INFO-VAX Sat, 10 Feb 2007 Volume 2007 : Issue 81 Contents: Re: Alpha to Integrity porting workshops - who's been? Re: Alpha to Integrity porting workshops - who's been? Re: Alpha to Integrity porting workshops - who's been? Re: Alpha to Integrity porting workshops - who's been? Re: Alpha to Integrity porting workshops - who's been? Re: Alpha to Integrity porting workshops - who's been? ANN: I64 versions of FREE and VMSTAR Anyone successfully proxying via SWS 2.11? Re: DST changes and VMS 7.2 Re: Help configuring VMS mail with Dynaccess.com Re: Help configuring VMS mail with Dynaccess.com Help wanted with Pathworks 5.0a Re: Help wanted with Pathworks 5.0a Re: Mark Daniel and/or Hein van den Huevel or anyone really Re: Mark Daniel and/or Hein van den Huevel or anyone really Re: Migrating C application from VMS to LINUX Re: Migrating C application from VMS to LINUX Re: Migrating C application from VMS to LINUX Re: Migrating C application from VMS to LINUX Re: Migrating C application from VMS to LINUX Re: Program looping Re: SPANNING BACKUP TAPES Re: SPANNING BACKUP TAPES Re: SPANNING BACKUP TAPES Re: SPANNING BACKUP TAPES Re: SPANNING BACKUP TAPES Re: SPANNING BACKUP TAPES Re: SPANNING BACKUP TAPES Re: SPANNING BACKUP TAPES Re: Virtual tape libraries Re: Virtual tape libraries Re: Virtual tape libraries ---------------------------------------------------------------------- Date: 9 Feb 2007 16:30:17 -0600 From: Kilgallen@SpamCop.net (Larry Kilgallen) Subject: Re: Alpha to Integrity porting workshops - who's been? Message-ID: In article , John Reagan writes: > sticker on it. At the last one I did in Atlanta, we had a guy all the > way from Alaska. I'm sure his system got there faster than he did. Perhaps that is the way he wanted it :-) ------------------------------ Date: 9 Feb 2007 05:47:01 -0800 From: twnews@kittles.com Subject: Re: Alpha to Integrity porting workshops - who's been? Message-ID: <1171028821.935598.188770@a75g2000cwd.googlegroups.com> twnews@kittles.com wrote: > I am curious if anyone here has been to one of the Alpha to Integrity > porting workshops (the ones where you bring home an Integrity server > to develop on). We are not a software company, but we have developed > an elaborate and good (IMHO) software system for our business. It is > a commercial quality product, it is just a home grown (RDB based) > system. We are scheduling one of our developers to go to the next > scheduled workshop. Someone at HP that is involved in this seems to > think that we are not eligible because we are not currently marketing > our software (from time to time we toy with the idea of productizing > it and selling it). At the last HPTF I thought that I spoke with > shops like ours that have attended one of these porting workshops. > > If you have attended one, what kind of work do you do (consultant, > employee of software developer, developer for end user company,...)? > Does anyone know more about the rules and intent of these workshops > than I do? I had the contact for the HP guy who runs this, but of > course now that I need it I have lost it. I am thinking that maybe I > need to talk to Leo (can not think of last name, but I think I have > the Leo right) that presents on VMS at the HPTF each year. He clearly > indicated to me that this would be a good thing for my company. (At > times like these I really regret that I am bad with names.) > > This work shop is an important part of our Alpha to Integrity upgrade > strategy. It would be a big disappointment to not be allowed to go. > Any help is appreciated. > > Thomas Wirt > Thanks are all of the quick responses! It turned out to be a non issue. I think that it was just one guy that did not realize this workshop servers more than one purpose. We are getting all the info we need now from the right people at HP. My developer is very much looking forward to the class. Thanks again Thomas Wirt ------------------------------ Date: Fri, 09 Feb 2007 14:12:55 GMT From: VAXman- @SendSpamHere.ORG Subject: Re: Alpha to Integrity porting workshops - who's been? Message-ID: <00A62FB2.F2A5B816@SendSpamHere.ORG> In article <1170962523.784400.198980@a34g2000cwb.googlegroups.com>, twnews@kittles.com writes: > > >I am curious if anyone here has been to one of the Alpha to Integrity >porting workshops (the ones where you bring home an Integrity server >to develop on). We are not a software company, but we have developed >an elaborate and good (IMHO) software system for our business. It is >a commercial quality product, it is just a home grown (RDB based) >system. We are scheduling one of our developers to go to the next >scheduled workshop. Someone at HP that is involved in this seems to >think that we are not eligible because we are not currently marketing >our software (from time to time we toy with the idea of productizing >it and selling it). At the last HPTF I thought that I spoke with >shops like ours that have attended one of these porting workshops. > >If you have attended one, what kind of work do you do (consultant, >employee of software developer, developer for end user company,...)? >Does anyone know more about the rules and intent of these workshops >than I do? I had the contact for the HP guy who runs this, but of >course now that I need it I have lost it. I am thinking that maybe I >need to talk to Leo (can not think of last name, but I think I have >the Leo right) that presents on VMS at the HPTF each year. He clearly >indicated to me that this would be a good thing for my company. (At >times like these I really regret that I am bad with names.) > >This work shop is an important part of our Alpha to Integrity upgrade >strategy. It would be a big disappointment to not be allowed to go. >Any help is appreciated. > >Thomas Wirt > You get a junker system anyway. Safe the company money you spend on the session, hotel, airfare, etc. and buy one with hardware support! -- VAXman- A Bored Certified VMS Kernel Mode Hacker VAXman(at)TMESIS(dot)COM "Well my son, life is like a beanstalk, isn't it?" ------------------------------ Date: Fri, 09 Feb 2007 10:54:38 -0500 From: JF Mezei Subject: Re: Alpha to Integrity porting workshops - who's been? Message-ID: <4c63c$45cc9955$cef8887a$21099@TEKSAVVY.COM> Question: From a user mode application point of view, do those porting workshops teach attendees anything ? Or is it more of a question of attendees playing on their IA64 box and a supervisor answering questions as they proceed with their compiles and may find the new compiler has differences ? I can see some teaching required for writing drivers and other kernel level stuff. And I can see some system management stuff sicne there are differences (especially with regards to EFI and managing EFI partition etc). I take it that an attendee is expected to bring his/her applications to the workshop and do the porting there ? If so, on what sort of media should the apps be brought ? ------------------------------ Date: Fri, 09 Feb 2007 19:41:40 GMT From: John Reagan Subject: Re: Alpha to Integrity porting workshops - who's been? Message-ID: VAXman- @SendSpamHere.ORG wrote: > In article <4c63c$45cc9955$cef8887a$21099@TEKSAVVY.COM>, JF Mezei writes: > >> >>Question: >> >>From a user mode application point of view, do those porting workshops >>teach attendees anything ? > > > Some short tutorials on the console interface. More than that. Tutorial on the console, talks about compiler differences (if any), about conditionalization code mistakes, etc. About half the time (perhaps a little more than half) is lab time where you get to play on the box in front of you. People like me walk around and answer questions, etc. > > > >>Or is it more of a question of attendees playing on their IA64 box and a >>supervisor answering questions as they proceed with their compiles and may >>find the new compiler has differences ? > > > They don't even play on their IA64 box. I played on a 1.5GHz box with 2GB; > HP sent me a 1.4GHz box with 512MB. It has changed since Brian has attended. Now you actually take the actual system home with you. You can physically take it or we pack it in the box (with you standing there) and we slap a 2nd day air DHL sticker on it. At the last one I did in Atlanta, we had a guy all the way from Alaska. I'm sure his system got there faster than he did. > > > >>I can see some teaching required for writing drivers and other kernel level >>stuff. And I can see some system management stuff sicne there are >>differences (especially with regards to EFI and managing EFI partition etc). >> >>I take it that an attendee is expected to bring his/her applications to the >>workshop and do the porting there ? If so, on what sort of media should the >>apps be brought ? Yes bring your code. CD/DVD/USB flash drive/etc. The HP folks often can connect their laptops to the internet. I've helped several folks FTP their ZIP files to my laptop, disconnect from the internet, connect a cable to my PC, and FTP it again to the I64 box. Works pretty well. > > > CD or DVD would be the ideal media. There were no tape drives and the boxes > were not connected to the internet. > > Save the money, buy a box with HW support. > Yes, the box you get has no support with it. We ask that you make sure you test it before we box it up. At Atlanta, one of the systems had a non-functional CD drive. We swapped it with a working one before we boxed it up. Also, the PAKs on the system have a short life span. If you are in DSPP, you can easily get more. But if you aren't in DSPP, you'll end up buying new PAKs at some point. But then again, you'd have to buy real PAKs if you bought a new box with HW support. -- John Reagan HP Pascal/{A|I}MACRO/COBOL for OpenVMS Project Leader Hewlett-Packard Company ------------------------------ Date: Fri, 09 Feb 2007 21:43:42 GMT From: VAXman- @SendSpamHere.ORG Subject: Re: Alpha to Integrity porting workshops - who's been? Message-ID: <00A62FF1.EC1094AF@SendSpamHere.ORG> In article , John Reagan writes: >> >> Save the money, buy a box with HW support. >> > >Yes, the box you get has no support with it. We ask that you make sure >you test it before we box it up. At Atlanta, one of the systems had a >non-functional CD drive. We swapped it with a working one before we >boxed it up. Also, the PAKs on the system have a short life span. If >you are in DSPP, you can easily get more. But if you aren't in DSPP, >you'll end up buying new PAKs at some point. But then again, you'd have >to buy real PAKs if you bought a new box with HW support. You could wind up with an HP boat anchor with leds. My next $3-4K is on this: http://www.apple.com/macpro/ 3 yr warranty is only $249. -- VAXman- A Bored Certified VMS Kernel Mode Hacker VAXman(at)TMESIS(dot)COM "Well my son, life is like a beanstalk, isn't it?" ------------------------------ Date: Fri, 09 Feb 2007 19:49:34 -0600 From: Hunter Goatley Subject: ANN: I64 versions of FREE and VMSTAR Message-ID: <45CD24AE.3090700@goatley.com> The following freeware has been updated in my archive to include binaries for OpenVMS I64: o FREE - Displays disk space usage in a brief display o VMSTAR - A tar implementation for VMS http://www.process.com/openvms/ ftp://ftp.process.com/vms-freeware/fileserv/vmstar.zip http://vms.process.com/ftp/vms-freeware/fileserv/vmstar.zip ftp://ftp.tmk.com/vms-freeware/fileserv/vmstar.zip http://www.tmk.com/ftp/vms-freeware/fileserv/vmstar.zip ftp://ftp.process.com/vms-freeware/fileserv/free.zip http://vms.process.com/ftp/vms-freeware/fileserv/free.zip ftp://ftp.tmk.com/vms-freeware/fileserv/free.zip http://www.tmk.com/ftp/vms-freeware/fileserv/free.zip Hunter ------ Hunter Goatley, Process Software, http://www.process.com/ PreciseMail Anti-Spam Gateway for OpenVMS, Tru64, Solaris, & Linux goathunter@goatley.com http://www.goatley.com/hunter/ ------------------------------ Date: Fri, 09 Feb 2007 22:58:19 GMT From: winston@SSRL.SLAC.STANFORD.EDU (Alan Winston - SSRL Central Computing) Subject: Anyone successfully proxying via SWS 2.11? Message-ID: <00A62FE3.330D566B@SSRL.SLAC.STANFORD.EDU> VMSers -- VMS 8.3 with update 1 Apache 2.11 (with update 1) All patches Hardware: DS20E (Alpha), RX2600 (Itanium) We were using proxy successfully under CSWS 1.3. Now with SWS 2.11, we have a problem. We have: LoadModule proxy_module modules/mod_proxy.exe LoadModule proxy_connect_module modules/mod_proxy_connect.exe LoadModule proxy_ftp_module modules/mod_proxy_ftp.exe LoadModule proxy_http_module modules/mod_proxy_http.exe ProxyRequests On ProxyVia On Order Deny,Allow Deny from all Allow from env=ssrlip (SSRLIP is defined elsewhere, and it's numeric IP addresses within our address range. This appears to work on allowing connections to the right things.) If you make either machine the proxy server and go to, eg, Google and click around, bad things happen: ----------------------------------------------------------------------- Proxy Error The proxy server received an invalid response from an upstream server. The proxy server could not handle the request GET http://groups.google.comproxy//groups.google.com/group/google-maps-api. Reason: DNS lookup failure for: groups.google.comproxy ----------------------------------------------------------------------- And indeed, that string for the GET is what appears in the location bar. It seems that the proxy module is appending "/proxy//locationstring" directly to the end of the http string. The logo shows as a broken picture; click on it and do "View Image" and you get http://www.google.comproxy//www.google.com/intl/en_all/images/logo.gif (A direct lookup to http://www.google.com/intl/en_all/images/logo.gif shows the logo, as expected.) This is a crucial problem for us because we're using our proxy server to get updates for things in our Internet Free Zone, like the SAN management station. We don't want to put them on the open Internet, and we don't want to have to bring in Windows or Linux boxes as part of our enterprise infrastructure just to run proxies. It worked on 1.3. Is there something I'm doing wrong? Does anybody have this working? Thanks, -- Alan ------------------------------ Date: Fri, 9 Feb 2007 18:13:20 -0500 From: "Neil Rieck" Subject: Re: DST changes and VMS 7.2 Message-ID: <45ccfe17$0$7466$9a6e19ea@news.newshosting.com> "Bob Koehler" wrote in message news:P6xSD31RtCEC@eisner.encompasserve.org... > In article <45cbac26$0$7464$9a6e19ea@news.newshosting.com>, "Neil Rieck" > writes: >> >> If you have any programs written in C/C++ or Java that make "date-time" >> calls, you need to have correct timezone support in the OS. This includes >> open source software like Apache, Samba (CIFS), GNV, etc. > > For these systems, the VMS C RTL will be fine with calls like > locatime(), since this returns the VMS clock, but not with > calls like gmtime() which convert to GMT. (The opposite of the > convention used by UNIX). Of course, this assumes you do something > to keep the VMS time correct (like a patched Multinet runing NTP). > [...snip...] > Like most things related to "C", the answer is almost never straight forward. The result returned by "localtime()" will also return fields TIMEZONE and OFFSET-FROM-GMT unless compiled with switch "/standard=ANSI89" or defining "_DECC_V4_SOURCE". Does enabling the compiler switch only disable theses two fields or are the others tweaked as well? A careful programmer will double check each implementation of the language. Neil Rieck Kitchener/Waterloo/Cambridge, Ontario, Canada. http://www3.sympatico.ca/n.rieck/ Neil Rieck Kitchener/Waterloo/Cambridge, Ontario, Canada. http://www3.sympatico.ca/n.rieck/links/cool_openvms.html ------------------------------ Date: Thu, 08 Feb 2007 09:36:13 -0500 From: "Richard B. gilbert" Subject: Re: Help configuring VMS mail with Dynaccess.com Message-ID: <45CB355D.2070009@comcast.net> Phillip Helbig---remove CLOTHES to reply wrote: > In article <1170913360.400794.261080@j27g2000cwj.googlegroups.com>, > mcbill20@yahoo.com writes: > > >>Well, now I have everything completely hosed. I can no longer receive >>SMTP mail either. When the SMTP connection is made, I get the >>following error: >> >>%%%%%%%%%%% OPCOM 7-FEB-2007 22:38:58.72 %%%%%%%%%%% >>Message from user TCPIP$SMTP on SHAGGY >>TCPIP-W-SMTP_UNRSLVMF, MAIL FROM:<> has >>unresolvable domain >> >>I think I am definitely missing something as far as configuration >>goes. When I do a "SHOW NAME" in TCPIP now I get the following: > > > This appears to be a nameserver problem. > > All NAT routers have their own nameserver for their own use which they > get from the provider via DHCP. Some of them will act as a nameserver > for stuff on the LAN (forwarding requests to their own name server). > Thus, use your NAT router as your name server in VMS. If that doesn't > work (and it depends on the NAT router), look into your NAT router and > see what nameserver(s) it is using and use that information to configure > your nameserver(s) on the VMS box. > This appears to be a rather strange NAT router to say the least. A router does not need DNS at all in order to function as a router! The typical NAT router configures itself via DHCP and, as part of the process, is given the addresses of two or three DNS servers. It also acts as a DHCP server and supplies these addresses to its clients. It makes no use of the DNS server addresses other than to pass them along to its clients. ------------------------------ Date: Fri, 09 Feb 2007 20:32:37 +0100 From: Paul Sture Subject: Re: Help configuring VMS mail with Dynaccess.com Message-ID: In article <45CC93E7.7060007@comcast.net>, "Richard B. gilbert" wrote: > Phillip Helbig---remove CLOTHES to reply wrote: > > In article <45CB355D.2070009@comcast.net>, "Richard B. gilbert" > > writes: > > > > > >>The typical NAT router configures itself via DHCP and, as part of the > >>process, is given the addresses of two or three DNS servers. It also > >>acts as a DHCP server and supplies these addresses to its clients. It > >>makes no use of the DNS server addresses other than to pass them along > >>to its clients. > > > > > > I have several NAT routers. All of them are DHCP clients with respect > > to the WAN, and can be DHCP servers with respect to the LAN (I don't use > > DHCP on my VMS machines, though). ALL of them pick up DNS servers when > > they are configured via DHCP. However, only SOME of them will in turn > > act as a nameserver for machines on the LAN. Thus, the ones that don't > > must be getting the DNS servers for some other reason. > > > > I would be somewhat surprised if ANY router did DNS lookups! All they > are supposed to do is supply the DNS server addresses when requested via > DHCP. When something on your LAN makes a DNS request and the server > address is your ISP's name server, I would expect the router to forward > the request and return the server's reply. > I believe it depends on the both router and what your ISP offers. With my first (dialup) router, you could enter the DNS addresses into the router itself. The ADSL router, which came later acted as a DNS relay, > There may be exceptions to this but I have never heard of one. Your > typical < $100 SOHO router simply stuffs outbound traffic into your WAN > whether it's ADSL, Cable, or tin cans and string. It also, of course, > admits inbound traffic from your WAN. Neither of those routers was sub $300. OK, prices have come down a lot since then, but that's probably the difference. -- Paul Sture ------------------------------ Date: Fri, 09 Feb 2007 22:40:17 +0100 From: Wilm Boerhout Subject: Help wanted with Pathworks 5.0a Message-ID: <45ccea41$0$10810$ba620dc5@nova.planet.nl> (This is about VMS 6.2 , UCX 4.0 and Pathworks 5.0a. Formal support for these products is non-existant) I helped a client today with moving his systems to a new location. Part of the move was that the servers (VAX, Windows, Unix) and the routers/switches became part of a new IP subnet (nnn.mmm.229.* --> nnn.mmm.76.*) So after the physical move I reconfigured UCX, providing it with the new IP-address, gateway, DNS, etc. numbers. After rebooting, access to/from the system was possible using telnet and ftp. Access from outside the system to the Pathworks file services was not possible: the Windows client could not find the network path \\server\service. Needless to say that this worked before the move. When starting up, the message "File Services use DECnet and TCP" appears on the console as before. No errors were found in the log files of any Pathworks modules. The file services show up as registered (PCSA SHOW FILE SERV /REG), but the only service that shows up with PCSA SHOW FILE SERV /ACT is the IPC service. Questions: 1. Do I need to tell Pathworks somehow about the changed IP address? Access from the client is via TCP/IP as before. Clients do not run DECnet anymore (they did not do so before the move either) 2. What is the magic incantation to make REGISTERED file services become ACTIVE? Thanks, Wilm Boerhout ------------------------------ Date: 9 Feb 2007 16:41:52 -0800 From: "Bobby" Subject: Re: Help wanted with Pathworks 5.0a Message-ID: <1171068112.157780.121890@p10g2000cwp.googlegroups.com> A few pieces of data that may help clarify that situation a bit more: Are the Windows clients on the same subnet as the VMS server? What version of Windows (95/98...)? Can the Windows client(s) connect to a Windows or Unix hosted LANMAN server sitting on the same subnet as the Pathworks server? Bobby ------------------------------ Date: 9 Feb 2007 20:42:41 +0100 From: peter@langstoeger.at (Peter 'EPLAN' LANGSTOeGER) Subject: Re: Mark Daniel and/or Hein van den Huevel or anyone really Message-ID: <45ccdcc1$1@news.langstoeger.at> In article <45CC9498.3CC0B3B7@spam.comcast.net>, David J Dachtera writes: >Richard Maher wrote: >> [snip] >> PPS. Why would one use HGFTP as opposed to whatever comes with your TCP/IP? > >Because UCX's FTP server does not work with IE. UCX's FTP server refuses to >provide NLST data in UN*X format, and IE chokes on anything but UN*X format. Wrong. TCPIP/UCX FTP server NLST _does_ provide data in U**X format, but M$ IE insists on requesting the long form LIST (which is according to the RFC/STD platform specific) and chokes then because it expects U**X format. >Stalemate. UCX Engineering refuses to fix the FTP server. Checkmate. No, M$ refuses to fix the IE. Firefox/Opera do it right... >So, either use HGFTP or migrate to Multinet or TCPWare. But only if you're strictly bound to M$IE as FTP client... -- 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: Fri, 09 Feb 2007 11:55:21 -0500 From: JF Mezei Subject: Re: Mark Daniel and/or Hein van den Huevel or anyone really Message-ID: David J Dachtera wrote: > Because UCX's FTP server does not work with IE. UCX's FTP server refuses to > provide NLST data in UN*X format, and IE chokes on anything but UN*X format. TCPIP Services privides NLIST in standard format. The problem is that IE does not use NLIST, it uses LIST and tries to parse the output of LIST despite the RFC stating it shouldn't be done. The end result may still be problems with interoperability between Microsoft and TCPIP services, but the blame should go to Microsoft. ------------------------------ Date: 9 Feb 2007 22:17:58 GMT From: bill@cs.uofs.edu (Bill Gunshannon) Subject: Re: Migrating C application from VMS to LINUX Message-ID: <534a8mF1r4fn4U1@mid.individual.net> In article , Kilgallen@SpamCop.net (Larry Kilgallen) writes: > In article <1171017147.288767.101230@a34g2000cwb.googlegroups.com>, "klaus_austria" writes: > >> I plan to migrate from VMS to Linux. I have lots of c applications >> with interprocess communication like global logs and sections. >> Do exist any identical tools, to migrate to Linux in an easy way? >> I have some SNA connections using the sna protocoll. Is there a chance >> in Unix too? >> What else could cause problems? > > To reach people who have done this you might be better off in a Linux > newsgroup. Why would you think that? I doubt anyone in a Linux newsgroup has even heard of VMS much less had any experience with it. Oh wait, that was just a brush-off to another VMS deserter. My mistake. 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: 9 Feb 2007 22:23:40 GMT From: bill@cs.uofs.edu (Bill Gunshannon) Subject: Re: Migrating C application from VMS to LINUX Message-ID: <534ajcF1r4fn4U2@mid.individual.net> In article <45CC978D.1060709@comcast.net>, "Richard B. gilbert" writes: > klaus_austria wrote: >> On Feb 9, 1:07 pm, Kilgal...@SpamCop.net (Larry Kilgallen) wrote: >> >>>In article <1171017147.288767.101...@a34g2000cwb.googlegroups.com>, "klaus_austria" writes: >>> >>> >>>>I plan to migrate from VMS to Linux. I have lots of c applications >>>>with interprocess communication like global logs and sections. >>>>Do exist any identical tools, to migrate to Linux in an easy way? >>>>I have some SNA connections using the sna protocoll. Is there a chance >>>>in Unix too? >>>>What else could cause problems? >>> >>>To reach people who have done this you might be better off in a Linux >>>newsgroup. >> >> >> thanks, i try in linux group >> > > Trivial programs are trivial to port. Real programs can present real > problems. VMS files typically have records. Unix/Linux files are > ordered streams of bytes and any internal structure is a figment of the > programmer's imagination! Do you use indexed files for anything? Have > fun porting that to Unix/Linux! There has been indexed file support for Unix (via external libraries) since I first started working with it. I wtill have copies of ISAM for the 3B1, 3B2, 3B5, etc. And I remember seeing the equivalent for SunOS/Solaris. I think there may even be a free (read real OpenSource) one as well. > > Do you use any VMS libraries not defined in the ANSI C Standard? Lots > of luck porting those! But then, they wouldn't really be C programs, would they? Another reason to avoid proprietary extensions. > > Are you sure of a return on your investment? If you are just porting to > be "politically correct" you may wind up with some big negative numbers > on the bottom line! I would guess he's porting because he (or his management) has seen the writting on the wall and is not going to wait till disaster strikes before looking at solutions. > > I'd take much the same position if you were porting in the other direction! Somehow, kowing the mindset here, I find that hard to believe. 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: Fri, 09 Feb 2007 13:59:38 -0500 From: =?ISO-8859-1?Q?Arne_Vajh=F8j?= Subject: Re: Migrating C application from VMS to LINUX Message-ID: <45ccc490$0$49200$14726298@news.sunsite.dk> klaus_austria wrote: > I plan to migrate from VMS to Linux. I have lots of c applications > with interprocess communication like global logs and sections. > Do exist any identical tools, to migrate to Linux in an easy way? Global memory exist in Linux too. You can use either shm* or mmap functions depending on your SysV or BSD preferences. Global locking is also possible using sem* functions. If you need cluster wide global locking then you will need something on top of out-of-the-box Linux. Arne ------------------------------ Date: Fri, 09 Feb 2007 12:46:37 -0700 From: Maverick Subject: Re: Migrating C application from VMS to LINUX Message-ID: klaus_austria wrote: > Hi! > > I plan to migrate from VMS to Linux. I have lots of c applications > with interprocess communication like global logs and sections. > Do exist any identical tools, to migrate to Linux in an easy way? > I have some SNA connections using the sna protocoll. Is there a chance > in Unix too? > What else could cause problems? Goto Amazon.com and there is a book that covers VMS/Linux. ------------------------------ Date: Fri, 09 Feb 2007 20:40:41 -0500 From: "Richard B. gilbert" Subject: Re: Migrating C application from VMS to LINUX Message-ID: <45CD2299.9030200@comcast.net> Bill Gunshannon wrote: > In article <45CC978D.1060709@comcast.net>, > "Richard B. gilbert" writes: > >>klaus_austria wrote: >> >>>On Feb 9, 1:07 pm, Kilgal...@SpamCop.net (Larry Kilgallen) wrote: >>> >>> >>>>In article <1171017147.288767.101...@a34g2000cwb.googlegroups.com>, "klaus_austria" writes: >>>> >>>> >>>> >>>>>I plan to migrate from VMS to Linux. I have lots of c applications >>>>>with interprocess communication like global logs and sections. >>>>>Do exist any identical tools, to migrate to Linux in an easy way? >>>>>I have some SNA connections using the sna protocoll. Is there a chance >>>>>in Unix too? >>>>>What else could cause problems? >>>> >>>>To reach people who have done this you might be better off in a Linux >>>>newsgroup. >>> >>> >>>thanks, i try in linux group >>> >> >>Trivial programs are trivial to port. Real programs can present real >>problems. VMS files typically have records. Unix/Linux files are >>ordered streams of bytes and any internal structure is a figment of the >>programmer's imagination! Do you use indexed files for anything? Have >>fun porting that to Unix/Linux! > > > There has been indexed file support for Unix (via external libraries) > since I first started working with it. I wtill have copies of ISAM > for the 3B1, 3B2, 3B5, etc. And I remember seeing the equivalent for > SunOS/Solaris. I think there may even be a free (read real OpenSource) > one as well. > > >>Do you use any VMS libraries not defined in the ANSI C Standard? Lots >>of luck porting those! > > > But then, they wouldn't really be C programs, would they? Another reason > to avoid proprietary extensions. > > >>Are you sure of a return on your investment? If you are just porting to >>be "politically correct" you may wind up with some big negative numbers >>on the bottom line! > > > I would guess he's porting because he (or his management) has seen the > writting on the wall and is not going to wait till disaster strikes > before looking at solutions. > > >>I'd take much the same position if you were porting in the other direction! > > > Somehow, kowing the mindset here, I find that hard to believe. > > bill > I have done my share of porting things to VMS. It can be a bitch. If you want to see how bad it can get, download the current version of the NTP reference implementation from http://ntp.org/ VMS Engineering did a port of a ten year old version. I doubt that it was easy. I tried to build the current version from source on VMS V7.2 without success. It wants Unixy things that VMS does not have. A piece of advice you may have heard once upon a time goes like this: Pick the application software you want to use and then buy the platform (hardware and software) that it runs on. If you are very lucky you may get a choice of platforms. In most cases you buy a PC and Windoze. ------------------------------ Date: Thu, 08 Feb 2007 14:20:22 +0100 From: Jur van der Burg <"vdburg at hotmail dot com"> Subject: Re: Program looping Message-ID: <45cb23a2$0$336$e4fe514c@news.xs4all.nl> There are two things you can do for a start: In V7.3-2 there's the PCS sda extension to collect pc samples. $ ana/sys SDA> PCS LOAD SDA> PCS START TRACE/PID=xxxxxx/TICK=0 SDA> PCS SHOW TRACE Later VMS versions can do more with PCS. Another easy thing to do is $ SET PROCESS/SUSPEND=KERNEL, and then use SDA to walk the stack to see what's going on. Jur. George Cornelius wrote: > In article <1170849523.384149.255050@m58g2000cwm.googlegroups.com>, "mb301@hotmail.com" writes: >> Have a cobol program that's looping does anyone have any ideas or >> pointer to find out where in the coding its looping? How about setting >> the process to DUMP? >> >> $ SET PROC/DUMP >> >> I could crash the system (with 1000 users I'll be very popular) >> generate a crash dump but what could I look at? > > Others have suggested ways based on watching the current PC, running with > debug on, or invasive techniques such as queueing a kernel AST to the process. > > But SDA (ANALYZE/SYSTEM) can look inside processes and is able to dump > the contents of any given virtual memory range, including the stacks. > > I use an initial SHOW PROCESS command, then SHOW CALL followed by repeated > SHOW CALL/NEXT commands, to trace back through the call stack. I don't know > how well this works on a looping process, since it may be that the process > context that SDA is examining changes between successive SHOW CALL > invocations, and I don't know what SDA would do if you were to suspend the > process beforehand. > > If you have a map file and compile listings you can locate individual > variables if you like. Just be aware that if optimization is turned > on the variable may at any point in time be out of date (maintained in > a register; value implied by other values; etc.), if it exists at all. > > NOTE: Look into SDA's SET OUTPUT command if you are going to generate > extensive output. > > -- > George Cornelius cornelius()eisner.decus.org > cornelius()mayo.edu ------------------------------ Date: 9 Feb 2007 10:55:59 -0800 From: "AEF" Subject: Re: SPANNING BACKUP TAPES Message-ID: <1171047359.700150.163840@h3g2000cwc.googlegroups.com> On Feb 9, 4:38 am, "R.A.Omond" wrote: > AEF wrote: > > On Feb 8, 12:31 pm, koeh...@eisner.nospam.encompasserve.org (Bob Koehler) wrote: > > >>In article <1170946079.566598.16...@v33g2000cwv.googlegroups.com>, "AEF" writes: > > >>>Have you tried BACKUP journal files? These take up less space than > >>>regular BACKUP listings. With BACKUP journal files you can run this: > > >>> $ BACKUP/LIST/JOURNAL=jounral-file/SELECT=missing-file > > >> I'm just curious. Would you findp utting the listing or journal on the > >> end of the last tape to be equivalent to the PKZIP aproach? You can > >> mount and access the last tape of a multi-volume set when you know > >> it's got what you want. > > > Yes, except that the pkzip method is more automated. (Note, I was just > > pointing out the probably motivation for the pkzip method.) > > > But in my situation I may not know which save sets the file is on, so > > I need the journal files on disk. Otherwise I'd have to mount the last > > tape of numerous save sets and my tapes are offsite! But this is just > > for my case. YMMV, as they say. > > Hmmm... just putting on my Devil's Advocate uniform (dons cape and hat). > > What happens when the *journal file* needs to span tapes (or the > PKZIP file needs to span floppies). You know it's going to > happen sometime ;-)- Hide quoted text - > > - Show quoted text - I keep mine on disk. Of course, disks fill up, too! Hmmm... What happens with pkzip when *its* directory won't fit on what free space is left on the last floppy? Maybe during the save operation is asks for another floppy and just puts the whole directory there even though it just put part of it on the current floppy that filled up. AEF ------------------------------ Date: 8 Feb 2007 08:21:08 -0600 From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) Subject: Re: SPANNING BACKUP TAPES Message-ID: In article <1170897182.050079.20850@q2g2000cwa.googlegroups.com>, "AEF" writes: > > I was talking about the opposite case in which there are multiple > tapes in a single save set. For that you need listings or a journal > file, especially if you don't know which save set your file is in! > If you do your backups via batch, like I do, the batch log will show when the tape got changed (the OPCOM request and response). This can be used as a clue to which tape files are on, so hopefully you will only have to generate partial listings to find the correct tape. Listing all the tapes isn't really faster than telling backup to search for an restore the file (/log will tell you whan that's done). Or, you can chew up massive amounts of disk space with listings generated during the backup. I've found that not to be cost effective. To get PKZIP-like, just generate the listing while the backup is runnning, then migrate the listing file to the end of the tape after the BACKUP is done. In my experience either: a) I've lost a disk and I need to run all the image backup and subsequent incrementals b) a user just deleted a file, knows it was old (therefor on the last image backup), or was modified in the last couple of days (therefor on that day's incremental); more likely the latter since I do weekly images (1 image, 6 incrementals maks a the odds 6:1). But what's really usefull is to have tapes that are much bigger than the used space so that the file must be on one of one or two tapes. ------------------------------ Date: 8 Feb 2007 08:24:00 -0600 From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) Subject: Re: SPANNING BACKUP TAPES Message-ID: In article <2ab6c$45ca7986$cef8887a$28777@TEKSAVVY.COM>, JF Mezei writes: > Bob Koehler wrote: >> If that was the purpose, you'ld be using shadowing. Or at least >> you'ld turn off CRCs and groups. > > > Shadowing is good for actual disk failure or multi node access to data even > if one node goes down. But it does not prevent a user from recovering from > a DELETE disk:[000000...]*.*;* You need to have a snapshot taken of your > disk from time to time. I'd count that the same as a disk failure without shadowing. It's going to take time. It's going to have CRCs and groups. But in this case the user's provileges are going to be reviewed. Generally nobody but I can make that mistake on my systems. ------------------------------ Date: 8 Feb 2007 08:26:13 -0600 From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) Subject: Re: SPANNING BACKUP TAPES Message-ID: In article , "R.A.Omond" writes: > > You don't need TAPESYS to guard against the inadvertant overwriting of > backup tapes. Sometimes I get the feeling that I'm the only one on > the planet who makes use of the very useful feature of Backup's > /Tape_Expiration qualifier. INITIALIZE /PROTECTION has saved more than one of my backups from other users of the tape drive. Now, just where is the UNIX equivalent? ------------------------------ Date: 8 Feb 2007 06:47:59 -0800 From: "AEF" Subject: Re: SPANNING BACKUP TAPES Message-ID: <1170946079.566598.16310@v33g2000cwv.googlegroups.com> On Feb 8, 9:21 am, koeh...@eisner.nospam.encompasserve.org (Bob Koehler) wrote: > In article <1170897182.050079.20...@q2g2000cwa.googlegroups.com>, "AEF" writes: > > > > > I was talking about the opposite case in which there are multiple > > tapes in a single save set. For that you need listings or a journal > > file, especially if you don't know which save set your file is in! > > If you do your backups via batch, like I do, the batch log will > show when the tape got changed (the OPCOM request and response). > This can be used as a clue to which tape files are on, so hopefully > you will only have to generate partial listings to find the correct > tape. > > Listing all the tapes isn't really faster than telling backup to > search for an restore the file (/log will tell you whan that's > done). > > Or, you can chew up massive amounts of disk space with listings > generated during the backup. I've found that not to be cost > effective. > > To get PKZIP-like, just generate the listing while the backup is > runnning, then migrate the listing file to the end of the tape after > the BACKUP is done. > > In my experience either: > a) I've lost a disk and I need to run all the image backup and > subsequent incrementals > b) a user just deleted a file, knows it was old (therefor on the > last image backup), or was modified in the last couple of days > (therefor on that day's incremental); more likely the latter > since I do weekly images (1 image, 6 incrementals maks a the > odds 6:1). > > But what's really usefull is to have tapes that are much bigger than > the used space so that the file must be on one of one or two tapes. Hi Bob, Here's my original statement: "Well, there is an advantage to the pkzip method. If you want to restore one file from your multi-floppy zip file, insertion of the last floppy to read the "directory" allows pkzip to tell you which floppy to put in next to recover that one file. (IIRC). With VMS BACKUP you can only do this is you already have a journal or listing file. And this is certainly better than the backup program that came ..." I never said to generate listings from the already-made tape. I only said what you would need to do with VMS BACKUP to get the advantage of the pkzip way. Just one sentence is the target of all your commentary. I see nothing wrong with my statement in light of your commentary. If you don't find it cost effective in your case, fine. I wasn't making a case for or against it. I was just saying what you would need to do with VMS BACKUP to find which volume (and save set!) a particular file is on. Have you tried BACKUP journal files? These take up less space than regular BACKUP listings. With BACKUP journal files you can run this: $ BACKUP/LIST/JOURNAL=jounral-file/SELECT=missing-file and it will tell you which tape and save set to use: $ BACK/LIST/JOUR=EQUITIES/SELE=FT010509E.RPT Listing of BACKUP journal Journal file SYS$SYSDEVICE:[FELDMAN]EQUITIES.BJL;1 on 8-FEB-2007 09:41:14.01 Save set EQUI-020517.BCK created on 17-MAY-2002 22:12:53.02 Volume number 1, volume label EQ01 [FT.ARC.EQUITIES]FT010509E.RPT;1 End of BACKUP journal Unfortunately, you can't use a wildcard in the journal-file-spec. AEF ------------------------------ Date: Fri, 09 Feb 2007 12:01:54 +0100 From: Paul Sture Subject: Re: SPANNING BACKUP TAPES Message-ID: In article , "R.A.Omond" wrote: > > You don't need TAPESYS to guard against the inadvertant overwriting of > backup tapes. Sometimes I get the feeling that I'm the only one on > the planet who makes use of the very useful feature of Backup's > /Tape_Expiration qualifier. > I forgot to add to my previous post that I've had good results with /EXACT_ORDER. For those not familiar with it: BACKUP /EXACT_ORDER Depending on the other qualifiers you specify on the command line, the /EXACT_ORDER qualifier allows you to perform the following actions: o Specify the exact order of tape volume labels that you want to use in a BACKUP operation. o Preserve the existing volume label on a tape. o Prevent previous volumes of a multivolume save operation from being overwritten. It does, in my experience, work as advertised wrt the last point. -- Paul Sture ------------------------------ Date: Fri, 09 Feb 2007 20:46:31 +0100 From: Paul Sture Subject: Re: SPANNING BACKUP TAPES Message-ID: In article , koehler@eisner.nospam.encompasserve.org (Bob Koehler) wrote: > In article <00A62F03.B0CCE9C4.15@tachysoft.com>, Wayne Sewell > writes: > > > > It's trivially easy to keep track of where files are with TAPESYS. It > > includes > > a history system that keeps track of every file in every saveset of every > > tape. > > The history is organized by file, so it will give you a list of the most > > recent > > n backups (where n is user defined) of a particular version of a particular > > file, returning label, saveset names and position on the tape (so that fast > > skip can be used). And you can simply select from a list and cause the > > restore > > of the file to be done automatically. > > Where is this history stored? Is there a rule of thumb for > estimating the space required? What happens if this storage device > fails? This is something which has always somewhat concerned me too, although I'm going back to the days before redundant hardware was available (and a bit later if you include affordable). > I'm quite familiar with how PKZIP does this, and somewhat familiar > with how Legato does this (mostly from dealing with failures). Were failures with Legato recoverable? -- Paul Sture ------------------------------ Date: Fri, 9 Feb 2007 18:52:17 -0500 From: Wayne Sewell Subject: Re: SPANNING BACKUP TAPES Message-ID: <00A63003.E2909A1D.9@tachysoft.com> >From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) >X-Newsgroups: comp.os.vms >Subject: Re: SPANNING BACKUP TAPES >Date: 9 Feb 2007 08:03:57 -0600 >In article <00A62F03.B0CCE9C4.15@tachysoft.com>, Wayne Sewell writes: >> >> It's trivially easy to keep track of where files are with TAPESYS. It includes >> a history system that keeps track of every file in every saveset of every tape. >> The history is organized by file, so it will give you a list of the most recent >> n backups (where n is user defined) of a particular version of a particular >> file, returning label, saveset names and position on the tape (so that fast >> skip can be used). And you can simply select from a list and cause the restore >> of the file to be done automatically. > > Where is this history stored? TAPESYS has a concept called a "history set". This is a user defined grouping of history from backups. At the time of the backup, the history set is specified. Multiple backups can feed into the same history set. A history set is a collection of indexed files containing the information about the savesets and files of all the backups that use that history set. All the files making up a history set are contained in a standard vms directory. By default the history tree is on the tapesys disk, but the system manager can place it anywhere, as long as it is accessible by all nodes doing backups and therefore creating history. Also, the directories used by history sets do not have to be in the same tree, but can be placed anywhere independently, again provided the directory is accessible to the backups using the history set. There is actually two kinds of history in tapesys. The history sets are organized by file and are used for individual file restores. The summaries are by saveset and are used for full disk restores. There are automatic restore mechanism for both modes. >Is there a rule of thumb for > estimating the space required? It is not easy. It depends entirely on the structure of the disks being backed up and the activity over time. As far as structure, some sites tend to have all the files on a disk in a very few top-level directories, for instance [user.*], where there are a shitload of users. Other sites have a huge number of top-level directories. These extremes have a great impact on the structure of the history set. The file portion of the history is organized by top-level directory; there is a pair of indexed files for each. Of course, this is more of an issue for speed, not for storage. Too many top-level directories in the target disk(s) means too many files in the history directory, and poor rms performance on the files opens. As far as activity, you have to remember that the disks being backed up are changing all the time. Newly created files of course mean new records in the history. Yet file deletions don't remove records, because that would defeat the purpose of history. All of the backups of those files have to expire first. So the history grows. On the other hand, tapesys keeps track of when a particular backup tape has been overwritten by a new backup. So that means all of the former backups on that tape have ceased to exist and that particular history is no longer valid. A periodic purge occurs and all overwritten backups are removed from history. When the entire list of backups for a particular file is gone, the entry for the file itself is removed. So after a purge, the history shrinks somewhat. It should be noted that the internal structure of the history files is designed to use the minimum amount of disk space. For instance, a hierarchical organization is used, reflecting the structure of the file system itself, so that the full path to a file (disk:[dir1.dir2.dir3.dir4]the.file;5) is not stored in the file record. Instead the full path is derived, as each record has a pointer to the containing record (file points to directory, directory points to containing directory, top-level directory points to disk). This eliminates a *huge* number of text strings, and allows the file records to be a fraction of the size. The following is the history for sys$sysdevice:[vms$common] on node MOE: Directory KEYKOP$DKA0:[TAPESYS_HISTORY.HIST_MOE] _VMS$COMMON.6R0;461 File ID: (42164,16,0) Size: 1272/1272 Owner: [TAPESYS] Created: 9-FEB-2007 03:04:56.01 Revised: 9-FEB-2007 03:04:56.01 (2) Expires: Backup: Effective: Recording: Accessed: Attributes: Modified: Linkcount: 1 File organization: Indexed, Prolog: 3, Using 2 keys Shelved state: Online Caching attribute: Writethrough File attributes: Allocation: 1272, Extend: 0, Maximum bucket size: 2, Global buffer count: 0, No version limit Record format: Fixed length 49 byte records Record attributes: None RMS attributes: None Journaling enabled: None File protection: System:RWED, Owner:RWED, Group:, World: Access Cntrl List: (IDENTIFIER=[TAPESYSNP],ACCESS=READ+WRITE+EXECUTE+DELETE) (IDENTIFIER=TAPESYS_USER,ACCESS=READ) Client attributes: None _VMS$COMMON.6R0_PTR;461 File ID: (42181,16,0) Size: 3280/3280 Owner: [TAPESYS] Created: 9-FEB-2007 03:05:02.22 Revised: 9-FEB-2007 03:05:02.22 (2) Expires: Backup: Effective: Recording: Accessed: Attributes: Modified: Linkcount: 1 File organization: Indexed, Prolog: 3, Using 1 key Shelved state: Online Caching attribute: Writethrough File attributes: Allocation: 3280, Extend: 0, Maximum bucket size: 2, Global buffer count: 0, No version limit Record format: Variable length, maximum 96 bytes, longest 0 bytes Record attributes: None RMS attributes: None Journaling enabled: None File protection: System:RWED, Owner:RWED, Group:, World: Access Cntrl List: (IDENTIFIER=[TAPESYSNP],ACCESS=READ+WRITE+EXECUTE+DELETE) (IDENTIFIER=TAPESYS_USER,ACCESS=READ) Client attributes: None Total of 2 files, 4552/4552 blocks. This is just the history for vms$common, but of course that is the bulk of a system disk. The two files together comprise the file history. The .6r0 file is the list of names used by the history (disk names, directory names, file names, etc.). The .6r0_ptr file is the actual history. It contains no names at all, merely points back to records in the other file. As you can see the records are pretty small. And note that the .6r0_ptr is variable. The record size is a max. The actual record can be far less than that. I have POINTERS set to the default of 14, so this means that these two files can keep up with 14 different backups of each of the files contained in the vms$common directory. Each file has a different counter and set of pointers to backups. >What happens if this storage device > fails? > It is possible to have backups go to redundant history sets in parallel, which can be on different disks. This is accomplish by specifying two history set names in the backup definition file. At twice the disk space of course. History sets are standard vms indexed files and can themselves be backed up and recorded inside history sets, including themselves. So it would be possible to use vms backup manually to restore the history sets to a different disk. Wayne =============================================================================== Wayne Sewell, Tachyon Software Consulting (281)812-0738 wayne@tachysoft.com http://www.tachysoft.com/www/tachyon.html and wayne.html =============================================================================== Curly:"Hey, I'm no mule." Moe:"No, your ears are too short!" ------------------------------ Date: 8 Feb 2007 08:12:19 -0600 From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) Subject: Re: Virtual tape libraries Message-ID: In article , Marc Van Dyck writes: > Has anyone out there already tried to backup one (or more) OpenVMS > systems with any kind of virtual tape library system ? Vax , Alpha, > Itanium ? Which system ? Which software on the server side ? And has > it been successful ? I'm interested in any experience that can be > shared in this area. I've backed up systems with Multinet's network tape support, but nothing really "virtual" (it went to a real tape drive). I've also accessed tape drive simulation via disk files in SIMH. I've not had any problems with either. ------------------------------ Date: Fri, 09 Feb 2007 09:51:08 -0600 From: David J Dachtera Subject: Re: Virtual tape libraries Message-ID: <45CC986C.8AEA6081@spam.comcast.net> Doug Kimball wrote: > > TAPESYS and JB for OpenVMS are compatible with HP's Virtual Library System > (VLS) 6000-series. > > Also, TAPESYS is a logical replacement for SLS, since TAPESYS has been > ported to Itanium/Integrity and also works with ACSLS and DCSC (an option > that can be chosen instead of our own free JB product). > > Please let me know if you would like to demo TAPESYS and JB for OpenVMS. It > could be the solution that would fit your needs. > > Thanks. > > Doug Please correspond with me privately. How to de-mung the Reply-to should be obvious. -- David J Dachtera dba DJE Systems http://www.djesys.com/ Unofficial OpenVMS Marketing Home Page http://www.djesys.com/vms/market/ Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/ Unofficial OpenVMS-IA32 Home Page: http://www.djesys.com/vms/ia32/ Unofficial OpenVMS Hobbyist Support Page: http://www.djesys.com/vms/support/ ------------------------------ Date: Fri, 9 Feb 2007 10:26:32 -0500 From: "Doug Kimball" Subject: Re: Virtual tape libraries Message-ID: <12sp4lgf6aaf389@corp.supernews.com> TAPESYS and JB for OpenVMS are compatible with HP's Virtual Library System (VLS) 6000-series. Also, TAPESYS is a logical replacement for SLS, since TAPESYS has been ported to Itanium/Integrity and also works with ACSLS and DCSC (an option that can be chosen instead of our own free JB product). Please let me know if you would like to demo TAPESYS and JB for OpenVMS. It could be the solution that would fit your needs. Thanks. Doug -- Doug Kimball Manager, Sales and Support Software Partners, Inc. 978-887-6409 tech_support@softwarepartners.com www.softwarepartners.com "Marc Van Dyck" wrote in message news:mn.3d5a7d7263d9de79.30579@brutele.be... > Has anyone out there already tried to backup one (or more) OpenVMS > systems with any kind of virtual tape library system ? Vax , Alpha, > Itanium ? Which system ? Which software on the server side ? And has > it been successful ? I'm interested in any experience that can be > shared in this area. > > Many thanks in advance, > > -- > Marc Van Dyck > > ------------------------------ End of INFO-VAX 2007.081 ************************