INFO-VAX Fri, 21 Dec 2007 Volume 2007 : Issue 698 Contents: Re: "snapshot" backup and HBVS Re: 20+ year old encrypted source code Re: COBOL: get buffered IO from process id Re: COBOL: get buffered IO from process id Re: COBOL: get buffered IO from process id Re: COBOL: get buffered IO from process id Re: COBOL: get buffered IO from process id Re: COBOL: get buffered IO from process id Re: IP standby query Re: IP standby query Re: IP standby query OT: Merry Christmas to c.o.v. ! Re: OT: Merry Christmas to c.o.v. ! Re: OT: Merry Christmas to c.o.v. ! Re: OT: Merry Christmas to c.o.v. ! Re: OT: Merry Christmas to c.o.v. ! OT: The Micro$oft Christmas Party... Samba Gains Legal Access to Microsoft Network File Protocols Re: Volume label. Re: Volume label. Re: Volume label. Re: Volume label. Re: Volume label. Re: Volume label. Re: Volume label. Re: Volume label. Re: Volume label. Re: Volume label. Re: WANTED: PDSDUMP information from VAX 780 (c1987) ---------------------------------------------------------------------- Date: Fri, 21 Dec 2007 11:25:04 -0700 From: Keith Parris Subject: Re: "snapshot" backup and HBVS Message-ID: Phillip Helbig---remove CLOTHES to reply wrote: >>However, if I have to go back to this nightly snapshot backup, then I >>have the choice betwwen a full shadow copy and living with one >>shadow-set, member, neither of which is acceptable. > > That should be: living with a one-member shadow set How about this: Place the 3rd member into the set, allow the Mini-Copy to complete, wait until a quiet time and do your DISMOUNT/POLICY=MINICOPY to remove it from the set, then, off-line with respect to production: 1) Mount the former 3rd member for read-only Files-11 access 2) Do a BACKUP/PHYSICAL from that disk to two more disk drives of at least the same size. 3) Mount these two additional drives as a two-member shadowset; while I haven't tested this specific instance recently I would expect a Full-Copy operation to result (at least with recent MOUNT versions/patches; prior to that in my tests the 2nd member would be included as a member of the shadowset and a Full-Merge initiated), but the time required for the Full-Copy will be minimized as the contents are already identical. As long as this nightly Full-Copy doesn't take more than an hour or two to complete, it seems like it might possibly meet your needs, since the Full-Copy would be occurring outside Production and during off-hours. And support for more than 3 members in a shadowset is not as improbable as it once was. ------------------------------ Date: Fri, 21 Dec 2007 10:29:31 -0700 From: Keith Parris Subject: Re: 20+ year old encrypted source code Message-ID: FrankS wrote: > I tried everything that has been mentioned here and suggestions sent > to me privately, all to no avail. Another compression option available around that timeframe was the DCX compression utility routines available in OpenVMS. They used Huffman encoding. See http://h71000.www7.hp.com/doc/83final/4493/4493pro_008.html ------------------------------ Date: Fri, 21 Dec 2007 02:43:48 -0800 (PST) From: ythefoe@gmail.com Subject: Re: COBOL: get buffered IO from process id Message-ID: <419b462d-5a37-4de4-8d39-b9364e1d2a6a@s19g2000prg.googlegroups.com> On 21 dec, 06:05, Hein RMS van den Heuvel wrote: > On Dec 20, 9:21=A0pm, Jim Duff wrote: > > > ythe...@gmail.com wrote: > > > Does someone know how to extract the buffered IO using the process > > Presuming that by "extract the buffered IO" you mean "get the buffered > > IO count", then here you go. =A0Obviously, you need to fill in the PID > > variable by some method, and you need appropriate privileges (either > > GROUP or WORLD) to request information on other processes. > > Jim shows a fine solution for the question asked. > If you wanted multiple process counters, then you may want to look at > SYS$GETJPI which is only a little harder to program but a lot more > efficient. > > Anyway.... > > WHY would a cobol program want to know BUFIO (count) ? > Do you even know what constritues a direct IO? > (folks who know exactly what bufio means, already know how to get > it :-) > What is the intended usage of the value obtained. > "what problem are you really trying to solve?" > > Cheers, > Hein. @ Hein: I need to rewrite an old program (due to loss of sources) that check the inactivity of users in a time frame... One of the parameters to check is the bufferid IO-count... @ Jim: thanks for your fast reply... I'm gonna try your example... ------------------------------ Date: Fri, 21 Dec 2007 07:26:48 -0500 From: "Ken Robinson" Subject: Re: COBOL: get buffered IO from process id Message-ID: <7dd80f60712210426xd498156j8c444c58595be6fe@mail.gmail.com> On Dec 21, 2007 5:43 AM, wrote: > > WHY would a cobol program want to know BUFIO (count) ? > > Do you even know what constritues a direct IO? > > (folks who know exactly what bufio means, already know how to get > > it :-) > > What is the intended usage of the value obtained. > > "what problem are you really trying to solve?" > @ Hein: I need to rewrite an old program (due to loss of sources) that > check the inactivity of users in a time frame... One of the parameters > to check is the bufferid IO-count... If you're re-writing an old program, it might be possible to re-write it in DCL and just forget trying to do it in COBOL. What else does the old program do? Ken ------------------------------ Date: Fri, 21 Dec 2007 04:37:59 -0800 (PST) From: Hein RMS van den Heuvel Subject: Re: COBOL: get buffered IO from process id Message-ID: <3605f06a-9ef7-4396-84f3-969de08b31cc@b1g2000pra.googlegroups.com> On Dec 21, 5:43=A0am, ythe...@gmail.com wrote: > On 21 dec, 06:05, Hein RMS van den Heuvel > > "what problem are you really trying to solve?" > @ Hein: I need to rewrite an old program (due to loss of sources) that > check the inactivity of users in a time frame... Ah, understood. Sort of. Inactivity detectors are those sophisticated programs which - lull management into a false sense of security, - piss off end users, - and increase the system activity in general by probing and waking up processes which are sleeping just fine - make some users create occasional false activity just to prevent from being zapped, - makes other users re-start stuff and re-open file and and re-load DB buffers and then go wait again. I forgot under what circumstances they did any good other than keeping programmers employed. I know of two credible reasons: - per user licenses (for crappy softwware) - modem pools (who still uses a modem?) Anyway... Odd language choice for such tool, but if all you have is a hammer,... use it. Sounds like you would also get the imagenam and maybe CPU time. So SYS$GETJPI could be easier for all than multiple LIB$GETJPI calls. Be sure to check out: OpenVMS Programming Concepts Manual E.7 COBOL Implementations http://www.itec.suny.edu/scsys/vms/OVMSDOC073/V73/5841/5841pro_089.html#4512= _vms_types_cobol Example 20-6 System Service Call in COBOL http://www.itec.suny.edu/scsys/vms/OVMSDOC073/V73/5841/5841pro_056.html#cobo= l_call_lang_ex Hein http://biophyrs.bp.dal.ca/OpenVMS/83final/5841/5841pro_059.html#cobol_call_l= ang_ex One of the parameters > to check is the bufferid IO-count... > > @ Jim: thanks for your fast reply... I'm gonna try your example...- Hide q= uoted text - > > - Show quoted text - ------------------------------ Date: Fri, 21 Dec 2007 04:42:42 -0800 (PST) From: Hein RMS van den Heuvel Subject: Re: COBOL: get buffered IO from process id Message-ID: <4f7f0dc0-6b72-4c8b-bd5c-d7d35d67b670@d21g2000prf.googlegroups.com> On Dec 21, 7:37=A0am, Hein RMS van den Heuvel wrote: > On Dec 21, 5:43=A0am, ythe...@gmail.com wrote: > > > On 21 dec, 06:05, Hein RMS van den Heuvel > Be sure to check out: > > OpenVMS Programming Concepts Manual http://www.itec.suny.... Oops, sorry for the non-hp-site Manual reference. I should have googled for : +cobol +sys$getjpi +site:hp.com Hein. ------------------------------ Date: Fri, 21 Dec 2007 13:11:27 GMT From: =?ISO-8859-1?Q?Jan-Erik_S=F6derholm?= Subject: Re: COBOL: get buffered IO from process id Message-ID: <3YOaj.1894$R_4.1450@newsb.telia.net> > I need to rewrite an old program (due to loss of sources) > that check the inactivity of users in a time frame... Don't do that. Download WATCHER instead : http://vms.process.com/scripts/fileserv/fileserv.com?WATCHER or : http://h71000.www7.hp.com/freeware/freeware80/watcher/ Regards, Jan-Erik. ------------------------------ Date: Fri, 21 Dec 2007 07:33:26 -0800 (PST) From: "David P. Murphy" Subject: Re: COBOL: get buffered IO from process id Message-ID: <201ebebe-fd30-48b6-9036-d6c09e8adc64@y5g2000hsf.googlegroups.com> On Dec 21, 7:37 am, Hein RMS van den Heuvel wrote: > Inactivity detectors are those sophisticated programs which > - lull management into a false sense of security, > - piss off end users, > - and increase the system activity in general by probing and waking up > processes which are sleeping just fine > - make some users create occasional false activity just to prevent > from being zapped, > - makes other users re-start stuff and re-open file and and re-load DB > buffers and then go wait again. Amen, brother. Don't forget the worst possible case, which actually happened to me once: the watcher did not take subprocesses into account, so it would constantly judge the parent to be inactive and would therefore kill it without warning . . . unless you consider the entire process tree disappearing to be a "warning". ok dpm ------------------------------ Date: Fri, 21 Dec 2007 01:56:24 -0800 (PST) From: issinoho Subject: Re: IP standby query Message-ID: <479bc350-a1c9-40b0-9085-d818267d28bd@y5g2000hsf.googlegroups.com> On Dec 20, 8:21 pm, JF Mezei wrote: > issinoho wrote: > > My question is: what constitutes the 'something'? What needs to happen on > > Node A such that node B gets the live IP address? > > Second question: is there any way of stimulating this swap from either of > > the nodes? > > For TCPIP services with just the alias strategy (no failsafe process), > it is a simple clusterwide lock. First node that gets the lock gets to > make the alias IP active. When that node goes out, the lock becomes free > and the next node then takes the lock and activates its alias. > > To simulate it, upi use the IFCONFIG > > I think it is > > $ifconfig WE0 -alias 10.0.0.50 which will disable the alias on that > node and let the next node take it. > > $ifconfig WE0 alias 10.0.0.50/16 ipmtu 1500 will re-activate the alias > on that node (will go on stand by mode if another node already has it). > > WE0 is the interface name (IFCONFIG -a to get the list of interfaces on > that node) Thanks, JF, that works a treat. ------------------------------ Date: Fri, 21 Dec 2007 06:06:36 -0500 From: JF Mezei Subject: Re: IP standby query Message-ID: <476b9e63$0$16152$c3e8da3@news.astraweb.com> >> $ifconfig WE0 alias 10.0.0.50/16 ipmtu 1500 will re-activate the alias >> on that node (will go on stand by mode if another node already has it). As an additional note, you can put that command in: SYS$STARTUP:TCPIP$SYSTARTUP.COM (site specific startup commands), as well as: $IFCONFIG :== "$tcpip$ifconfig.exe" before invoking it. This will define that alias whenever tcpip services gets started. (this is the simpler IP failover scheme not the one that is managed by the failsafe process. Note also that this would happen to the first node booting which may not be the prefered node to get the IP after a power failure for instance. One would have to hunt the lock table to figure out the name of the lock taken by a node when it assumes an alias in order to then write a program to figure out which node has the alias at the moment (and then decide if you should be taking it and telling that node to give it up). ------------------------------ Date: Fri, 21 Dec 2007 03:54:44 -0800 (PST) From: "Bart.Zorn@gmail.com" Subject: Re: IP standby query Message-ID: <8bfb1375-b51e-4dd9-91f7-136a3eb33a45@e23g2000prf.googlegroups.com> The command $ ifconfig -a reports, among other things, which node and interface holds the active address! Bart Zorn On Dec 21, 12:06 pm, JF Mezei wrote: > >> $ifconfig WE0 alias 10.0.0.50/16 ipmtu 1500 will re-activate the alias > >> on that node (will go on stand by mode if another node already has it). > > As an additional note, you can put that command in: > > SYS$STARTUP:TCPIP$SYSTARTUP.COM (site specific startup commands), as > well as: > > $IFCONFIG :== "$tcpip$ifconfig.exe" > before invoking it. > > This will define that alias whenever tcpip services gets started. (this > is the simpler IP failover scheme not the one that is managed by the > failsafe process. > > Note also that this would happen to the first node booting which may not > be the prefered node to get the IP after a power failure for instance. > > One would have to hunt the lock table to figure out the name of the lock > taken by a node when it assumes an alias in order to then write a > program to figure out which node has the alias at the moment (and then > decide if you should be taking it and telling that node to give it up). ------------------------------ Date: Fri, 21 Dec 2007 03:48:53 -0500 From: JF Mezei Subject: OT: Merry Christmas to c.o.v. ! Message-ID: <2d171$476b7deb$cef8887a$17422@TEKSAVVY.COM> Merry Christmas and Happy New Year to all readers of c.o.v. (1) May this festive season bring you lots of snow (2), joy, happiness in your family. May the next 12 months bring you success with your challenges and endeavours. Disclaimer: (1) If you are offended by the expression "Merry Christmas", then substitute it with "Happy Holidays". If your culture's calendar doesn't end on Dec 31, then disregard the "Happy New Year" expression. No offense meant to anyone for whom those wishes may be derogatory. (2) For those in southern hemisphere, read "sun and fun at the beach" instead of "snow". No animals were harmed in the making of the announcement. Any use of the word "man" or "mankind" is meant to apply equally to all sexes. This message has a very low carbon footprint. Only one fart was produced while typing it. \ ------------------------------ Date: Fri, 21 Dec 2007 05:16:15 -0800 (PST) From: ultradwc@gmail.com Subject: Re: OT: Merry Christmas to c.o.v. ! Message-ID: <812296be-a361-47f8-89ae-ab7fe9bab7f6@i3g2000hsf.googlegroups.com> On Dec 21, 3:48=A0am, JF Mezei wrote: > Merry Christmas and Happy New Year to all readers of c.o.v. (1) > > May this festive season bring you lots of snow (2), joy, happiness in > your family. > > May the next 12 months bring you success with your challenges and > endeavours. > > > Disclaimer: > > (1) If you are offended by the expression "Merry Christmas", then > substitute it with "Happy Holidays". If your culture's calendar doesn't > end on Dec 31, then disregard the "Happy New Year" expression. No > offense meant to anyone for whom those wishes may be derogatory. > > (2) For those in southern hemisphere, read "sun and fun at the beach" > instead of "snow". > > No animals were harmed in the making of the announcement. > > Any use of the word "man" or "mankind" is meant to apply equally to all > sexes. > > This message has a very low carbon footprint. Only one fart was produced > while typing it. > \ > Merry Christmas ... and the holiday is Christmas, and if that offends you then go to work Tuesday ... ------------------------------ Date: Fri, 21 Dec 2007 13:36:06 GMT From: =?ISO-8859-1?Q?Jan-Erik_S=F6derholm?= Subject: Re: OT: Merry Christmas to c.o.v. ! Message-ID: ultradwc@gmail.com wrote: > and the holiday is Christmas, and if that offends you > then go to work Tuesday ... Or Monday, which is the main "Christmas" over here... :-) And of course, over here "Christmas" have very little to do with anything that happend to happen 2008 years ago... :-) ------------------------------ Date: Fri, 21 Dec 2007 06:32:04 -0800 (PST) From: IanMiller Subject: Re: OT: Merry Christmas to c.o.v. ! Message-ID: <5595e467-138e-44c2-8337-a4bdb255d840@n20g2000hsh.googlegroups.com> Happy Eid ul-Adha (yesterday?) Happy Winter Solstice (tomorrow) BBC web site - always handy :-) http://www.bbc.co.uk/religion/tools/calendar/month.shtml?200712 May 2008 bring more relevant content to this newsgroup. ------------------------------ Date: Fri, 21 Dec 2007 06:46:04 -0800 (PST) From: AEF Subject: Re: OT: Merry Christmas to c.o.v. ! Message-ID: <51e3d732-ea17-49ca-b828-bc2814671411@l1g2000hsa.googlegroups.com> On Dec 21, 9:32 am, IanMiller wrote: > Happy Eid ul-Adha (yesterday?) > Happy Winter Solstice (tomorrow) > > BBC web site - always handy :-) > > http://www.bbc.co.uk/religion/tools/calendar/month.shtml?200712 > > May 2008 bring more relevant content to this newsgroup. Then please ask those who bash people who are so quick to say RTFM in response even to questions that are answered more easily and quickly than scolding the poster with RTFM to stop it!!! If this newsgroup were flooded with FAQ's, I could see a need for RTFM- scolding. But I don't really see such a flood here! Scolding posters will not help increase the use of VMS! So we get several types of posts: Obnoxious spam like the MI5 stuff. Off-topic posts by regular members. "FAQ's" which are scolded by nasty chants of "RTFM". Highly involved technical questions about the very complex innards of some particular strange device most of us (well, at least I [!]) haven't heard of. OK, I exaggerate, but if we insist on only the last type of post there will be very few posts indeed. Answering the FAQ's with a gentle mention of the fine manuals and other fine resources would be more appropriate, IMO. AEF ------------------------------ Date: Fri, 21 Dec 2007 15:43:20 GMT From: VAXman- @SendSpamHere.ORG Subject: OT: The Micro$oft Christmas Party... Message-ID: Merry Christmas to all of the regulars and lurkers save for one here on comp.os.vms. There's no office Christmas Party here in the VAXcave but I did light up my John Wisniewski Spirit Award with my 200mW DPSS laser for a bit of holiday ambience on my desk. I'll post a photo before the day's out. However, I thought I would share some of the Micro$oft Christmas Party scuttlebutt with you... http://www.tmesis.com/M$xmas.html -- 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, 21 Dec 2007 18:44:36 +0000 From: "Main, Kerry" Subject: Samba Gains Legal Access to Microsoft Network File Protocols Message-ID: All, This might be of interest to this newsgroup. http://www.eweek.com/article2/0,1895,2239112,00.asp "Samba and the Software Freedom Law Center sign an agreement with Microsoft= that gives them full access to Microsoft's server protocols." "On Dec. 20, the Samba Group and the Software Freedom Law Center announced = a deal with Microsoft that places all of Microsoft's network protocols needed for progr= ams to work with Windows Server into the hands of the newly formed Protocol Freedom Informat= ion Foundation. The PFIF is a U.S.-based nonprofit corporation. It will make Microsoft's se= rver network protocol documentation available to open-source developers such as Samba, which crea= tes programs for Windows Server interoperability, and private companies. This informatio= n is provided under an NDA (nondisclosure agreement) and developers must agree to the NDA= before gaining access to the documentation." [snip .. see url for rest of article] 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, 21 Dec 2007 02:39:42 -0800 (PST) From: Bob Gezelter Subject: Re: Volume label. Message-ID: <952eb3b2-dc61-4afc-ac85-a590a2423385@i3g2000hsf.googlegroups.com> On Dec 20, 4:04 pm, Fred Bach wrote: > Bob Gezelter wrote: > > > I also published a series of columns on the use of logical names on > > OpenVMS.org. The first of these "The OpenVMS Consultant: Logical Names > > (Part 1)", is available athttp://www.openvms.org/stories.php?story=02/09/24/5441505 > > Bob, > > Nice material. I learned something. Thanks. > > Part 3 has no forwarding URL to Part 4. > > .. fred bach .. Fred, I am glad that the material was useful. The Technical Journal article was particularly fun to write and diagram. I have forwarded your report of a missing (hyper)link to Ken Farmer at OpenVMS.org. - Bob Gezelter, http://www.rlgsc.com ------------------------------ Date: Fri, 21 Dec 2007 02:42:44 -0800 (PST) From: Bob Gezelter Subject: Re: Volume label. Message-ID: <6ffc9bc5-d82b-4487-8112-8e00b2016b4b@q77g2000hsh.googlegroups.com> On Dec 20, 10:34 am, AEF wrote: > On Dec 20, 7:42 am, Bob Gezelter wrote: > > > > > On Dec 19, 2:12 pm, Chuck Aaron wrote: > > > > I have a ds10 with 2 disks. One is named alphasys which is the > > > system disk and one named userdisk (which equates to user$disk). > > > Is there a way to add an alias name to user$disk so that when > > > an application wants to access web$disk it will access the > > > user$disk? > > > > Thanks in advance. > > > Chuck, > > > As has been noted, logical names are the answer to this need. For > > safety's sake, my standard recommendation to clients with a situation > > such as this is to define a logical name WEB$DISK in terms of the > > already existent name for USER$DISK making use of DCL Lexical > > functions, to wit: > > > $ ASSIGN [various qualifiers as appropriate] 'F > > $TRNLNM("USER$DISK")' WEB$DISK > > That depends on his particular needs, no? Some might instead need > something like > > $ DEFINE /SYSTEM/EXEC WEB$DISK USER$DISK: > > so that reDEFINEing USER$DISK would automatically reDEFINE both > logical names. It depends on your what your site needs. > > Be sure to include the colon in the equivalence name in this command > (you should *NOT* use the colon inside the F$TRNLNM function as in Bob > Gezelter's example above)! Otherwise you may get problems with certain > commands (like SET DEFAULT). You should always include the trailing > colon in the equivalence name whenever it is a disk or another logical > name that ultimately translates to a generalized file specification. > [...] > > > - Bob Gezelter,http://www.rlgsc.com > > AEF AEF, Or not, as the case may be. Defining the equivalence name to point to a logical name means that the evaluation of WEB$DISK is deferred till it is actually used. If it is pointing to an area used by the WWW server, it can be confusing for it to follow another part of one's logical name context. - Bob Gezelter, http://www.rlgs.com ------------------------------ Date: Fri, 21 Dec 2007 05:36:16 -0800 (PST) From: AEF Subject: Re: Volume label. Message-ID: On Dec 21, 5:42 am, Bob Gezelter wrote: > On Dec 20, 10:34 am, AEF wrote: > > > > > On Dec 20, 7:42 am, Bob Gezelter wrote: > > > > On Dec 19, 2:12 pm, Chuck Aaron wrote: > > > > > I have a ds10 with 2 disks. One is named alphasys which is the > > > > system disk and one named userdisk (which equates to user$disk). > > > > Is there a way to add an alias name to user$disk so that when > > > > an application wants to access web$disk it will access the > > > > user$disk? > > > > > Thanks in advance. > > > > Chuck, > > > > As has been noted, logical names are the answer to this need. For > > > safety's sake, my standard recommendation to clients with a situation > > > such as this is to define a logical name WEB$DISK in terms of the > > > already existent name for USER$DISK making use of DCL Lexical > > > functions, to wit: > > > > $ ASSIGN [various qualifiers as appropriate] 'F > > > $TRNLNM("USER$DISK")' WEB$DISK > > > That depends on his particular needs, no? Some might instead need > > something like > > > $ DEFINE /SYSTEM/EXEC WEB$DISK USER$DISK: > > > so that reDEFINEing USER$DISK would automatically reDEFINE both > > logical names. It depends on your what your site needs. > > > Be sure to include the colon in the equivalence name in this command > > (you should *NOT* use the colon inside the F$TRNLNM function as in Bob > > Gezelter's example above)! Otherwise you may get problems with certain > > commands (like SET DEFAULT). You should always include the trailing > > colon in the equivalence name whenever it is a disk or another logical > > name that ultimately translates to a generalized file specification. > > [...] > > > > - Bob Gezelter,http://www.rlgsc.com > > > AEF > > AEF, > > Or not, as the case may be. > > Defining the equivalence name to point to a logical name means that > the evaluation of WEB$DISK is deferred till it is actually used. So why is that a bad thing? With SYS$MANAGER, SYS$SYSROOT isn't evaluated until it is used, e.g. > If it > is pointing to an area used by the WWW server, it can be confusing for > it to follow another part of one's logical name context. I don't understand this comment. If I do SHOW LOG WEB$DISK and I get USER$DISK, then I know that WEB$DISK is supposed to be USER$DISK. If I use your method, then all I see upon running SHOW LOG is the final translation and I then have no clue that it is supposed to be whatever USER$DISK is. For example, consider SYS$SYSROOT, SYS$SPECIFIC, and SYS $COMMON. By running SHOW LOGICAL on all of these, it is not clear whether SYS$SPECIFIC is the same as the first equivalence name of SYS $SYSROOT by design or coincidence. In fact, people have several times here asked why SYS$SYSROOT wasn't defined to be SYS$SPECIFIC, SYS $COMMON. (Can you answer that?) Just to emphasize, which method you want to use will depend on your needs. > > - Bob Gezelter,http://www.rlgs.com AEF ------------------------------ Date: Fri, 21 Dec 2007 05:40:25 -0800 (PST) From: IanMiller Subject: Re: Volume label. Message-ID: <3a0edecf-b96a-461f-80ec-69c9bb3598c8@e4g2000hsg.googlegroups.com> On Dec 21, 10:39 am, Bob Gezelter wrote: > On Dec 20, 4:04 pm, Fred Bach wrote: > > > Bob Gezelter wrote: > > > > I also published a series of columns on the use of logical names on > > > OpenVMS.org. The first of these "The OpenVMS Consultant: Logical Names > > > (Part 1)", is available athttp://www.openvms.org/stories.php?story=02/09/24/5441505 > > > Bob, > > > Nice material. I learned something. Thanks. > > > Part 3 has no forwarding URL to Part 4. > > > .. fred bach .. > > Fred, > > I am glad that the material was useful. The Technical Journal article > was particularly fun to write and diagram. > > I have forwarded your report of a missing (hyper)link to Ken Farmer at > OpenVMS.org. > > - Bob Gezelter,http://www.rlgsc.com The links between parts 1 to 5 have now been fixed ------------------------------ Date: Fri, 21 Dec 2007 05:57:05 -0800 (PST) From: AEF Subject: Re: Volume label. Message-ID: On Dec 21, 5:42 am, Bob Gezelter wrote: > On Dec 20, 10:34 am, AEF wrote: > > > > > On Dec 20, 7:42 am, Bob Gezelter wrote: > > > > On Dec 19, 2:12 pm, Chuck Aaron wrote: > > > > > I have a ds10 with 2 disks. One is named alphasys which is the > > > > system disk and one named userdisk (which equates to user$disk). > > > > Is there a way to add an alias name to user$disk so that when > > > > an application wants to access web$disk it will access the > > > > user$disk? > > > > > Thanks in advance. > > > > Chuck, > > > > As has been noted, logical names are the answer to this need. For > > > safety's sake, my standard recommendation to clients with a situation > > > such as this is to define a logical name WEB$DISK in terms of the > > > already existent name for USER$DISK making use of DCL Lexical > > > functions, to wit: > > > > $ ASSIGN [various qualifiers as appropriate] 'F > > > $TRNLNM("USER$DISK")' WEB$DISK > > > That depends on his particular needs, no? Some might instead need > > something like > > > $ DEFINE /SYSTEM/EXEC WEB$DISK USER$DISK: > > > so that reDEFINEing USER$DISK would automatically reDEFINE both > > logical names. It depends on your what your site needs. > > > Be sure to include the colon in the equivalence name in this command > > (you should *NOT* use the colon inside the F$TRNLNM function as in Bob > > Gezelter's example above)! Otherwise you may get problems with certain > > commands (like SET DEFAULT). You should always include the trailing > > colon in the equivalence name whenever it is a disk or another logical > > name that ultimately translates to a generalized file specification. > > [...] > > > > - Bob Gezelter,http://www.rlgsc.com > > > AEF > > AEF, > > Or not, as the case may be. > > Defining the equivalence name to point to a logical name means that > the evaluation of WEB$DISK is deferred till it is actually used. If it > is pointing to an area used by the WWW server, it can be confusing for > it to follow another part of one's logical name context. > > - Bob Gezelter,http://www.rlgs.com You mistyped your Web address! This one is for sale. ------------------------------ Date: Fri, 21 Dec 2007 10:22:43 -0500 From: JF Mezei Subject: Re: Volume label. Message-ID: <476bda4f$0$22049$c3e8da3@news.astraweb.com> AEF wrote: >> Defining the equivalence name to point to a logical name means that >> the evaluation of WEB$DISK is deferred till it is actually used. > > So why is that a bad thing? With SYS$MANAGER, SYS$SYSROOT isn't > evaluated until it is used, e.g. If you truly want to emulate the logical that is created when you mount the disk, then I would put in a identical logical name definition. consider the difference between: > $define/system/exec/trans=(conc,terminal) $mydisk 'f$trnlnm("$DISK2") > $define/system/exec/trabs=(conc) $mydisk $DISK2 in cases where you want to: > $define/system/trans=(conc,term) $myroot 'f$trnlnm("$mydisk")'[myroot.] The first case will work. The second case won't work because $myroot will point to $disk2 which won't translate to whatever disk drive it is pointing to. ------------------------------ Date: Fri, 21 Dec 2007 08:00:13 -0800 (PST) From: AEF Subject: Re: Volume label. Message-ID: On Dec 21, 10:22 am, JF Mezei wrote: > AEF wrote: > >> Defining the equivalence name to point to a logical name means that > >> the evaluation of WEB$DISK is deferred till it is actually used. > > > So why is that a bad thing? With SYS$MANAGER, SYS$SYSROOT isn't > > evaluated until it is used, e.g. > > If you truly want to emulate the logical that is created when you mount > the disk, then I would put in a identical logical name definition. What if you don't? > > consider the difference between: > > > $define/system/exec/trans=(conc,terminal) $mydisk 'f$trnlnm("$DISK2") > > $define/system/exec/trabs=(conc) $mydisk $DISK2 You forgot to include the trailing colon in the second command. > > in cases where you want to: > > > $define/system/trans=(conc,term) $myroot 'f$trnlnm("$mydisk")'[myroot.] > > The first case will work. The second case won't work because $myroot > will point to $disk2 which won't translate to whatever disk drive it is > pointing to. The second case won't work because you forgot to include the trailing colon and because you didn't omit "term" in the command that defines $MYROOT. *** Always include the trailing colon (after apostrophe substitution, if any) in equivalence names that are file-oriented devices or logical names which ultimately translate to any valid subset of a full file- spec. *** (Seems to me I heard this somewhere before -- possibly even in this thread!) If you look at SYS$SYSROOT, the SYS$COMMON portion of it drops the "term" part for this very reason. Notice also that SYS$COMMON is followed by a trailing colon. $ SHOW LOG SYS$SYSROOT/FUL "SYS$SYSROOT" [exec] = "EISNER$DRA5:[SYS0.]" [concealed,terminal] (LNM$SYSTEM _TABLE) = "SYS$COMMON:" ^^^^^^^^^^^^^----- no concealed or terminal for this one (terminal being the relevant item here) 1 "SYS$COMMON" [exec] = "EISNER$DRA5: [SYS0.SYSCOMMON.]" [concealed,terminal] (L NM$SYSTEM_TABLE) $ AEF ------------------------------ Date: Fri, 21 Dec 2007 12:17:46 -0500 From: JF Mezei Subject: Re: Volume label. Message-ID: <476bf54e$0$25334$c3e8da3@news.astraweb.com> AEF wrote: > The second case won't work because you forgot to include the trailing > colon and because you didn't omit "term" in the command that defines > $MYROOT. Which is my point. If you want to define a logical that behaves exactly like that of the logical created when you mount a drive, you need to have it translate directly because that logical might be used in situations where it is translated to form a new "translation=terminal" logical. ------------------------------ Date: Fri, 21 Dec 2007 09:38:12 -0800 (PST) From: AEF Subject: Re: Volume label. Message-ID: On Dec 21, 12:17 pm, JF Mezei wrote: > AEF wrote: > > The second case won't work because you forgot to include the trailing > > colon and because you didn't omit "term" in the command that defines > > $MYROOT. > > Which is my point. If you want to define a logical that behaves exactly > like that of the logical created when you mount a drive, you need to > have it translate directly because that logical might be used in > situations where it is translated to form a new "translation=terminal" > logical. First of all, when you DEFINE or logical name you should ***ALWAYS***, ***ALWAYS***, ***ALWAYS*** include a trailing colon (after apostrophe substitution) in the equivalence name if it is a file-oriented device or a logical name that iteratively translates to any valid subset (or all) of a full file-spec. Period. (For batch queues -- and perhaps some of the other types of logical names -- you SHOULDN'T include a trailing colon!) Second, what if you don't and what difference does it make anyway other than what I already pointed out? Nothing you posted contradicts what I originally said: It depends on what your site's needs are. I do agree that if you want A you need to do A. But if you want B, you have to do B, which is what I already said before. Interestingly, in the case of SYS$SYSROOT, both A *and* B are used! Like I said, it depends on your needs. AEF Reminder: When you DEFINE or logical name you should ***ALWAYS***, ***ALWAYS***, ***ALWAYS*** include a trailing colon (after apostrophe substitution) in the equivalence name if it is a file-oriented device or a logical name that iteratively translates to any valid subset (or all) of a full file-spec. Period. (For batch queues -- and perhaps some of the other types of logical names -- you SHOULDN'T include a trailing colon!) In case you missed it the first two times: Reminder: When you DEFINE or logical name you should ***ALWAYS***, ***ALWAYS***, ***ALWAYS*** include a trailing colon (after apostrophe substitution) in the equivalence name if it is a file-oriented device or a logical name that iteratively translates to any valid subset (or all) of a full file-spec. Period. (For batch queues -- and perhaps some of the other types of logical names -- you SHOULDN'T include a trailing colon!) ------------------------------ Date: Fri, 21 Dec 2007 09:58:14 -0800 (PST) From: AEF Subject: Re: Volume label. Message-ID: <8fdb786d-3a7d-410b-baa8-147273e266b1@e10g2000prf.googlegroups.com> On Dec 21, 12:38 pm, AEF wrote: > On Dec 21, 12:17 pm, JF Mezei wrote: > > > AEF wrote: > > > The second case won't work because you forgot to include the trailing > > > colon and because you didn't omit "term" in the command that defines > > > $MYROOT. > > > Which is my point. If you want to define a logical that behaves exactly > > like that of the logical created when you mount a drive, you need to > > have it translate directly because that logical might be used in > > situations where it is translated to form a new "translation=terminal" > > logical. > > First of all, when you DEFINE or logical name you should ***ALWAYS***, > ***ALWAYS***, ***ALWAYS*** include a trailing colon (after apostrophe > substitution) in the equivalence name if it is a file-oriented device > or a logical name that iteratively translates to any valid subset (or > all) of a full file-spec. Period. (For batch queues -- and perhaps > some of the other types of logical names -- you SHOULDN'T include a > trailing colon!) > > Second, what if you don't and what difference does it make anyway > other than what I already pointed out? > > Nothing you posted contradicts what I originally said: It depends on > what your site's needs are. > > I do agree that if you want A you need to do A. But if you want B, you > have to do B, which is what I already said before. Interestingly, in > the case of SYS$SYSROOT, both A *and* B are used! Like I said, it > depends on your needs. > > AEF > > Reminder: When you DEFINE or logical name you should ***ALWAYS***, > ***ALWAYS***, ***ALWAYS*** include a trailing colon (after apostrophe > substitution) in the equivalence name if it is a file-oriented device > or a logical name that iteratively translates to any valid subset (or > all) of a full file-spec. Period. (For batch queues -- and perhaps > some of the other types of logical names -- you SHOULDN'T include a > trailing colon!) > > In case you missed it the first two times: > > Reminder: When you DEFINE or logical name you should ***ALWAYS***, > ***ALWAYS***, ***ALWAYS*** include a trailing colon (after apostrophe > substitution) in the equivalence name if it is a file-oriented device > or a logical name that iteratively translates to any valid subset (or > all) of a full file-spec. Period. (For batch queues -- and perhaps > some of the other types of logical names -- you SHOULDN'T include a > trailing colon!) Oops! That should be "When you DEFINE *a* logical name...", of course. Sorry. From your example > > > $define/system/exec/trans=(conc,terminal) $mydisk 'f$trnlnm("$DISK2") > > > $define/system/exec/trabs=(conc) $mydisk $DISK2 > > > > in cases where you want to: > > > > > $define/system/trans=(conc,term) $myroot 'f$trnlnm("$mydisk")'[myroot.] So it's really the difference between (A) $mydisk = dka0:[myroot.] conc,term and (B) $mydisk = $disk2:[myroot.] conc The key difference is that if you later redefine $disk2 to something else, A will still point to dka0: but B will be redirected to [myroot.] on the new equivalence name of $disk2. In some cases you may want A and in others you may want B. That's all I was trying to say. AEF ------------------------------ Date: Fri, 21 Dec 2007 16:17:14 GMT From: Rob Brown Subject: Re: WANTED: PDSDUMP information from VAX 780 (c1987) Message-ID: On Mon, 3 Dec 2007, Silver Bells wrote: > But if translators are not available,I would need to know the method > of compression the PDSDUMP used. I could write my own (I did that > for years at CDC in the 80's.) I spent a few moments with google. Are and <...&P=20118> relevant? -- Rob Brown b r o w n a t g m c l d o t c o m G. Michaels Consulting Ltd. (780)438-9343 (voice) Edmonton (780)437-3367 (FAX) http://gmcl.com/ ------------------------------ End of INFO-VAX 2007.698 ************************