negotiations with the *HOPE* that it can be rescued. Westjet has already spent $30 million on that project. Also announced is increase in contract with navitaire (the old supplier) to last until at least 2008 which would include some software upgrades. So with good luck, they might be just one year late, but chances are that this project will be abandonned and Westjet will probably seek an existing software solution on "legacy" platforms that work right away. Now, you can look at the narrow "software" issue where the company might end up losing 30 million dollars in a failed project. When you look at SAP for instance, when those project fail, the company just continues to use their old software which works fine. Often SAP was being implemented because someone decided SAP was the way to go for the future, not because there was an immediate need for it. But in the case of Westjet, the failure of that project prevents Westjet from linking up with other carriers before the lucrative summer and postpones probably by a full year such a link up, resulting in loss of revenus, probable re-adjusting of any ad campaign they had set out for this summer, re-adjusting of schedules since there won't be any linkups etc. And it also forces the other airlines with whom Westjet would have linked up to also review their plans. So, here is a good example where going for the latest and greatest is a big risk for software projects with real impact in the company's ability to implement its immediate core business needs. > http://yahoo.reuters.com/misc/PrinterFriendlyPopup.aspx?type=comktNews&storyID=urn%3anewsml%3areuters.com%3a20070110%3aMTFH52496_2007-01-10_20-01-27_N10250815 ------------------------------ Date: 11 Jan 2007 07:40:42 -0600 From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) Subject: Re: Star Trek and $CONFIGURE GALAXY Message-ID: In article <45A5812B.2090002@comcast.net>, John writes: > That may be true however would I endeavor to write GALAXY code one a > single CPU? > I think not - I could only imagine stating to management, well it worked > in test... > It would be an unrealistic test at that - limited testing or career ending? Because it may be cheaper to get you a single CPU to edit, compile, and test with Galaxy capabilities and then finish the test in a multi-CPU Galaxy. Enabling Galaxy enables features that you can't access otherwise, even for a simple test. ------------------------------ Date: 11 Jan 2007 09:25:26 -0800 From: "AEF" Subject: Re: Strategy for common DECNET proxy files in cluster ? Message-ID: <1168536326.449642.24730@p59g2000hsd.googlegroups.com> Joe Bloggs wrote: > On Wed, 10 Jan 2007 03:41:20 -0500, JF Mezei > wrote: > > > >What is the reason or having two files ? > >NET$PROXY.DAT;1 NETPROXY.DAT;1 > > > iirc, net$proxy.DAT is Phase V, netproxy.dat Phase IV. >From AUTHORIZE HELP: CREATE /PROXY Creates and initializes the network proxy authorization file, NETPROXY.DAT. On VAX systems, the primary network proxy authorization file is NET$PROXY.DAT. The file NETPROXY.DAT is maintained for compatibility. NOTE Do not delete NETPROXY.DAT because DECnet Phase IV and many layered products still use it. Format CREATE/PROXY In the System Manager's Manual SECTION 6.9.4 it says: Deleting either NETPROXY.DAT or NET$PROXY.DAT will result in incorrect functioning of proxy access on systems running DECnet Phase IV. Also, many applications such as ALL-IN-1 rely on NETPROXY.DAT. In section 6.9.3 it says that in a mixed-version cluster to use VAX VMS v6.1 or later or Alpha VMS v6.2 or later to ensure that NET$PROXY.DAT is updated with correct proxy information. AEF ------------------------------ Date: 11 Jan 2007 07:44:32 -0600 From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) Subject: Re: VMS732_UPDATE-V0900 fails Message-ID: In article <45A45056.874D645@spam.comcast.net>, David J Dachtera writes: > %PCSI-I-PRCOUTPUT, output from subprocess follows ... > %LIBRAR-F-WRITEERR, error writing DISK$ALPHASYS:[SYS0.SYSCOMMON.][SYSLIB]STARLET > .OLB;1 > -SYSTEM-F-EXQUOTA, process quota exceeded > About the only ones I didn't blow through the roof were WSextent > and Pgflquota. The first quota I would suspect in this case is PGFLQUOTA. ------------------------------ Date: Thu, 11 Jan 2007 12:49:39 +0000 From: "R.A.Omond" Subject: Would anyone have a copy of the Availability Manager V2.5 ? Message-ID: Gentle people, I'm desperately searching for a copy of the Availability Manager V2.5 Installation kits (in particular the Data Analyser). I can't use the latest version (V2.6) because the Java stuff requires VMS 7.3-1 (or higher), and this is to run on a VMS 7.2-2 system. Many thanks for any help on this. Roy Omond Blue Bubble Ltd. ------------------------------ End of INFO-VAX 2007.022 ************************