INFO-VAX Sat, 26 Jan 2008 Volume 2008 : Issue 52 Contents: Re: Anyone interested in building a vms-like OS? Re: Anyone interested in building a vms-like OS? Re: Anyone interested in building a vms-like OS? RE: Davos Re: Davos Re: DEGXA Gigabit card special Re: DEGXA Gigabit card special Re: Installing MX 4.2 Re: Using disk space on NAS? ---------------------------------------------------------------------- Date: 26 Jan 2008 12:47:55 GMT From: billg999@cs.uofs.edu (Bill Gunshannon) Subject: Re: Anyone interested in building a vms-like OS? Message-ID: <600ofrF1p13heU1@mid.individual.net> In article <479AA84A.59921D0A@spam.comcast.net>, David J Dachtera writes: > Bill Gunshannon wrote: >> >> In article , >> Keith Parris writes: >> > Bill Gunshannon wrote: >> >> Some of them. But the kernel has more than it's share of warts. >> >> Just as an example, which kernel did they choose as their starting >> >> point? A generic kernel or one with the NSA security patches? >> > >> > Start the FreeVMS project over with a *bsd kernel, and Bill G. will be >> > its biggest supporter. ;-) >> >> And just how would that be VMS any more than the current one is? It would >> be BSD (like OSX) with a VMSy look and feel. If you want to clone VMS take >> the VMS spec and clone it. Anything else will not be VMS. And considering >> how much is not published, I would not hold my breath waiting for a VMS clone. > > So the old Internals and Data Structures book is no good? > A start maybe, but insuficient to actually clone an OS. 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: 26 Jan 2008 14:33:27 GMT From: billg999@cs.uofs.edu (Bill Gunshannon) Subject: Re: Anyone interested in building a vms-like OS? Message-ID: <600ulnF1o9telU1@mid.individual.net> In article <479acd0c$0$3470$c3e8da3@news.astraweb.com>, JF Mezei writes: > One needs to put some thoughts into this. > > > What unique portions of VMS are strictly kernel and impossible to > implement at higher levels ? Everything that makes VMS what it is are at the kernel level. The rest (like DCL and RMS) are just applications that could be put on any OS. > > > Right now, much of VMS consists of unix applications that are ported to > VMS. (just about everything tcpip related for instance). Are you interested in running VMS or are the applications the important thing? > > > You need to consider that VMS' application portfolio isn't evolving much > anymore. Much of any new stuff requires to "fight" the Unix square peg , > trying to fit in into the VMS round hole. (for instance, Apache with > fork/process management). Same question as above!! > > Now, imagine if VMS system services were ported to Unix, along with lock > manager and the logical name subsystem. You would be running Unix. I thought it was VMS that was what everyone wanted? There is absolutely no reason why things like DCL and RMS and lock managers can not be put onto Unix systems. For that matter, even the deeper stuff like ACL's could be done. The onyl reason it has not been done up to his point is that no one on the UNix side is consider any of it of value or worth the effort. > > You could then run "native" Unix applications without needing a porting > effort. You could also move VMS applications to Unix. Which, of course, would not be running VMS but running that dreaded Unix thing everyone rails against here. :-) > > Sure, you lose some of the VMS security features and need to adapt. But > in the end, you would end up with a vibrant system that receives all new > applications while still giving you access to many VMS features that you > like. And the rest of the Unix world might take notice and integrate > them too. Actually, no they wouldn't. If they cared, they would have already done it. Using ACL's as an example again, do you even realize how many new filesystems have been grafted into various Unix flavors since the introduction of Filesystem ACL's? And still, no one has bothered creating one for Unix. > > At this point in time, I feel that the only way for VMS to survive is to > donate its organs to the oopen source world and see its organs > implemented in Unix. > > Barring the donation of VMS' organs to Unix, one could consider > something akin to Cygwin that would provide a mini VMS environment on a > Unix platform that would allow VMS apps to be compiled and linked > against VMS libraries. If you want to know what I think the best solution is, (well, you probably don't, but I'll say it anyway) I think a group of people who are interested in the ability to run VMS programs and have the security and stability VMS offers should take something like NetBSD or OpenBSD and create a new tree. In this tree they should start by going over the code with a fine tooth comb and eliminating any of the well documented C pitfalls that are still present (my guess is that lint would point out piles of things that should be fixed in the name of "good programming practice"). One could even take it a step further and re-write all of it in some other language than C (Ada maybe?) a module at a time until it was all converted. Then Add the stuff that NSA came up with for "secure Linux" (how's that for an oxymoron) like Mandatory Access Controls. And then start adding the things that make VMS what it is. A clone of DCL. A clone of RMS. A filesystem with the same features as the VMS filesystem (ACL's, versioning, etc.) The basic functioning system is there. Take the ball and run with it. The result could be a true replacement for a dying VMS and a greatly improved Unix. Sounds like a win-win situation to me. 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: 26 Jan 2008 16:32:57 GMT From: billg999@cs.uofs.edu (Bill Gunshannon) Subject: Re: Anyone interested in building a vms-like OS? Message-ID: <6015lpF1p135jU1@mid.individual.net> In article <479B4F5B.1080409@comcast.net>, "Richard B. Gilbert" writes: > Bill Gunshannon wrote: >> In article <479acd0c$0$3470$c3e8da3@news.astraweb.com>, >> JF Mezei writes: >> >>>One needs to put some thoughts into this. >>> >>> >>>What unique portions of VMS are strictly kernel and impossible to >>>implement at higher levels ? >> >> >> Everything that makes VMS what it is are at the kernel level. The >> rest (like DCL and RMS) are just applications that could be put on >> any OS. >> >> >>> >>>Right now, much of VMS consists of unix applications that are ported to >>>VMS. (just about everything tcpip related for instance). >> >> >> Are you interested in running VMS or are the applications the important >> thing? >> >> >>> >>>You need to consider that VMS' application portfolio isn't evolving much >>>anymore. Much of any new stuff requires to "fight" the Unix square peg , >>>trying to fit in into the VMS round hole. (for instance, Apache with >>>fork/process management). >> >> >> Same question as above!! >> >> >>>Now, imagine if VMS system services were ported to Unix, along with lock >>>manager and the logical name subsystem. >> >> >> You would be running Unix. I thought it was VMS that was what everyone >> wanted? There is absolutely no reason why things like DCL and RMS and >> lock managers can not be put onto Unix systems. For that matter, even >> the deeper stuff like ACL's could be done. The onyl reason it has not >> been done up to his point is that no one on the UNix side is consider >> any of it of value or worth the effort. >> >> >> >>>You could then run "native" Unix applications without needing a porting >>>effort. You could also move VMS applications to Unix. >> >> >> Which, of course, would not be running VMS but running that dreaded >> Unix thing everyone rails against here. :-) >> >> >>>Sure, you lose some of the VMS security features and need to adapt. But >>>in the end, you would end up with a vibrant system that receives all new >>>applications while still giving you access to many VMS features that you >>>like. And the rest of the Unix world might take notice and integrate >>>them too. >> >> >> Actually, no they wouldn't. If they cared, they would have already >> done it. Using ACL's as an example again, do you even realize how >> many new filesystems have been grafted into various Unix flavors >> since the introduction of Filesystem ACL's? And still, no one has >> bothered creating one for Unix. >> > > Not true! Solaris has had ACLs since, at least, Solaris 8. That was at > least five years ago! Is it the same as VMS? No. Is it as good? Who > knows? My need for/use of either has been absolutely minimal. --------------------------- Under Solaris, the ACL type can be one of the following: user group mask other default_user default_group default_mask default_other --------------------------- I guess I mistook it as it doesn't seem to offer any more granularity than normal user permissions. I suppose they handle group in a more elegant way, but nothing about this is what I was used to for ACL's and that long before I saw my first VMS machine. 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: Sat, 26 Jan 2008 16:41:51 +0000 From: "Main, Kerry" Subject: RE: Davos Message-ID: > -----Original Message----- > From: JF Mezei [mailto:jfmezei.spamnot@vaxination.ca] > Sent: January 25, 2008 6:14 PM > To: Info-VAX@Mvb.Saic.Com > Subject: Re: Davos > > Main, Kerry wrote: > > > "Mark Hurd, chief executive officer of the world's largest technology > company, > > doesn't chase the public limelight. He doesn't go to high-profile > international > > business conferences like the upcoming World Economic Forum in Davos, > > > While Carly had no management value and was all glamour, you need to > take a serious look at the business guest list for Davos. A lot of > serious CEOs there. Just because one goes to Davos doesn't mean that > one > intends to be extremely public, visible and dancing on tables. > > There is a lot of business and informal meetings being conducted. And > it > is also a high profile event where you can accept some serious business > interviews with serious media. > > These allow a CEO to show to the world that he is a real leader. And > when done properly, does a lot of good to the corporation's image. > > While Hurd doesn't *need* to be there, it is not necessarily a bad > thing > to be there and there can be a lot of value. > > And NOT being there gives the impression that he wasn't invited and > thus > not seen as a world business leader. And that isn't so good for HP. > > Just because La Carly exagerated her importance in the world doesn't > mean that going to Davos and behaving like a mature leader doesn't have > value. Come on now .. for whatever reason, you are pushing your own agenda here. The fact is that approx 99% of the worlds CEO's are not at Davos because they realize that it is primarily photo op's, rubbing shoulders with big names and/or trying to leverage some time to promote their company. Mark Hurd was just named BW "Businessman of the Year" and a large part of that was due to his not trying to be a media star and simply focussing on the $104B technology company business he is in charge of. Regards Kerry Main Senior Consultant HP Services Canada Voice: 613-254-8911 Fax: 613-591-4477 kerryDOTmainAThpDOTcom (remove the DOT's and AT) OpenVMS - the secure, multi-site OS that just works. ------------------------------ Date: Sat, 26 Jan 2008 12:14:53 -0500 From: "Richard B. Gilbert" Subject: Re: Davos Message-ID: <479B6A8D.2090601@comcast.net> Main, Kerry wrote: >>-----Original Message----- >>From: JF Mezei [mailto:jfmezei.spamnot@vaxination.ca] >>Sent: January 25, 2008 6:14 PM >>To: Info-VAX@Mvb.Saic.Com >>Subject: Re: Davos >> >>Main, Kerry wrote: >> >> >>>"Mark Hurd, chief executive officer of the world's largest technology >> >>company, >> >>>doesn't chase the public limelight. He doesn't go to high-profile >> >>international >> >>>business conferences like the upcoming World Economic Forum in Davos, >> >> >>While Carly had no management value and was all glamour, you need to >>take a serious look at the business guest list for Davos. A lot of >>serious CEOs there. Just because one goes to Davos doesn't mean that >>one >>intends to be extremely public, visible and dancing on tables. >> >>There is a lot of business and informal meetings being conducted. And >>it >>is also a high profile event where you can accept some serious business >>interviews with serious media. >> >>These allow a CEO to show to the world that he is a real leader. And >>when done properly, does a lot of good to the corporation's image. >> >>While Hurd doesn't *need* to be there, it is not necessarily a bad >>thing >>to be there and there can be a lot of value. >> >>And NOT being there gives the impression that he wasn't invited and >>thus >>not seen as a world business leader. And that isn't so good for HP. >> >>Just because La Carly exagerated her importance in the world doesn't >>mean that going to Davos and behaving like a mature leader doesn't have >>value. > > > Come on now .. for whatever reason, you are pushing your own agenda here. > > The fact is that approx 99% of the worlds CEO's are not at Davos because > they realize that it is primarily photo op's, rubbing shoulders with big > names and/or trying to leverage some time to promote their company. > > Mark Hurd was just named BW "Businessman of the Year" and a large part of > that was due to his not trying to be a media star and simply focussing > on the $104B technology company business he is in charge of. > Technology company??? I thought HP's major revenue source these days was ink and toner! ------------------------------ Date: 26 Jan 2008 12:52:56 GMT From: billg999@cs.uofs.edu (Bill Gunshannon) Subject: Re: DEGXA Gigabit card special Message-ID: <600op8F1p13heU2@mid.individual.net> In article <479a98b1$0$15767$c3e8da3@news.astraweb.com>, JF Mezei writes: > David Turner wrote: >> Got some refurbished ones in for $309 >> That is dealer to dealer pricing >> 12 months warranty > > Are these Q-BUS cards compatible with an all mighty Microvax II ? > >:-) :-) :-) :-) :-) :-) That;s funny. Just after marking it read I thought I should have asked if they were QBUS. Of course, I would have put it in my PDP-11 before even thinking about puting it into one of my VAXen. :-) It is supported by Ultrix-11, right? :-) 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: Sat, 26 Jan 2008 07:20:13 -0800 (PST) From: "johnhreinhardt@yahoo.com" Subject: Re: DEGXA Gigabit card special Message-ID: On Jan 25, 9:17 pm, JF Mezei wrote: > David Turner wrote: > > Got some refurbished ones in for $309 > > That is dealer to dealer pricing > > 12 months warranty > > Are these Q-BUS cards compatible with an all mighty Microvax II ? > > :-) :-) :-) :-) :-) :-) Sorry. It's a PCI card suitable for a semi-mighty Alpha (Is plural Alphen?) And, yeah, I know you were kidding. :-P ------------------------------ Date: Sat, 26 Jan 2008 10:37:05 +0000 (UTC) From: david20@alpha2.mdx.ac.uk Subject: Re: Installing MX 4.2 Message-ID: In article <479a62a0$1@news.langstoeger.at>, eplan@langstoeger.at (Peter 'EPLAN' LANGSTOEGER) writes: >In article , =?ISO-8859-15?Q?Jan-Erik_S=F6derholm?= writes: >>Tom Linden wrote: >>> On Fri, 25 Jan 2008 08:02:11 -0800, Jan-Erik Söderholm >>> wrote: >>> >>>> I though I'd install MX on my little box over here. >>>>... >>>> Found a kit at last on the Freeware-4 kit.... >> >>... >>>> %VMSINSTAL-E-INSFAIL, The installation of MX V4.2 has failed. >>>> >>>> Jan-Erik. >>> >>> Why not use 5.4? >> >>Yes, why not ? > >Because, since V5, MX was no longer free. (IIRC, I paid $499 then) >It got Anti-SPAM features (which were ahead of time and worth every price). > >>Link ? >>URL ? > >I could offer the kits, but I can't offer a license. That's the duty of Matt. > > http://www.madgoat.com/ > >But unfortunately, his website is down, and maybe his mailbox is as well. >I do not think that he offers licenses now (at least, that was my impression >when I last browsed his webserver) but I also don't think, that he declared >MX V5 now as freeware (eg. he could offer a PAK here in the newsgroup for all). >So, MX V4 is unfortunately still the only official MX freeware for you. > >But there is TCPware/Multinet/TCPIP, all in hobbyist versions, as U know... > PMDF is also available from Process free for hobbyist use See http://www.process.com/openvms/hobbyist.html then just click on the License terms and PAKS link and accept the license - you'll then be asked to fill in your details and what software you want a license for. David Webb Security team leader CCSS Middlesex University >-- >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: Sat, 26 Jan 2008 23:04:47 +1100 From: Gremlin Subject: Re: Using disk space on NAS? Message-ID: <13pm8f1hb8tnfc8@corp.supernews.com> Hi Zane JF has had (some) success using VMS as an NFS server for connecting his Mac to it. I believe that others may have connected a VMS client to an NFS share. However, for more than 2 years I have *not* been able to connect a VMS client (up to v8.2) to an NFS share served by Linux (Ubuntu, RedHat and SuSE) or Windows Server (2000, 2003 and 2003R2). This group and ITRC can't provide assistance and the ITRC thread(s) about this have many people with the same problem but noone with a solution. It appears that the TCPIP stack from HP may be broken/sub-optimal or it may be a VMS peculiarity - I certainly don't know and this group and ITRC don't appear to either. So, best of luck and if you can get a VMS client to connect and use files from an NFS server on Linux/Windows, please let me know how you did it - and perhaps the ITRC thread(s) as well!!! healyzh@aracnet.com wrote: > I've been researching a building a NAS device for at home. Unless something > better comes to light I'll be using FreeNAS. It handles ftp, rsync, NFS, > AFP (Appletalk), iSCSI, CIFS/SMB, and one or two other things. > > My current home VMS system is an XP1000/667 running OpenVMS 8.3 with 2GB > RAM, 100Mbit Ethernet, DLT7000 tape drive, and the following disks; 18GB > system disk, 50GB Scratch, 36GB Web, and 36GB User. The web and user disks > are volume shadowed. The problem I have is that these are 10,000RPM SCSI > drives running in external JBOD enclosures in my dining room. In other > words they generate some definite heat and noise. I mainly use the system > as a Web, email, and database server. The light fileserver duties it has > been performed in the past will move to the NAS. > > Looking around iSCSI appears to be just about perfect for my needs, however, > except for an EAK on Itanium that doesn't exist right now (I gather it will > be in 8.4). This basically leaves NFS. > > Does anyone have any experience using either plain Unix NFS space from VMS, > or using logical disk containers on Unix NFS space? What kind of problems > should I expect? While 100Mbit isn't that fast, it is probably fast enough > for what I am doing with the system, and if this works out, I might invest > in a GigE controller. > > Zane > ------------------------------ End of INFO-VAX 2008.052 ************************