1 INFO-VAX	Mon, 15 Aug 2005	Volume 2005 : Issue 454       Contents:% Re: Help Needed: Tape Backup Save Set % Re: Help Needed: Tape Backup Save Set % Re: Help Needed: Tape Backup Save Set % Re: Help Needed: Tape Backup Save Set % Re: Help Needed: Tape Backup Save Set % Re: Help Needed: Tape Backup Save Set % Re: Help Needed: Tape Backup Save Set % Re: Help Needed: Tape Backup Save Set % Re: Help Needed: Tape Backup Save Set % Re: Help Needed: Tape Backup Save Set % Re: Help Needed: Tape Backup Save Set % Re: Help Needed: Tape Backup Save Set % Re: Help Needed: Tape Backup Save Set % Re: Help Needed: Tape Backup Save Set % Re: Help Needed: Tape Backup Save Set  Re: making unzip.exe a command Re: my 2 centsD Re: Problems with updating firmware on SDLT1 drives using LTT on VMS Re: SAMBA for VMS  Re: SAMBA for VMS  Re: SAMBA for VMS  Re: SAMBA for VMS  Re: SAMBA for VMS  Re: SAMBA for VMS 7 Re: TCPIP Services: SMTP bug/wrong logic with 450 codes 7 Re: TCPIP Services: SMTP bug/wrong logic with 450 codes 7 Re: TCPIP Services: SMTP bug/wrong logic with 450 codes   F ----------------------------------------------------------------------  % Date: Mon, 15 Aug 2005 00:39:15 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> . Subject: Re: Help Needed: Tape Backup Save Set, Message-ID: <43001C48.17E1DA1E@teksavvy.com>  
 AEF wrote:- > Is there some reason you can't just try it?   A I know it is supposed to work. But when I wrote my procedures, I  B ALLOCATE MUA0: to make sure that the device remains in my process'A control between multiple invocations of BACKUP. So I just want to  understand what really happens.   G > Now just why in the world would the first backup command dismount the F > tape? And why wouldn't it be at the end of the tape when it is done?    H Because it wants to release the device so that another user can use it ?  G I don't want user X to mount a tape and not release it for hours on end E because the BACKUP command left the tape mounted and the user doesn't F know that once his backups are done, he needs to specifically dismountE it to release the device/tape and let other users use the tape drive.    ------------------------------  % Date: Mon, 15 Aug 2005 07:39:32 -0500 ( From: Wayne Sewell <wayne@tachysoft.com>. Subject: Re: Help Needed: Tape Backup Save Set/ Message-ID: <00A484F4.F0C038D4.3@tachysoft.com>   & >Date: Mon, 15 Aug 2005 00:39:15 -0400. >From: JF Mezei <jfmezei.spamnot@teksavvy.com> >X-Newsgroups: comp.os.vms/ >Subject: Re: Help Needed: Tape Backup Save Set    >AEF wrote: . >> Is there some reason you can't just try it? > B >I know it is supposed to work. But when I wrote my procedures, I C >ALLOCATE MUA0: to make sure that the device remains in my process' B >control between multiple invocations of BACKUP. So I just want to  >understand what really happens.  M Not necessary.  Mounting does an implied allocate, even if done automatically I by backup without an explicit mount command, and backup does not dismount ; unless told to do so.  So it does what you want by default.    > H >> Now just why in the world would the first backup command dismount theG >> tape? And why wouldn't it be at the end of the tape when it is done?  >  > I >Because it wants to release the device so that another user can use it ?   H Digital assumed, and quite rightly so, that most people would want to doK multiple backups in one shot.  It's a hell of a lot more efficient than the M dismount/remount thing, especially since backup would have to scan to the end O of the tape on every backup and the wait would get longer and longer with every K backup as the amount of data on the tape increased.  That would be a lot of * unnecessary wear on tape drives and media.   > H >I don't want user X to mount a tape and not release it for hours on endF >because the BACKUP command left the tape mounted and the user doesn'tG >know that once his backups are done, he needs to specifically dismount F >it to release the device/tape and let other users use the tape drive.  I Training issue.  If they don't know how to use backup, they should not be L allowed to use backup, or any other function that can impact the system as a whole or affect other users.    O Probably at most sites, individual users don't back up their own stuff anyway.  K Instead the system manager does image backups of all the disks on a regular 4 schedule, often with incremental backups in between.  N BTW, use of the /release qualifier on the last backup would cause an automaticO rewind/dismount/deallocate, as long as an explicit allocate had not been done.  N Better yet, backup lets go of the drive when it is time to start the recording6 pass, if any, making the drive available even quicker.  O =============================================================================== N Wayne Sewell, Tachyon Software Consulting  (281)812-0738   wayne@tachysoft.com; http://www.tachysoft.com/www/tachyon.html and wayne.html    O =============================================================================== P Jake Blues:"You traded the Caddy for a microphone? ...... Okay, I can buy that."   ------------------------------    Date: 15 Aug 2005 08:26:37 -0700( From: "Chris Allen" <ca.allen@gmail.com>. Subject: Re: Help Needed: Tape Backup Save SetC Message-ID: <1124119597.326006.134530@o13g2000cwo.googlegroups.com>   C Not being stubborn, just being cautious.  As I know next to nothing D about DECNet I fear making a mistake that could result in a securityD hole.  The VMS servers here run on a hardened network, and it is top( priority that they are properly secured.   ------------------------------    Date: 15 Aug 2005 08:57:11 -0700$ From: "AEF" <spamsink2001@yahoo.com>. Subject: Re: Help Needed: Tape Backup Save SetC Message-ID: <1124121431.103549.239110@g47g2000cwa.googlegroups.com>    Wayne Sewell wrote: ' > >From: "AEF" <spamsink2001@yahoo.com>  > >X-Newsgroups: comp.os.vms1 > >Subject: Re: Help Needed: Tape Backup Save Set  >  > >Wayne Sewell wrote:* > >> >From: "AEF" <spamsink2001@yahoo.com> > >> >X-Newsgroups: comp.os.vms 4 > >> >Subject: Re: Help Needed: Tape Backup Save Set& > >> >Date: 12 Aug 2005 15:58:34 -0700 > >> > >> > > >> >Chris Allen wrote:E > >> >> Well everything worked fine this time around.  I must of been K > >> >> mounting it wrong before.  Here's a copy of my backup .com file and M > >> >> some terminal output.  It really baffles me that "BACKUP/LIST" worked L > >> >> this time around, because I'm almost positive I tried it before withM > >> >> and without /FOR for mount.  But I MUST have not since it worked now. N > >> >> This time around I also tried DIR and BACKUP/LIST with and without theO > >> >> /MEDIA=COMPACT and it made no difference.  If I restore and forget this  > >> > > >> >J > >> >/MEDIA=COMPACT only applies when writing to tape. If you use it withC > >> >BACKUP, it only applies to tapes that BACKUP actually mounts.  > >>8 > >> And on top of that, if you use /rewind.  See below. > >  > > B > >If you use rewind, then it re-inits the tape, re-mounts it, andH > >compaction is therefore enabled as I said in the post you're quoting. > H > But using /rewind *without* backup mounting the tape will also turn onQ > compaction.  That's in direct opposition to your statement.  You obviously mean 6 > an implied mount that is part of /rewind processing. > P > You are assuming backup actually does a formal dismount/init/remount sequence,P > using the standard system services, if you use /rewind and the tape is alreadyO > mounted foreign.  I would have to look at the backup source to be sure, but I I > am not sure that backup actually does this.  It is not required.  It is I > possible to initialize a tape without it as long as the tape is mounted Q > foreign.   You can do a rewind qio to the drive, which takes it to the absolute N > beginning of the tape, even ahead of the label (not possible with a standardP > ansi mount).  Then issue write qios to write the header records (three of themO > IIRC), including the VOL1 record containing the label, followed by three tape E > marks to create the end-of-tape marker.  At this point, the tape is P > indistinguishable from one that an init was just performed on, so backup couldO > simply go on with its usual gag of writing the saveset at EOT no matter what.    The label may have changed.   P > And the tape was *not* dismounted/remounted.  I do this inside tapesys to makeP > absolutely sure the tape is initialized before vms backup is even executed.  IQ > haven't looked at this part of the backup source code lately, but they could do P > it that way if they wanted.  No need to call sys$mount or sys$init_vol, and itJ > would be a hell of a lot faster.  Again, would need to check the source.  F Well, something has to tell the system what the possibly new label is.C Isn't that what mounting a tape does? It tells the system, "Hey, we + have a tape loaded. And the label is BLAH."     , Here's what typically happens on my systems:   $ MOU/FOR MKA0/ %MOUNT-I-MOUNTED, TEST mounted on _FTMKT3$MKA0: 3 $ BACK/IGNO=LABEL *.COM MKA0:1.BCK/REW/LABED=NEWTST G %BACKUP-I-LBLOVRWRITE, volume label TEST         overwritten, new label 	 is NEWTST   1 %MOUNT-I-MOUNTED, NEWTST mounted on _FTMKT3$MKA0:  $     C It looks like it MOUNTs it to me! (VAX/VMS V6.1) It explicilty says F "NEWTST mounted" so I concluded that this means that BACKUP MOUNTs the tape.   $ It depends what you mean by "mount"!  	 >---o---<   D I just checked what happens if you don't change the label. You don't get any MOUNT message!  C OK. Then let me ask this: Why can you change the label on a mounted # tape with BACKUP but not with INIT?    [...]   Q > =============================================================================== P > Wayne Sewell, Tachyon Software Consulting  (281)812-0738   wayne@tachysoft.com: > http://www.tachysoft.com/www/tachyon.html and wayne.htmlQ > =============================================================================== R > Jake Blues:"You traded the Caddy for a microphone? ...... Okay, I can buy that."   AEF    ------------------------------    Date: 15 Aug 2005 08:59:37 -0700$ From: "AEF" <spamsink2001@yahoo.com>. Subject: Re: Help Needed: Tape Backup Save SetC Message-ID: <1124121577.224829.261460@z14g2000cwz.googlegroups.com>    Chris Allen wrote:E > Not being stubborn, just being cautious.  As I know next to nothing F > about DECNet I fear making a mistake that could result in a securityF > hole.  The VMS servers here run on a hardened network, and it is top* > priority that they are properly secured.     Nitpick alert!!!   It's spelled DECnet.   :-|        &-)   AEF    ------------------------------    Date: 15 Aug 2005 09:09:36 -0700$ From: "AEF" <spamsink2001@yahoo.com>. Subject: Re: Help Needed: Tape Backup Save SetC Message-ID: <1124122176.359203.321490@o13g2000cwo.googlegroups.com>    JF Mezei wrote:  > AEF wrote:/ > > Is there some reason you can't just try it?  > B > I know it is supposed to work. But when I wrote my procedures, ID > ALLOCATE MUA0: to make sure that the device remains in my process'C > control between multiple invocations of BACKUP. So I just want to ! > understand what really happens.  > I > > Now just why in the world would the first backup command dismount the H > > tape? And why wouldn't it be at the end of the tape when it is done? >  > J > Because it wants to release the device so that another user can use it ?  G All you had to do was run a backup and then SHOW DEVICE tape-drive/FULL 0 and this would have answered all your questions.   > I > I don't want user X to mount a tape and not release it for hours on end G > because the BACKUP command left the tape mounted and the user doesn't H > know that once his backups are done, he needs to specifically dismountG > it to release the device/tape and let other users use the tape drive.   G As others have said: training issue (or social engineering, or teaching  users how to share, whatever)    ------------------------------    Date: 15 Aug 2005 09:33:47 -0700( From: "Chris Allen" <ca.allen@gmail.com>. Subject: Re: Help Needed: Tape Backup Save SetC Message-ID: <1124123627.411221.134510@g44g2000cwa.googlegroups.com>   % hehe man you guys are rough on noobs.    ------------------------------    Date: 15 Aug 2005 09:31:37 -0700( From: "Chris Allen" <ca.allen@gmail.com>. Subject: Re: Help Needed: Tape Backup Save SetC Message-ID: <1124123497.454391.150870@f14g2000cwb.googlegroups.com>   F Thanks for the help.  I'll have to really crunch down and read as much? documentation as I can, when I get the time.  (If I ever do...)    ------------------------------    Date: 15 Aug 2005 09:36:09 -0700( From: "Chris Allen" <ca.allen@gmail.com>. Subject: Re: Help Needed: Tape Backup Save SetB Message-ID: <1124123769.518750.44880@g49g2000cwa.googlegroups.com>  H Interesting...  Have they fixed this in newer versions?  I'm running 7.1   ------------------------------    Date: 15 Aug 2005 09:40:19 -0700( From: "Chris Allen" <ca.allen@gmail.com>. Subject: Re: Help Needed: Tape Backup Save SetB Message-ID: <1124124019.292578.88510@f14g2000cwb.googlegroups.com>  A I'll try it with /FILE.   I'm almost sure it's the case that Norm D mentioned, because my directory focus at the time was a search list.   ------------------------------    Date: 15 Aug 2005 12:10:16 -0500 From: briggs@encompasserve.org. Subject: Re: Help Needed: Tape Backup Save Set3 Message-ID: <n+QAIXVuLxY$@eisner.encompasserve.org>   Z In article <00A484F4.F0C038D4.3@tachysoft.com>, Wayne Sewell <wayne@tachysoft.com> writes:J > Digital assumed, and quite rightly so, that most people would want to doM > multiple backups in one shot.  It's a hell of a lot more efficient than the O > dismount/remount thing, especially since backup would have to scan to the end Q > of the tape on every backup and the wait would get longer and longer with every M > backup as the amount of data on the tape increased.  That would be a lot of , > unnecessary wear on tape drives and media.  F Of course, Digital in its infinite wisdom then gutted the intelligenceA from BACKUP and/or the DLT drivers.  When appending a save set to E DLT tape when the tape is already positioned at logical EOV, it ended B up rewinding the tape, skipping forward (record by record IIRC) toE reposition itself at EOV and only then did it write the appended save ! set.  Ugly n^2 performance issue.   + Though I think it has since been addressed.   C DEC had to jump through some pretty nasty hoops in order to support  mount verification on tape.    	John Briggs   ------------------------------    Date: 15 Aug 2005 10:17:23 -0700$ From: "AEF" <spamsink2001@yahoo.com>. Subject: Re: Help Needed: Tape Backup Save SetC Message-ID: <1124126243.377787.171630@g14g2000cwa.googlegroups.com>    Chris Allen wrote:' > hehe man you guys are rough on noobs.    You gotta learn somehow! &-)   ------------------------------    Date: 15 Aug 2005 10:15:42 -0700$ From: "AEF" <spamsink2001@yahoo.com>. Subject: Re: Help Needed: Tape Backup Save SetC Message-ID: <1124126142.478236.257620@z14g2000cwz.googlegroups.com>    Chris Allen wrote:C > I'll try it with /FILE.   I'm almost sure it's the case that Norm F > mentioned, because my directory focus at the time was a search list.  E Yes, upon further reflection I believe you are right. My recolleciton E was that this usually happens when you have serveral search list-type # defaults as a list argument to DIR.    For example:  +    $ DIR SYS$MANAGER:LOGIN.COM,SYS$STARTUP:   E for which you don't even have to be SET DEFAULT to a search list-type  default.  G But later when I was back at work I found that your case is indeed most G likely due to being SET DEFAULT to SYS$MANAGER. The fact that the dates D and sizes of your files were exactly the same further supports that.   ------------------------------    Date: 15 Aug 2005 12:39:51 -0500- From: Kilgallen@SpamCop.net (Larry Kilgallen) . Subject: Re: Help Needed: Tape Backup Save Set3 Message-ID: <dTFbHPkC99au@eisner.encompasserve.org>   N In article <1124119597.326006.134530@o13g2000cwo.googlegroups.com>, "Chris All" h en" <ca.allen@gmail.com> writes:E > Not being stubborn, just being cautious.  As I know next to nothing F > about DECNet I fear making a mistake that could result in a securityF > hole.  The VMS servers here run on a hardened network, and it is top* > priority that they are properly secured.  F DECnet Phase IV on VMS is one of the most readily secured network APIs3 I have seen.  My list of things to check starts at:   R http://ljk.com/ljk/LJK_SECURITY_DOCUMENTATION/ljk_security_d_016.html#tests_decnet   The PDF version at  L http://ljk.com/ljk/LJK_SECURITY_DOCUMENTATION/LJK_SECURITY_LJKS-REF-V029.PDF   is a little easier to read.   D For sites subject to FISMA/FIPS 200/NIST 800-53, I recommend testingD all those DECnet elements, even if you are doing it with DCL command procedures.    ------------------------------    Date: 15 Aug 2005 10:40:43 -0700( From: "Chris Allen" <ca.allen@gmail.com>. Subject: Re: Help Needed: Tape Backup Save SetB Message-ID: <1124127642.999531.43060@z14g2000cwz.googlegroups.com>  G >    What's the /ful for? That only applies if you use /LIST. Check the 
 >    docs.  E Not sure how I missed that one.  And thanks for your suggestion about F the qualifiers I'll definantly order them correctly.  I'm usually veryG picky about that in other shell scripting, I just didn't notice/realize  they were out of order.   L >    Also, you might want to add /BLOCK=32256; or, if you don't care to everL >    copy a save set to disk as a save set, /BLOCK=65536 or whatever the max) >    is, assuming it works on your drive.   A The previous admin before me had /BLOCK=65536 set (he like me had F limited VMS knowledge).  So I looked it up in the hardcopy manuals, itF says this... "If the block size is set to a large value for a save setE on magnetic tape, it is possible for the magnetic tape to run off its E reel or for a large number of write errors to be logged."  Granted we E aren't using 9 track or something that can "run off it's reel" but "a A large numer of write errors" made me worry.  After reading that I @ wasn't comfortable using his block size setting so I removed it,F figuring the default was safer even if I lost a little bit of space on: the tape.   32,256 is the default block size for Files-11?  G Also I'm using a TZ89 DLT 35/70 tape drive.  These are new tape drives, @ before we were using 4mm DAT (the 65,536 was set for the old DDS drives)    ------------------------------  % Date: Mon, 15 Aug 2005 14:37:07 +0300 4 From: Mike Rechtman <michael.rechtman.nospam@hp.com>' Subject: Re: making unzip.exe a command & Message-ID: <4300A893.2E0F3787@hp.com>   Bill Gunshannon wrote: >    ...    >  > Even I am not optomistic. C                 ^^^^^^^^^^  hopefully expectant eye-doctor? (sorry,  couldn't resist)   Mike   >  > bill >  > --L > Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesF > bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. > University of Scranton   |@ > Scranton, Pennsylvania   |         #include <std.disclaimer.h>   --  E --------------------------------------------------------------------- E Usual disclaimer: All opinions are mine alone, perhaps not even that. ? Mike Rechtman                            *rechtman@tzora.co.il* F Kibbutz Tzor'a.                          Voice (home): 972-2-9908337  B   "20% of a job takes 80% of the time, the rest takes another 80%"E ---------------------------------------------------------------------    ------------------------------  % Date: Sun, 14 Aug 2005 22:14:53 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com>  Subject: Re: my 2 cents , Message-ID: <42FFFA7C.FBA7478B@teksavvy.com>  " susan_skonetski@hotmail.com wrote: > My feelings only  C And thank you for sharing them. Knowing how you feel is important.    > > He cares more about customers than Hollywood and I like that  0 Reference to previous CEO ???  I like that ! :-)    I > I am more concerned that people that are not customers may have tainted # > the water where VMS is concerned.   G Have you read the book from Lou Gerstner ? One of the things he changed F when he got the helm of IBM is trying to fidn out why some people wereE NOT customers.  Finding what causes a captive customer to continue to D pay doesn't improve the company. Finding out why some people chose aG competitor's product/services allows you to fix what is wrong with your > company and win those non-customers over to the customer side.  D Your note made it sound like non-customers are not worth any effort.E That is a wrong mentality to have. The opportunity to have discussion D with non-customers should be seen as being more valuable than having! contacts with existing customers.   H > out that they are not even purchasing customers.  He is a numbers guy,1 > you need to be willing to include your numbers.   A Then that is wrong. If someone takes the time to write to Hurd to H explain what is wrong with HP's handling of VMS and this is why they areN no longer VMS customers, that is worth a LOT more than some existing customer.      A > points in it, unfortunately these good points are surrounded by I > paragraphs of rants of how VMS has been mistreated as an orphaned child A > for the last decade.  And that essentially the companies senior H > management is imbeciles / morons, with the IQ of ice cubes and a waste
 > of skin   C Sue, this has to be taken in the context of HP's mishandling of VMS B between June 25 2001 And May 7th 2002 and beyond. Remember that HPB bought Compaq on the heals of the Alpha murder and then refused toF discuss plans for VMS during the period between merger announcement onC Sept 7th and May7th, followed by the infamous Stallard memos all of F which gave a strong impression that HP didn't care about VMS' past and@ wanted to continue the very "plan of record" that such distrust.  F The fact that customers would point to VMS' past mishandling is a goodF message to HP that HP needs to make verty public its goal to take trueF onwership of VMS and steer it to success instead of just stating "plan$ of record" with no change in policy.    G And the fact that Hollywood didn't react to cries from VMS folks asking G HP for changes in handling of VMS increased the fears by customers that A HP had same intentions of slowly winding down VMS over the years.   E Now, if Hurd really intends to give VMS a chance, the same applies to 1 him: he must send a loud and precise message that E he acknowledges the past mistakes with VMS and that HP would now take G VMS seriouesly and fully intent to levereage its potential and grow its . installed base. (and loud message = marketing)  B The message from customers and ex-customers hs been consistant forE years, yet the owner of VMS has not reacted to it. And as long as the ? owner of VMS reacts to this message, the message will continue.   H Now, from your point of view, you should be going to HP upper managementH and explain why customers are sending this message and that this message> has been the same for years because the onwer of VMS has neverE acknowledged that the lack of marketing  (either explicit or implicit H though press releases etc) is a big stumbling bliock to prevent VMS from$ growing (or stopping its shrinkage).    D > Not being a senior manager myself, I tried to think how they think3 > about the VMS community.  What is our reputation?   F If they cared about VMS as a product with profit and growth potential,H each successive owner of VMS would have taken those complaints seriouslyB about the previou owner's mistakes and made changes. Instead, bothC Compaq and HP have said "we'll continue the plan of record" sending K customers the wrong message that they plan on continue to de-emphasise VMS.     G > speakers I know don't mind hard questions, in fact they like them. As E > long as they are in their scope of responsibility they are happy to  > answer them.  C If a person with VMS interest asks a "why aren't you marketing VMS" C question in a technical forum to someone who isn't in a position to ? answer, YOU shoudl go back to upper management and explain that E customers are so desperate to see upper management deal with VMS in a D positive sense that they catch any representative of HP tyhey can toA send the message that VMS needs marketing. A sign of desperation.   F The problem is simple. Simple solutions. Yet, HP refuses to acknoledge) and fix this simple problem of marketing.    ------------------------------    Date: 15 Aug 2005 01:01:31 -0700* From: "MDPlatts" <martin.platts@cdl.co.uk>M Subject: Re: Problems with updating firmware on SDLT1 drives using LTT on VMS C Message-ID: <1124092891.022908.126710@f14g2000cwb.googlegroups.com>   F I just download the firmware file (*.FRM) from the following URL, pickF your tape drive and then any of the individual vartieties and then anyC really of the operating systems (windows usually is more up to date D than OpenVMS) from the pages after that - then put the FRM file into4 the FIRMWARE directory created when you install LTT.  k http://h20000.www2.hp.com/bizsupport/TechSupport/Product.jsp?prodTypeId=12169&prodCatId=304613&locale=en_US    ------------------------------   Date: 15 Aug 2005 14:33:48 GMT( From: bill@cs.uofs.edu (Bill Gunshannon) Subject: Re: SAMBA for VMS, Message-ID: <3mbnecF164928U1@individual.net>  - In article <Hi1Me.1$N96.339@news.uswest.net>, 9 	"Michael D. Ober" <obermd.@.alum.mit.edu.nospam> writes: - > http://www.pi-net.dyndns.org/anonymous/jyc/  > M > Based on the Samba 2.2.8 source.  Versions for Alpha and VAX.  It's best if F > you compile it from the source, a-la Unix.  VMS 7.3 has the required; > compilers and linkers as part of the system distribution.  >   F Thanks for the pointer, but it has become somewhat moot at this point.F Something I did in trying to get the other versions running has causedH the system to no longer boot.  Add to that the fact that for some reasonE neither of the CDROM's are visible to the VAX anymore so I can't boot H from the CD and re-install.  And this close to the start of the semester0 I am unlikely to have free time to play with it.   bill   --  J Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>       ------------------------------  % Date: Mon, 15 Aug 2005 08:05:26 -0600 6 From: "Michael D. Ober" <obermd.@.alum.mit.edu.nospam> Subject: Re: SAMBA for VMS- Message-ID: <Hi1Me.1$N96.339@news.uswest.net>   + http://www.pi-net.dyndns.org/anonymous/jyc/   K Based on the Samba 2.2.8 source.  Versions for Alpha and VAX.  It's best if D you compile it from the source, a-la Unix.  VMS 7.3 has the required9 compilers and linkers as part of the system distribution.    Mike.   5 "Bill Gunshannon" <bill@cs.uofs.edu> wrote in message & news:3m4uv9F159sepU1@individual.net..., > I'm back looking for a little help, again. > B > Can someone point me at a SAMBA distribution for the VAX that is@ > relatively simple to install?  So far I have found three.  TheA > first two didn't seem to put things in the right subdirectories @ > so installing failed and the third came as an ODS file which I > have no idea what to do with.  > A > I would like to have SAMBA running for the upcoming semester in A > order to make it easy for students to move files on and off the 
 > VMS system.  >  > bill >  > --  L > Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesF > bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. > University of Scranton   |@ > Scranton, Pennsylvania   |         #include <std.disclaimer.h> >    ------------------------------  % Date: Mon, 15 Aug 2005 10:37:21 -0400 ' From: Glenn Everhart <Everhart@gce.com>  Subject: Re: SAMBA for VMS, Message-ID: <eYWdnbtBI-SNNJ3eRVn-iQ@rcn.net>  F Equivalently, connect to a virtual disk. If using the LDdriver or someH variants of xrdriver (sigtapes, freeware) the file can be fragmented. If# using vddriver, make it contiguous.   C The CD writer in cdrtools (sigtapes) doesn't care what filename you K have. If using a pc, rename the file from .ods to .iso to let some of their = software have a clue, if it can't be told to open "any file".   J The sigtapes have had versions of samba including the V2.8 that was postedG online, in addition to the then-most-recent sources posted by the samba F working groups. John Malmberg's V2.2-12 port is on the Fall 2004 tapes also.    Glenn Everhart   Bill Gunshannon wrote:- > In article <42FD6DB1.FF70697D@comcast.net>, 7 > 	David J Dachtera <djesys.nospam@comcast.net> writes:  >  >>David J Dachtera wrote:  >> >>>Bill Gunshannon wrote:  >>> . >>>>I'm back looking for a little help, again. >>>>D >>>>Can someone point me at a SAMBA distribution for the VAX that isB >>>>relatively simple to install?  So far I have found three.  TheC >>>>first two didn't seem to put things in the right subdirectories B >>>>so installing failed and the third came as an ODS file which I! >>>>have no idea what to do with.  >>> B >>>The .ODS file is disk image. Burn it to a CD-R and you will get" >>>something that can be MOUNTed.  >>E >>Alternately, you can LD CONNECT it to an LD device, and MOUNT that.  >  > A > Yeah, I could.  If I had a clue what that actually meant or how  > to do it.  >  > bill >    ------------------------------  % Date: Mon, 15 Aug 2005 10:46:47 -0400 ' From: Glenn Everhart <Everhart@gce.com>  Subject: Re: SAMBA for VMS, Message-ID: <1ZednZqwgfTbNp3eRVn-rA@rcn.net>  L Thanks for the comment, but LDdriver was not one of mine. I did VDdriver andO some others, but LDdriver was done internally at DEC by Ton D. (sorry, I forget Q the spelling of his last name). The old RSX11M/M+ VDdriver was Ralph Stamerjohn's N thing, though I and others did various tweaks. Several of the tweaks from bothR VDdriver variants for VMS and the 11M flavors got duplicated in LDdriver (possiblyO independently devised; I don't know whether Ton ever even knew of the existence M of VDdriver, FDdriver (aka xrdriver aka a few other names; it used a separate P process to do actual i/o, so could encrypt, decompress, use tape as disk, handleP fragmented files, do remote disk over decnet or ip, do various shadowing tricks,L etc. etc. using different host processes). LDdriver pulls a set of functionsM together as a single facility, instead of having a host of slightly different O virtual disk drivers for different stuff with only a few documents to guide the " unwary user about which was which.L    I believe that lddriver does have some limits on how badly fragmented itsL container files can be but don't actually know. If you have a file that mustH be treated as a disk and is in hundreds or thousands of fragments (ecch)I then something like the old frdriver would be your best bet, since it has I a real user mode process doing real file read/write operations...slow but 
 functional...   , But hopefully that doesn't happen too often.   Glenn Everhart everhart@gce.com   Dan Foster wrote:   d > In article <VO5$2o0o8q68@eisner.encompasserve.org>, Larry Kilgallen <Kilgallen@SpamCop.net> wrote: > ; >>"LD" does not seem to be a command on my VMS V7.3 system.  >  > D > I think LD was bundled with VMS starting with V7.3-1? I'm not sure) > LDDRIVER ever shipped with OpenVMS/VAX.  >  > F >>Perhaps you meant to specify that some software beyond VMS should be >>installed first. >  > D > The Freeware CD does include LDDRIVER and works great on Alpha. No% > idea if it works with VAX (or I64).  > J > I seem to recall that LDDRIVER was a Glenn Everhart creation (and a very > nicely done one, at that). >  > -Dan   ------------------------------  % Date: Mon, 15 Aug 2005 10:54:42 -0400 ' From: Glenn Everhart <Everhart@gce.com>  Subject: Re: SAMBA for VMS, Message-ID: <VK6dnTJi3cm9MJ3eRVn-qA@rcn.net>   David J Dachtera wrote:    > Bill Gunshannon wrote: > 5 >>In article <VO5$2o0o8q68@eisner.encompasserve.org>, 9 >>        Kilgallen@SpamCop.net (Larry Kilgallen) writes:  >>c >>>In article <42FD6DB1.FF70697D@comcast.net>, David J Dachtera <djesys.nospam@comcast.net> writes:  >>>  >>>>David J Dachtera wrote:  >>>> >>>>>Bill Gunshannon wrote:  >>>>> 0 >>>>>>I'm back looking for a little help, again. >>>>>>F >>>>>>Can someone point me at a SAMBA distribution for the VAX that isD >>>>>>relatively simple to install?  So far I have found three.  TheE >>>>>>first two didn't seem to put things in the right subdirectories D >>>>>>so installing failed and the third came as an ODS file which I# >>>>>>have no idea what to do with.  >>>>> D >>>>>The .ODS file is disk image. Burn it to a CD-R and you will get# >>>>>something that can be MOUNTed.  >>>>G >>>>Alternately, you can LD CONNECT it to an LD device, and MOUNT that.  >>> < >>>"LD" does not seem to be a command on my VMS V7.3 system.G >>>Perhaps you meant to specify that some software beyond VMS should be  >>>installed first.  >>H >>See what I mean!  If someone with Larry's experience doesn't know what( >>this means, how could I possibly know? >  > 2 > Have you read the most recent VMS Tech. Journal? > H > Sneak preview: Having been unsupported freeware for some years now, LD > is about to become supported!  > F > Support is "latent" in V7.3-2 and V8.2. The driver and API / controlI > program are there, but the verb isn't. The V8.0 LD kit, as indicated by J > the Tech Journal article, is not only compatible back to at least V7.2-2H > (I think), but does some really, REALLY neat stuff! ...like allowing aI > portion of an unMOUNTed volume to be treated as a virtual volume (whole , > new take on the "partitioning" paradigm!). > A vddriver has had that for some years too...specify /lbn and /len. D I (finally) got round to adding the access call to asnvd in the lateF 90s so that /foreign mounting the underlying disk was not necessary toM spin it up and get it to go online. It is a neat thing though and much better G packaged by lddriver than vddriver ever was. Also adds a good many more  tweaks..   ------------------------------  % Date: Mon, 15 Aug 2005 11:03:28 -0400 ' From: Glenn Everhart <Everhart@gce.com>  Subject: Re: SAMBA for VMS, Message-ID: <J9adnXGsspOzMp3eRVn-3A@rcn.net>  D There aren't many 2 letter combos starting in D that are not alreadyC used for some type of disk or other. Maybe some of the old ones can H be reused. I did that with a couple of my old drivers, reusing DF and DTI for peripherals (rf11 disk or dectape) that aren't too common (to say the F least) today.  DT was at least used for a tape pretending to be a diskM though, so as not to dishonor the memory of a wonderful peripheral, the TU56. H Seems to me the DC disk (RC11) is also available in the same sense. RF11I was small enough (256kb or so) but RC11 size was something like 32k words H (memory gets fuzzy). I can't imagine anybody botherint to connect such a beast today.  F If you have sources, changing driver name is pretty easy; just be sureF in addition to the obvious places to get the mscp_media_id field reset right.   Glenn Everhart   Stanley F. Quayle wrote:  2 > On 13 Aug 2005 at 23:17, David J Dachtera wrote: > E >>Having been unsupported freeware for some years now, LD is about to  >>become supported!  >  > F > Before it becomes supported, can the device type be changed so that E > the virtual disks come up as "D" something?  To see the disks on a  B > system, I do a "SHOW DEVICE D", and this command will miss them  > all... >  > --Stan Quayle  > Quayle Consulting Inc. >  > ----------/ > Stanley F. Quayle, P.E. N8SQ  +1 614-868-1363 5 > 8572 North Spring Ct., Pickerington, OH  43147  USA 2 > stan-at-stanq-dot-com       http://www.stanq.com+ > "OpenVMS, when downtime is not an option"  >  >    ------------------------------  % Date: Mon, 15 Aug 2005 14:46:32 +0300 4 From: Mike Rechtman <michael.rechtman.nospam@hp.com>@ Subject: Re: TCPIP Services: SMTP bug/wrong logic with 450 codes& Message-ID: <4300AAC8.78CFF6B4@hp.com>   JF Mezei wrote:  > J > The VMS SMTP server fails to handle greylisting by the receiving server.C > (VAX TCPIP Services 5.3-2). And it gives the used erroneous error P > message, and logs don't show the real error message unless you enable tracing. > % > The standard (RFC 2821) stipulates:  > ; > "450 Requested mail action not taken: mailbox unavailable   >          (e.g., mailbox busy)" > = > More specifically for 400 seried messages: (page 42 of RFC)  > - > "4yz   Transient Negative Completion reply"  > 1 > Which means: this isn't fatal, try again later.  > I > 450  is used by smart SMTP servers with greylisting implemented to tell G > the sending SMTP server to retry in 30 minutes of more. (spam senders  > don't bother retrying).  > [ > =========================================================================================     A This has been fixed in the latest ECO (ECO 5 IIRC) for TCPIP V5.4 H I don't know what to suggest for your version, (5.3-xx) Can you upgrade?   Mike.      --  E --------------------------------------------------------------------- E Usual disclaimer: All opinions are mine alone, perhaps not even that. ? Mike Rechtman                            *rechtman@tzora.co.il* F Kibbutz Tzor'a.                          Voice (home): 972-2-9908337  B   "20% of a job takes 80% of the time, the rest takes another 80%"E ---------------------------------------------------------------------  -----BEGIN GEEK CODE BLOCK-----  Version: 3.1: GCM/CS d(-)pu s:+>:- a++ C++ U-- L-- W++ N++ K? w--- V+++$6 PS+ PE-- t 5? X- tv-- b+ DI+ D-- G e++ h--- r+++ y+++@ ------END GEEK CODE BLOCK------    ------------------------------    Date: 15 Aug 2005 14:17:58 +01006 From: peter@langstoeger.at (Peter 'EPLAN' LANGSTOEGER)@ Subject: Re: TCPIP Services: SMTP bug/wrong logic with 450 codes, Message-ID: <4300a416$1@news.langstoeger.at>  ] In article <4300AAC8.78CFF6B4@hp.com>, Mike Rechtman <michael.rechtman.nospam@hp.com> writes: B >This has been fixed in the latest ECO (ECO 5 IIRC) for TCPIP V5.4I >I don't know what to suggest for your version, (5.3-xx) Can you upgrade?   ; Only if you (finally) deliver TCPIP V5.4 (and V5.5) for VAX    --   Peter "EPLAN" LANGSTOEGER % Network and OpenVMS system specialist  E-mail  peter@langstoeger.atF A-1030 VIENNA  AUSTRIA              I'm not a pessimist, I'm a realist   ------------------------------  % Date: Mon, 15 Aug 2005 16:38:37 +0300 4 From: Mike Rechtman <michael.rechtman.nospam@hp.com>@ Subject: Re: TCPIP Services: SMTP bug/wrong logic with 450 codes& Message-ID: <4300C50D.50EF12DD@hp.com>    Peter 'EPLAN' LANGSTOEGER wrote: > _ > In article <4300AAC8.78CFF6B4@hp.com>, Mike Rechtman <michael.rechtman.nospam@hp.com> writes: D > >This has been fixed in the latest ECO (ECO 5 IIRC) for TCPIP V5.4K > >I don't know what to suggest for your version, (5.3-xx) Can you upgrade?  > = > Only if you (finally) deliver TCPIP V5.4 (and V5.5) for VAX  >  > -- > Peter "EPLAN" LANGSTOEGER ' > Network and OpenVMS system specialist  > E-mail  peter@langstoeger.atH > A-1030 VIENNA  AUSTRIA              I'm not a pessimist, I'm a realist  # Sorry, missed the "VAX" in the O.P. G If you have support (or if you know of someone who has, and who has the G same problem) perhaps you should get a request for back-porting the fix  into the system. (Point your support person at:6 http://customer-escalations.rose.hp.com/ccp_index.htm D From there should click on  "Customer Commitment Request Form" which takes you here: L http://customer-escalations.rose.hp.com/CCP/CustomerCommitmentRequest1.8.docE Then fill it out and send it to the relevant recipient.  In this case  that; would be "OpenVMS requests" and/or refer to PTR 70-5-2813 )   E Also, quoting some of the replies when the request arose for a fix on  V5.4:   B      Meanwhile, a good workaround would be to specify an Alternate Gateway.H     If you send your outbound mail via your (or your ISP's) SMTP server,6     it's then up to that server how to handle retries.   Passing the buck, if you can...    Mike   --  E --------------------------------------------------------------------- E Usual disclaimer: All opinions are mine alone, perhaps not even that. ? Mike Rechtman                            *rechtman@tzora.co.il* F Kibbutz Tzor'a.                          Voice (home): 972-2-9908337  B   "20% of a job takes 80% of the time, the rest takes another 80%"E ---------------------------------------------------------------------  -----BEGIN GEEK CODE BLOCK-----  Version: 3.1: GCM/CS d(-)pu s:+>:- a++ C++ U-- L-- W++ N++ K? w--- V+++$6 PS+ PE-- t 5? X- tv-- b+ DI+ D-- G e++ h--- r+++ y+++@ ------END GEEK CODE BLOCK------    ------------------------------   End of INFO-VAX 2005.454 ************************