1 INFO-VAX	Sat, 20 Aug 2005	Volume 2005 : Issue 463       Contents: Re: 'Nuff said Re: 'Nuff said Re: 'Nuff said RE: 'Nuff said Re: 'Nuff said Re: 'Nuff said RE: 'Nuff said Re: 'Nuff said Re: 'Nuff said 7.3 supported ethernet Re: 7.3 supported ethernet@ AlphaServer 4100 CPU speed upgrade and/or backplane speed rangesD RE: AlphaServer 4100 CPU speed upgrade and/or backplane speed rangesD Re: AlphaServer 4100 CPU speed upgrade and/or backplane speed rangesD Re: AlphaServer 4100 CPU speed upgrade and/or backplane speed ranges: Re: How many files can you have in a VMS directory without: Re: How many files can you have in a VMS directory without Re: Image restore fails  Re: Image restore fails  Re: microVax3100-40 cpu number Re: microVax3100-40 cpu number. OSU Webserver and NETSERVER.LOG files question2 Re: OSU Webserver and NETSERVER.LOG files question2 Re: OSU Webserver and NETSERVER.LOG files question2 Re: OSU Webserver and NETSERVER.LOG files question2 Re: OSU Webserver and NETSERVER.LOG files question2 Re: OSU Webserver and NETSERVER.LOG files question( Re: Packet Trace Facility in DEC TCP/IP? Re: The SAMBA Saga at an end!  Re: The SAMBA Saga at an end! ! Re: What is the official status ? ! Re: What is the official status ? P Re: [TCPware V5.6-2,KERBEROS V2.1-72,VMS V7.3-2] TCPIP$IPC_SHR         SHRIDMISMK Re: [TCPware V5.6-2,KERBEROS V2.1-72,VMS V7.3-2] TCPIP$IPC_SHR  SHRIDMISMAT J Re: [TCPware V5.6-2,KERBEROS V2.1-72,VMS V7.3-2] TCPIP$IPC_SHR SHRIDMISMATJ Re: [TCPware V5.6-2,KERBEROS V2.1-72,VMS V7.3-2] TCPIP$IPC_SHR SHRIDMISMAT  F ----------------------------------------------------------------------  % Date: Fri, 19 Aug 2005 14:09:23 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com>  Subject: Re: 'Nuff said , Message-ID: <43062050.8B1822A8@teksavvy.com>   Bill Gunshannon wrote:  I > Security by obscurity is not security, it is a false sense of security.     H Security by obscurity doesn't prevent break in. It reduces the odds of a	 break in.   F Think of a boat with a hole. Security by obscurity reduces the size ofC the hole. Using very well know networking increases the size of the  hole.     C However, with decnet, the advantage doesn't come from obscurity, it G comes from it being different from TCPIP and not widely accessible from  the internet.     C Consider Windows with its ports 138 445 et all. It opens Windows to # attacks from anywhere in the world.   H But DECNET isn't TCPIP, so all the tools hackers have are rather useless to try to crack into DECNET.    F If DECNET were widely used on the internet, hackers would scan for VMSG machines and then try every possible object name and number to see what  they could do.    C The big difference is that with VMS, it is fairly easy to see which F DECNET objects are defined, which one require authentication and which= don't (for instance PHONE is open, but FAL isn't by default).   D On Windows, it seems next to impossible to see what ports Windows isH listening to. Just yesterday, read an article about Windows Vista testerF discovering that by default Windows has some peer to peer port enabledF and it scans the network for other machines with whom it can peer.  SoD again, MS sends out products with some ports enabled under the tableC without any obvious way to list those ports, knwo what they do etc.   D The Zotob is even more interesting. It seems that the guilty port isH enabled or disabled based on an obscure registry variable being set to 0 or 1.     C So in a sense, Windows is insecure through obscurity because system G managers don't have the tools to see what the windows machine is really E listening to. (a port scanner from another machine may tell you which H port is being listened to, but it won't tell you what portion of windows is listening to that port).    ------------------------------  % Date: Fri, 19 Aug 2005 14:23:55 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com>  Subject: Re: 'Nuff said , Message-ID: <430623B4.1DC56FB0@teksavvy.com>   "Main, Kerry" wrote:H > Nope - As I stated, "obscurity + real security" is imho, a good way toJ > further increase security. Having good security like Mosler safe exampleH > you provided but with the Mosler plans available on the Internet would? > not enhance the overall security of the Mosler safe solution.     C Technically speaking, the Mosler plans are on the internet. If some H locksmith in Germany has plans of how to break into the safe, then thoseQ plans must be considered to be out in the open. (albeit not in wide circulation).   B If any locksmith can get those plans, then I just need to become a: locksmith and get the plans and then breakinto some vault.  G This is very similar to OPA0: being weak on a system because anyone who G knows VMS can get into a system without a username/password via OPA0: . + So physical security of OPA0: is important.    ------------------------------    Date: 19 Aug 2005 13:26:14 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler)  Subject: Re: 'Nuff said 3 Message-ID: <z+RK8Scbj1M+@eisner.encompasserve.org>   Z In article <00A4881D.9A1DB16A.1@tachysoft.com>, Wayne Sewell <wayne@tachysoft.com> writes: > N > Yes, decnet is superior when going from vms to vms.  For instance, FAL makes& > file transfers extremely easy to do. >   E    FAL (and DNA) also makes transfering files between non-VMS systems E    extreemly easy to do.  I do miss those days when everything I had, 4    DEC and otherwise, ran somebody's Phase IV stack.   ------------------------------  % Date: Fri, 19 Aug 2005 14:45:41 -0400 ' From: "Main, Kerry" <Kerry.Main@hp.com>  Subject: RE: 'Nuff said R Message-ID: <FD827B33AB0D9C4E92EACEEFEE2BA2FB6B201A@tayexc19.americas.cpqcorp.net>   > -----Original Message-----9 > From: JF Mezei [mailto:jfmezei.spamnot@teksavvy.com]=20  > Sent: August 19, 2005 2:24 PM  > To: Info-VAX@Mvb.Saic.Com  > Subject: Re: 'Nuff said  >=20 > "Main, Kerry" wrote:A > > Nope - As I stated, "obscurity + real security" is imho, a=20 
 > good way to B > > further increase security. Having good security like Mosler=20 > safe example> > > you provided but with the Mosler plans available on the=20 > Internet wouldA > > not enhance the overall security of the Mosler safe solution.  >=20 >=20E > Technically speaking, the Mosler plans are on the internet. If some B > locksmith in Germany has plans of how to break into the safe,=20 > then those@ > plans must be considered to be out in the open. (albeit not=20 > in wide circulation).  >=20D > If any locksmith can get those plans, then I just need to become a< > locksmith and get the plans and then breakinto some vault. >=20  B I would assume that Mosler only releases those plans to a very fewF trusted and authorized i.e. police checked/cleared partners. One would< also assume that these binders are themselves kept in safes.  	 [Snip ..]    Regards   
 Kerry Main Senior Consultant  HP Services Canada Voice: 613-592-4660  Fax: 613-591-4477  kerryDOTmainAThpDOTcom (remove the DOT's and AT)=20  4 OpenVMS - the secure, multi-site OS that just works.   ------------------------------  % Date: Fri, 19 Aug 2005 15:24:06 -0400 - From: William Webb <william.w.webb@gmail.com>  Subject: Re: 'Nuff said 7 Message-ID: <8660a3a1050819122469bbdf3f@mail.gmail.com>   2 On 8/19/05, Main, Kerry <Kerry.Main@hp.com> wrote: > > -----Original Message-----8 > > From: JF Mezei [mailto:jfmezei.spamnot@teksavvy.com]! > > Sent: August 19, 2005 2:24 PM  > > To: Info-VAX@Mvb.Saic.Com  > > Subject: Re: 'Nuff said  > >  > > "Main, Kerry" wrote:@ > > > Nope - As I stated, "obscurity + real security" is imho, a > > good way to A > > > further increase security. Having good security like Mosler  > > safe example= > > > you provided but with the Mosler plans available on the  > > Internet wouldC > > > not enhance the overall security of the Mosler safe solution.  > >  > > G > > Technically speaking, the Mosler plans are on the internet. If some A > > locksmith in Germany has plans of how to break into the safe,  > > then those? > > plans must be considered to be out in the open. (albeit not  > > in wide circulation).  > > F > > If any locksmith can get those plans, then I just need to become a> > > locksmith and get the plans and then breakinto some vault. > >  >=20D > I would assume that Mosler only releases those plans to a very fewH > trusted and authorized i.e. police checked/cleared partners. One would> > also assume that these binders are themselves kept in safes. >=20 > [Snip ..]  >=20	 > Regards  >=20 > Kerry Main > Senior Consultant  > HP Services Canada > Voice: 613-592-4660  > Fax: 613-591-4477  > kerryDOTmainAThpDOTcom > (remove the DOT's and AT)  >=206 > OpenVMS - the secure, multi-site OS that just works. >=20  < If they were Mosler safes, would that be recursive security?   : ^ )    WWWebb   --=20 C NOTE: This email address is only used for noncommerical VMS-related  correspondence. C All unsolicited commercial email will be deemed to be a request for 8 services pursuant to the terms and conditions located at# http://bellsouthpwp.net/w/e/webbww/    ------------------------------  % Date: Fri, 19 Aug 2005 15:10:57 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com>  Subject: Re: 'Nuff said , Message-ID: <43062EB7.EF0EA5D6@teksavvy.com>   "Main, Kerry" wrote:   D > I would assume that Mosler only releases those plans to a very fewH > trusted and authorized i.e. police checked/cleared partners. One would> > also assume that these binders are themselves kept in safes.    R assume, assume, assume ! Not a word that should be part of any security programme.  B This is similar to the paper airline tickets issues. A blank paperB ticket is akin to a signed blank check and anyone could then writeG themselves an airline ticket and airlines would be forced to honour it. C As a result, all travel agents are required to keep the blank paper  tickets in a safe.  H The rule are there. But that doesn't mean that they are followed 100% ofC the time and the possibility exists that some paper ticket stock is  stolen.     H You either close all possible hols in the security, or you become a risk; manager and then decide if a weakness is acceptable or not.   G For instance: do you really need to encrypt a dialup modem connection ? D What are the odds that someone will be listening on the phone line ?  F If you use that line at specific times of the day to transfer billionsH of dollars and you know that there are industrial spies around, then you? may require that line to be encrypted. But if you use that line H sporadically to access another system to type a few commands to check onG its health, then the odds are quite low that someone would want to work 6 hard to tap that phone line to see what you're typing.   ------------------------------  % Date: Fri, 19 Aug 2005 17:00:09 -0400 # From: "Dan Allen" <dallen@nist.gov>  Subject: RE: 'Nuff said : Message-ID: <JFEPKAPBPMDFDBOIANGDAEGKGLAA.dallen@nist.gov>  P The standard issue DOD safes (Mosler) we had circa 1975-85 (approved for storageK of documents up to and including TS as long as they were inside our secured N perimeter) contained a security rating label inside the front drawer. IIRC theP ratings on that label were about 30 min for fire, 15 min for "manipulated" entryP and 0 min for "forced" entry. I guess the Mosler tech was out of C4 ;-) I'd alsoO suggest that hiding said safe would increase it's effective security rating for H manipulated and forced entry if measured from the time our perimeter wasI breached. Obscurity always ENHANCES security. That's why the duty station O schedules for war ships and subs was Top Secret and locked in that Mosler safe.    Dan    > -----Original Message-----. > From: Main, Kerry [mailto:Kerry.Main@hp.com]' > Sent: Friday, August 19, 2005 2:46 PM  > To: Info-VAX@Mvb.Saic.Com  > Subject: RE: 'Nuff said  >  >  > > -----Original Message-----8 > > From: JF Mezei [mailto:jfmezei.spamnot@teksavvy.com]! > > Sent: August 19, 2005 2:24 PM  > > To: Info-VAX@Mvb.Saic.Com  > > Subject: Re: 'Nuff said  > >  > > "Main, Kerry" wrote:@ > > > Nope - As I stated, "obscurity + real security" is imho, a > > good way to A > > > further increase security. Having good security like Mosler  > > safe example= > > > you provided but with the Mosler plans available on the  > > Internet wouldC > > > not enhance the overall security of the Mosler safe solution.  > >  > > G > > Technically speaking, the Mosler plans are on the internet. If some A > > locksmith in Germany has plans of how to break into the safe,  > > then those? > > plans must be considered to be out in the open. (albeit not  > > in wide circulation).  > > F > > If any locksmith can get those plans, then I just need to become a> > > locksmith and get the plans and then breakinto some vault. > >  > D > I would assume that Mosler only releases those plans to a very fewH > trusted and authorized i.e. police checked/cleared partners. One would> > also assume that these binders are themselves kept in safes. >  > [Snip ..]  > 	 > Regards  >  > Kerry Main > Senior Consultant  > HP Services Canada > Voice: 613-592-4660  > Fax: 613-591-4477  > kerryDOTmainAThpDOTcom > (remove the DOT's and AT)  > 6 > OpenVMS - the secure, multi-site OS that just works. >    ------------------------------  % Date: Fri, 19 Aug 2005 18:01:05 -0700 % From: DeanW <dean.woodward@gmail.com>  Subject: Re: 'Nuff said 6 Message-ID: <3f119ada05081918016b99a6e@mail.gmail.com>  : On 8/19/05, JF Mezei <jfmezei.spamnot@teksavvy.com> wrote:D > This is similar to the paper airline tickets issues. A blank paperD > ticket is akin to a signed blank check and anyone could then writeI > themselves an airline ticket and airlines would be forced to honour it. E > As a result, all travel agents are required to keep the blank paper  > tickets in a safe.  B Last time I got on an airplane, I'd printed the tickets / boarding- passes at home. Did one of us miss something?    ------------------------------  % Date: Fri, 19 Aug 2005 22:49:51 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com>  Subject: Re: 'Nuff said , Message-ID: <43069A3B.44AC3F0E@teksavvy.com>   DeanW wrote:D > Last time I got on an airplane, I'd printed the tickets / boarding/ > passes at home. Did one of us miss something?    not really.   E Traditional paper tickets were like cash or traveller's cheques.  One D you boarded the aircraft, they lifted a ticket stub. That stub wouldA then be sent to a processing centre (often in india) where ticket < reconciliation was made and money/commissions were settled.   F This was very costly to airlines because it required a lot of manpowerL and  especially time before money actually got assigned to the right people.  F (consider you buy a ticket from british airways to london as first legH and qantas to sydney as second leg of the ticket).  Someone actually had? to calculate how much money British airways owed Qantas, deduct F commissions to british airways (or to travel agent etc etc), all basedD on what was WRITTEN on the ticket as well as void the ticket number.  H Your reservations were just a convenience for capacity planning purposes and not used for accounting.  H Airlines have, over the past few years been moving to electronic ticketsH where the paper you print on is worthless and all reconciliation is done by computer.  G It removes the need to keep paper stock in a safe, and more importantly F eliminates the costly and time consuming paper reconciliation process.G Your reservation in the computer is now the authoritative record of the  financial transaction.  C In the past, the airline could lose your reservation, but with your A paper ticket, they were still obligated to carry you. Now, if the 4 airline loses your reservation, you're up the creek.  E And as time progresses, airlines are increasing the flexibility to do $ electronic tickets between airlines.   ------------------------------  # Date: Sat, 20 Aug 2005 00:24:30 GMT / From: Adam Stouffer <adam_stouffer@hotmail.com>  Subject: 7.3 supported ethernet & Message-ID: <2LuNe.49$N22.33@trndny07>  H Anyone have a list of supported ethernet cards other than the DE500 and J tulip cards? Or does that cover it ;) I'm gathering parts for a PC164 box.     Adam   ------------------------------  % Date: Sat, 20 Aug 2005 01:37:10 -0400 ' From: Dave Froble <davef@tsoft-inc.com> # Subject: Re: 7.3 supported ethernet 0 Message-ID: <11gdg4iggnpoq43@corp.supernews.com>   Adam Stouffer wrote:J > Anyone have a list of supported ethernet cards other than the DE500 and L > tulip cards? Or does that cover it ;) I'm gathering parts for a PC164 box. >  >  > Adam  I There are the DE600 cards, and then the 4-port cards, don't remember the   number.    --  4 David Froble                       Tel: 724-529-04504 Dave Froble Enterprises, Inc.      Fax: 724-529-0596> DFE Ultralights, Inc.              E-Mail: davef@tsoft-inc.com 170 Grimplin Road  Vanderbilt, PA  15486    ------------------------------  % Date: Fri, 19 Aug 2005 17:57:34 -0500 . From: Duncan Brown <brown_du@eisner.decus.org>I Subject: AlphaServer 4100 CPU speed upgrade and/or backplane speed ranges 2 Message-ID: <o7adnXy1JuJ9_pveRVn-1g@speakeasy.net>  A I have a hobbyist AS4100 5/400 4-processor machine running in my  D basement.  (Hey, if you're going to do it, might as well go all the G way...)  It's been running fine for a few years, though it was only in  I the last year or so that I added the 4th processor board, thanks to ebay.   I I recently snagged, all in one ebay auction, a set of 4 AS4100 5/533 CPU  F boards.  I installed them and they all tested fine, but after running H for an hour or so, the machine crashed and hung.  I rebooted and it ran G fine for another day and a half or so and then crashed and hung.  SO I  G started looking to see if I'd done something wrong in installing these   boards.   C A bunch of Google searching later, it looks like maybe there are 3  H flavors of 4100 system: 5/300; 5/400 and 5/466; 5/533 and 5/600.  I see G mention of different backplane speeds as regards intermixing processor  I speeds, but it's not clear if the speed is determined by the chassis, or  F by the cards plugged into it.  Is that the big difference between the I flavors of 4100s?  Is there any way to change a jumper or an oscillator,  F and successfully use my 5/533 boards?  Or should they be working, and ( there's something else going wrong here?  I (I'll mention that when I put the 5/400 boards back in, CPU03 decided to  A fail Test 11 so I simply removed it and haven't investigated any  K farther... but I mention it, in case it's somehow a clue to my problems...)   F Anyone out there know these really obscure details on the AlphaServer I 4100?  Should I be spending my efforts and money a nice used DS20 system  E instead?  (Then how would I hook in my dual-redundant HSJ50 cabinet?   But I digress...)    Duncan   ------------------------------  % Date: Fri, 19 Aug 2005 20:24:46 -0400 ' From: "Main, Kerry" <Kerry.Main@hp.com> M Subject: RE: AlphaServer 4100 CPU speed upgrade and/or backplane speed ranges R Message-ID: <FD827B33AB0D9C4E92EACEEFEE2BA2FB6B203A@tayexc19.americas.cpqcorp.net>   > -----Original Message-----: > From: Duncan Brown [mailto:brown_du@eisner.decus.org]=20 > Sent: August 19, 2005 6:58 PM  > To: Info-VAX@Mvb.Saic.Com A > Subject: AlphaServer 4100 CPU speed upgrade and/or backplane=20  > speed ranges >=20E > I have a hobbyist AS4100 5/400 4-processor machine running in my=20 H > basement.  (Hey, if you're going to do it, might as well go all the=20? > way...)  It's been running fine for a few years, though it=20  > was only in=20> > the last year or so that I added the 4th processor board,=20 > thanks to ebay.  >=20< > I recently snagged, all in one ebay auction, a set of 4=20 > AS4100 5/533 CPU=20 J > boards.  I installed them and they all tested fine, but after running=20A > for an hour or so, the machine crashed and hung.  I rebooted=20  > and it ran=20 ? > fine for another day and a half or so and then crashed and=20  > hung.  SO I=20: > started looking to see if I'd done something wrong in=20 > installing these=20 	 > boards.  >=20G > A bunch of Google searching later, it looks like maybe there are 3=20 > > flavors of 4100 system: 5/300; 5/400 and 5/466; 5/533 and=20 > 5/600.  I see=20A > mention of different backplane speeds as regards intermixing=20  > processor=20A > speeds, but it's not clear if the speed is determined by the=20  > chassis, or=20J > by the cards plugged into it.  Is that the big difference between the=20A > flavors of 4100s?  Is there any way to change a jumper or an=20  > oscillator,=20J > and successfully use my 5/533 boards?  Or should they be working, and=20* > there's something else going wrong here? >=20B > (I'll mention that when I put the 5/400 boards back in, CPU03=20 > decided to=20 E > fail Test 11 so I simply removed it and haven't investigated any=20 @ > farther... but I mention it, in case it's somehow a clue to=20 > my problems...)  >=20J > Anyone out there know these really obscure details on the AlphaServer=20A > 4100?  Should I be spending my efforts and money a nice used=20  > DS20 system=20I > instead?  (Then how would I hook in my dual-redundant HSJ50 cabinet?=20  > But I digress...)  >=20 > Duncan >=20   Duncan,    Hey, nice system!    :-)   H I also have an Alpha 4100 dual cpu EV56 600Mhz system (VMS V8.2) at homeA that I hope to bring online shortly. Electrical bills are what is H holding me back right now, so I will likely be looking for creative ways4 to offset this somewhat, but that's another story...   A few misc- * - what version of OpenVMS and patch level?B - what are board p/n's and versions (likely B3002-AB or B3004-AA?), - what type of memory do you have installed?6 - what type of graphics console do you have installed?( - are any errors logged in the errorlog?> - if running a recent version of OpenVMS, post the results of: $ show cpu/full  $ Show cpu/all9 - are you running the latest firmware? If not, check out: 7 ftp://ftp.digital.com/pub/DEC/Alpha/firmware/index.html @ ftp://ftp.digital.com/pub/DEC/Alpha/firmware/readmes/as4x00.html  H Here are some Alpha pointers. Drop me a line directly and I can fwd someD PDF files that may be of assistance (if not immediately, perhaps for later).   1 http://h18002.www1.hp.com/alphaserver/archive/=20 H http://h18002.www1.hp.com/alphaserver/options/as4100/as4100_options.html   Regards   
 Kerry Main Senior Consultant  HP Services Canada Voice: 613-592-4660  Fax: 613-591-4477  kerryDOTmainAThpDOTcom (remove the DOT's and AT)=20  4 OpenVMS - the secure, multi-site OS that just works.   ------------------------------  % Date: Fri, 19 Aug 2005 19:46:30 -0500 . From: Duncan Brown <brown_du@eisner.decus.org>M Subject: Re: AlphaServer 4100 CPU speed upgrade and/or backplane speed ranges 2 Message-ID: <3_SdncpkvNv14JveRVn-ig@speakeasy.net>   Main, Kerry wrote:  J > I also have an Alpha 4100 dual cpu EV56 600Mhz system (VMS V8.2) at homeC > that I hope to bring online shortly. Electrical bills are what is J > holding me back right now, so I will likely be looking for creative ways6 > to offset this somewhat, but that's another story...  F Electrical bills, shmelectrical bills!  I'm ComEd's favorite customer!  
 > A few misc- , > - what version of OpenVMS and patch level?  I Version 7.3-1, no patches, straight off the CD (I think the one from the  C hobbyist program, or maybe I bought a set on ebay...)  That's been   running fine for some time now.   D > - what are board p/n's and versions (likely B3002-AB or B3004-AA?)  D The 400MHz boards that have been running forever are B3004-AA.  The = 533MHz boards are B3005-CA  All 4 are identical in each case.   . > - what type of memory do you have installed?  ( I have 2 512-MB sets, for a total of 1GB  8 > - what type of graphics console do you have installed?  9 The standard cheap card (Number 9?) running into a VRT16.   * > - are any errors logged in the errorlog?  I Ever since analyze/error went away I haven't bothered to dig up whatever  H it is that I need to look at the error log (DECevent?)  There's nothing D on the console and nothing in Operator.log  I just find the machine ? wedged at the console level.  Should I go investigate DECevent?   @ > - if running a recent version of OpenVMS, post the results of: > $ show cpu/full  > $ Show cpu/all  I I assume this is only useful if I do it for the 533MHz boards, which are  I not in there right now?  I did do a SHOW CPU when they were in there and  H it showed them as 5/533 boards, and all up and running properly.  I can ( see that /FULL provides a lot more info.    ; > - are you running the latest firmware? If not, check out: 9 > ftp://ftp.digital.com/pub/DEC/Alpha/firmware/index.html B > ftp://ftp.digital.com/pub/DEC/Alpha/firmware/readmes/as4x00.html  H I think I am.  I never liked the goofy firmware numbering system, but I I know that when I updated the boards after installing them it showed they  F were now at version 5.70-0 of AlphaBIOS... is that the number you are I looking for?  That was the (already ancient) latest firmware update that  E I got when I last upgraded VMS.  Also, at the console level, all the  I 533MHz boards showed an SROM version of 3.0, while my 400 MHz boards are   a mixture of 2.0 and 1.1.    > J > Here are some Alpha pointers. Drop me a line directly and I can fwd someF > PDF files that may be of assistance (if not immediately, perhaps for	 > later).  > 1 > http://h18002.www1.hp.com/alphaserver/archive/  J > http://h18002.www1.hp.com/alphaserver/options/as4100/as4100_options.html  B I had perused there but hadn't seen anything that got down to the  nitty-gritty level I need.   Thanks,  Duncan   ------------------------------  % Date: Sat, 20 Aug 2005 01:35:54 -0400 ' From: Dave Froble <davef@tsoft-inc.com> M Subject: Re: AlphaServer 4100 CPU speed upgrade and/or backplane speed ranges 0 Message-ID: <11gdg2bj6oerg13@corp.supernews.com>   Duncan Brown wrote:  > Main, Kerry wrote: > K >> I also have an Alpha 4100 dual cpu EV56 600Mhz system (VMS V8.2) at home D >> that I hope to bring online shortly. Electrical bills are what isK >> holding me back right now, so I will likely be looking for creative ways 7 >> to offset this somewhat, but that's another story...  >  > H > Electrical bills, shmelectrical bills!  I'm ComEd's favorite customer! >  >> A few misc-- >> - what version of OpenVMS and patch level?  >  > K > Version 7.3-1, no patches, straight off the CD (I think the one from the  E > hobbyist program, or maybe I bought a set on ebay...)  That's been  ! > running fine for some time now.  > E >> - what are board p/n's and versions (likely B3002-AB or B3004-AA?)  >  > F > The 400MHz boards that have been running forever are B3004-AA.  The ? > 533MHz boards are B3005-CA  All 4 are identical in each case.  > / >> - what type of memory do you have installed?  >  > * > I have 2 512-MB sets, for a total of 1GB > 9 >> - what type of graphics console do you have installed?  >  > ; > The standard cheap card (Number 9?) running into a VRT16.  > + >> - are any errors logged in the errorlog?  >  > K > Ever since analyze/error went away I haven't bothered to dig up whatever  J > it is that I need to look at the error log (DECevent?)  There's nothing F > on the console and nothing in Operator.log  I just find the machine A > wedged at the console level.  Should I go investigate DECevent?  > A >> - if running a recent version of OpenVMS, post the results of:  >> $ show cpu/full >> $ Show cpu/all  >  > K > I assume this is only useful if I do it for the 533MHz boards, which are  K > not in there right now?  I did do a SHOW CPU when they were in there and  J > it showed them as 5/533 boards, and all up and running properly.  I can * > see that /FULL provides a lot more info. >  > < >> - are you running the latest firmware? If not, check out:: >> ftp://ftp.digital.com/pub/DEC/Alpha/firmware/index.htmlC >> ftp://ftp.digital.com/pub/DEC/Alpha/firmware/readmes/as4x00.html  >  > J > I think I am.  I never liked the goofy firmware numbering system, but I K > know that when I updated the boards after installing them it showed they  H > were now at version 5.70-0 of AlphaBIOS... is that the number you are K > looking for?  That was the (already ancient) latest firmware update that  G > I got when I last upgraded VMS.  Also, at the console level, all the  K > 533MHz boards showed an SROM version of 3.0, while my 400 MHz boards are   > a mixture of 2.0 and 1.1.  >  >>K >> Here are some Alpha pointers. Drop me a line directly and I can fwd some G >> PDF files that may be of assistance (if not immediately, perhaps for 
 >> later). >>2 >> http://h18002.www1.hp.com/alphaserver/archive/ K >> http://h18002.www1.hp.com/alphaserver/options/as4100/as4100_options.html  >  > D > I had perused there but hadn't seen anything that got down to the  > nitty-gritty level I need. > 	 > Thanks,  > Duncan  H Like you said, if you're going to run a system, might as well make it a  good one.  :-)  E This is a wild ass guess.  I'm not up to speed on some of the larger  C systems.  Would the power supplies be an issue?  Might the 533 MHz  9 boards need more/different power than the 400 MHz boards?    --  4 David Froble                       Tel: 724-529-04504 Dave Froble Enterprises, Inc.      Fax: 724-529-0596> DFE Ultralights, Inc.              E-Mail: davef@tsoft-inc.com 170 Grimplin Road  Vanderbilt, PA  15486    ------------------------------  % Date: Fri, 19 Aug 2005 14:13:37 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> C Subject: Re: How many files can you have in a VMS directory without , Message-ID: <4306214E.D9DF177C@teksavvy.com>  
 AEF wrote:@ > Well, regardless of all that, won't this only affect the firstE > invocation of DELETE.EXE? I mean, you have a script that invokes it E > repeatedly, each invocation right after the other. Wouldn't that be G > enough to keep it in physical memory, thereby giving only soft faults   > from the second invocation on?    G Getting the code into memory mapped to your process is one thing. But I ? suspect that image activation involves far more than just that.    ------------------------------    Date: 19 Aug 2005 13:23:46 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) C Subject: Re: How many files can you have in a VMS directory without 3 Message-ID: <95x4AQTwdUbg@eisner.encompasserve.org>   T In article <CZVD9ZIN1uGh@eisner.encompasserve.org>, briggs@encompasserve.org writes: > J > Is that summary reasonably correct or has my internals knowledge decayed > too badly? >   E    That was a good summary.  Since its installed /header_resident the G    image header istelf should be in RAM.  On an Alpha INSTALL does have G    options which will keep some code pages in RAM, but they're not used H    for DELETE.  And DELETE is so tiny that very few systems can be found$    with such a small free page list.   ------------------------------  % Date: Fri, 19 Aug 2005 14:16:49 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com>   Subject: Re: Image restore fails, Message-ID: <4306220E.BCEEB003@teksavvy.com>   norm.raphael@metso.com wrote: D > I thought the primary idea of backward-compatibility would be most/ > inviolate in a backup restore implementation.     F It was a bug in backup. Hence the patch to fix the 6.x backup version.A BACKUP 7.2 has no problem with earlier version of image backups.    H So it isn't a sign of bad design. Just a sign of poor quality control in the 6.x timeframe.   ------------------------------  % Date: Fri, 19 Aug 2005 14:37:52 -0400  From: norm.raphael@metso.com  Subject: Re: Image restore failsQ Message-ID: <OF33808C35.D06A1FD2-ON85257062.006614C7-85257062.00665806@metso.com>   H JF Mezei <jfmezei.spamnot@teksavvy.com> wrote on 08/19/2005 02:16:49 PM:   > norm.raphael@metso.com wrote: F > > I thought the primary idea of backward-compatibility would be most1 > > inviolate in a backup restore implementation.  >  > H > It was a bug in backup. Hence the patch to fix the 6.x backup version.B > BACKUP 7.2 has no problem with earlier version of image backups. > J > So it isn't a sign of bad design. Just a sign of poor quality control in > the 6.x timeframe.   Okay.   H And as was alluded to earlier in the thread, the Installation-CD versionF would not work and cannot readily be patched, so when backup is fixed,I ECO'd, etc.  then a new OpenVMS installation CD should be issued as well, H yes?  I also hate it when there is bad or erroneous documentation on theE distribution CD because I cannot fix it (short of burning a new one).    ------------------------------    Date: 19 Aug 2005 12:03:40 -0700( From: John.Martin_At_Home@BTInternet.com' Subject: Re: microVax3100-40 cpu number C Message-ID: <1124478220.382847.264290@o13g2000cwo.googlegroups.com>   ) John.Martin_At_Home@BTInternet.com wrote:  > A simple question ? A > A need my microVAX3100-40 cpu number in order to get my openVMS  > licences. D > How do you find it out? - Is it the multidigit number separated by1 > hyphens ever 2 digits displayed by the console?   E Many thanks to all clarification. For Cpu number read Serial Number -  got it!    ------------------------------  % Date: Sat, 20 Aug 2005 02:02:01 +0200 - From: Alex van Denzel <vandenzel@hotmail.com> ' Subject: Re: microVax3100-40 cpu number 7 Message-ID: <430672f9$0$83709$dbd4b001@news.wanadoo.nl>   ) John.Martin_At_Home@BTInternet.com wrote:  > A simple question ? A > A need my microVAX3100-40 cpu number in order to get my openVMS  > licences. D > How do you find it out? - Is it the multidigit number separated by1 > hyphens ever 2 digits displayed by the console?  >   @ The nn-nn-nn-nn-nn-nn number (where nn is a hexadecimal number) 7 displayed on the console would be the ethernet addresss    -- Alex.    ------------------------------    Date: 19 Aug 2005 19:24:26 -07001 From: nothome@spammers.are.scum (Malcolm Dunnett) 7 Subject: OSU Webserver and NETSERVER.LOG files question , Message-ID: <1XYH5SKTBGfV@malvm9.mala.bc.ca>  > I'm running OSU webserver 3.10 on VMS 7.3-2. I have a question> about the creation of NETSERVER.LOG files. It appears that the> "permanent" DECNET server processes create a new netserver.log; file about every 5 minutes or so. This creates a whole mess @ of netserver.log files in the home directory. Is there a logical; or something that controls this ( eg can I get them to just / create a new logfile once a day or something? )    ------------------------------  % Date: Fri, 19 Aug 2005 23:23:26 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> ; Subject: Re: OSU Webserver and NETSERVER.LOG files question , Message-ID: <4306A218.23EDF8C4@teksavvy.com>   Malcolm Dunnett wrote:@ > about the creation of NETSERVER.LOG files. It appears that the@ > "permanent" DECNET server processes create a new netserver.log= > file about every 5 minutes or so. This creates a whole mess / > of netserver.log files in the home directory.     0 try DEFINE/SYSTEM NETSERVER$TIMEOUT "delta time"  C When your OSU script has finished, control returns to netserver.exe D which theh check the logical to see how long it should stick around,G waiting for another decnet request to arrive. The default is 5 minutes.    ------------------------------  # Date: Sat, 20 Aug 2005 03:19:58 GMT L From: winston@SSRL.SLAC.STANFORD.EDU (Alan Winston - SSRL Central Computing); Subject: Re: OSU Webserver and NETSERVER.LOG files question 6 Message-ID: <00A48883.D58F97D6@SSRL.SLAC.STANFORD.EDU>  ` In article <1XYH5SKTBGfV@malvm9.mala.bc.ca>, nothome@spammers.are.scum (Malcolm Dunnett) writes: > ? >I'm running OSU webserver 3.10 on VMS 7.3-2. I have a question ? >about the creation of NETSERVER.LOG files. It appears that the ? >"permanent" DECNET server processes create a new netserver.log < >file about every 5 minutes or so. This creates a whole messA >of netserver.log files in the home directory. Is there a logical < >or something that controls this ( eg can I get them to just0 >create a new logfile once a day or something? )  D If I understand what's going on correctly, rather than the existing E DECnet server processes creating a new log every five minutes, what's H happening is that unused server processes are timing out and going away,6 and new processes are spinning up, with new log files.  ; You could increase the lifetime of unused server processes.    -- Alan    ------------------------------    Date: 19 Aug 2005 20:36:58 -07001 From: nothome@spammers.are.scum (Malcolm Dunnett) ; Subject: Re: OSU Webserver and NETSERVER.LOG files question , Message-ID: <TWGfOEXvbE$N@malvm9.mala.bc.ca>  6 In article <00A48883.D58F97D6@SSRL.SLAC.STANFORD.EDU>,R     winston@SSRL.SLAC.STANFORD.EDU (Alan Winston - SSRL Central Computing) writes: > F > If I understand what's going on correctly, rather than the existing G > DECnet server processes creating a new log every five minutes, what's J > happening is that unused server processes are timing out and going away,8 > and new processes are spinning up, with new log files. >   O    That doesn't appear to be what's happening, for example sho dev/files shows:   > SERVER_00B0     22402159  [HTTP_SERVER_3-10A]NETSERVER.LOG;139  &  SHO PROC/ACCO for that process shows:  G 19-AUG-2005 20:29:37.78   User: WWW_SERVER       Process ID:   22402159 L                           Node: MALVM9           Process name: "SERVER_00B0"   Accounting information: A  Buffered I/O count:  12867288  Peak working set size:      45072 A  Direct I/O count:     8763460  Peak virtual size:         251296 A  Page faults:         25355191  Mounted volumes:                0   Images activated:      201596)  Elapsed CPU time:          0 02:37:55.66 )  Connect time:              3 12:08:09.95     Soft CPU Affinity: off  $  a directory of that log file shows:  > NETSERVER.LOG;139             File ID:  (61602,36,0)          8 Size:           91/160        Owner:    [WWW,APACHE$WWW]# Created:    19-AUG-2005 20:10:05.85 ' Revised:    19-AUG-2005 20:10:05.85 (0)  Expires:    <None specified>  Backup:     <No backup recorded> Effective:  <None specified> Recording:  <None specified> Accessed:   <None specified> Attributes: <None specified> Modified:   <None specified>
 Linkcount:  1  File organization:  Sequential Shelved state:      Online    Caching attribute:  WritethroughX File attributes:    Allocation: 160, Extend: 0, Global buffer count: 0, No version limitI Record format:      VFC, 2 byte header, maximum 0 bytes, longest 90 bytes / Record attributes:  Print file carriage control  RMS attributes:     None Journaling enabled: None; File protection:    System:RWED, Owner:RWED, Group:, World:  Access Cntrl List:  None Client attributes:  None    L  So I appear to have a server process which has been around for 3.5 days butI just created a new NETSERVER.LOG file 20 minutes ago. The timing seems to K vary, could it be that it creates a new log file after it's been called "X" ? number of times? If so, can I define a different value for "X"?   = > You could increase the lifetime of unused server processes.  >   >   I have the NETSERVER$SERVERS_WWW_SERVER defined as 10, so if> I understand correctly the first 10 servers processes that get; created should never go away. The fact that this particular = process has been around for over 3 days seems to support that  theory.    ------------------------------  % Date: Sat, 20 Aug 2005 00:39:02 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> ; Subject: Re: OSU Webserver and NETSERVER.LOG files question , Message-ID: <4306B3CC.F7D94D14@teksavvy.com>   One thing you may wish to do:   F When you see an apparantly iddle netserver SERVER_xxxx process runningR under the OSU username, doing show proc/cont should show it running netserver.exe.  E If it isn't running it, it would mean that it is still executing your > script and thus not ready to receive a new connection (so next4 connection would result in a new one being created).   ------------------------------    Date: 19 Aug 2005 22:35:55 -07001 From: nothome@spammers.are.scum (Malcolm Dunnett) ; Subject: Re: OSU Webserver and NETSERVER.LOG files question , Message-ID: <6GSg7sFIDAQ2@malvm9.mala.bc.ca>  - In article <4306B3CC.F7D94D14@teksavvy.com>,  3     JF Mezei <jfmezei.spamnot@teksavvy.com> writes:  > One thing you may wish to do:  > H > When you see an apparantly iddle netserver SERVER_xxxx process runningT > under the OSU username, doing show proc/cont should show it running netserver.exe. > G > If it isn't running it, it would mean that it is still executing your @ > script and thus not ready to receive a new connection (so next6 > connection would result in a new one being created).      It is running NETSERVER.EXE  :    I looked at a bunch of the log files and one consistant< thing is that they each have served 25 connect requests - so> it appears there's a setting somewhere that tells NETSERVER toB open a new log file after serving each 25th request - the question9 remains whether or not I can modify that setting somehow.    ------------------------------  % Date: Fri, 19 Aug 2005 13:44:33 -0400 $ From: "PEN" <paul.nuneznosp@mhp.com>1 Subject: Re: Packet Trace Facility in DEC TCP/IP? , Message-ID: <de55q4$brb$1@hplms2.hpl.hp.com>  	 Hi Chris,   K "Christopher Story" <ke6rwj@spam-eater-remove-me-msn.com> wrote in message  " news:snlNe.613$cO6.251@fe04.lga...C > Is there a packet trace facility in DEC TCP/IP like the Multinet   > "TCPDUMP"? >  > Thanks >  > Chris  >  >   L Prior to v5.4 there was only TCPTRACE which produces text output of capture  files (/output).  L As of 5.4 and later, there's also a tcpdump utility which is much more akin ! to other tcpdump implementations.    TCPTRACE is a DCL command.  2 tcpdump is not a dcl command.  But if you execute:  $ $ @sys$startup:tcpip$define_commands  + a foreign command/symbol is defined for it.   @ See $ help tcpdump for more info on what switches are available.  J Capture files from both utilities can be viewed using Ethereal.  However, I concerning the text files created by TCPTRACE, if you ever get a message  H from Ethereal indicating it can't read the file (can't recall the exact J message), edit the capture file using your favorite VMS editor and remove H the lines at the beginning which begin  TCPIPTRACE-W-BUFFERSFULL (don't  remove the blank lines).   HTH,   Paul   ------------------------------   Date: 19 Aug 2005 17:46:57 GMT$ From: "Doc." <Doc@openvms-rocks.com>& Subject: Re: The SAMBA Saga at an end!1 Message-ID: <Xns96B7C7C4D4E71Doc@212.100.160.126>   K On Fri, 19 Aug 2005 16:17:07 GMT, bill@cs.uofs.edu (Bill Gunshannon) wrote  1 message news:3mmf03F178rj5U1@individual.net . . .   $ > Thanks to everyone for their help.  K I, and I'm sure others here, are glad to see you've got your stuff working.   G Now, whilst you're probably the first to admit you don't use VMS often  = enough to be adept with it, was it really that bad to set up?   K I do hope you get your potential license problems sorted, for one I'd like  L to think you can present the VMS system to students as something that "just  works".      Doc. --  F OpenVMS:    Eight out of ten hackers prefer *other* operating systems.F             http://vmsbox.cjb.net          http://vmsbox.cjb.net/docs/   ------------------------------   Date: 19 Aug 2005 18:19:07 GMT( From: bill@cs.uofs.edu (Bill Gunshannon)& Subject: Re: The SAMBA Saga at an end!, Message-ID: <3mmm4rF17g602U1@individual.net>  1 In article <Xns96B7C7C4D4E71Doc@212.100.160.126>, ' 	"Doc." <Doc@openvms-rocks.com> writes: M > On Fri, 19 Aug 2005 16:17:07 GMT, bill@cs.uofs.edu (Bill Gunshannon) wrote  3 > message news:3mmf03F178rj5U1@individual.net . . .  > % >> Thanks to everyone for their help.  > M > I, and I'm sure others here, are glad to see you've got your stuff working.  > I > Now, whilst you're probably the first to admit you don't use VMS often  ? > enough to be adept with it, was it really that bad to set up?   F Yes.  Forgetting all the stuff I didn't know, the amount of stuff thatG just didn't work was more than enough to make most of the people I know C just give up.   Actually, to be fair, VMS is no problem setting up. F After trashing the system with one of the early SAMBA versions I triedC I was able to sort out my CDROM Hardware problem and re-install VMS B in just a couple of hours (including all the compilers I put on inE case I need them :-)  UNZIP, VIM and SAMBA, now that is another thing  altogether!    > M > I do hope you get your potential license problems sorted, for one I'd like  N > to think you can present the VMS system to students as something that "just 	 > works".   D Nice thought, but no more true with VMS than with anythng else.  AllE depends on how good a job was done on whatever piece your working on  A at any given point in time.  For the most part, I expect what the C students will be doing to work just fine.  Mostly COBOL, DECWindows " and maybe a little playing around.   bill   --  J Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>       ------------------------------    Date: 19 Aug 2005 16:21:48 -0700 From: bob@instantwhip.com * Subject: Re: What is the official status ?B Message-ID: <1124493708.803980.44800@g43g2000cwa.googlegroups.com>  G I just got mail that said 5.7 is due out soon and will be itanium ready  ...    ------------------------------    Date: 19 Aug 2005 16:24:38 -0700 From: bob@instantwhip.com * Subject: Re: What is the official status ?B Message-ID: <1124493878.437520.31300@o13g2000cwo.googlegroups.com>  , here it is in their summer update letter ...  6 http://www.process.com/tcpip/2005Newsletters/10-2.html   ------------------------------    Date: 19 Aug 2005 22:10:39 +01006 From: peter@langstoeger.at (Peter 'EPLAN' LANGSTOEGER)Y Subject: Re: [TCPware V5.6-2,KERBEROS V2.1-72,VMS V7.3-2] TCPIP$IPC_SHR         SHRIDMISM , Message-ID: <430658df$1@news.langstoeger.at>  i In article <6.1.2.0.2.20050819133657.023ebc48@raptor.psccos.com>, Dan O'Reilly <dano@process.com> writes:  >Peter - > L >At this time, you can't use Kerberos V2.1-72 on a TCPware system, so you'llH >have to regress back to the previous version (2.0-6).  We're working onN >getting an ECO together to allow this, but it will be at least a week or two, >if not longer.   K Ok. In the meantime I found out that I already had this problem in Feb 2005 H and solved it by stepping back to V2.0-6 in Mar 2005 (and shame on me, II forgot it - over my holidays - but could now remember after I checked the > PCSI history and the installation logfiles on all my systems).  H Since then at least 2 SSH ECOs came out, but obviously didn't solve this problem.   So, thanks for helping.  But two questions remain: " * Will this new fix be a SSH ECO ?H * What is the reason for the incompatibility ? (What part of KERBEROS is used by TCPware and why)   --   Peter "EPLAN" LANGSTOEGER % Network and OpenVMS system specialist  E-mail  peter@langstoeger.atF A-1030 VIENNA  AUSTRIA              I'm not a pessimist, I'm a realist   ------------------------------  % Date: Fri, 19 Aug 2005 16:21:37 -0600 % From: Dan O'Reilly <dano@process.com> T Subject: Re: [TCPware V5.6-2,KERBEROS V2.1-72,VMS V7.3-2] TCPIP$IPC_SHR  SHRIDMISMATA Message-ID: <6.1.2.0.2.20050819162046.022eaaf8@raptor.psccos.com>   7 At 03:10 PM 8/19/2005, Peter 'EPLAN' LANGSTOEGER wrote: G >In article <6.1.2.0.2.20050819133657.023ebc48@raptor.psccos.com>, Dan  $ >O'Reilly <dano@process.com> writes:
 > >Peter - > > N > >At this time, you can't use Kerberos V2.1-72 on a TCPware system, so you'llJ > >have to regress back to the previous version (2.0-6).  We're working onL > >getting an ECO together to allow this, but it will be at least a week or  > two, > >if not longer.  > L >Ok. In the meantime I found out that I already had this problem in Feb 2005I >and solved it by stepping back to V2.0-6 in Mar 2005 (and shame on me, I J >forgot it - over my holidays - but could now remember after I checked the? >PCSI history and the installation logfiles on all my systems).  > I >Since then at least 2 SSH ECOs came out, but obviously didn't solve this 	 >problem.  >  >So, thanks for helping. >But two questions remain:# >* Will this new fix be a SSH ECO ? I >* What is the reason for the incompatibility ? (What part of KERBEROS is  >used by TCPware and why)   J The incompatibility is in the UCX$IPC_SHR module, so at least that will beI part of the ECO.  We're not sure that SSH will need an ECO at this point, 6 but we're doing further examination to determine that.   ------J +-------------------------------+----------------------------------------+J | Dan O'Reilly                  |  "There are 10 types of people in this |J | Principal Engineer            |   world: those who understand binary   |J | Process Software              |   and those who don't."                |J | http://www.process.com        |                                        |J +-------------------------------+----------------------------------------+   ------------------------------  # Date: Fri, 19 Aug 2005 19:41:02 GMT , From: Wayne Morrison <Wayne.Morrison@hp.com>S Subject: Re: [TCPware V5.6-2,KERBEROS V2.1-72,VMS V7.3-2] TCPIP$IPC_SHR SHRIDMISMAT 3 Message-ID: <iBqNe.10660$ZG3.3629@news.cpqcorp.net>     Peter 'EPLAN' LANGSTOEGER wrote:  L > Since upgrading KERBEROS (you know, you should keep your security productsK > always as current as possible ;-) from V2.0-6 (comes with OpenVMS V7.3-2) M > to V2.1-72 (downloaded from HPQ) my TCPware SSH client&server unfortunately E > is no longer working (no problem on my TCPIP [V5.4 ECO 5] systems).  > M > Please note, that I've KERBEROS only installed, but not used [explicitely].  > B > Error is, SSH server simply doesn't work (hangs forever) and the% > SSH client shows the likely cause:   > 7 > %DCL-W-ACTIMAGE, error activating image TCPIP$IPC_SHR K > -CLI-E-IMGNAME, image file DSA0:[SYS0.SYSCOMMON.][TCPWARE]UCX$IPC_SHR.EXE < > -SYSTEM-F-SHRIDMISMAT, ident mismatch with shareable image >  > $ AN/IM TCPIP$IPC_SHR shows  > + >                 image name: "UCX$IPC_SHR" > >                 image file identification: "5.6-2 PATCH 4.0"5 >                 image file build identification: "" 9 >                 link date/time:  8-APR-2005 22:14:42.41 1 >                 linker identification: "A13-02"  > L > which is quite right as I installed some TCPware and VMS ECOs at this day.I > (OTOH, I installed SSH ECO V6 yesterday and this .EXE didn't change !?)  > K > If nobody has an explaination and hopefully a (fast) fix for this, I will I > need to downgrade KERBEROS (from V2.1-72 to V2.0-6) again (which I hate I > to do because KERBEROS is referenced by the plaform product OPENVMS and @ > therefor can't be that easily uninstalled and installed again) > 
 > Many TIA  F The problem is unlikely to be the Kerberos upgrade.  Kerberos doesn't G include UCX$IPC_SHR (or any other TCP/IP images) in the kit.  Kerberos  I is linked against TCP/IP sharables, but it doesn't ship them on the kit.  I   Also, there were only minor differences between Kerberos V2.0 and V2.1  F - mostly bug fixes.  The new version we're working on now will change D the base from MIT 1.2.x to 1.4.x, and will be a significant upgrade.  E I'm told by our TCP/IP team that Process Software ships a version of  H UCX$IPC_SHR.EXE with TCPware for compatibility purposes.  Since you say 1 you installed a TCPware ECO, I'd start with them.    	Wayne Morrison, CISSP$ 	Kerberos for OpenVMS Project Leader   ------------------------------    Date: 19 Aug 2005 22:42:06 +01006 From: peter@langstoeger.at (Peter 'EPLAN' LANGSTOEGER)S Subject: Re: [TCPware V5.6-2,KERBEROS V2.1-72,VMS V7.3-2] TCPIP$IPC_SHR SHRIDMISMAT , Message-ID: <4306603e$1@news.langstoeger.at>  b In article <iBqNe.10660$ZG3.3629@news.cpqcorp.net>, Wayne Morrison <Wayne.Morrison@hp.com> writes:4 >The problem is unlikely to be the Kerberos upgrade.  1 Sorry to disappoint you. But this is proofed now. B I had the problem already in Feb 2005 and solved it by downgrading KERBEROS to V2.0-6 again.  And PSC also acked it today.  H >                                                      Kerberos doesn't = >include UCX$IPC_SHR (or any other TCP/IP images) in the kit.   ? Of course. But TCPware delivers UCX$IPC_SHR and now there is an @ incompatibily with something TCPware uses and KERBEROS upgraded.  J >                                                                Kerberos J >is linked against TCP/IP sharables, but it doesn't ship them on the kit.   J Yup. I could understand if KERBERIZED Telnet - which I don't use or care -H or so does no longer work until PSC upgrades UCX$IPC_SHR as well - whichJ will take a couple of weeks - but I can't understand why TCPware's own SSHJ does no longer work. What part of KERBEROS is TCPware's SSH using and why.  J >  Also, there were only minor differences between Kerberos V2.0 and V2.1 G >- mostly bug fixes.  The new version we're working on now will change  E >the base from MIT 1.2.x to 1.4.x, and will be a significant upgrade.   I But obviously, an (shareable) image version number was increased, right ?   K And hopefully your next KERBEROS version is not a requirement for something 5 (like VMS V8.3) until PSC supports it with TCPware...   F >I'm told by our TCP/IP team that Process Software ships a version of I >UCX$IPC_SHR.EXE with TCPware for compatibility purposes.  Since you say  2 >you installed a TCPware ECO, I'd start with them.  I The problem is not the recent TCPware SSH ECO. It is a problem because so I far no TCPware ECO fixes the problem with UCX$IPC_SHR & KERBEROS V2.1-72.    --   Peter "EPLAN" LANGSTOEGER % Network and OpenVMS system specialist  E-mail  peter@langstoeger.atF A-1030 VIENNA  AUSTRIA              I'm not a pessimist, I'm a realist   ------------------------------   End of INFO-VAX 2005.463 ************************