INFO-VAX Sat, 18 Aug 2007 Volume 2007 : Issue 451 Contents: Re: Fun with bugs Re: Fun with bugs Re: Fun with bugs Re: Fun with bugs Re: Fun with bugs Re: Fun with bugs Re: Fun with bugs Getting sftp to work in batch mode Re: Getting sftp to work in batch mode Re: Getting sftp to work in batch mode Re: incoming SMTP 255 character wrap Re: Intel marginalizing Itanium even faster than expected? Re: Intel marginalizing Itanium even faster than expected? Re: Intel marginalizing Itanium even faster than expected? Re: Intel marginalizing Itanium even faster than expected? Re: Intel marginalizing Itanium even faster than expected? RE: Intel marginalizing Itanium even faster than expected? RE: Intel marginalizing Itanium even faster than expected? RE: Intel marginalizing Itanium even faster than expected? Looking for SEDT source code Re: Looking for SEDT source code Re: Looking for SEDT source code Re: Looking for SEDT source code Re: Looking for SEDT source code Re: Looking for SEDT source code Re: Looking for SEDT source code Re: Looking for SEDT source code SSH login welcome message? Re: SSH login welcome message? RE: SSH login welcome message? Re: Wonderful things happen to an OS when it has an internal champion RE: Wonderful things happen to an OS when it has an internal champion RE: Write locked file ---------------------------------------------------------------------- Date: Fri, 17 Aug 2007 14:25:55 -0400 From: John Reagan Subject: Re: Fun with bugs Message-ID: DeanW wrote: > FUNCTION MyFunc ( > VAR ExtFlag as BOOLEAN ) : BOOLEAN; > > VAR MyFlag : BOOLEAN; > > BEGIN > IF ( ExtFlag = TRUE ) THEN MyFlag := True; > IF NOT ( MyFlag ) THEN BEGIN > {...Do some stuff} > END; > MyFunc := MyFlag; > END; > I have another theory... A recent (ie, in the past 3-4 years) change to GEM for more agressive value propagation could result in this behavior (ie, apparant FALSE on Alpha; apparant TRUE on I64). Dean didn't say how old the Alpha compiler was... If you look at the fetch of MyFlag in the IF control expression, the state of MyFlag is either TRUE (via the THEN part above) or UNDEFINED. Since fetching an undefined variable is illegal (easy for us to say, eh?) the only legal value that MyFlag can hold is TRUE. If you look at the generated code, it is possible that we replaced the fetch of MyFlag with TRUE. -- John Reagan OpenVMS Pascal/Macro-32/COBOL Project Leader Hewlett-Packard Company ------------------------------ Date: Fri, 17 Aug 2007 13:28:15 -0700 From: Doug Phillips Subject: Re: Fun with bugs Message-ID: <1187382495.109957.167630@r23g2000prd.googlegroups.com> On Aug 17, 1:25 pm, John Reagan wrote: > DeanW wrote: > > FUNCTION MyFunc ( > > VAR ExtFlag as BOOLEAN ) : BOOLEAN; > > > VAR MyFlag : BOOLEAN; > > > BEGIN > > IF ( ExtFlag = TRUE ) THEN MyFlag := True; > > IF NOT ( MyFlag ) THEN BEGIN > > {...Do some stuff} > > END; > > MyFunc := MyFlag; > > END; > > I have another theory... A recent (ie, in the past 3-4 years) change to > GEM for more agressive value propagation could result in this behavior > (ie, apparant FALSE on Alpha; apparant TRUE on I64). Dean didn't say > how old the Alpha compiler was... > > If you look at the fetch of MyFlag in the IF control expression, the > state of MyFlag is either TRUE (via the THEN part above) or UNDEFINED. > Since fetching an undefined variable is illegal (easy for us to say, > eh?) the only legal value that MyFlag can hold is TRUE. If you look at > the generated code, it is possible that we replaced the fetch of MyFlag > with TRUE. Assuming the conventional TRUE=1 FALSE=0, why would TRUE be chosen for the UNDEFINED value? It would seem, if TRUE or FALSE are the only choices (UNDEFINED has no representation) that NOT TRUE would be a more reasonable expectation for UNDEFINED; or the compiler would just clear them rather than set them. If the compiler/run-time *is* smart enough to understand UNDEFINED, and an uninitialized variable is tested, then NOT TRUE and NOT FALSE would be true, and TRUE or FALSE would be false. Just curious. ------------------------------ Date: Fri, 17 Aug 2007 16:47:49 -0400 From: John Reagan Subject: Re: Fun with bugs Message-ID: Doug Phillips wrote: > On Aug 17, 1:25 pm, John Reagan wrote: > >>If you look at the fetch of MyFlag in the IF control expression, the >>state of MyFlag is either TRUE (via the THEN part above) or UNDEFINED. >>Since fetching an undefined variable is illegal (easy for us to say, >>eh?) the only legal value that MyFlag can hold is TRUE. If you look at >>the generated code, it is possible that we replaced the fetch of MyFlag >>with TRUE. > > > Assuming the conventional TRUE=1 FALSE=0, why would TRUE be chosen for > the UNDEFINED value? It would seem, if TRUE or FALSE are the only > choices (UNDEFINED has no representation) that NOT TRUE would be a > more reasonable expectation for UNDEFINED; or the compiler would just > clear them rather than set them. > > If the compiler/run-time *is* smart enough to understand UNDEFINED, > and an uninitialized variable > is tested, then NOT TRUE and NOT FALSE would be true, and TRUE or > FALSE would be false. > > Just curious. > I didn't say value, I said state. BOOLEAN variable has 3 states (TRUE, FALSE, and UNDEFINED). Only when the state isn't UNDEFINED can one trust the value in memory. We didn't pick TRUE for UNDEFINED, we picked the only store (albeit a conditional one) earlier in the code. The compiler doesn't know if the variable *IS* undefined at the fetch, only that it *MIGHT* be undefined. -- John Reagan OpenVMS Pascal/Macro-32/COBOL Project Leader Hewlett-Packard Company ------------------------------ Date: Fri, 17 Aug 2007 14:19:32 -0700 From: Doug Phillips Subject: Re: Fun with bugs Message-ID: <1187385572.866590.233810@i13g2000prf.googlegroups.com> On Aug 17, 3:47 pm, John Reagan wrote: > Doug Phillips wrote: > > On Aug 17, 1:25 pm, John Reagan wrote: > > >>If you look at the fetch of MyFlag in the IF control expression, the > >>state of MyFlag is either TRUE (via the THEN part above) or UNDEFINED. > >>Since fetching an undefined variable is illegal (easy for us to say, > >>eh?) the only legal value that MyFlag can hold is TRUE. If you look at > >>the generated code, it is possible that we replaced the fetch of MyFlag > >>with TRUE. > > > Assuming the conventional TRUE=1 FALSE=0, why would TRUE be chosen for > > the UNDEFINED value? It would seem, if TRUE or FALSE are the only > > choices (UNDEFINED has no representation) that NOT TRUE would be a > > more reasonable expectation for UNDEFINED; or the compiler would just > > clear them rather than set them. > > > If the compiler/run-time *is* smart enough to understand UNDEFINED, > > and an uninitialized variable > > is tested, then NOT TRUE and NOT FALSE would be true, and TRUE or > > FALSE would be false. > > > Just curious. > > I didn't say value, I said state. BOOLEAN variable has 3 states (TRUE, > FALSE, and UNDEFINED). Only when the state isn't UNDEFINED can one > trust the value in memory. > Sorry. Please substitute the word "state" for the one appearance of the word "value" in my question. > We didn't pick TRUE for UNDEFINED, we picked the only store (albeit a > conditional one) earlier in the code. I understand now. I've been burned by that in the past. I don't think a compiler should "pick" anything, especially from a conditioned store. Since UNDEFINED is a valid state, then my next-to-last statement is applicable: An UNDEFINED variable is neither TRUE nor FALSE. > > The compiler doesn't know if the variable *IS* undefined at the fetch, > only that it *MIGHT* be undefined. > That's o.k, but something somewhere apparently made an assumption. My first thought when I read the OP was the same as Bob Gezelter's (that doesn't happen too often:-); that it's probably an artifact. You're the guy, though, so I'll believe whatever you say. ------------------------------ Date: Fri, 17 Aug 2007 18:09:06 -0400 From: John Reagan Subject: Re: Fun with bugs Message-ID: Doug Phillips wrote: > I understand now. I've been burned by that in the past. I don't think > a compiler should "pick" anything, especially from a conditioned > store. Since UNDEFINED is a valid state, then my next-to-last > statement is applicable: An UNDEFINED variable is neither TRUE nor > FALSE. Yes, UNDEFINED is a valid state. However, it is illegal to fetch a variable whose state is UNDEFINED. And since illegal programs have undefined behavior, substituting the prior conditional store for the variable fetch is just as good as fetching the random bits from memory (and is faster since it saves a memory fetch :-) ) All legal programs must have gone through the THEN part and placed TRUE into the variable. Why fetch it again if we already know what the value will be? You're > the guy, though, so I'll believe whatever you say. Then leave a bag with small, unmarked bills next to my motorcycle in the Spitbook parking lot... -- John Reagan OpenVMS Pascal/Macro-32/COBOL Project Leader Hewlett-Packard Company ------------------------------ Date: Fri, 17 Aug 2007 15:35:34 -0700 From: Doug Phillips Subject: Re: Fun with bugs Message-ID: <1187390134.016474.16270@q3g2000prf.googlegroups.com> On Aug 17, 5:09 pm, John Reagan wrote: > Doug Phillips wrote: > > I understand now. I've been burned by that in the past. I don't think > > a compiler should "pick" anything, especially from a conditioned > > store. Since UNDEFINED is a valid state, then my next-to-last > > statement is applicable: An UNDEFINED variable is neither TRUE nor > > FALSE. > > Yes, UNDEFINED is a valid state. However, it is illegal to fetch a > variable whose state is UNDEFINED. And since illegal programs have > undefined behavior, substituting the prior conditional store for the > variable fetch is just as good as fetching the random bits from memory > (and is faster since it saves a memory fetch :-) ) All legal programs > must have gone through the THEN part and placed TRUE into the variable. > Why fetch it again if we already know what the value will be? > Ok, I understand. So what if there had been another conditional statement in the code that set it to FALSE? Do you FIFI or LIFO? Do the fetch? ...? Shouldn't there be a compile or run-time error or warning on an illegal operation (or is there one)? (I'm not trying to be a pest; I don't have Pascal, I don't have access to the internals, and I have always been fascinated by the hows and whys of compilers. Must be my BAL beginnings:-) > You're > > > the guy, though, so I'll believe whatever you say. > > Then leave a bag with small, unmarked bills next to my motorcycle in the > Spitbook parking lot... > I don't think I can get even one of the bills into a *small* bag, and then there's that tattoo... :-| ------------------------------ Date: Fri, 17 Aug 2007 21:34:04 -0700 From: Joshua Lehrer Subject: Re: Fun with bugs Message-ID: <1187411644.178201.265940@g4g2000hsf.googlegroups.com> On Aug 17, 6:48 am, FrankS wrote: > On Aug 17, 2:06 am, Joshua Lehrer wrote: > > > > > > > Here is a very simple C++ program that crashes on Itanium and runs > > fine under ALPHA: > > > #include > > > int main(int argc, char **argv) { > > try { > > struct TestTypeInfo { > > ~TestTypeInfo() { > > typeid(int).name(); > > } > > } t; > > throw (int)1; > > } catch (int) { > > } > > > } > > > While unwinding the stack due to an exception you can not attempt to > > fetch the demangled typename of a builtin type/pointer. > > > -J > > I'm pretty sure the Alpha -> Itanium porting notes mention the > (significant) differences in how stack unwinding should be handled.- Hide quoted text - > > - Show quoted text - So? That isn't the issue here. Notice that the exception is caught. Thus, C++ requires the stack based objects to be destroyed, and they are, on both platforms. The difference is that the Itanium typeid system uses exceptions and if you are already handling an exception, you have multiple exceptions, and things can go south real fast. It's just an example of how porting from Alpha to Itanium is not as simple as compile, link, run. -J ------------------------------ Date: Fri, 17 Aug 2007 18:27:38 -0000 From: Ken Robinson Subject: Getting sftp to work in batch mode Message-ID: <1187375258.713156.70330@r29g2000hsg.googlegroups.com> I've read all the previous posts on the subject as well as the posts in the IRTC and I still can't get this to work. I think I know why sftp is so secure -- no one can get it to work. Environment: VMS: HP TCP/IP Services for OpenVMS Alpha Version V5.4 - ECO 6 on a AlphaServer 4X00 5/400 4MB running OpenVMS V7.3-2 Remover Server: HPUX (I think) running SSH-2.0-OpenSSH_4.0 I've sent the public key and the admin on the unix machine converted it to the OpenSSH format Here's part of the output obtained by doing: $ sftp -v "-B" tt: user@remote.host.domain debug: Connecting to remote.host.domain, port 22... (SOCKS not used) debug: Ssh2/SSH2.C:2872: Entering event loop. debug: Ssh2Client/SSHCLIENT.C:1608: Creating transport protocol. debug: SshAuthMethodClient/SSHAUTHMETHODC.C:101: Added "publickey" to usable methods. debug: SshAuthMethodClient/SSHAUTHMETHODC.C:101: Added "password" to usable methods. debug: Ssh2Client/SSHCLIENT.C:1649: Creating userauth protocol. debug: client supports 2 auth methods: 'publickey,password' debug: SshUnixTcp/SSHUNIXTCP.C:1378: using local hostname my.host.domain debug: Ssh2Common/SSHCOMMON.C:546: local ip = xxx.xxx.xxx.xxx, local port = 49328 debug: Ssh2Common/SSHCOMMON.C:548: remote ip = yyy.yyy.yyy.yyy, remote port = 22 debug: SshConnection/SSHCONN.C:2290: Wrapping... debug: Remote version: SSH-2.0-OpenSSH_4.0 debug: OpenSSH: Major: 4 Minor: 0 Revision: 0 debug: Ssh2Transport/TRCOMMON.C:1828: All versions of OpenSSH handle kex guesses incorrectly. debug: Ssh2Transport/TRCOMMON.C:2268: lang s to c: `', lang c to s: `' debug: Ssh2Transport/TRCOMMON.C:2333: c_to_s: cipher aes128-cbc, mac hmac-sha1, compression none debug: Ssh2Transport/TRCOMMON.C:2336: s_to_c: cipher aes128-cbc, mac hmac-sha1, compression none debug: Remote host key found from database. debug: Ssh2Common/SSHCOMMON.C:347: Received SSH_CROSS_STARTUP packet from connection protocol. debug: Ssh2Common/SSHCOMMON.C:397: Received SSH_CROSS_ALGORITHMS packet from connection protocol. debug: server offers auth methods 'publickey,password,keyboard- interactive'. debug: Ssh2AuthPubKeyClient/AUTHC-PUBKEY.C:1682: adding keyfile "/ $2$DKC0/myusername/ssh2/myusername-MY_HOST_DOMAIN" to candidates debug: Ssh2AuthPubKeyClient/AUTHC-PUBKEY.C:1682: adding keyfile "/ $2$DKC0/myusername/ssh2/id_rsa_2048_a" to candidates debug: server offers auth methods 'publickey,password,keyboard- interactive'. debug: server offers auth methods 'publickey,password,keyboard- interactive'. debug: Ssh2AuthClient/SSHAUTHC.C:377: Method 'publickey' disabled. debug: server offers auth methods 'publickey,password,keyboard- interactive'. debug: Ssh2AuthPasswdClient/AUTHC-PASSWD.C:261: In Batchmode, so we're not asking the user for password. debug: Ssh2AuthClient/SSHAUTHC.C:377: Method 'password' disabled. The line that says "debug: Ssh2AuthClient/SSHAUTHC.C:377: Method 'publickey' disabled." is the one that concerns me. Why is it being disabled? How can I debug this further? Why does setting up sftp to use public keys have to be so difficult? :-) Thanks in advance. Ken ------------------------------ Date: Fri, 17 Aug 2007 17:00:20 -0500 (CDT) From: sms@antinode.org (Steven M. Schweda) Subject: Re: Getting sftp to work in batch mode Message-ID: <07081717002092_20200296@antinode.org> From: Ken Robinson > VMS: > HP TCP/IP Services for OpenVMS Alpha Version V5.4 - ECO 6 > on a AlphaServer 4X00 5/400 4MB running OpenVMS V7.3-2 > > Remover Server: > HPUX (I think) running SSH-2.0-OpenSSH_4.0 Around here: alp $ tcpip show version HP TCP/IP Services for OpenVMS Alpha Version V5.4 - ECO 6 on a COMPAQ Professional Workstation XP1000 running OpenVMS V7.3-2 sol# uname -a SunOS sol 5.10 Generic_118822-25 sun4u sparc SUNW,Ultra-60 sol# ssh -V Sun_SSH_1.1, SSH protocols 1.5/2.0, OpenSSL 0x0090704f dy $ uname -a HP-UX dy B.11.11 U 9000/785 2012616114 unlimited-user license dy $ ssh -V OpenSSH_4.3p2-hpn, OpenSSL 0.9.7i 14 Oct 2005 HP-UX Secure Shell-A.04.30.006, HP-UX Secure Shell version > I've sent the public key and the admin on the unix machine converted > it to the OpenSSH format I have (almost) the same stuff in my ".ssh" directory on both my Solaris system and my HP-UX system, including appropriately converted keys from the VMS system. > Here's part of the output obtained by doing: > $ sftp -v "-B" tt: user@remote.host.domain > [... noise ...] > The line that says "debug: Ssh2AuthClient/SSHAUTHC.C:377: Method > 'publickey' disabled." is the one that concerns me. Why is it being > disabled? How can I debug this further? Why does setting up sftp to > use public keys have to be so difficult? :-) In the ssh world, "disabled" is code for "didn't work". First, I'd start with plain ssh, not sftp. Second, it works great here from anywhere to enywhere, _except_ I haven't yet found a way to get into the HP-UX system without a password. With "ssh -vvv sol", the VMS system says: debug: Ssh2AuthPubKeyClient/AUTHC-PUBKEY.C:1920: Public key authentication was s uccessful. HP-UX and Solaris say: debug3: sign_and_send_pubkey debug2: we sent a publickey packet, wait for reply debug1: Authentication succeeded (publickey). However, trying to get into HP-UX: VMS says: debug: server offers auth methods 'publickey,password,keyboard-interactive'. debug: Ssh2AuthClient/SSHAUTHC.C:377: Method 'publickey' disabled. HP-UX and Solaris say: debug3: sign_and_send_pubkey debug2: we sent a publickey packet, wait for reply debug1: Authentications that can continue: publickey,password,keyboard-interacti ve Not having tried this on my "new" HP-UX system until just now, I don't know why it fails, but I'm guessing that it's some configuration problem and/or quirk with the HP-UX sshd. Can anyone from anywhere get into your HP-UX system using public-key authorization, or is it something VMS-specific? (I can do anything _but_ into HP-UX, including from the same HP-UX system.) Wasn't that helpful? ------------------------------------------------------------------------ Steven M. Schweda sms@antinode-org 382 South Warwick Street (+1) 651-699-9818 Saint Paul MN 55105-2547 ------------------------------ Date: Fri, 17 Aug 2007 18:25:00 -0500 (CDT) From: sms@antinode.org (Steven M. Schweda) Subject: Re: Getting sftp to work in batch mode Message-ID: <07081718250083_20200296@antinode.org> > However, trying to get into HP-UX: > [...] Well, _my_ problem was bad permissions on my HP-UX "~/.ssh" directory. Knowing nothing, I killed the sshd, ran it manually with debug ("/usr/sbin/sshd -ddd"), and saw the permissions complaint in the output. (umask = 0 is often not good. Who sets this stuff?) A real HP-UX administrator would probably know enough to look in /var/adm/syslog/syslog.log for stuff like this: Aug 17 15:36:14 dy sshd[4593]: Authentication refused: bad ownership or modes for directory /home/sms/.ssh ------------------------------------------------------------------------ Steven M. Schweda sms@antinode-org 382 South Warwick Street (+1) 651-699-9818 Saint Paul MN 55105-2547 ------------------------------ Date: Fri, 17 Aug 2007 14:31:09 -0400 From: JF Mezei Subject: Re: incoming SMTP 255 character wrap Message-ID: Spud Demon wrote: >> Any way to get SMTP to accept long header lines without wrapping, or at > least wrap-before-the-previous-whitespace? MAIL has a limit of 255 characters per line when transmitting messages between DECNET connected nodes. Work to improve the protocol stopped a century ago because VMS management weren't shown powerpoint presentations showing customer demand for changes to MAIL (even though they are really needed). In order to ensure the SMTP software could deliver mail to any node on a network, I assume that it was built with that limitation in mind and thus folds lines at 255. (at least if it added a blank space on the new line, it would have made the header compliant and the folding would have worked). The only possibility I can see is to look at the SMTP documentation. You can declare in some config files that a certain set of nodes are to be "local" delivery. Consider a cluster consisting of nodes Cognac, Brandy, Rhum, and Whiskey, with the SMTP server on Rhum. If you tell it that all those nodes are local, then a message to chef@Brandy.distillery.com will be delivered to user "chef" on Rhum, instead of using decnet mail to deliver it to user chef on "Brandy". (since in a cluster, one assumes user files are accessible on any node with the same UAF file, this works. By telling the smtp software that Brandy is to be delivered o Rhum, it means that the smtp server on rhum can use callable mail to deliver to chef on rhum and thus not use decnet. But I do not think this will make a difference, but should VMS management ever se some powerpoint presentation telling them to provide funding to fix this problem, I suspect it would apply only to local deliveries. ------------------------------ Date: Fri, 17 Aug 2007 14:38:44 -0400 From: JF Mezei Subject: Re: Intel marginalizing Itanium even faster than expected? Message-ID: <70003$46c5eb36$cef8887a$24194@TEKSAVVY.COM> Main, Kerry wrote: > What if the guru who maintains this > code leaves, is on vacation or becomes a disgruntled employee? You really shouldn't be bringing this argument up because VMS is far more vulnerable than Linux in the current context. What if FredK leaves and there is nobody else left to handle all the X windows issues, the drivers to interface to the various video cards (which had to be reverse engineered with a lot of talent) etc etc ???? (And I have to assume FredK has done far more work than I know about). And many other experienced folks have already left. Since VMS is in a downward trend in terms of staffing, and the more talented/experienced gurus are generally the first to leave, VMS is far more vulnerable. In an open source community, when one person leaves, others can jump in more easily because of the community aspect where many peole have seen the work of the guru that has left. You can easily teach a monkey to access source code and press a PF key to start the build process and deliver an executable. But you can't teach a young newbie indian how to start to reverse engineer a windows video drivers and build an equivalent Alpha/IA64 driver that feeds 8086 code to the hardware. ------------------------------ Date: Fri, 17 Aug 2007 14:42:14 -0400 From: JF Mezei Subject: Re: Intel marginalizing Itanium even faster than expected? Message-ID: <99120$46c5ec07$cef8887a$24194@TEKSAVVY.COM> Ron Johnson wrote: > Gah! I agree with Kerry!!!! Don't worry. That is not as bad/dangerous as "agreeing with JF". The former is just an oddity. The latter has the potential to cause disruptions in the fabric of space/time :-) ------------------------------ Date: Fri, 17 Aug 2007 14:42:26 -0400 From: John Reagan Subject: Re: Intel marginalizing Itanium even faster than expected? Message-ID: JF Mezei wrote: the more > talented/experienced gurus are generally the first to leave, VMS is far > more vulnerable. > I think I was just insulted! :-) :-) :-) -- John Reagan OpenVMS Pascal/Macro-32/COBOL Project Leader Hewlett-Packard Company ------------------------------ Date: Fri, 17 Aug 2007 16:31:30 -0500 From: Ron Johnson Subject: Re: Intel marginalizing Itanium even faster than expected? Message-ID: On 08/17/07 13:42, JF Mezei wrote: > Ron Johnson wrote: >> Gah! I agree with Kerry!!!! > > > Don't worry. That is not as bad/dangerous as "agreeing with JF". The > former is just an oddity. The latter has the potential to cause > disruptions in the fabric of space/time :-) Then what happens when one agrees with Boob? -- Ron Johnson, Jr. Jefferson LA USA Give a man a fish, and he eats for a day. Hit him with a fish, and he goes away for good! ------------------------------ Date: Fri, 17 Aug 2007 16:34:36 -0500 From: Ron Johnson Subject: Re: Intel marginalizing Itanium even faster than expected? Message-ID: On 08/17/07 13:42, John Reagan wrote: > JF Mezei wrote: > the more >> talented/experienced gurus are generally the first to leave, VMS is >> far more vulnerable. >> > > I think I was just insulted! :-) :-) :-) Worse: I agree with JF. Do you hear The Fabric ripping? -- Ron Johnson, Jr. Jefferson LA USA Give a man a fish, and he eats for a day. Hit him with a fish, and he goes away for good! ------------------------------ Date: Fri, 17 Aug 2007 21:56:05 +0000 From: "Main, Kerry" Subject: RE: Intel marginalizing Itanium even faster than expected? Message-ID: > -----Original Message----- > From: JF Mezei [mailto:jfmezei.spamnot@vaxination.ca] > Sent: August 17, 2007 2:39 PM > To: Info-VAX@Mvb.Saic.Com > Subject: Re: Intel marginalizing Itanium even faster than expected? > > Main, Kerry wrote: > > > What if the guru who maintains this > > code leaves, is on vacation or becomes a disgruntled employee? > > > You really shouldn't be bringing this argument up because VMS is far > more vulnerable than Linux in the current context. > The note was in reference to Bill inferring (or my understanding of his ref= erence) that a Customer "maintain their own kernel".. Since he made that statement,= he has clarified that he was not talking about Customers maintaining their own ker= nel code. Nobody twiddles OpenVMS kernel code except VMS Engineering and the same goe= s for HP-UX, AIX, Solaris etc. Regards Kerry Main Senior Consultant HP Services Canada Voice: 613-592-4660 Fax: 613-591-4477 kerryDOTmainAThpDOTcom (remove the DOT's and AT) OpenVMS - the secure, multi-site OS that just works. ------------------------------ Date: Fri, 17 Aug 2007 20:04:24 -0500 From: "Paul Raulerson" Subject: RE: Intel marginalizing Itanium even faster than expected? Message-ID: <000801c7e133$b7f9c690$27ed53b0$@com> > -----Original Message----- > From: Main, Kerry [mailto:Kerry.Main@hp.com] > Sent: Thursday, August 16, 2007 8:27 AM > To: Info-VAX@Mvb.Saic.Com > Subject: RE: Intel marginalizing Itanium even faster than expected? > > > -----Original Message----- > > From: Paul Raulerson [mailto:paul@raulersons.com] > > Sent: August 16, 2007 12:57 AM > > To: Main, Kerry; Info-VAX@Mvb.Saic.Com > > Subject: RE: Intel marginalizing Itanium even faster than expected? > > > > > > > > Regarding userland code: there's no law that says that userland > > must > > > > be open source. (After all, Oracle makes a pretty penny selling > > > > licenses for RDBMSs that run on Linux.) > > > > > > > > > > Sure they do, but for them the platform does not matter as they > > charge > > > the > > > same for Oracle on Windows, Linux, OpenVMS, Solaris, HP-UX etc. The > > > only > > > real difference is in the area of multi-socket systems. > > > > > > The challenge is how to increase the scalability of Oracle on Linux > > > when > > > changes are required at the kernel level. That is one area where > IBM > > > (DB2/AIX) > > > and Microsoft (SQL Server) have a big advantage over Oracle ie. > they > > > control > > > the OS kernel and the DB. > > > > > > [sidebar note - it was stated at one time that one of the primary > > > design > > > factors that go into Windows kernel is how to make SQL Server run > > > better.] > > > > > > Do you think this is lost on Oracle? > > > > > > > Well, not to belabor the point, but it is not a major exercise to add > a > > module to the kernel, > > even one that makes major changes in behavior. Or to issue a special > > purpose > > version of Linux > > with a modified kernel. > > > > IBM does this with tape drivers, DB/2, etc. > > > > This rather makes the idea of a very stable controlled kernel > > attractive to > > companies > > like IBM. Especially when they sell mainframe systems because of > Linux. > > :) > > > > -Paul > > > > Based on what you have stated, while adding drivers specific to your > specific > hardware or ISV package may not be that hard, the challenge is adding > parts to > the kernel which will impact more than your code e.g. clustering, SMP, > multi-threading, scheduling, locking etc. These are the areas that > typically > get major changes when one looks to optimize a kernel for additional > scalability, > stability, new features etc. > > That's when the fun begins regarding who's wishes get priority. > > Especially when you have many different ego's with different agendas > involved. > > Reference: > http://tinyurl.com/lrecx (May, 2006) > http://tinyurl.com/preview.php?num=lrecx (preview) > > "Andrew Morton, the lead maintainer of the Linux production kernel, is > worried > that an increasing number of defects are appearing in the 2.6 version > and is > considering drastic action to resolve it. > > "I believe the 2.6 kernel is slowly getting buggier. It seems we're > adding bugs > at a higher rate than we're fixing them," Morton said in a talk at the > LinuxTag > conference in Wiesbaden, Germany, on Friday." > Same is often true in any fast developing system. More and more hardware specific drivers are added every day to Linux, and every one of them has the potential to wreak havoc on the system. This is not true on systems like VMS or z/OS; hardware that goes in either uses existing drivers, or most often, it does not go in at all. -Paul > Regards > > > Kerry Main > Senior Consultant > HP Services Canada > Voice: 613-592-4660 > Fax: 613-591-4477 > kerryDOTmainAThpDOTcom > (remove the DOT's and AT) > > OpenVMS - the secure, multi-site OS that just works. > > ------------------------------ Date: Fri, 17 Aug 2007 20:13:02 -0500 From: "Paul Raulerson" Subject: RE: Intel marginalizing Itanium even faster than expected? Message-ID: <000e01c7e134$ecc87e60$c6597b20$@com> > -----Original Message----- > From: Ron Johnson [mailto:ron.l.johnson@cox.net] > Sent: Thursday, August 16, 2007 12:18 PM > To: Info-VAX@Mvb.Saic.Com > Subject: Re: Intel marginalizing Itanium even faster than expected? > > On 08/16/07 12:04, Main, Kerry wrote: > [snip] > > > > The point I was making is that it is very difficult for a company > who's primary > > focus is something other than OS platform support to "maintain their > own kernel" > > as you called it. > > > > Once you start twiddling bits in the kernel on your own, you are > opening all > > sorts of potential future compatibility issues with new releases, > security > > patches, custom drivers etc. > > Gah! I agree with Kerry!!!! > > (Does that mean I'm wrong?) > > -- > Ron Johnson, Jr. > Jefferson LA USA > Naw- just coming to your senses. ------------------------------ Date: Fri, 17 Aug 2007 22:36:37 +0100 From: "Paul McIlfatrick" Subject: Looking for SEDT source code Message-ID: Back on 22nd November 2005 Lee Roth posted the following article in the comp.unix.solaris newsgroup: > Briefly: I have permission from the original author of SEDT (Anker > Berg-Sonne) to make my modified-for-Solaris binary + source files > available. > > If you've never used the DEC EDT editor I certainly don't suggest you > begin to do so now, but if you are an old-time DEC user like me that > has the EDT editor burned deep into your brain's ROM, you may want to > give this powerful EDT-on-steroids editor a look. > > I had to do a couple of minor tweaks to get the sources to compile, but > I have created a binary that runs on Solaris 7, 8 and 9 (Sparc) and a > separate binary that (at least) Solaris 7 on Intel platforms. I have > not yet tested the binary on Solaris 10 (Sparc). > > Additional details at http://easy48.com/sedt/ > > Check out the above link... if interested, please respond to this > posting. > > Thanks, > > Lee Roth A few days ago I posted an article to the comp.unix.solaris newsgroup asking if Lee Roth had ever made the modified-for-Solaris binary + source files available. I got three replies, each trying to be funny about painting the numlock key gold. In Alan Roth's article he mentioned the http://easy48.com/sedt/ web page and it still exists. Some days ago I sent an e-mail to Lee at the address he gives on that page but so far there has been no reply. As someone who started out on VMS back in 1983 and got used to programming DCL and using EDT, I have never got used to vi on Solaris after my company moved to Sun machines around 1996. Do any of you people who read this newsgroup know: 1) is the SEDT source available somewhere on the Internet for download (Google search didn't turn up anything)? or 2) is there SEDT Solaris binary available (I know this is a VMS newsgroup!)? Thanks Paul McIlfatrick ------------------------------ Date: Fri, 17 Aug 2007 21:49:49 GMT From: VAXman- @SendSpamHere.ORG Subject: Re: Looking for SEDT source code Message-ID: <1Koxi.310$0y3.172@newsfe12.lga> In article , "Paul McIlfatrick" writes: > > >Back on 22nd November 2005 Lee Roth posted the following article in the >comp.unix.solaris newsgroup: > >> Briefly: I have permission from the original author of SEDT (Anker >> Berg-Sonne) to make my modified-for-Solaris binary + source files >> available. >> >> If you've never used the DEC EDT editor I certainly don't suggest you >> begin to do so now, but if you are an old-time DEC user like me that >> has the EDT editor burned deep into your brain's ROM, you may want to >> give this powerful EDT-on-steroids editor a look. >> >> I had to do a couple of minor tweaks to get the sources to compile, but >> I have created a binary that runs on Solaris 7, 8 and 9 (Sparc) and a >> separate binary that (at least) Solaris 7 on Intel platforms. I have >> not yet tested the binary on Solaris 10 (Sparc). >> >> Additional details at http://easy48.com/sedt/ >> >> Check out the above link... if interested, please respond to this >> posting. I believe SEDT appeared on DECUS library distributions. I do not know though how easy/difficult it would be to add the necessary built tweaks to get it to compile for your particular unix environment. I've just taken to accepting vi and know enough to get around. If I do need something more elaborate, I have emacs in EDT mode to fall back on. If you do happen to get a working version of SEDT for OS X, let me know. -- VAXman- A Bored Certified VMS Kernel Mode Hacker VAXman(at)TMESIS(dot)COM "Well my son, life is like a beanstalk, isn't it?" http://tmesis.com/drat.html ------------------------------ Date: Fri, 17 Aug 2007 16:08:16 -0700 From: David Mathog Subject: Re: Looking for SEDT source code Message-ID: Paul McIlfatrick wrote: > 1) is the SEDT source available somewhere on the Internet for download > (Google search didn't turn up anything)? I couldn't find it either. It was apparently written by "Anker Berg-Sonne". That doesn't seem like it's going to be a very common name, so I tried finding him (sorry if Anker is a her) instead. Which led to this: http://forums.flyesl.com/topic.asp?TOPIC_ID=292 and then this: http://www.zoominfo.com/people/Berg-Sonne_Anker_530520001.aspx which looks like the right guy, as he was listed as having been at Digital. They list him as being employed here: http://www.ltcq.com/ but a bit more searching indicated he's now at eternalventures.com which sounds vaguely fatal, hopefully not the result of participation in that soaring club! http://center.spoke.com/info/p87hB25/AnkerBerg-Sonne The eternalventures web site has nothing, but whois at least gave some contact info: http://www.networksolutions.com/whois/results.jsp?domain=eternalventures.com It doesn't list Anker though. Email the contact, maybe he can tell you where Anker is. Ironically, I was within about 200 feet of the that San Francisco address last week. Yahoo "people search" shows that there's also an Anker Berg-Sonne who's 59 and lives (or lived) in Stow, MA and has an unlisted phone number. So I put this into google: "anker berg-sonne" STOW and found this: http://www.decedout.org/events.htm which lists a street address for him. (That's a DEC reunion page.) There's a directory on the decedout site, and it looks like it has a current link to Anker, but I'm not a member and so could not be sure. That's enough sleuthing for me for one day. Good thing his name wasn't John Smith! David Mathog ------------------------------ Date: Fri, 17 Aug 2007 23:21:55 GMT From: VAXman- @SendSpamHere.ORG Subject: Re: Looking for SEDT source code Message-ID: In article , David Mathog writes: > > >Paul McIlfatrick wrote: > >> 1) is the SEDT source available somewhere on the Internet for download >> (Google search didn't turn up anything)? > >I couldn't find it either. It was apparently written by "Anker >Berg-Sonne". That doesn't seem like it's going to be a very common >name, so I tried finding him (sorry if Anker is a her) >instead. Which led to this: > >http://forums.flyesl.com/topic.asp?TOPIC_ID=292 > >and then this: > >http://www.zoominfo.com/people/Berg-Sonne_Anker_530520001.aspx > >which looks like the right guy, as he was listed as having been at >Digital. They list him as being employed here: > >http://www.ltcq.com/ > >but a bit more searching indicated he's now at eternalventures.com which >sounds vaguely fatal, hopefully not the result of participation >in that soaring club! > >http://center.spoke.com/info/p87hB25/AnkerBerg-Sonne > >The eternalventures web site has nothing, but whois at least gave >some contact info: > >http://www.networksolutions.com/whois/results.jsp?domain=eternalventures.com > >It doesn't list Anker though. Email the contact, maybe he can tell >you where Anker is. > >Ironically, I was within about 200 feet of the that San Francisco >address last week. > >Yahoo "people search" shows that there's also an Anker Berg-Sonne who's >59 and lives (or lived) in Stow, MA and has an unlisted phone number. >So I put this into google: "anker berg-sonne" STOW > >and found this: > >http://www.decedout.org/events.htm > >which lists a street address for him. (That's a DEC reunion page.) >There's a directory on the decedout site, and it looks like it has >a current link to Anker, but I'm not a member and so could not be sure. > >That's enough sleuthing for me for one day. Good thing his name wasn't >John Smith! > >David Mathog Ah! The wealth of info one can conjure up from the internet! -- VAXman- A Bored Certified VMS Kernel Mode Hacker VAXman(at)TMESIS(dot)COM "Well my son, life is like a beanstalk, isn't it?" http://tmesis.com/drat.html ------------------------------ Date: Fri, 17 Aug 2007 17:37:24 -0600 From: Dan O'Reilly Subject: Re: Looking for SEDT source code Message-ID: <6.1.2.0.2.20070817173411.022a06d0@192.168.0.11> Hmmmmmmmmmmm...long ago in a galaxy far, far away (must have been, oh, 1984 or so), I actually ported SEDT to RSX-11M-PLUS on PDP-11's using a public-domain C, if I remember correctly. I can look around and see if I still have the source code someplace. If need be, can somebody out there read a BRU magtape? At 03:36 PM 8/17/2007, Paul McIlfatrick wrote: >Back on 22nd November 2005 Lee Roth posted the following article in the >comp.unix.solaris newsgroup: > > > Briefly: I have permission from the original author of SEDT (Anker > > Berg-Sonne) to make my modified-for-Solaris binary + source files > > available. > > > > If you've never used the DEC EDT editor I certainly don't suggest you > > begin to do so now, but if you are an old-time DEC user like me that > > has the EDT editor burned deep into your brain's ROM, you may want to > > give this powerful EDT-on-steroids editor a look. > > > > I had to do a couple of minor tweaks to get the sources to compile, but > > I have created a binary that runs on Solaris 7, 8 and 9 (Sparc) and a > > separate binary that (at least) Solaris 7 on Intel platforms. I have > > not yet tested the binary on Solaris 10 (Sparc). > > > > Additional details at http://easy48.com/sedt/ > > > > Check out the above link... if interested, please respond to this > > posting. > > > > Thanks, > > > > Lee Roth > >A few days ago I posted an article to the comp.unix.solaris newsgroup asking >if Lee Roth had ever made the modified-for-Solaris binary + source files >available. I got three replies, each trying to be funny about painting the >numlock key gold. > >In Alan Roth's article he mentioned the http://easy48.com/sedt/ web page and >it still exists. Some days ago I sent an e-mail to Lee at the address he >gives on that page but so far there has been no reply. > > >As someone who started out on VMS back in 1983 and got used to programming >DCL and using EDT, I have never got used to vi on Solaris after my company >moved to Sun machines around 1996. > > >Do any of you people who read this newsgroup know: > >1) is the SEDT source available somewhere on the Internet for download >(Google search didn't turn up anything)? > >or > >2) is there SEDT Solaris binary available (I know this is a VMS newsgroup!)? > > >Thanks > > >Paul McIlfatrick ------ +-------------------------------+----------------------------------------+ | Dan O'Reilly | "There are 10 types of people in this | | Principal Engineer | world: those who understand binary | | Process Software | and those who don't." | | http://www.process.com | | +-------------------------------+----------------------------------------+ ------------------------------ Date: 17 Aug 2007 19:27:48 -0500 From: burley.not-this@encompasserve-or-this.org (Graham Burley) Subject: Re: Looking for SEDT source code Message-ID: In article <1Koxi.310$0y3.172@newsfe12.lga>, VAXman- @SendSpamHere.ORG writes: > I believe SEDT appeared on DECUS library distributions. I do not know > though how easy/difficult it would be to add the necessary built tweaks > to get it to compile for your particular unix environment. Just had a quick look on DECUServe, there's a library submission and various stuff on the SIG tapes. This looks most relevant, it includes C sources: ftp://ftp.encompasserve.org/decus/vms/sig_tapes/vmslt00a/net/sedt/ (see sedt_main.txt and sedsrc.zip) ------------------------------ Date: Fri, 17 Aug 2007 18:55:22 -0700 From: William Subject: Re: Looking for SEDT source code Message-ID: <1187402122.650703.272960@r34g2000hsd.googlegroups.com> I have a version of microemacs with built-in EDT emulation at http://williambader.com/pat/emacs.tar.bz2 It has features similar to SEDT: EDT keypad, multiple buffers, insert files in the current buffer (including file name completion), large files (limited only by virtual memory, can edit binary files), spawn commands (stdin from the keyboard or from a buffer, stdout to the screen or captured in a buffer), keyboard macros, repeat prefixes on commands, cmode with auto-indenting, paren matching, user-definable macros, etc. I have built it under Solaris, Linux, FreeBSD, SCO, and OpenVMS. William Bader http://williambader.com/ ------------------------------ Date: Fri, 17 Aug 2007 22:38:32 -0400 From: Glenn Everhart Subject: Re: Looking for SEDT source code Message-ID: <46C65BA8.3080008@gce.com> David Mathog wrote: > Paul McIlfatrick wrote: > >> 1) is the SEDT source available somewhere on the Internet for download >> (Google search didn't turn up anything)? > > I couldn't find it either. It was apparently written by "Anker > Berg-Sonne". That doesn't seem like it's going to be a very common > name, so I tried finding him (sorry if Anker is a her) > instead. Which led to this: > > http://forums.flyesl.com/topic.asp?TOPIC_ID=292 > > and then this: > > http://www.zoominfo.com/people/Berg-Sonne_Anker_530520001.aspx > > which looks like the right guy, as he was listed as having been at > Digital. They list him as being employed here: > > http://www.ltcq.com/ > > but a bit more searching indicated he's now at eternalventures.com which > sounds vaguely fatal, hopefully not the result of participation > in that soaring club! > > http://center.spoke.com/info/p87hB25/AnkerBerg-Sonne > > The eternalventures web site has nothing, but whois at least gave > some contact info: > > http://www.networksolutions.com/whois/results.jsp?domain=eternalventures.com > > > It doesn't list Anker though. Email the contact, maybe he can tell > you where Anker is. > > Ironically, I was within about 200 feet of the that San Francisco > address last week. > > Yahoo "people search" shows that there's also an Anker Berg-Sonne who's > 59 and lives (or lived) in Stow, MA and has an unlisted phone number. > So I put this into google: "anker berg-sonne" STOW > > and found this: > > http://www.decedout.org/events.htm > > which lists a street address for him. (That's a DEC reunion page.) > There's a directory on the decedout site, and it looks like it has > a current link to Anker, but I'm not a member and so could not be sure. > > That's enough sleuthing for me for one day. Good thing his name wasn't > John Smith! > > David Mathog SEDT has been on the sigtapes several times, with sources. However I do not know how much work it takes to get those working on newer machines. Incidentally I did find more recently a version of Pete Siemsen's old tecoc which had been recompiled for windows. It works very well, but alas lacks the VT macro so vtedit and the teco EDT emulator cannot run in it far as I know. Glenn Everhart (The EDT emulator in Teco VTEDIT is really a thing of beauty btw.) ------------------------------ Date: Fri, 17 Aug 2007 22:49:03 -0500 From: "Paul Raulerson" Subject: SSH login welcome message? Message-ID: <000001c7e14a$b9150320$2b3f0960$@com> This is a multipart message in MIME format. ------=_NextPart_000_0001_01C7E120.D03EFB20 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit What file(s) hold the welcome message for SSH logins? -Paul ------=_NextPart_000_0001_01C7E120.D03EFB20 Content-Type: text/html; charset="us-ascii" Content-Transfer-Encoding: quoted-printable

What file(s) hold the welcome message for SSH = logins?

-Paul

------=_NextPart_000_0001_01C7E120.D03EFB20-- ------------------------------ Date: Fri, 17 Aug 2007 23:04:37 -0500 (CDT) From: sms@antinode.org (Steven M. Schweda) Subject: Re: SSH login welcome message? Message-ID: <07081723043708_20200296@antinode.org> From: "Paul Raulerson" > This is a multipart message in MIME format. Why? > What file(s) hold the welcome message for SSH logins? Google broken? HTML manuals unavailable? http://groups.google.com/group/comp.os.vms/browse_thread/thread/42b16fa49dddadf4/8c6f3c71727b6cf5 > ------=_NextPart_000_0001_01C7E120.D03EFB20 > Content-Type: text/html; > charset="us-ascii" > Content-Transfer-Encoding: quoted-printable Thanks again. ------------------------------------------------------------------------ Steven M. Schweda sms@antinode-org 382 South Warwick Street (+1) 651-699-9818 Saint Paul MN 55105-2547 ------------------------------ Date: Sat, 18 Aug 2007 00:28:05 -0500 From: "Paul Raulerson" Subject: RE: SSH login welcome message? Message-ID: <000801c7e158$8e5c8c30$ab15a490$@com> This is a multipart message in MIME format. ------=_NextPart_000_0009_01C7E12E.A5868430 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit WHY: It is 12:11 am and I am tired. Question: Why did you waste all the effort being nasty? If you don't know the answer, just don't answer. Better yet: Why am I wasting all this time? I keep trying to make OpenVMS work and be a viable alternative to a mainframe platform. When people like yourself waste so much time and energy being negative and nasty, it really makes life unpleasant. Important: Do I really want to submit my customers to dealing with this kind of attitude? Especially if I get into a position where I cannot support them and they need support from someone else? Those are even better questions. I'm going to bed. And yes, this is formatted in HTML. I'm too tired to try and format it in plain text, and don't particularly feel like being polite to you right now. Go get a better e-mail client or newsreader. -Paul > -----Original Message----- > From: Steven M. Schweda [mailto:sms@antinode.org] > Sent: Friday, August 17, 2007 11:05 PM > To: Info-VAX@Mvb.Saic.Com > Subject: Re: SSH login welcome message? > > From: "Paul Raulerson" > > > This is a multipart message in MIME format. > > Why? > > > What file(s) hold the welcome message for SSH logins? > > Google broken? HTML manuals unavailable? > > http://groups.google.com/group/comp.os.vms/browse_thread/thread/42b16fa > 49dddadf4/8c6f3c71727b6cf5 > > > > ------=_NextPart_000_0001_01C7E120.D03EFB20 > > Content-Type: text/html; > > charset="us-ascii" > > Content-Transfer-Encoding: quoted-printable > > Thanks again. > > ----------------------------------------------------------------------- > - > > Steven M. Schweda sms@antinode-org > 382 South Warwick Street (+1) 651-699-9818 > Saint Paul MN 55105-2547 ------=_NextPart_000_0009_01C7E12E.A5868430 Content-Type: text/html; charset="us-ascii" Content-Transfer-Encoding: quoted-printable

WHY:      It is 12:11 = am and I am tired.

 

Question:   Why did you waste all the = effort being nasty?  

If = you don't know the answer, just don't answer.

 

Better yet: Why am I wasting all this = time?

        &= nbsp;   I keep trying to make OpenVMS work and be a viable =

alternative to a mainframe platform. When people = like

yourself waste so much time and energy being = negative

and nasty, it really makes life unpleasant. =

        &nbs= p;

Important:  Do I really want to submit my customers to dealing =

with this kind of attitude?

Especially if I get into a position where

I = cannot support them and they need support

from = someone else?

 

        &nbs= p;   Those are even better questions. =

 

I'm going to bed.

 

And yes, this is formatted in HTML. I’m = too tired to try and format it

in plain text, and don’t particularly feel = like being polite to you right now.

Go get a better e-mail client or newsreader. =

 

 

-Paul

  =       

 

> -----Original Message-----

> From: Steven M. Schweda = [mailto:sms@antinode.org]

> Sent: Friday, August 17, 2007 11:05 PM

> To: Info-VAX@Mvb.Saic.Com

> Subject: Re: SSH login welcome message?

>

> From: "Paul Raulerson" <paul@raulersons.com>

>

> > This is a multipart message in MIME = format.

>

>    Why?

>

> > What file(s) hold the welcome message = for SSH logins?

>

>    Google broken?  HTML = manuals unavailable?

>

> = http://groups.google.com/group/comp.os.vms/browse_thread/thread/42b16fa

> 49dddadf4/8c6f3c71727b6cf5

>

>

> > = ------=3D_NextPart_000_0001_01C7E120.D03EFB20

> > Content-Type: text/html;

> >   = charset=3D"us-ascii"

> > Content-Transfer-Encoding: = quoted-printable

>

>    Thanks again.

>

> = -----------------------------------------------------------------------

> -

>

>    Steven M. Schweda           =     sms@antinode-org

>    382 South Warwick Street        (+1) 651-699-9818

>    Saint Paul  MN  55105-2547

------=_NextPart_000_0009_01C7E12E.A5868430-- ------------------------------ Date: Fri, 17 Aug 2007 16:40:11 -0500 From: Ron Johnson Subject: Re: Wonderful things happen to an OS when it has an internal champion Message-ID: <%Aoxi.13080$Zk5.1467@newsfe23.lga> On 08/17/07 08:03, Andrew wrote: [snip] > > While it may well be true that Linux and Windows lack reliable > workload management the same cannot be said for UNIX which has > supported this for some time. Solaris has Zones/Containers hooked into > workload management, while both HP-UX and AIX also have support for > workload managers although without the container type features of > Solaris. > > All the major commercial UNIX platforms allow administrators to > restrict the amount of resource used by an application to a share of > the system ensuring that all apps get the service they need. In the > case of Solaris the apps can be running in a container which looks > like another instance of the OS but in fact isn't. > > BSD Jails are very similar to Solaris Zones/Containers. Linux *does*, though, have CPU percentage limiting software. http://cpulimit.sourceforge.net/ BTW, how does one limit CPU percentage in VMS? -- Ron Johnson, Jr. Jefferson LA USA Give a man a fish, and he eats for a day. Hit him with a fish, and he goes away for good! ------------------------------ Date: Fri, 17 Aug 2007 22:08:01 +0000 From: "Main, Kerry" Subject: RE: Wonderful things happen to an OS when it has an internal champion Message-ID: > -----Original Message----- > From: Ron Johnson [mailto:ron.l.johnson@cox.net] > Sent: August 17, 2007 5:40 PM > To: Info-VAX@Mvb.Saic.Com > Subject: Re: Wonderful things happen to an OS when it has an internal > champion > > On 08/17/07 08:03, Andrew wrote: > [snip] > > > > While it may well be true that Linux and Windows lack reliable > > workload management the same cannot be said for UNIX which has > > supported this for some time. Solaris has Zones/Containers hooked > into > > workload management, while both HP-UX and AIX also have support for > > workload managers although without the container type features of > > Solaris. > > > > All the major commercial UNIX platforms allow administrators to > > restrict the amount of resource used by an application to a share of > > the system ensuring that all apps get the service they need. In the > > case of Solaris the apps can be running in a container which looks > > like another instance of the OS but in fact isn't. > > > > BSD Jails are very similar to Solaris Zones/Containers. > > Linux *does*, though, have CPU percentage limiting software. > http://cpulimit.sourceforge.net/ > > BTW, how does one limit CPU percentage in VMS? > > -- Class scheduler: $ mcr sysman Sysman> help class (see ADD, DELETE, MODIFY, SHOW, SUSPEND) .. ADD The ADD command creates a new scheduling class. The class scheduler provides the ability to limit the amount of CPU time that a system's users receive by placing users in scheduling classes. Each class is assigned a percentage of the overall system CPU time. As the system runs, the combined set of users in a class is limited to the percentage of CPU execution time allocated to their class. Users might get some additional CPU time if the qualifier /WINDFALL is enabled for their scheduling class. Enabling the qualifier /WINDFALL allows the system to give a small amount of CPU time to a scheduling class when the scheduling class's allotted time has been depleted, but a free CPU is available. Other ways of managing workloads include: - setting process priorities - restricting processes to specific cpu's (can also be performance enhancem= ent as caching has better chance of not being flushed) Regards Kerry Main Senior Consultant HP Services Canada Voice: 613-592-4660 Fax: 613-591-4477 kerryDOTmainAThpDOTcom (remove the DOT's and AT) OpenVMS - the secure, multi-site OS that just works. ------------------------------ Date: Fri, 17 Aug 2007 20:05:15 -0500 From: "Paul Raulerson" Subject: RE: Write locked file Message-ID: <000901c7e133$d66df2e0$8349d8a0$@com> Use the fstat() call and just look up the process ID it returns. -Paul > -----Original Message----- > From: Mike Minor [mailto:mminorhsd@earthlink.net] > Sent: Thursday, August 16, 2007 8:28 AM > To: Info-VAX@Mvb.Saic.Com > Subject: Write locked file > > I have a situation where a .txt file is in a write locked state, but I > can't > find the process that has the file locked. I have tried the following > commands to no avail: > > show device/files /system > show device/files /nosystem > showdevice/files user1 > > All show a list of open files on the system, but the file in question > does > not show up. Does anyone have any ideas about how to identify the > process > that has the file locked? > > Thank you in advance > > Mike Minor > ------------------------------ End of INFO-VAX 2007.451 ************************