INFO-VAX Thu, 04 Jan 2007 Volume 2007 : Issue 7 Contents: Re: Current nodename nomenclatures DECNet over IP pointers Re: DECNet over IP pointers Re: DECterm special escape sequences ? Re: HP wins new VMS contract with justice dept? Re: HP wins new VMS contract with justice dept? Re: Quorum disk removal Re: Relationship between node numbers, alloclass etc Re: US Military bans HTML in emails Re: What method(s) to connect to VMS from Mac OS X through X11? Re: What method(s) to connect to VMS from Mac OS X through X11? Re: What method(s) to connect to VMS from Mac OS X through X11? Re: What method(s) to connect to VMS from Mac OS X through X11? Re: ZX6000 server with hung system firmware Re: ZX6000 server with hung system firmware Re: ZX6000 server with hung system firmware Re: ZX6000 server with hung system firmware ---------------------------------------------------------------------- Date: Wed, 03 Jan 2007 15:44:25 -0800 From: "Tom Linden" Subject: Re: Current nodename nomenclatures Message-ID: On Wed, 03 Jan 2007 07:37:17 -0800, Jilly wrote: > > wrote in message > news:emjfcl$der$1@south.jnrs.ja.net... >> In article >> = , >> norm.raphael@metso.com writes: >>> This is a multipart message in MIME format. >>> --=3D_alternative 004A14048525724D_=3D >>> Content-Type: text/plain; charset=3D"US-ASCII" >>> >>> david20@alpha2.mdx.ac.uk wrote on 12/23/2006 06:27:55 AM: >>> >>>> In article , JF Mezei >>>> writes: >>>> >Since I am adding yet another node to my cluster, the time has com= e = >>>> for >>> one >>>> >of the most difficult tasks for humanity: finding a nodename :-) >>>> > >>>> >With VMS 8.3 on alpha and 7.3 on VAX, I take it SCSNODENAME and = >>>> DECNET >>> node >>>> >names are still restricted to 6 characters ? >>>> > >>>> The SCSNODE SYSGEN parameter is still restricted to 6 characters >>>> >>>> From an ALPHA VMS 8.3 >>>> >>>> SYSGEN> HELP SYS_PARAMETER SCSNODE >>>> >>>> SCSNODE >>>> >>>> SCSNODE specifies the name of the computer. This parameter i= s = >>>> not >>>> dynamic. >>>> >>>> Specify SCSNODE as a string of up to six characters. Enclose= = >>>> the >>>> string in quotation marks. >>>> >>>> NOTE >>>> >>>> The maximum size of six characters is strictly enforced. >>>> SYSBOOT truncates the value of SCSNODE if the size of the= >>>> system parameter is set to more than six characters. >>>> >>>> >>>> >Would multinational characters be permitted in nodenames now ? >>>> > >>>> >>>> I don't think so. >>>> >>>> >>>> >score: >>>> > 2 Microvax IIs (owned by me in 1989) >>>> > 1 Vaxstation 3100-30 >>>> > 2 VAX 4000-200 >>>> > 1 VAX 4000-600 >>>> > 2 Alpha DS10Ls >>>> > >>>> >I am sure there are hobbyists with many more machines than that. B= ut >>> that >>>> >is one new machine (for me) on average every 2.1 years ! >>>> >>>> Just pick a naming scheme that includes a large number of relativel= y >>> short >>>> names. My systems are currently named after elements :- >>>> >>>> So the VMS compatible node names could include all of >>>> >>>> HELIUM >>>> SODIUM >>>> BARIUM >>>> RADIUM >>>> IRON >>>> OSMIUM >>>> COBALT >>>> NICKEL >>>> COPPER >>>> SILVER >>>> GOLD >>>> ZINC >>>> BORON >>>> CARBON >>>> INDIUM >>>> TIN >>>> LEAD >>>> OXYGEN >>>> IODINE >>>> NEON >>>> ARGON >>>> XENON >>>> RADON >>>> CERIUM >>>> ERBIUM >>>> CURIUM >>>> >>>> >>>> I think it unlikely I'll ever have that many VMS systems at home. >>> >>> If you used the one- or two-character symbols for the elements, >>> you would not need to restrict the list :-) . >>> >> True but then you might just as well go the whole hog and name them >> >> AA >> AB >> AC >> . >> . >> . >> ZY >> ZZ >> >> >> :) >> >> >> Other popular naming schemes tend to be >> >> Gods and creatures from Mythology >> >> Major and Minor planets and Moons >> >> Animals eg Lion, Bear, Tiger etc >> >> Trees eg PINE, ELM , OAK, BIRCH , LARCH, WILLOW etc >> (Though I think that might be a bit restrictive for the OP since I do= n't >> think >> there are that many trees with names of 6 characters or less). >> >> David Webb >> Security team leader >> CCSS >> Middlesex University >> > > Here at the CSC we have a fondness for beer ;*) thus most all of our = = > node > names are easy to remember ;*) > > Unless your fondness for beer causes you to forget:-) -- = Using Opera's revolutionary e-mail client: http://www.opera.com/mail/ ------------------------------ Date: Wed, 3 Jan 2007 11:49:13 -0800 From: DeanW Subject: DECNet over IP pointers Message-ID: <3f119ada0701031149x488b4d75n395b6bd902dce85e@mail.gmail.com> I'm having a little DECNet over IP issue that has me baffled. Given four machines (all with DECnet Plus) at two sites , connected by VPN: SUBNET 192.168.1.0/24 Node1: DECnet 1.3 - IP ...3 Node2: Decnet 1.5 - IP ...5 SUBNET 192.168.2.0/24 NodeA: DECnet 1.2 - IP ...2 NodeB: DECnet 1.4 - IP ...4 Node2 is fairly new. Appropriate users are proxied across nodes as desired. think I broke the configuration on NodeA when I was trying to remember how to set it up to work over IP. Name service is just (Local, Registry). Right now, everything works, except NodeA, which used to work and now doesn't work at all. I'm assuming I've munged up the configuration when I was poking at it. What I get when I try to look at another host is: NodeA $ DIR Node1:: %DIRECT-E-OPENIN, error opening Node1::*.*;* as input -RMS-E-FND, ACP file or directory lookup failed -SYSTEM-F-UNREACHABLE, remote node is not currently reachable Ideally, it'd be easiest to blow off all of Node A's DECnet configuration and start over- I've done a NET$CONFIGURE ADVANCED and selected "perform an entire configuration", to no avail- but I'm sure I don't know all of what needs to be done, and at this point I'm afraid I'd make things worse if I missed something. Help? -- Dean Woodward =o&o dean.woodward@gmail.com ------------------------------ Date: Thu, 04 Jan 2007 03:13:54 GMT From: John Santos Subject: Re: DECNet over IP pointers Message-ID: DeanW wrote: > I'm having a little DECNet over IP issue that has me baffled. Given > four machines (all with DECnet Plus) at two sites , connected by VPN: > > SUBNET 192.168.1.0/24 > Node1: DECnet 1.3 - IP ...3 > Node2: Decnet 1.5 - IP ...5 > > SUBNET 192.168.2.0/24 > NodeA: DECnet 1.2 - IP ...2 > NodeB: DECnet 1.4 - IP ...4 > > Node2 is fairly new. Appropriate users are proxied across nodes as > desired. think I broke the configuration on NodeA when I was trying to > remember how to set it up to work over IP. Name service is just > (Local, Registry). Right now, everything works, except NodeA, which DECnet-plus-over-IP requires the Domain name service, IIRC. (though using the IP address instead of the node name *might* work.) I don't have DECdns, so that might change things, but I have (Local,Domain) for directory services. (This is the first question asked by NET$CONFIGURE.COM under Option 2 (change naming information), just to be sure we're on the same page.) HTH. > used to work and now doesn't work at all. I'm assuming I've munged up > the configuration when I was poking at it. What I get when I try to > look at another host is: > > NodeA $ DIR Node1:: > %DIRECT-E-OPENIN, error opening Node1::*.*;* as input > -RMS-E-FND, ACP file or directory lookup failed > -SYSTEM-F-UNREACHABLE, remote node is not currently reachable > > Ideally, it'd be easiest to blow off all of Node A's DECnet > configuration and start over- I've done a NET$CONFIGURE ADVANCED and > selected "perform an entire configuration", to no avail- but I'm sure > I don't know all of what needs to be done, and at this point I'm > afraid I'd make things worse if I missed something. > > Help? > -- John Santos Evans Griffiths & Hart, Inc. 781-861-0670 ext 539 ------------------------------ Date: Thu, 28 Dec 2006 06:17:43 GMT From: Roger Ivie Subject: Re: DECterm special escape sequences ? Message-ID: On 2006-12-28, JF Mezei wrote: > There really should be some central "official" document that lists all the > escape sequences supported by DECterm above and beyond what a VT200 dupports. I hear ya. I've fought a few futile battles with DECterm myself. I see occasional references to a document called "DECterm Programming Release Notes", part number EK-DTERM-RN-001, but have not encountered a copy of the document. I see a couple of folks who say they've been looking through the PuTTY source code for escape sequences because PuTTY tends to support the DEC sequences. I don't know if this particular sequence is among them (in my case, I was interesting in particulars of its ReGIS implementation, so things like PuTTY and xterm wouldn't help). Looks like xterm using ^[[2t for to minimize and ^[[1t to unminimize. I don't know if they'll work for DECterm; I don't have one handy at the moment. I got the xterm sequence from: http://rtfm.etla.org/xterm/ctlseq.html -- roger ivie rivie@ridgenet.net ------------------------------ Date: 3 Jan 2007 19:40:17 GMT From: bill@cs.uofs.edu (Bill Gunshannon) Subject: Re: HP wins new VMS contract with justice dept? Message-ID: <502f51F1dicm0U1@mid.individual.net> In article <1167852612.067466.192700@i12g2000cwa.googlegroups.com>, bob@instantwhip.com writes: > it says the os being used isn't mentioned, so it has to be vms ... > > http://www.theinquirer.net/default.aspx?article=36691 > And here we go again!!! "to provide workstations, desktops and laptops" And where exactly do you see VMS in this? Looks like another Windows win with a little Linux thrown in just for good measure. "Prism Pointa will provide services" A quick Google search returns nothing called "Prism Pointa". bill -- Bill Gunshannon | de-moc-ra-cy (di mok' ra see) n. Three wolves bill@cs.scranton.edu | and a sheep voting on what's for dinner. University of Scranton | Scranton, Pennsylvania | #include ------------------------------ Date: 3 Jan 2007 23:36:30 GMT From: bill@cs.uofs.edu (Bill Gunshannon) Subject: Re: HP wins new VMS contract with justice dept? Message-ID: <502svuF1d3fc2U1@mid.individual.net> In article <12poemarjhejf75@corp.supernews.com>, "C.W.Holeman II" writes: > "Bill Gunshannon" wrote in message > news:502f51F1dicm0U1@mid.individual.net... >> In article <1167852612.067466.192700@i12g2000cwa.googlegroups.com>, >> bob@instantwhip.com writes: >>> it says the os being used isn't mentioned, so it has to be vms ... >>> >>> http://www.theinquirer.net/default.aspx?article=36691 >>> >> >> And here we go again!!! >> >> "to provide workstations, desktops and laptops" >> >> And where exactly do you see VMS in this? Looks like another Windows >> win with a little Linux thrown in just for good measure. >> >> "Prism Pointa will provide services" >> >> A quick Google search returns nothing called "Prism Pointa". > > Google "GTSI Prism" produces a number of matches. I guess we can assume the "Pointa" in the Inquireer article was a typo. No surprise. But, to get back to Boob's original claim. Looks like no chance of VMS having anything to do with this deal either. No surprise there either. bill -- Bill Gunshannon | de-moc-ra-cy (di mok' ra see) n. Three wolves bill@cs.scranton.edu | and a sheep voting on what's for dinner. University of Scranton | Scranton, Pennsylvania | #include ------------------------------ Date: Wed, 3 Jan 2007 10:23:49 -0500 From: "Jilly" Subject: Re: Quorum disk removal Message-ID: <459bca86$1@usenet01.boi.hp.com> "Jeff Goodwin" wrote in message news:459aecce$0$2264$4c368faf@roadrunner.com... > We have a two node cluster with a quorum disk and we're about to add a > third node. I want to eliminate the quorum disk. I'd like to do this as > a rolling reboot as this is a 7x24 environment. I'd like to modify > DISK_QUORUM (and EXPECTED_VOTES when complete) on a node and reboot it. > When I'm done with all nodes, will the quorum disk be entirely forgotten > by all cluster members? I'm a little leery as I think that once finished, > all the nodes will be looking for the other nodes to supply the quorum > disk votes. > > Has anyone removed a quorum disk via a rolling reboot method? > > -Jeff > Yes it will be forgotten after all nodes that once had DISK_QUORUM set are rebooted. ------------------------------ Date: Wed, 3 Jan 2007 10:23:00 -0500 From: "Jilly" Subject: Re: Relationship between node numbers, alloclass etc Message-ID: <459bca54$1@usenet01.boi.hp.com> "JF Mezei" wrote in message news:c76d9$458e9c26$cef8887a$1921@TEKSAVVY.COM... > It is well known that the > SCSSYSTEMID = DECNET AREA*1024 + DECNET NODE NUMBER. > > > But the rest seems to be at the discretion of the system manager. > > Is is worth the trouble of trying to synchronise the DECNET node number > with the TCPIP address ? > > (eg: decnet node 1.10, TCPIP would be xx.xx.xx.10 ?) > > Is it also worth trying to synchronise the ALLOCLASS with a node number > for each node ? > > (eg: if you are decnet node 10, your alloclass would also be 10 ?) > > > > Also, in a cluster, the oldest node is the one to let a booting node in or > not. And it remembers nodes that have left the cluster. > > If I change the SCSYSTEMID on node VELO and then reboot that node, will it > be allowed back into the cluster ? > > What about if I keep the SCSYSTEMID the same, but change the SCSNODE ? > > > Is it correct to state that allocation class doesn't matter for cluster > membership, it only is used to name devices ? (eg: I can change alloclass > and then reboot - the existing nodes will see both the old devices (marked > unavailable) and the new ones with the new alloclass ? There are many possible ALLOCLASS schemes and prefrences vary. Where most folks get into trouble is when implementing Port Allocation Classes or PAC. I have always advocated the following setup for ALLOCLASS when using PAC 1) If any node in a cluster needs PAC then set all the cluster nodes with PACs 2) Each node is assigned a base number that is a multiple of 10 3) Set all HSx allcation classes to 256 in a decreasing fashion 4) If the node does not serve HSx based devices then the SYSGEN ALLOCLASS = BN otherwise the SYSGEN ALLOCLASS = HSx allocation class for which it will be a disk server 5) Identify all multi host SCSI busses and number the individual busses from 2 to 9 6) All SCSI ports connected to a multihost bus will have their PAC = bus number 7) Assign a port number to all other system port where DQA=1, DQB=2, PKA=3, PKB=4 etc. 8) Set ports PAC to base number + port number This should make management of cluster devices much easier when you have to use port allocation classes. Unfortunately you cannot change ALLOCLASS or PAC and have the other systems in the cluster magically change thier UCBs to reflect this nodes changes. In some cases you will get new UCBs (thus new devices) that may or may not work and in some cases you won't get new UCBs and the original ones may or may not work. Any change in any allocation class in the cluster requires a cluster reboot (roling reboot is fine). ------------------------------ Date: 4 Jan 2007 03:49:16 GMT From: bill@cs.uofs.edu (Bill Gunshannon) Subject: Re: US Military bans HTML in emails Message-ID: <503bpsF1ed599U1@mid.individual.net> In article <1167873822.299476.210410@s34g2000cwa.googlegroups.com>, "AEF" writes: > Bill Gunshannon wrote: >> In article <1167832240.731568.243390@a3g2000cwd.googlegroups.com>, >> "AEF" writes: >> > Arne Vajhøj wrote: >> >> AEF wrote: >> >> > Arne Vajhøj wrote: >> >> >> David J Dachtera wrote: >> >> >>> Bill Gunshannon wrote: >> >> >>>> Not in the real world. Good enough doesn't cut it when someone higher >> >> >>>> than you says, "The corporate standard is MS Word." >> >> >>> ...until the(ir next) multi-billion-dollar outage due to malware. Then, the >> >> >>> higher-ups face turn-over while the workers bees burn the midnight oil to clean >> >> >>> up the mess. >> >> >> There were some real bad incidents 3-5 years ago. >> >> >> >> >> >> No move from Windows then. >> >> >> >> >> >> Now the MIS departments has tightened security. >> >> > >> >> > You mean like requiring 6-character passwords to now be "complex"? >> >> > Yeah, that'll stop 'em!!! ;-) >> >> > >> >> > OK, maybe they're actually doing some more useful things. >> >> >> >> More as in: >> >> min 8 characters >> >> min 1 uppercase >> >> min 1 lowercase >> >> min 1 digit >> >> min 1 punctuation >> >> >> >> Or as in email scanners that removes all EXE, BAT etc. from attached >> >> ZIP files in inbound email. >> >> >> >> Arne >> > It turns out that you get a lot more bang for the buck by requiring >> > longer passwords. >> >> Not sure where it is in the approval process but we talked about >> allowing entire sentences (paragraphs?) as passwords and eliminating >> the need for non-alpha characters. >> >> > Complex passwords are not that much harder to crack. >> >> Not if you are in a position to use brute-force. Most computers today >> have all the horsepower needed to just try every possible character. >> After all, there are less than 128. :-) > Not sure what you're saying here. You open with 'Not', then seem to > agree with me. I was agreeing with you. I was stressing that there are a max of 128 characters and if you have the encrypted string so that you can try every possible combination against it, even complex passwords can be broken. (Think distributed cracking on thousands of machines!) > Anyway, according to the article referenced, anything 10 > or more chars long, NOT complex, is all but uncrackable. I find that hard to believe. The amount of computing resources available today is mind boggling. Look at the size of the cluster used to render the graphics for the movie "Ice Age". Or, what about all those machines doing SETI@Home. And what about all the machines that have been infected with the virus dujour. What if instead of spreading SPAM they were all coordinated to break a password? > I think he should have offered $1000 instead of $100, though. >> >> > Most characters will be lowercase. >> >> Not always. :-) > True, but you don't need to crack everyone's password. >> >> > Puncutation will almost certainly be >> > limited to periods, hyphens, and commas. > Arne listed punctuation as a required character, so I was commenting on > that. Well, that was probably a slight mis-statement. We can blame language as I don't think English is his first. :-) What the chart should say is: * It must be at least 10 characters * It must contain at least 2 special characters: !@#$%^&*_-+=':;., * It must contain at least 2 numbers * It must contain at least 2 uppercase and 2 lowercase letters * It must not be one of your last 10 passwords. * It is case sensitive >> >> I do not now and have never (up to this point, at least) used any of >> those characters. > You'd have to if the password software enforced it according to the > rules given by Arne above! See actual rules just above. >> > This greatly reduces the total >> > number of possible combinations comapred to a random character for each >> > character. Hackers already know this trick. >> >> The biggest safeguard is not providing a way to use a brute force attack. >> Unfortunately, this brings up another whole issue. Use of brute force >> to cause a DOS. > Well, I assume here that the hacker has somehow obtained a backup tape > or something similar. Well, I also assume that one can't quickly try > millions of combinations over the network, especially if your target > has break-in evasion set up. Thus my comment about DOS. >> >> > If you sit down and calculate it, you'll find that complex passwords >> > aren't worth the trouble (I'll post some numerical examples later when >> > I have more time). Some say that users will write down passwords >> > anyway. (So why lock your door? Burglars will get in anyway!) I say >> > *more* users will write down complex passwords and they'll hate it a >> > lot more than adding a few characters to the minimum length. >> >> I have no statistics to back it up, but I don't agree. I think >> people who are likely to write down their password are as likely >> to do it no matter what the length. Unless you let them use >> something that is easily guessed anyway. > Well, it's certainly not going to reduce the number of people who write > down passwords! >> > You'll >> > probably also get more help desk calls for complex-password resets. >> > Here's an article from infoworld (I can't find the original articles >> > right now, I'll post them later). >> > http://www.infoworld.com/archives/emailPrint.jsp?R=printThis&A=/article/06/11/10/46OPsecadvise_1.html >> > I've seen passwords with zeros for O's and 3's for E's. What hacker >> > could break through this fortress of security? Trying zeros for O's and >> > 3's for E's? What hacker ever think of that? >> >> With your toungue pushed so far into your cheek be careful you don't >> bite it. :-) > Or choke on it!!! > Yeah, I overdid it. I was thinking of a Seinfeld shtick in his book, > "Sein Languange" where he talks about the stupid security things people > do, like park in NYC with a TV in the back seat and cover it with a > sweater. "It's just a sweater, except that it's square and has an > antenna sticking out of it", he says. Or putting your wallet down by > the toes in your sneakers when you go to the beach. "What criminal mind > can penetrate this fortress of security?" he says. Etc. > Additionally, I admit that my posts this morning were poorly written. I > was in a hurry and perhaps need to catch up on sleep! >> >> > This is like a mild speed >> > bump where as increasing the length a few characters is more like huge >> > mountain. It's like putting your wallet in the toe of your sneaker as >> > you go into the water at the beach. Yes, it incerases the total >> > possible number of passwords, but not by much. See the link I provided >> > for more detail. >> >> Thus the reason for the suggestion above. I hope it gets approved soon. > On further thought, many people would capitalize the first letter and > use a number for the last char. That actually *reduces* the possible > number of combinations! > It's x**L, and L is your friend. Don't understand this. The suggestion I was talking about was allowing really long passwords like sentences. Even given that they would be made up of dictionary words, I think they would be very difficult to break. Of course, the better way is no password. Using my CAC is the safest way. Then you have to steal my CAC, know what username it actually works with and have access to a machine that will accept it. bill -- Bill Gunshannon | de-moc-ra-cy (di mok' ra see) n. Three wolves bill@cs.scranton.edu | and a sheep voting on what's for dinner. University of Scranton | Scranton, Pennsylvania | #include ------------------------------ Date: Thu, 04 Jan 2007 01:52:38 GMT From: VAXman- @SendSpamHere.ORG Subject: Re: What method(s) to connect to VMS from Mac OS X through X11? Message-ID: <00A61301.913F5D6F@SendSpamHere.ORG> In article <1167839953.107057.69930@n51g2000cwc.googlegroups.com>, "johnhreinhardt@yahoo.com" writes: > > >Okay all you Mac OS X and VMS fans. What is your way of connecting the >two using X11 on OS X to DECWindows/CDE on VMS? Do you use XDM or >fix up a set of scripts/DCL procedures or something else? > eXodus and ssh -- VAXman- A Bored Certified VMS Kernel Mode Hacker VAXman(at)TMESIS(dot)COM "Well my son, life is like a beanstalk, isn't it?" ------------------------------ Date: Thu, 04 Jan 2007 02:58:54 GMT From: John Santos Subject: Re: What method(s) to connect to VMS from Mac OS X through X11? Message-ID: sean@obanion.us wrote: > johnhreinhardt@yahoo.com wrote: > >>Okay all you Mac OS X and VMS fans. What is your way of connecting the >>two using X11 on OS X to DECWindows/CDE on VMS? Do you use XDM or >>fix up a set of scripts/DCL procedures or something else? > > > I have not tried to DCL script this, yet, but these are the manual > steps I use. > > For X windows, the free but optionaly installed X11 application ( > http://www.apple.com/macosx/features/x11/ ) is doing well with the > DECWindows VUE$MASTER (FileView). But geting the ssh access working > was not obvious. Since I had a hard time finding all of this in any > one place, I figured I better document it here... > > 1) Enable X11 TCP ports in System Preferences under Sharing in the > Firewall tab by adding a new entry with TCP Port Numbers 6000-6999. I > named my new entry "X Windows" > > Or disable the OS X Firewall, but that's not recomended. > > 2) Enable the X11 TCP listen function on OS X from the shell command > line using: > %defaults write com.apple.x11 nolisten_tcp -boolean false > % defaults read com.apple.x11 > "NSWindow Frame x11_apps" = "439 500 486 332 0 0 1280 832 "; > . > . > . > "nolisten_tcp" = 0; > > You should only need to do this once. This can be done from the X11 > xterm window that automatically opens when you launch X11. > > 3) Launch the X11 application, which by default is installed in the > Utilites folder, an xterm window will automatically open, and use xhost > > to enable access: > % xhost + > > So far I've had to do this each time I launch X11, but I expect there > is a scriptable way... > > 4) Connect through ssh enabling X 11 tunneling: > %ssh -X @ xhost already added> > > This strongly implies that ssh v2 is enabled on the VMS system in > question. > > 5) Under VMS, use the Set Display command as expected: > $ set disply/create/transport=tcpip/node= name that matches the xhost already added> > I believe the /node= IP address should be the address (or name) of the Mac, not the VMS host specified in the xhost command in step 3. (cut & pasted from step 4 without changing it?) > All the instructions that I found say the using ssh with -X will > automaticaly set the X window server (which in this case is the OS X > system: remember that X windows has the graphics display at the server) > > envronment variable or something to that effect, but not on VMS... > > And I hear somebody asking: "If X is coming back through ssh, why open > the TCP X windows ports?" > I don't know... > > 6) At this point I use: > $ mcr vue$master > because FileView lets me create subprocesses for DCL windows and the > DECWindows EVE editor, among others. > I tried decw$clock, and it popped up nicely. I should save this post for future reference. Thanks. > > Hope this helps! > > > Sean > -- John Santos Evans Griffiths & Hart, Inc. 781-861-0670 ext 539 ------------------------------ Date: 3 Jan 2007 19:51:18 -0800 From: "johnhreinhardt@yahoo.com" Subject: Re: What method(s) to connect to VMS from Mac OS X through X11? Message-ID: <1167882676.036553.92240@51g2000cwl.googlegroups.com> VAXman-@SendSpamHere.ORG wrote: > In article <1167839953.107057.69930@n51g2000cwc.googlegroups.com>, "johnhreinhardt@yahoo.com" writes: > > > > > >Okay all you Mac OS X and VMS fans. What is your way of connecting the > >two using X11 on OS X to DECWindows/CDE on VMS? Do you use XDM or > >fix up a set of scripts/DCL procedures or something else? > > > > eXodus and ssh > > -- > VAXman- A Bored Certified VMS Kernel Mode Hacker VAXman(at)TMESIS(dot)COM > > "Well my son, life is like a beanstalk, isn't it?" What about this program made you feel paying for it was worthwhile? The price to me would be $270 which is higher than I'd like, but less than I've paid for other programs. I don't mind if it's worth it and the company support is good. John H. Reinhardt ------------------------------ Date: Wed, 3 Jan 2007 21:49:21 -0600 (CST) From: sms@antinode.org (Steven M. Schweda) Subject: Re: What method(s) to connect to VMS from Mac OS X through X11? Message-ID: <07010321492190_2020028F@antinode.org> sean@obanion.us wrote: > to enable access: > % xhost + > > So far I've had to do this each time I launch X11, but I expect there > is a scriptable way... "man xhost". Just put the names of the hosts who should be allowed to access X display "n" into "/etc/Xn.hosts" (one name per line). "/etc/X0.hosts" is the most likely candidate. "ssh -X" seems to be effective, too. ------------------------------------------------------------------------ Steven M. Schweda sms@antinode-org 382 South Warwick Street (+1) 651-699-9818 Saint Paul MN 55105-2547 ------------------------------ Date: Wed, 03 Jan 2007 11:05:40 -0800 From: "Tom Linden" Subject: Re: ZX6000 server with hung system firmware Message-ID: On Wed, 03 Jan 2007 10:35:59 -0800, Malcolm Dunnett wrote: > "Robert Deininger" wrote in message > news:rdeininger-0201072203590001@dialup-4.233.149.88.dial1.manchester1.level3.net... >> >> I don't know of any simple way for you to fix a corrupted firmware ROM >> on >> the ZX6000. (If your system actually has a corrupted ROM, and not some >> other problem. The diagnostics can be misleading. In particular, if >> the >> monarch (primary) CPU doesn't come out of its reset state and talk to >> the >> Baseboard Management Controller (BMC), it can look like a bad FW ROM. > > I've tried swapping the primary and secondary CPUs and mixing and > matching > their power modules. No luck. I've tried different memory and a new power > supply. I've removed all the PCI cards. The results are always the same. > > The Standby LED is on, the BMC led is flashing but the FW led is off. > > The last event in the event log says: No events were received from > system > firmware > Data: FRB2/Hang in POST failure > > >> If you are offered an Integrity server for sale, please check that it >> will >> at least boot to EFI. If it won't, don't expect it to be easy to fix >> without specialized tools or expensive parts. > > Well this is an Ebay purchase , you pays your money and you takes your > chances :-) That's why I use Paypal and buy the insurance. > > -- Using Opera's revolutionary e-mail client: http://www.opera.com/mail/ ------------------------------ Date: Thu, 04 Jan 2007 03:33:06 GMT From: rdeininger@mindspringdot.com (Robert Deininger) Subject: Re: ZX6000 server with hung system firmware Message-ID: In article , "Tom Linden" wrote: >On Wed, 03 Jan 2007 10:35:59 -0800, Malcolm Dunnett > wrote: > >> "Robert Deininger" wrote in message >> news:rdeininger-0201072203590001@dialup-4.233.149.88.dial1.manchester1.level3.net... >>> >>> I don't know of any simple way for you to fix a corrupted firmware ROM >>> on >>> the ZX6000. (If your system actually has a corrupted ROM, and not some >>> other problem. The diagnostics can be misleading. In particular, if >>> the >>> monarch (primary) CPU doesn't come out of its reset state and talk to >>> the >>> Baseboard Management Controller (BMC), it can look like a bad FW ROM. >> >> I've tried swapping the primary and secondary CPUs and mixing and >> matching >> their power modules. No luck. I've tried different memory and a new power >> supply. I've removed all the PCI cards. The results are always the same. >> >> The Standby LED is on, the BMC led is flashing but the FW led is off. >> >> The last event in the event log says: No events were received from >> system >> firmware >> Data: FRB2/Hang in POST failure The odds of corrupted firmware seem higher than the odds of both CPUs failing. So corrupted firmware seems the most likely cause. But 1 bad CPU can affect the other. This is a long shot, not a common failure mode. (It's vaguely possible mismatched CPUs could do this also.) Have you tried removing the secondary CPU completely? If that still doesn't work, swap CPUs and try again. I don't know if any of these systems had the ROM in a socket. If so, you might be able to re-program it outside of the system. Alas, most Integrity servers have soldered ROMs by the time they go into production. ------------------------------ Date: Thu, 04 Jan 2007 05:07:39 GMT From: "Malcolm Dunnett" Subject: Re: ZX6000 server with hung system firmware Message-ID: "Robert Deininger" wrote in message news:rdeininger-0301072233540001@dialup-4.233.128.51.dial1.manchester1.level3.net... | The odds of corrupted firmware seem higher than the odds of both CPUs | failing. So corrupted firmware seems the most likely cause. | | But 1 bad CPU can affect the other. This is a long shot, not a common | failure mode. (It's vaguely possible mismatched CPUs could do this | also.) Have you tried removing the secondary CPU completely? If that | still doesn't work, swap CPUs and try again. Yep, I've tried single CPU configurations with each of the CPUs and I've tried both power modules on the single CPU, no difference. I thought of the "both CPUs dead" but I agree that seems less likely than bad firmware. If I go into command mode and do a DF command it finds all the components. I'm thinking it shouldn't call out the CPU if it was bad or if its power module was no good. ------------------------------ Date: Thu, 04 Jan 2007 05:42:37 GMT From: rdeininger@mindspringdot.com (Robert Deininger) Subject: Re: ZX6000 server with hung system firmware Message-ID: In article , "Malcolm Dunnett" wrote: >"Robert Deininger" wrote in message >news:rdeininger-0301072233540001@dialup-4.233.128.51.dial1.manchester1.level3.net... > >| The odds of corrupted firmware seem higher than the odds of both CPUs >| failing. So corrupted firmware seems the most likely cause. >| >| But 1 bad CPU can affect the other. This is a long shot, not a common >| failure mode. (It's vaguely possible mismatched CPUs could do this >| also.) Have you tried removing the secondary CPU completely? If that >| still doesn't work, swap CPUs and try again. > > Yep, I've tried single CPU configurations with each of the CPUs and I've >tried both power modules on the single CPU, no difference. I thought of the >"both CPUs dead" but I agree that seems less likely than bad firmware. If >I go into command mode and do a DF command it finds all the components. I'm >thinking it shouldn't call out the CPU if it was bad or if its power module >was no good. No, DF will find components even if they are mostly dead. ------------------------------ End of INFO-VAX 2007.007 ************************