INFO-VAX Wed, 07 Mar 2007 Volume 2007 : Issue 132 Contents: Re: Connecting IDE drives to a XP1000 Re: Connecting IDE drives to a XP1000 Re: DNS- What I'm NOT doing wrong DVD on zx2000 Re: NTP newbie questions Re: NTP newbie questions Re: NTP newbie questions Re: NTP newbie questions Re: OT: Ownership of Ethernet Re: OT: Ownership of Ethernet Re: OT: Ownership of Ethernet Re: OT: Proposed additions to the PDP11 instruction set Re: OT: Proposed additions to the PDP11 instruction set Re: OT: Proposed additions to the PDP11 instruction set Re: OT: Proposed additions to the PDP11 instruction set Re: OT: Proposed additions to the PDP11 instruction set Remove All print jobs from a Print Queue Re: SAMBA on OpenVMS with OS X client Re: SAMBA on OpenVMS with OS X client Re: SAMBA on OpenVMS with OS X client Re: SWAT using Samba v3 on VMS8.2 Re: SWAT using Samba v3 on VMS8.2 Re: SYSMAN IO SET EXCLUDE question Re: TCPIP, non-existent users, and backscatter-spam Re: Time zone/DST change question. Re: Time zone/DST change question. Re: Time zone/DST change question. Re: VMS Client to Windows Storage Server x64 R2 ---------------------------------------------------------------------- Date: Wed, 07 Mar 2007 08:04:50 +0100 From: JOUKJ Subject: Re: Connecting IDE drives to a XP1000 Message-ID: <9dc53$45ee6412$82a13c9d$8432@news2.tudelft.nl> Stephen Hoffman wrote: > JOUKJ wrote: > >> I'm playing a little with IDE drives on my XP1000. >> When I do a "show device" from the running system I see the following >> devices DQA0,DQA1,DQB1,DQB0, which suggest that 2 IDE busses are >> present. However opening up the machine I can find only one (for DQA). >> I see a row of connectors : (SCSI,Floppy,IDE) all for different plugs. >> >> Where do I have to look for the secondary IDE bus? > > > > DQ looks for or knows how many IDE buses that can be connected off the > southbridge, and not at how many devices are connected off the > southbridge. This means you'll regularly see two or more of the DQ > devices (DQA0: and DQA1: on the first bus and DQB0: and DQB1: on the > second) marked as offline on many systems. DQ doesn't differentiate > from "not pinned" and "not connected". > > IIRC, there are a few Alpha systems around where DQA0: and DQA1: are > offline and are either not pinned or not connected, and DQB0: and DQB1: > are. > > The code modification within DQ that would delete any of these > dangling UCBs isn't particularly difficult, but AFAIK that's apparently > never made it "above the fold" and never been tested and shipped by HP. > (And in the department of "no good deed goes unpunished", removing > these dangling UCBs might well derail some application somewhere that > always expects to find them. Simple changes compiled with an installed > base leads to situations becoming Not Simple.) The delete-if-offline > change is probably a dozen lines of Macro32 code, give or take. There > are various SYS$DQDRIVER ECO kits, however. > > The AlphaStation XP1000 boxes I've worked with have all had one IDE > bus connection. There's no DMA path on this box, either. It's PIO. > (I've been told Linux also used the PIO path for this particular > southbridge, though I've not confirmed that.) > > There's an additional wrinkle in the SRM console found on the > AlphaStation XP1000, in that it requires specific device configurations > from ATAPI devices in order to bootstrap. Many current DVD recorders do > not present that, and SRM can't deal with these. There are recording > devices that do boot, and one that has been referenced here in the past > is the Plextor PX-716A series drives. There are undoubtedly others. > Off-hand, I don't know if HP has a supported half-height recorder that > is bootable in this box. > > Newer versions of SRM do not have this limitation, but the most > current version of SRM for the AlphaStation XP1000 appears to be frozen > at an older release. (I dug into this a while back, and it appears to > be a bootstrap-time callback into the SRM firmware that's failing. I > didn't run a bus trace to figure out why.) Newer SRM seems rather more > tolerant of the default ATAPI device settings for the connected CD-R/RW > or DVD+R/RW device. > > There are some related media-recording discussions at: > > http://64.223.189.234/node/28 > > And if somebody has an AlphaStation XP1000 with a second IDE -- that > 54-25090-01 revision F series motherboard mentioned else-thread -- I'd > certainly be interested in hearing about that. I've not encountered one. > > Hoff > Thanks for the explanation. So of the 4 devices on most machines only 2 can be used. Since I'm playing with a satelite in a cluster booting from the bootmember, the bootability of the device is presently not a problem. Jouk ------------------------------ Date: Wed, 07 Mar 2007 08:57:17 +0100 From: JOUKJ Subject: Re: Connecting IDE drives to a XP1000 Message-ID: <2b556$45ee705d$82a13c9d$28736@news1.tudelft.nl> Stephen Hoffman wrote: > > And if somebody has an AlphaStation XP1000 with a second IDE -- that > 54-25090-01 revision F series motherboard mentioned else-thread -- I'd > certainly be interested in hearing about that. I've not encountered one. > > Hoff > I opened up another (newer) XP1000 and found the second connector next to the first one. So the config: SCSI FLOPPY IDE(DQAx) IDE(DQBx) Jouk ------------------------------ Date: Wed, 07 Mar 2007 11:01:21 -0600 From: Hunter Goatley Subject: Re: DNS- What I'm NOT doing wrong Message-ID: Marcin 'Rambo' Roguski wrote: >> I remmember a vnc client, not a vnc server. > > Yup, in deed, just viewer, not server- although it > shouldn't be impossible to port the server. I not > the one who's gonna do it (just yet), though... > David North and I did the VNC Viewer port to VMS. We couldn't do the VNC server because it relied on X features that VMS doesn't have (or didn't at the time). -- 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: Wed, 07 Mar 2007 05:13:16 -0800 From: "Tom Linden" Subject: DVD on zx2000 Message-ID: Have following, HP TCP/IP Services for OpenVMS Industry Standard 64 Version V5.6 on an HP zx2000 (900MHz/1.5MB) running OpenVMS V8.3 Disk $5$DQA0: (ITANIC), device type HL-DT-STDVD-ROM GDR8160B, Does anyone know if Plextor 716A or 740A will work in place of above? ------------------------------ Date: Wed, 7 Mar 2007 12:45:30 +0000 (UTC) From: david20@alpha2.mdx.ac.uk Subject: Re: NTP newbie questions Message-ID: In article <5e877$45ee5de6$cef8887a$794@TEKSAVVY.COM>, JF Mezei writes: >Here is what I would like to do: get one node in my lan to NTP synchronise with >a trusted external source, and then get other nodes, switches, modems to >synchronise with that VMS node. > > >In the configuration file, I know I need > >server ntp.externalclock.com lines for a recommended 3 external servers. > >But the doc isn't too clear on on to enable that node to be a server so it can >feed its time to local nodes. There is mention of adding a "fudge" line that >allows this node to become a server if there are no others available. > >So what configuration lines do I need to have that node fetch time to an >extrenal source (assume the external source is stratum 2), and what lines do I >need to allow this node to serve time to other nodes on the lan ? > The VMS node needs server ntp.external1.com server ntp.external2.com server ntp.external3.com To guard against your VMS node being down you should ideally have two more ntp servers on your network (if the VMS node is VMS1 and the other two are VMS2 and VMS3) then the tcpip$ntp.conf file should also contain peer vms2.yourdomain peer vms3.yourdomain And the other ntp servers should be setup in a similar manner. These systems will then get their times from the external ntp servers and will themselves be setup as your internal ntp servers (they will be one stratum level higher than the external ntp servers). Your switches, other nodes on the lan etc can then be setup to have server vms1.yourdomain server vms2.yourdomain server vms3.yourdomain in their ntp configuration files and will get their time from your internal ntp servers. David Webb Security team leader CCSS Middlesex University > ------------------------------ Date: Wed, 07 Mar 2007 11:06:51 -0500 From: JF Mezei Subject: Re: NTP newbie questions Message-ID: david20@alpha2.mdx.ac.uk wrote: > The VMS node needs > server ntp.external1.com > server ntp.external2.com > server ntp.external3.com OK. Thanks. So basically, the NTP software is, by default acting as a server, correct ? It wasn't too clear in the documentation. ------------------------------ Date: Wed, 07 Mar 2007 09:18:48 -0700 From: Jeff Campbell Subject: Re: NTP newbie questions Message-ID: <1173284410_23067@sp6iad.superfeed.net> JF Mezei wrote: > Here is what I would like to do: get one node in my lan to NTP > synchronise with a trusted external source, and then get other nodes, > switches, modems to synchronise with that VMS node. > > > In the configuration file, I know I need > > server ntp.externalclock.com lines for a recommended 3 external servers. > > But the doc isn't too clear on on to enable that node to be a server so > it can feed its time to local nodes. There is mention of adding a > "fudge" line that allows this node to become a server if there are no > others available. > > So what configuration lines do I need to have that node fetch time to an > extrenal source (assume the external source is stratum 2), and what > lines do I need to allow this node to serve time to other nodes on the > lan ? Simply specify your VMS machine's IP address (or name if DNS is available) as the NTP server on the other machines you want to synchronize. The 'fudge' lines at the bottom of TCPIP$NTP.CONF will let the VMS machine act as a reference source if connectivity to the normal NTP servers is lost. The VMS machine becomes a stratum 8 server when those lines are in effect. Jeff ----== Posted via Newsfeeds.Com - Unlimited-Unrestricted-Secure Usenet News==---- http://www.newsfeeds.com The #1 Newsgroup Service in the World! 120,000+ Newsgroups ----= East and West-Coast Server Farms - Total Privacy via Encryption =---- ------------------------------ Date: Wed, 7 Mar 2007 17:02:31 +0000 (UTC) From: david20@alpha2.mdx.ac.uk Subject: Re: NTP newbie questions Message-ID: In article , JF Mezei writes: >david20@alpha2.mdx.ac.uk wrote: >> The VMS node needs >> server ntp.external1.com >> server ntp.external2.com >> server ntp.external3.com > >OK. Thanks. So basically, the NTP software is, by default acting as a server, >correct ? Yes. David Webb Security team leader CCSS Middlesex University >It wasn't too clear in the documentation. ------------------------------ Date: Wed, 07 Mar 2007 06:08:18 -0500 From: Bob Willard Subject: Re: OT: Ownership of Ethernet Message-ID: Bob Harris wrote: > In article <3f1bb$45edc873$cef8887a$11937@TEKSAVVY.COM>, > JF Mezei wrote: > > >>It is my understanding that Ethernet was developped jointly by Digital, >>Intel and Xerox. >> >>Recently, I got a new ADSL modem and in it, it had a mention that Ethernet >>was a trademark of Xerox corporation. >> >>If it were truly a "trademarked" product, shouldn't the caption mention >>Intel and HP as well as Xerox ? >> >>(This was a Thompson Speedtouch modem). >> >>It was my understanding that Ethernet set out, right from the start, to be >>an open standard without any royalties. Was that the case ? > > > My memory says that the initial ethernet ideas came from Xerox. > > Fleshing out that idea into a robust commercial product is where > Xerox joined forces with intel and Digital. > > Or at least that is my memory of things from reading things like > EE Times and ComputerWorld at the time. > > Bob Harris Some basic ethernet ideas were inspired by, if not directly based on, AlohaNet. Surf's up, dudes. -- Cheers, Bob ------------------------------ Date: Wed, 07 Mar 2007 13:05:27 +0100 From: Paul Sture Subject: Re: OT: Ownership of Ethernet Message-ID: In article <87ps7lhbsw.fsf@k9.prep.synonet.com>, Paul Repacholi wrote: > JF Mezei writes: > > > It is my understanding that Ethernet was developped jointly by > > Digital, Intel and Xerox. > > > Recently, I got a new ADSL modem and in it, it had a mention that > > Ethernet was a trademark of Xerox corporation. > > > If it were truly a "trademarked" product, shouldn't the caption > > mention Intel and HP as well as Xerox ? > > > (This was a Thompson Speedtouch modem). > > > It was my understanding that Ethernet set out, right from the start, > > to be an open standard without any royalties. Was that the case ? > > It was. Xerox did the initial development, then joined forces with > intel and dev to publish the spec and push it as an open standard. > The only fee was was for purchasing an address block AIR. A concise early history is here: http://www2.rad.com/networks/2001/ethernet/hist.htm "The IEEE 802.3 standard is for a CSMA/CD LAN. Ethernet is a specific product that almost implements this standard (Ethernet differs from the standard in one header field). The IEEE 802.3 standard has an interesting history. In 1973 Xerox Corporations Palo Alto Research Center began the development of a bus  topology LAN. Later, in 1976, carrier sensing was added, and Xerox built a 2.94 Mbps  CSMA/CD system to connect over 100 personal workstations on a 1 km cable. Xerox Ethernet was so successful, that in 1980 Digital Equipment Corporation, Intel Corporation and Xerox had released a de facto standard  for a 10 Mbps Ethernet, informally called DIX Ethernet (for the initials of the 3 companies). This standard was a basis for the IEEE 802.3. In 1983 IEEE 802.3 10Base5 was approved, in 1986 10Base2, in 1991 10BaseT, in 1994-1995 10BaseF. In 1995 100 Mbps Ethernet was released, and in 1998-1999 Gigabit Ethernet." -- Paul Sture ------------------------------ Date: 7 Mar 2007 14:23:25 GMT From: bill@cs.uofs.edu (Bill Gunshannon) Subject: Re: OT: Ownership of Ethernet Message-ID: <55806tF241m8eU1@mid.individual.net> In article <87649dgvnk.fsf@k9.prep.synonet.com>, Paul Repacholi writes: > Paul Repacholi writes: > >> JF Mezei writes: >> >>> It is my understanding that Ethernet was developped jointly by >>> Digital, Intel and Xerox. >> >>> Recently, I got a new ADSL modem and in it, it had a mention that >>> Ethernet was a trademark of Xerox corporation. >> >>> If it were truly a "trademarked" product, shouldn't the caption >>> mention Intel and HP as well as Xerox ? >> >>> (This was a Thompson Speedtouch modem). >> >>> It was my understanding that Ethernet set out, right from the start, >>> to be an open standard without any royalties. Was that the case ? >> >> It was. Xerox did the initial development, then joined forces with >> intel and dev to publish the spec and push it as an open standard. >> The only fee was was for purchasing an address block AIR. > > Well, Uni of Honolulu did the original `Aloha Net' with radio, and then > a 1Mb over cable CSMA/CD version. Xerox then the `Xerox Wire' 3Mb then > the DIX 10Mb we all know and love. > Where did ARCNet fall in this big picture? I know very few people offered it, but I do remember seeing it before my first ethernet. 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: 7 Mar 2007 08:04:23 -0600 From: lederman@encompasserve.org (B. Z. Lederman) Subject: Re: OT: Proposed additions to the PDP11 instruction set Message-ID: In article , m.kraemer@gsi.de (Michael Kraemer) writes: > In article <555j5eF21d8nnU1@mid.individual.net>, bill@cs.uofs.edu (Bill > Gunshannon) writes: >> In article , >> sol gongola writes: >> > JF Mezei wrote: >> >> Saw this as a "quote of the day" on a web site... >> >> >> >> Proposed Additions to the PDP-11 Instruction Set: >> >> >> >> BBW Branch Both Ways >> >> BEW Branch Either Way >> >> BBBF Branch on Bit Bucket Full >> >> BH Branch and Hang >> >> BMR Branch Multiple Registers >> >> BOB Branch On Bug >> >> BPO Branch on Power Off >> >> BST Backspace and Stretch Tape >> >> CDS Condense and Destroy System >> >> CLBR Clobber Register >> >> CLBRI Clobber Register Immediately >> >> CM Circulate Memory >> >> CMFRM Come From -- essential for truly structured programming >> >> CPPR Crumple Printer Paper and Rip >> >> CRN Convert to Roman Numerals >> > >> > I remember some of these, and others from my old IBM OS/360 days. >> >> How about the old Motorola Instruction - HCF -- Halt and Catch Fire :-) >> > > I'd add > > XMB: Exclusive Maybe Some of us are still trying to implement the XPI command. (eXecute Programmer Immediate) -- B. Z. Lederman. My personal opinions. ------------------------------ Date: 7 Mar 2007 08:10:26 -0600 From: lederman@encompasserve.org (B. Z. Lederman) Subject: Re: OT: Proposed additions to the PDP11 instruction set Message-ID: Someone mentioned the "Rewind Printer" instruction. In real life, there was one that was very close. The IBM 1620 (a machine many of us started on) usually had a card reader and card punch. It also had two halt buttons on the front console. The "Halt at End of Instruction" would usually halt the machine as indicated. However, in some cases, you could have an instruction that would never complete, such as a branch to self. In that case, the "Halt in the Middle of an Instruction" button was guaranteed by IBM to always halt the machine immediately, even if it was in the middle of reading an instruction (which was 12 characters long). Someone discovered that you could issue a "read on the card punch" instruction. This would cause a card to go through the punch without being punched, but also without being read. Since no data was returned, the punch was instructed to read another card. This would continue as the punch ran through it's stack of cards as fast as possible, at which point it would ask for more cards. The Halt Immediately button would not stop it, you had to power down the machine (which usually took 20 minutes or more to restore). Every time IBM sent a new support person out to the University, someone would ask him if there was any instruction that could not be halted by pressing the button. Being carefully trained by IBM, he would of course always say that there was no such instruction. He soon learned othewise. -- B. Z. Lederman. My personal opinions. ------------------------------ Date: Wed, 7 Mar 2007 11:31:51 -0500 From: norm.raphael@metso.com Subject: Re: OT: Proposed additions to the PDP11 instruction set Message-ID: lederman@encompasserve.org (B. Z. Lederman) wrote on 03/07/2007 09:10:26 AM: > Someone mentioned the "Rewind Printer" instruction. > > In real life, there was one that was very close. > > The IBM 1620 (a machine many of us started on) usually had a > card reader and card punch. It also had two halt buttons on the > front console. The "Halt at End of Instruction" would usually > halt the machine as indicated. However, in some cases, you could > have an instruction that would never complete, such as a branch > to self. In that case, the "Halt in the Middle of an > Instruction" button was guaranteed by IBM to always halt the > machine immediately, even if it was in the middle of reading an > instruction (which was 12 characters long). > > Someone discovered that you could issue a "read on the card > punch" instruction. This would cause a card to go through the > punch without being punched, but also without being read. Since > no data was returned, the punch was instructed to read another > card. This would continue as the punch ran through it's stack of > cards as fast as possible, at which point it would ask for more > cards. The Halt Immediately button would not stop it, you had to > power down the machine (which usually took 20 minutes or more to > restore). > > Every time IBM sent a new support person out to the > University, someone would ask him if there was any instruction > that could not be halted by pressing the button. Being carefully > trained by IBM, he would of course always say that there was no > such instruction. He soon learned othewise. > > -- > B. Z. Lederman. My personal opinions. I was once exposed to a CDC-3600 where if you had, for example, a branch to itself instruction, modified by a zero-offset, executing, you would be hung inside an I-cycle as the instruction modified itself and rebranched ad infinitum. The cure, posted on a large sign near the machine, was, if the Interrupt button did not halt the loop, press the pause button, set up in the I-register (which was, since this machine had a 48-bit word and an octal display) a specific 16-character-octal number that corresponded to a bank-return jump to an illegal address, and press start. This crashed the machine. SOP ------------------------------ Date: 7 Mar 2007 10:41:12 -0800 From: davidc@montagar.com Subject: Re: OT: Proposed additions to the PDP11 instruction set Message-ID: <1173292872.008372.207430@64g2000cwx.googlegroups.com> On Mar 6, 10:28 am, b...@cs.uofs.edu (Bill Gunshannon) wrote: > How about the old Motorola Instruction - HCF -- Halt and Catch Fire :-) I thought that the HCF instruction was inspired on some old IBM hardware where if the CPU actually halted, the discrete logic gates stayed in their halted state and overheated. During normal CPU operation, the gates cycled sufficiently to prevent overheating. I do remember about a HCF type instruction on some later Motorola microprocessor. One instruction that has been removed from hardware due to office politics is the TSM (Trap Secretary and Mount) instruction. ------------------------------ Date: 7 Mar 2007 18:50:36 GMT From: bill@cs.uofs.edu (Bill Gunshannon) Subject: Re: OT: Proposed additions to the PDP11 instruction set Message-ID: <558frrF242mvjU1@mid.individual.net> In article <1173292872.008372.207430@64g2000cwx.googlegroups.com>, davidc@montagar.com writes: > On Mar 6, 10:28 am, b...@cs.uofs.edu (Bill Gunshannon) wrote: > >> How about the old Motorola Instruction - HCF -- Halt and Catch Fire :-) > > I thought that the HCF instruction was inspired on some old IBM > hardware where if the CPU actually halted, the discrete logic gates > stayed in their halted state and overheated. During normal CPU > operation, the gates cycled sufficiently to prevent overheating. I do > remember about a HCF type instruction on some later Motorola > microprocessor. It may have been more urban legend than fact, but the rumor many, many moons ago was that the M6809 (and associated family) had a non-documented instruction that was intended for some kind of testing that locked the processor in a tight loop with all external connections disabled. It would then loop so fast it would overheat and burn itself up. And, with all externals disabled there was no way to break out of this other than shutting off the power. If, of course, you realized in time that it was getting ready to self-destruct. :-) It is believeable as there were other things with test-mode commands that were rather destructive. Like the test-data generate capability in the WD8003 network card. Could completely fill a 10 meg ethernet. Anyone familiar with the other hardware and system of that period will now that many of them could not deal with that much traffic and usually crashed. :-) I learned about it just after an engineer in LA brought down the corporate network while trying to torture test a new bridge design but having forgot to disconnect the lab network from the live network before doing so, :-) 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: 7 Mar 2007 06:48:20 -0800 From: "brucebrown100@yahoo.com" Subject: Remove All print jobs from a Print Queue Message-ID: <1173278900.794168.276010@p10g2000cwp.googlegroups.com> Hi All VMS 7.3.2 What is a quick command to remove/delete all pending print jobs from a VMS print queue? Thank You Bruce ------------------------------ Date: Wed, 07 Mar 2007 12:22:51 GMT From: VAXman- @SendSpamHere.ORG Subject: Re: SAMBA on OpenVMS with OS X client Message-ID: <00A64411.E11860C9@SendSpamHere.ORG> In article , Paul Sture writes: > > >In article <00A64361.972B879E@SendSpamHere.ORG>, > VAXman- @SendSpamHere.ORG wrote: > >> In article , Paul Sture >> writes: >> > >> > >> >In article <00A641D1.5601C314@SendSpamHere.ORG>, >> > VAXman- @SendSpamHere.ORG wrote: >> > >> >> Here's what I'm seeing. >> >> >> >> A share mapped to an ODS-2 drive is no problem. I can open, edit and >> >> close >> >> a file with TextEdit. I can also open a terminal session and do all that >> >> I >> >> want to do from the command line. A share mapped to an ODS-5 volume is >> >> the >> >> problematic one. I can, like with the mapped ODS-2 share, do most >> >> anything >> >> I want from the command line; however, I cannot save any file editted >> >> using >> >> TextEdit. I find files of the form .dat### on the share. I am thinking >> >> it >> >> has something to do with the ODS-5 filenaming semantics and the renaming >> >> of >> >> the temporary file .dat### that TextEdit creates. >> >> >> >> In both cases, the protections that are displayed look identical: >> >> >> >> [~] % cd /Volumes >> >> [/Volumes] % ls -ls >> >> 8 lrwxr-xr-x 1 root admin 1 Mar 3 21:52 Hard Drive -> / >> >> 32 drwx------ 1 vaxman admin 16384 Dec 31 1969 squyrm ; >> >> ODS-2 >> >> 32 drwx------ 1 vaxman admin 16384 Dec 31 1969 squyrm-web ; >> >> ODS-5 >> >> >> > >> >My equivalent listing has today's date and times. "Dec 31 1969" being >> >before the Unix start date, could this be a cause or by-product of your >> >problem? >> > >> >Another thought: from a post you made here last week, you were looking >> >at DECC$EFS_* logicals. Do you have those logicals present in the target >> >environment? >> > >> >On my system, both the NMBD and SMDB_BGnnnn processes show "Parse Style: >> >Traditional" and "Case Lookup: Blind". >> >> >> That was for David Jones's OSU web server. I wanted to modify the code >> for the DIRECTORY server so that it would display: >> >> This^_filename^_has^_spaces.txt >> >> as >> >> This filename has spaces.txt >> >> As well as other non-traditional OpenVMS filenaming characters without >> the ^ escape. > >Sorry, I wasn't asking why you were using 'em *, but if you had the >relevant logicals defined where your Samba processes could see them. > >Ditto for any SET PROCESS/PARSE=EXTENDED commands lurking in the login >command procedure(s) that your Samba processes execute. > >* but thanks for letting me know anyway. Surely! ;) None of the logicals are defined anywhere. When I did define them and I started up the OSU web server everything went awry with the server. I finally used: decc$feature_set_value(decc$feature_get_index("DECC$EFS_CHARSET"),1,n); decc$feature_set_value(decc$feature_get_index("DECC$EFS_CASE_PRESERVE"),1,n); where 'n' is 0:off and 1:on around the decc$translate_vms(filename) to get the proper translation within the DIRSERV_MST.C source. As for PROCESS/PARSE=EXTENDED, there are none in any (SY)LOGIN.COM file. I am completely perplex as to why I can use the TextEdit on an ODS-2 share and not on the ODS-5 share. -- VAXman- A Bored Certified VMS Kernel Mode Hacker VAXman(at)TMESIS(dot)COM "Well my son, life is like a beanstalk, isn't it?" ------------------------------ Date: Wed, 07 Mar 2007 12:27:57 GMT From: VAXman- @SendSpamHere.ORG Subject: Re: SAMBA on OpenVMS with OS X client Message-ID: <00A64412.97B5FF2B@SendSpamHere.ORG> In article , Bob Harris writes: > > >In article >, > Paul Sture wrote: > >> In article <00A641D1.5601C314@SendSpamHere.ORG>, >> VAXman- @SendSpamHere.ORG wrote: >> >> > Here's what I'm seeing. >> > >> > A share mapped to an ODS-2 drive is no problem. I can open, edit and close >> > a file with TextEdit. I can also open a terminal session and do all that I >> > want to do from the command line. A share mapped to an ODS-5 volume is the >> > problematic one. I can, like with the mapped ODS-2 share, do most anything >> > I want from the command line; however, I cannot save any file editted using >> > TextEdit. I find files of the form .dat### on the share. I am thinking it >> > has something to do with the ODS-5 filenaming semantics and the renaming of >> > the temporary file .dat### that TextEdit creates. >> > >> > In both cases, the protections that are displayed look identical: >> > >> > [~] % cd /Volumes >> > [/Volumes] % ls -ls >> > 8 lrwxr-xr-x 1 root admin 1 Mar 3 21:52 Hard Drive -> / >> > 32 drwx------ 1 vaxman admin 16384 Dec 31 1969 squyrm ; >> > ODS-2 >> > 32 drwx------ 1 vaxman admin 16384 Dec 31 1969 squyrm-web ; >> > ODS-5 >> > >> >> My equivalent listing has today's date and times. "Dec 31 1969" being >> before the Unix start date, could this be a cause or by-product of your >> problem? >> >> Another thought: from a post you made here last week, you were looking >> at DECC$EFS_* logicals. Do you have those logicals present in the target >> environment? >> >> On my system, both the NMBD and SMDB_BGnnnn processes show "Parse Style: >> Traditional" and "Case Lookup: Blind". > >31-Dec-1969 is generally a by product of a date value of zero (0) >with a minus timezone offset, such as -5 for EST or -8 for PST. On the shares mounted on OS X, mine seem to pick up a date value as soon as I access them. -- 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: Wed, 07 Mar 2007 18:35:58 +0100 From: Paul Sture Subject: Re: SAMBA on OpenVMS with OS X client Message-ID: In article <00A64411.E11860C9@SendSpamHere.ORG>, VAXman- @SendSpamHere.ORG wrote: > I am completely perplex as to why I can use the TextEdit on an ODS-2 > share and not on the ODS-5 share. I've reproduced the problem, and I am willing to bet that the ODS-2/ODS-5 thing is just a red herring. Consider the following: My Samba share is directory [STUREP], and STUREP is the account I set up for Samba (only NETMBX,TMPMBX privileges, and unique UIC Group number) However, I had a subdirectory in there "zk.dir" with wo:r access, belonging to another user. I couldn't see any of the files in it from either Finder or the CLI, as they had no world access. Navigate there from the OS X CLI and I could create test.txt. Open that in TextEdit, change then try to save, and we get Couldn¹t save document as ³test.txt² in folder ³zk². Give zk.dir World:RW on the VMS side, and TextEdit can then save the file. Better still, remove world:write, and change the owner of zk.dir to [STUREP]. I think that should get you going. If not, please let me know. -- Paul Sture ------------------------------ Date: 7 Mar 2007 04:10:49 -0800 From: "dky" Subject: Re: SWAT using Samba v3 on VMS8.2 Message-ID: <1173269449.124295.143020@v33g2000cwv.googlegroups.com> On Mar 5, 5:27 pm, Dan Foster wrote: > In article <1173096835.192402.101...@p10g2000cwp.googlegroups.com>, dky wrote: > > > PS: > > Can someone provide a shell account on an IA64 box running VMS 8.3, I > > can continue working onSamba. I am no longer with HP and hence no > > access to a VMS box. > > If it doesn't require privileges for porting efforts, you could sign up > for a HP TestDrive account which has VMS V8.3 on I64 systems and access > is provided for this kind of purpose (amongst others). I have a testdrive account. The last I checked, it did not have MMS, compiler etc... that I would require for development. I will need just enough privileges to define system wide logicals and someone need to create an account with few identifiers. This is required _now_ as we added UNIX group support which was missing in CRTL. Using the newly added group support, we can work towards ACL support. -dky ------------------------------ Date: Wed, 7 Mar 2007 08:10:37 -0600 (CST) From: sms@antinode.org (Steven M. Schweda) Subject: Re: SWAT using Samba v3 on VMS8.2 Message-ID: <07030708103739_2020028F@antinode.org> From: "dky" > On Mar 5, 5:27 pm, Dan Foster wrote: > > In article <1173096835.192402.101...@p10g2000cwp.googlegroups.com>, dky wrote: > > > > > PS: > > > Can someone provide a shell account on an IA64 box running VMS 8.3, I > > > can continue working onSamba. I am no longer with HP and hence no > > > access to a VMS box. > > > > If it doesn't require privileges for porting efforts, you could sign up > > for a HP TestDrive account which has VMS V8.3 on I64 systems and access > > is provided for this kind of purpose (amongst others). > > I have a testdrive account. The last I checked, it did not have MMS, > compiler etc... that I would require for development. And when was that? td183 $ mms /id %MMS-I-IDENT, MMS T3.8 © Copyright 2006 Hewlett-Packard Development Company, L.P td183 $ cc /vers HP C V7.2-022 on OpenVMS IA64 V8.3 SYS$MANAGER:WELCOME.TXT claims much/many more. > I will need just > enough privileges to define system wide logicals and someone need to > create an account with few identifiers. [...] I'd advise sending e-mail to testdrive@hp.com to discuss such requests. The new case-sensitive MMS may be expected to cause some problems for the case-careless user, but it's probably an improvement. ------------------------------------------------------------------------ Steven M. Schweda sms@antinode-org 382 South Warwick Street (+1) 651-699-9818 Saint Paul MN 55105-2547 ------------------------------ Date: Wed, 07 Mar 2007 11:13:08 -0500 From: JF Mezei Subject: Re: SYSMAN IO SET EXCLUDE question Message-ID: <130e5$45eee4ac$cef8887a$27376@TEKSAVVY.COM> Peter 'EPLAN' LANGSTOeGER wrote: > But obviously doesn't work (at least with IDE). > Have you contacted HP? It has worked fine for me until I rebooted one of my 2 alphas with the patch. as a proof that it is working: $ mc sysman io show exclude %SYSMAN-I-OUTPUT, command execution on node BIKE %SYSMAN-I-IOEXCLUDE, the current permanent exclusion list is: DQA1,DQB0,DQB1,DVA0 show dev $10$d Device Device Error Volume Free Trans Mnt Name Status Count Label Blocks Count Cnt $10$DQA0: (BIKE) Mounted 0 FREEWHEEL 27822640 563 4 ------------------------------ Date: Wed, 7 Mar 2007 16:41:47 +0000 (UTC) From: david20@alpha2.mdx.ac.uk Subject: Re: TCPIP, non-existent users, and backscatter-spam Message-ID: In article , helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) writes: >In some threads here recently, there has been some discussion of >non-existent users, how to handle them, backscatter spam etc. > >Rule 1: If it is spam, don't send a bounce. > >Rule 2: A lot of spam is sent to non-existent users, so for a long time >now I have been rejecting these (NOT bouncing them). This got rid of a >lot of spam, but there is still enough so that I am now routing email >through an external MX server which scans it for spam and, if spam is >detected, rejects the messages (it doesn't bounce them). So far, so >good. > >Problem: Spam messages to non-existent users below the spam threshold >are still rejected by me, but not to the original sender, but rather to >my external MX server. This server then issues a bounce. That is not >what I want! (I can reduce the number of these bounces by letting the >external server reject spam, but I can't eliminate them completely, >since some are below the spam threshold.) Can I somehow via SMTP >protocol tell the external server NOT to bounce in case I reject it with >"user unknown"? > >Other people who don't receive email directly on VMS must have this >problem. If the external server has a list of valid email addresses >everywhere, fine, but this is impractical. > >It seems that in this case the lesser evil would be a catchall mail >account on the VMS side. Don't use it (rather, reject the messages) if >one gets email directly. However, if one is not getting email directly >(or if one wants to be really safe), use it so that bounces aren't >generated for non-existent users by the external MX server. > In which case you are not providing any feedback to valid users who just mistyped the email address. Unfortunately having an intermediate mail server which doesn't have full information on all addresses so it can reject during the smtp dialogue is the cause of these types of problems. The only real solution is to remove that intermediate system or to provide it with full information. (This isn't a problem for most of the ISP's users since they are just receiving mail into the mail account provided by the ISP and aren't having it delivered from their ISP to their own mailserver which has lots of accounts which the ISP doesn't know about). It appears from previous posts that with the latest version of DEC TCPIP services it will correctly reject most invalid addresses but seems to not cope when the addresses are enclosed in quotes. That is what needs fixing. Then you need to be able to run a content-scanning anti-spam product with DEC TCPIP services. If you can afford it then PMAS from process will run on VMS. To work with DEC TCPIP services it would have to be run in pass through proxy mode either on a separate box in front of the DEC TCPIP mail server or on the same box. (If run on the same box you would have to run DEC TCPIP service's mailserver on a port other than port 25). In pass through proxy mode it connects to the backend mailserver during the SMTP dialogue to determine if the account exists and rejects if it doesn't. It can also be configured to reject spam and viruses during the smtp dialogue. The alternative would be to investigate the state of Spamassassin for VMS. I haven't looked at this for years - not since we started using PMAS - hence am not sure what the current state is. I'm not sure how you would tie this into DEC TCPIP services' mail server you might have to be content with just delivering the mail and then using DELIVER to run spamassassin on the delivered mail message. For hobbyists and those who can afford the commercial price I would always recommend PMDF and PMAS from process (see http://www.process.com). In the past I would then have recommended MX (http://www.madgoat.com) for those organisations looking for a commercial solution but looking for something cheaper than PMDF. Unfortunately as I mentioned before it looks like Madgoat is no longer offering MX for sale to new customers. The other alternatives might be to get TCPWARE or Multinet as a TCPIP stack instead of DEC TCPIP services. David Webb Security team leader CCSS Middlesex University >I can't think of any feature HP TCPIP for VMS needs more than this one. > ------------------------------ Date: Wed, 07 Mar 2007 08:57:58 -0500 From: "Richard B. gilbert" Subject: Re: Time zone/DST change question. Message-ID: <45EEC4E6.3020202@comcast.net> Neil Rieck wrote: > > "JF Mezei" wrote in message > news:80b26$45ee2c59$cef8887a$19152@TEKSAVVY.COM... > >> The release notes for VMS83A_UPDATE-V0200 which include the TZ update >> mention that from now on, those west of GMT will in in a positive time >> differential instead of negative as in the past. >> >> However, after intalling the kit (but not yet rebooting), I executed >> the UTC$TIME_ZONE_SETUP procedure as the release notes said, but when >> selecting eastern canada time zone it still told me that my time zone >> (eastern canada) is -5 in winter and -4 in summer. >> > A very weird thing happened to me last week when I was testing some > application software on an Oracle-8 system. For reasons I won't go into > here, we are in possession of some third-party UNIX application software > with baked-in DST rules "and" there is not enough time to install an > upgrade. > > So I installed the Solaris-8 patches and was then developing a procedure > so that 2:00 AM on Mar-11 we were going to manually change the timezone > from Canada/Eastern to just plain old GMT-4. (cuz we won't get the new > s/w loaded until mid-summer and I didn't want the baked-in s/w to flip > over to EDT on April-1). > > Now here is the weird part, while testing the procedure on a captive > system I learned than GMT-4 actually produced GMT+4. Peering inside file > "/usr/share/lib/zoneinfo/GMT-4" actually revealed "GMT+4" and checking > file "GMT+4" revealed "GMT-4". This is the only UNIX box doing this so I > just assumed that someone screwed up at Sun. > > Now it's kind of weird that people have been confusing this stuff all > along. While it is true that EST (local) is the same as GMT minus 5, > program data we send to our ExpressVu satellites is always expressed in > GMT which is 5 hours ahead of EST (this is just a matter of perspective > I guess). > > I guess I had better check the release notes you mentioned. > "GMT-4" is ambiguous! It can mean an offset from GMT or can be the name of a time zone. The time zone is what I think of as GMT+4; e.g. 4 hours East of Greenwich. ------------------------------ Date: Wed, 7 Mar 2007 09:59:40 -0500 From: "Neil Rieck" Subject: Re: Time zone/DST change question. Message-ID: <45eec6cc$0$16303$88260bb3@free.teranews.com> "Richard B. gilbert" wrote in message news:45EEC4E6.3020202@comcast.net... > Neil Rieck wrote: > [...snip...] > > "GMT-4" is ambiguous! It can mean an offset from GMT or can be the name > of a time zone. The time zone is what I think of as GMT+4; e.g. 4 hours > East of Greenwich. > You may be right correct but I am still confused by this latest POSIX compliancy wrinkle. When you look at documentation like this: http://www.acad.bg/ftp/usr/share/lib/zoneinfo/etcetera http://docs.sun.com/app/docs/doc/806-5189/6je7g4sr4?a=view http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=4976879 http://src.opensolaris.org/source/xref/onnv/onnv-gate/usr/src/cmd/zic/etcetera you would be tempted to use GMT+5 and GMT+4 to move between EST and EDT time. But for years we have always thought about these time zones as "GMT minus 5" and "GMT minus 4" Renaming these timezones seems confuse things. Neil Rieck Kitchener/Waterloo/Cambridge, Ontario, Canada. http://www3.sympatico.ca/n.rieck/links/cool_openvms.html http://www3.sympatico.ca/n.rieck/links/openvms_demos.html -- Posted via a free Usenet account from http://www.teranews.com ------------------------------ Date: Wed, 7 Mar 2007 10:10:03 -0500 From: "Neil Rieck" Subject: Re: Time zone/DST change question. Message-ID: <45eec93c$0$16290$88260bb3@free.teranews.com> "Richard B. gilbert" wrote in message news:45EEC4E6.3020202@comcast.net... > Neil Rieck wrote: [...snip...] > > "GMT-4" is ambiguous! It can mean an offset from GMT or can be the name > of a time zone. The time zone is what I think of as GMT+4; e.g. 4 hours > East of Greenwich. > To make things even more confusing, this Solaris-8 file: /usr/share/lib/zoneinfo/GMT-4 contains the string GMT+4 while this file: /usr/share/lib/zoneinfo/Etc/GMT-4 contains the string GMT-4 I've always thought that adopting industry standards was a good idea (until now). An whether we like it or not, those decisions eventually affect OpenVMS. Neil Rieck Kitchener/Waterloo/Cambridge, Ontario, Canada. http://www3.sympatico.ca/n.rieck/ -- Posted via a free Usenet account from http://www.teranews.com ------------------------------ Date: Wed, 07 Mar 2007 16:19:40 +0100 From: Paul Sture Subject: Re: VMS Client to Windows Storage Server x64 R2 Message-ID: In article , "Michael D. Ober" wrote: > I'm close to having the VMS NFS Client working with the WSS NFS Server. > Since I know this configuration has been asked about before, when I get it, > I'll post the configuration for both ends of the link. As is happens, I have got the VMS NFS Client working with the OS X NFS Server. When you get the Windows flavour working, it might be useful to compare notes here, at least from the VMS side. -- Paul Sture ------------------------------ End of INFO-VAX 2007.132 ************************