1 INFO-VAX	Mon, 24 May 2004	Volume 2004 : Issue 287       Contents: Re: 83x08 Re: CDE - redirecting complete desktop to another system Re: CSWS 2.0 Source Code?  Re: CSWS 2.0 Source Code?  Re: Dave Cutler and VMS  Re: Dave Cutler and VMS  RE: Dave Cutler and VMS  Re: Dave Cutler and VMS  Re: Dave Cutler and VMS 0 DSNLINK ECO database appears to be being updated Re: OpenVMs Cluster  Re: OpenVMs Cluster  Re: OpenVMs Cluster  Paypal Scams... 0 Previous Post - Upgrading Cluster Communications4 Re: Previous Post - Upgrading Cluster Communications- Re: Puzzling LIB$SPAWN problem with DEC BASIC - Re: Puzzling LIB$SPAWN problem with DEC BASIC - Re: Puzzling LIB$SPAWN problem with DEC BASIC  Re: Request for new SMTP SDLT 320 Tape Drive Issues Sortwork file query  Re: Sortwork file query  Re: Sortwork file query  Re: Sortwork file query / Upgrading Cluster from FDDI to Gigabit Ethernet 3 Re: Upgrading Cluster from FDDI to Gigabit Ethernet 3 Re: Upgrading Cluster from FDDI to Gigabit Ethernet 3 Re: Upgrading Cluster from FDDI to Gigabit Ethernet 3 Re: Upgrading Cluster from FDDI to Gigabit Ethernet % Re: Vaxstation 3100 and graphics card % Re: Vaxstation 3100 and graphics card ) Re: You'll never guess what HP advertised ) Re: You'll never guess what HP advertised   F ----------------------------------------------------------------------  % Date: Mon, 24 May 2004 11:37:14 -0500 / From: Chris Scheers <chris@applied-synergy.com>  Subject: Re: 83x0 3 Message-ID: <40B224BA.6CCD92A5@applied-synergy.com>    "Roert G. Schaffrath" wrote:  > G > I had always wanted to try taking two CPU's from a spare 8350 and put E > them into another 8350 for a total of four CPU's and see what would H > happen.  Unfortunately, when I had access to that many 82xx CPU cards,H > VMS V5.0 was not available.  By the time VMS V5.0 came out, I was at aG > different job and did not have to ability to perform that experiment. H > Performance would have been lousy on the BI bus but it would have been > an interesting machine.   G The early documentation for SMP support in VMS referenced two machines: 2 an 8800 with two CPUs and an 8370 with seven CPUs.  @ It appears that any 8300 or 8350 with three or more CPUs will be identified by VMS as an 8370.   G ----------------------------------------------------------------------- $ Chris Scheers, Applied Synergy, Inc.  C Voice: 817-237-3360            Internet: chris@applied-synergy.com     Fax: 817-237-3074    ------------------------------    Date: 24 May 2004 01:19:21 -0700. From: martinkirby12@yahoo.co.uk (Martin Kirby)A Subject: Re: CDE - redirecting complete desktop to another system < Message-ID: <224291b.0405240019.4b157eba@posting.google.com>  \ Paul Sture <nospam@sture.homeip.net> wrote in message news:<2h6b1iF9inkgU1@uni-berlin.de>...  H > It's been many years since I did this, but I'm trying to create a CDE + > desktop on another system (running OS X).  >  > ...  > # > However, when I do the following:  > 2 > $ set display /create /transport=tcpip /node=mac1 > $ @sys$common:[cde$defaults.system.bin]xsession  >    Try:  2 $set display /create/transport=tcpip/node=mac/exec $run sys$system:decw$startlogin   F The exec mode is needed so the ws device remains after logging out for use by the login process.   C You need to make sure the display server has allowed foreign window 	 managers.   B You need to have the APPS side of DECwindows started on the clientE system which will happen if it has a graphics card. Not sure what you 9 can do if there is no graphics card in the client system.   F DECW$STARTLOGIN has a slightly odd way of finding the display it looks for:  + Job logical DECW$LOGIN_DISPLAY (since V1.3)  Job logical DECW$DISPLAY Process logical DECW$DISPLAY  F As setting the display creates a process logical if you do this from aE window that already has a job logical then you need to override that, 9 or use DECW$LOGIN_DISPLAY if on DECwindows V1.3 or later.   F If the display server has access control enabled and the client systemE is in the controlling host list then, as noted in the discussion with B XDM, V1.3 and V1.3-1 have a security problem that means it doesn't work.    Martin Kirby   ------------------------------  % Date: Mon, 24 May 2004 07:29:24 -0700 - From: Kaleb Pederson <kpederson@mail.ewu.edu> " Subject: Re: CSWS 2.0 Source Code?7 Message-ID: <200405240729.24547.kpederson@mail.ewu.edu>   @ On Saturday 22 May 2004 6:04 am, david20@alpha2.mdx.ac.uk wrote: [snip]I > Exactly why should any webserver including Apache really care about how  > long a file really is ?   M RFC 2616 Section 14.13  (http 1.1) defines a content-length header that must  J send the total length of the document.  In other words, to be a compliant L webserver, it must.  There are a number of other uses and, IIRC, the length ( must be defined in other places as well.   --Kaleb    ------------------------------  # Date: Mon, 24 May 2004 15:49:29 GMT ) From: "Rick Barry" <richard.barry@hp.com> " Subject: Re: CSWS 2.0 Source Code?0 Message-ID: <dQosc.2140$gn.553@news.cpqcorp.net>  G The accurate determination of file size is just one issue. Meta-data in J record formats such as variable-length cause problems when Apache 2.0 mapsJ the file in memory for performance reasons. The meta-data get read as dataE which, of course, is wrong.  We either figure out a way to filter the , meta-data or not map certain record formats.  
 Rick Barry OpenVMS System Software Group  Hewlett Packard Company 
 Nashua, NH  + <david20@alpha2.mdx.ac.uk> wrote in message # news:c8nj5f$phh$1@news.mdx.ac.uk... 5 > In article <Tg4THFxrP1uz@eisner.encompasserve.org>, D clubley@remove_me.eisner.decus.org-Earth.UFP (Simon Clubley) writes:8 > >In article <2h6usbF9geo5U1@uni-berlin.de>, Paul Sture! <nospam@sture.homeip.net> writes:  > >>L > >> Basic question here, which I've been meaning to ask since this question > >> arose.  > >>I > >> Doesn't Windows use CRLF instead of LF line terminators? Apache does  run B > >> on Windows too, even though it might not be mainstream there. > >> > > K > >That's not the real problem. The problem is that neither Windows or Unix L > >filesystems have files with filesystem internal data within them; unlike,I > >for example, RMS files in variable length record format, which contain  record" > >length information within them. > > J > >This means, for example, that when you ask the filesystem how long is a file, L > >and you get a reply of 1234 bytes, you don't know how much of this length isL > >user data and how much is internal RMS data, unless you use a file format that5 > >doesn't have internal RMS data, such as Stream-LF.  > > I > Exactly why should any webserver including Apache really care about how  long a > file really is ?L > I believe IMAP and possibly SAMBA have requirements to be able to tell theF > clients how large files are but they are perfectly happy with a best guess.I > Other webservers (including CSWS 1.3) are able to deal with this issue. I > Indeed with CGI generated data the webserver doesn't know how much data  there 9 > is until it has finished generating and tansmitting it.  > J > >Unless the Apache port is explictly looking for Stream-LF format files, thenJ > >Stream format files should also work. Does anyone know if they actually do ? > > B > I believe it is just looking for stream-LF but I could be wrong. > J > If it is just looking for stream-lf and this philosophy of just catering for L > the filetypes of the system where the code was first developed spread then you J > could end up with a future where all the files you want to webserve must beF > stream-lf but any you also want to access via a program developed on8 > windows (including pathworks) would have to be stream. > 3 > (Is windows actually stream or is it stream-cr ?)  >  >  >  > David Webb > VMS and Unix team leader > CCSS > Middlesex University > 	 > >Simon.  > >  > >-- > > >Simon Clubley, clubley@remove_me.eisner.decus.org-Earth.UFPI > >SCO: Proudly pushing Microsoft down to #2 on the list of most disliked 	 companies    ------------------------------  % Date: Mon, 24 May 2004 10:39:44 -0400 + From: Steve Lionel <Steve.Lionel@intel.com>   Subject: Re: Dave Cutler and VMS8 Message-ID: <4624b0hbg8hav4nb77p1mb78ja5f035n6j@4ax.com>  I On Sun, 23 May 2004 16:47:30 GMT, John Reagan <john.reagan@hp.com> wrote:   P >> I should know this. But it's been so long.  I can't remember if VAXELN PascalQ >> was a port of VAX Pascal or if it was a separate implementation.  I'm inclined I >> to say it was a port because I know it used a port of my VAX Pascal V2 5 >> run-time library.  But I can't be certain of this.  > J >No VAXELN Pascal was not a port of VAX Pascal.  Cutler and McLaren would I >probably rather eat glass.  Like all the other Cutler compilers, VAXELN   >Pascal was VCG-based.  J Indeed,  That;'s why I was somewhat unsure.  They did adopt the VAX Pascal RTL, though.   Steve    ------------------------------  % Date: Mon, 24 May 2004 10:50:41 -0400 + From: Steve Lionel <Steve.Lionel@intel.com>   Subject: Re: Dave Cutler and VMS8 Message-ID: <j924b0pqts9st956n8d958ua4anj801u9k@4ax.com>  I On 22 May 2004 07:21:13 GMT, "Dave Weatherall" <djw-nothere@nospam.nohow>  wrote:  : >> I was the project leader for VAXELN Ada for five years. > B >That must make for some interesting discussions. Fortran v Ada v , >Pascal, or maybe its been done to death :-)  L I'm not interested in such comparisons. I like all three languages, each hasJ its strengths and weaknesses.  I do have a particular fondness for Ada andN wish it had been more popular.  It was refreshing to work with a language thatM was actually "designed" rather than "thrown together". I was fortunate enough M to spend a week at Ada designer Jean Ichbiah's company Alsys in France, while * they did an early test of VAX Ada in 1984.  N Fortran 90 tried to copy some of the elegance of Ada, but it didn't quite come out right... Oh well...   H My philosophy is that you should use the language you know which is mostI suitable for the problem.  I've written "system" type applications in all E three languages, which some people expressed surprise at - especially J regarding Fortran, which has an undeserved reputation of unsuitability for such things.  = Don't get me started on C or C++, though... I hate them both.    Steve    ------------------------------  % Date: Mon, 24 May 2004 08:03:20 -0700 # From: "Tom Linden" <tom@kednos.com>   Subject: RE: Dave Cutler and VMS9 Message-ID: <NDEMLKKEBOIFBMJLCECIOELLDEAA.tom@kednos.com>      -----Original Message-----4   From: Steve Lionel [mailto:Steve.Lionel@intel.com]$   Sent: Monday, May 24, 2004 7:51 AM   To: Info-VAX@Mvb.Saic.Com "   Subject: Re: Dave Cutler and VMS    K   On 22 May 2004 07:21:13 GMT, "Dave Weatherall" <djw-nothere@nospam.nohow>    wrote:  <   >> I was the project leader for VAXELN Ada for five years.   > C   >That must make for some interesting discussions. Fortran v Ada v .   >Pascal, or maybe its been done to death :-)  :   I'm not interested in such comparisons. I like all three   languages, each has L   its strengths and weaknesses.  I do have a particular fondness for Ada andB   wish it had been more popular.  It was refreshing to work with a   language that >   was actually "designed" rather than "thrown together". I was   fortunate enoughA   to spend a week at Ada designer Jean Ichbiah's company Alsys in    France, while ,   they did an early test of VAX Ada in 1984.  >   Fortran 90 tried to copy some of the elegance of Ada, but it   didn't quite come    out right... Oh well...   J   My philosophy is that you should use the language you know which is mostK   suitable for the problem.  I've written "system" type applications in all G   three languages, which some people expressed surprise at - especially L   regarding Fortran, which has an undeserved reputation of unsuitability for   such things.  H Most of Primos was written in Fortran, though in later years they used a version of PL/I   ?   Don't get me started on C or C++, though... I hate them both.    Ditto.     Steve      --- (   Incoming mail is certified Virus Free.<   Checked by AVG anti-virus system (http://www.grisoft.com).A   Version: 6.0.680 / Virus Database: 442 - Release Date: 5/9/2004    --- & Outgoing mail is certified Virus Free.: Checked by AVG anti-virus system (http://www.grisoft.com).? Version: 6.0.680 / Virus Database: 442 - Release Date: 5/9/2004    ------------------------------  # Date: Mon, 24 May 2004 15:45:44 GMT - From: JF Mezei <jfmezei.spamnot@teksavvy.com>   Subject: Re: Dave Cutler and VMS@ Message-ID: <a0900283b88bb3fe25ecc8da017e32da@news.teranews.com>   Steve Lionel wrote: J > My philosophy is that you should use the language you know which is most > suitable for the problem.     J A language which interfaces well with the operating system's libraries and6 data structures is also better than one which doesn't.  5 For instance, VAX COBOL has native descriptor support   2 	CALL BAKE_CAKE USING RECIPE_STRING BY DESCRIPTOR.    M C doesn't have such integrated descriptor support. It is in many aspects just U a glorified assembler with standard words instead of architecture specific mnemonics.    ------------------------------  % Date: Mon, 24 May 2004 22:22:16 +0800 , From: Paul Repacholi <prep@prep.synonet.com>  Subject: Re: Dave Cutler and VMS- Message-ID: <87y8nh6h2f.fsf@prep.synonet.com>   4 Bob Willard <BobwBSGS@TrashThis.comcast.net> writes:  C > In addition to all of the software for which Dave is properly and A > highly regarded, he and his DecWest gang also created the first  > uVAX: uVAX-I, a ~0.3 VUP VAX.   D > Dave's opinion of the VMS gang and what he saw as DecEast politicsF > was certainly not improved by the fact that DecWest had a stable and> > shippable uVAX for roughly a year before it had VMS support.  M >    {I ran a uVAX-I under VMS v3.x in my house for years before I gave in to H >     the household inevitable, in the form of a 486/33 under Win v3.1.}  E Yeah, but Seahorse needed lots of work to get VMS to work on it for a A one shot deal. The uV-IIs and follow ons hopefully would not do a < Venus, so it was only a small time window for the I to ship.   That aside, I still want one :)    --  < Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda. @                                              West Australia 6076* comp.os.vms,- The Older, Grumpier Slashdot. Raw, Cooked or Well-done, it's all half baked.F EPIC, The Architecture of the future, always has been, always will be.   ------------------------------  % Date: Mon, 24 May 2004 09:04:08 -0400  From: norm.raphael@metso.com9 Subject: DSNLINK ECO database appears to be being updated Q Message-ID: <OF28DC2CFE.4E456F0F-ON85256E9E.00478498-85256E9E.0047B3E1@metso.com>   @ The DSNLINK ECO database appears to be being brought up-to-date.   ->- Subj:   26 new DSNlink articles in 1 database   " ECO-Summary database, 26 articles:H   [OpenVMS] VMS731_FIBRE_SCSI-V0600 Facility Patch Kit For OpenVMS ALPHA V7.3-1H   [OpenVMS] VMS732_FIBRE_SCSI-V0300 Facility Patch Kit For OpenVMS ALPHA V7.3-2C   [OpenVMS] VMS731_F11X-V0300 Facility Kit For OpenVMS Alpha V7.3-1 B   [OpenVMS] VMS731_EV7-V0200 Facility Kit For OpenVMS Alpha V7.3-1F   [OpenVMS] VMS731_GRAPHICS-V0200 Facility Patch Kit For OpenVMS ALPHA V7.3-1K   [OpenVMS] VMS731_INSTAL-V0100 Facility Patch Kit For OpenVMS ALPHA V7.3-1 H   [OpenVMS] VMS731_IPC-V0100 Facility Patch Kit For OpenVMS ALPHA V7.3-1H   [OpenVMS] VMS731_LAN-V0700 Facility Patch Kit For OpenVMS ALPHA V7.3-1H   [OpenVMS] VMS731_LAN-V0800 Facility Patch Kit For OpenVMS ALPHA V7.3-1H   [OpenVMS] VMS731_LAN-V0900 Facility Patch Kit For OpenVMS ALPHA V7.3-1H   [OpenVMS] VMS731_LAN-V1000 Facility Patch Kit For OpenVMS ALPHA V7.3-1K   [OpenVMS] VMS731_LIBRTL-V0100 Facility Patch Kit For OpenVMS ALPHA V7.3-1  ..   <-   Any knowledgeable comments?    -Norm    ------------------------------    Date: 24 May 2004 00:23:52 -0700" From: jutta.reichel@gmx.de (jutta) Subject: Re: OpenVMs Cluster; Message-ID: <ec7e473.0405232323.d3e446d@posting.google.com>   k Michael Austin <maustin@firstdbasource.com> wrote in message news:<40AB8F02.96F7A6C9@firstdbasource.com>... E > The relevant information would include the model number of the "DEC G > UNIX" system.  Sounds like a much older system (PAL code 1.xxx).  The ) > model number would tell us a lot more..  >    Thanks for the answer.  B Yes, we have very old systems. We don't make any update neither in@ hardware or software. But we need the old systen for the old 2-D CAD-System for about one year.= Model of the Server and the other cllients: DEC Alpha 255/300 = Model of the new client: Digiltal personal workstation 433au.   3 We have got another old Workstations :DEC 400 4/233  or  DEC 3000/500 Jutta P -------------------------------------------------------------------------------- > jutta wrote: > 8 > > We have a OpenVMS (Alpha) Cluster with 4 Satellites.G > > One of them is defect. We took another Client with Operating System @ > > DEC UNIX. We changed the Hardware Address on the Server with > > Cluster_config. ) > > The boot parameter on the new Client:  > > >>> os_type OpenVMS  > > >>> ew0_protocols MOP  > > >>> boot_dev ewa0.0.0.14.0 > > >>> boot_osflags 0,0 > > >>> ewa0_inet_init  bootp   > > >>> pal VMS PALcode V1.20-14 > > >>> version V6-9-7 > > 9 > > We try to boot the new Client and we got the message: 	 > > >>> b   > > boot ewa0.0.0.3.0 -flags 0,0 > > Trying MOP boot  > > .... bootstrap failure > >  > > one of the other clients: ! > >  >>> pal  VMS PALcode V5.65.2  > >  >>> version V6.9-4  > > 4 > > Do we need to do somethings more on the server ?" > > Is the firmware the wrong one? > >  > > Who can help us? > > 
 > >  Jutta	 > > Jutta    ------------------------------    Date: 24 May 2004 00:34:07 -0700" From: jutta.reichel@gmx.de (jutta) Subject: Re: OpenVMs Cluster; Message-ID: <ec7e473.0405232334.7b80553@posting.google.com>    Thanks for your answer.   ,  ewa0_mode  Twisted Pair and that's correct.  F I don't know which Version DECnet we have. It is a very old system and9 and I don't do anything with OpenVMS in the last 4 jears.   D replay/enable on the Server take no effect on the server. I think it don't know the new client.( mc ncp show known nodes characteristics:    Remote node =   2.32 (POALF4), Hardware address         = 00-00-F8-75-6C-7F" Load file                = APB.EXE2 Load Assist Agent        = SYS$SHARE:NISCS_LAA.EXE/ Load Assist Parameter    = POALF2$DKA0:<SYS11.>    Remote node =   2.33 (POALF5) , Hardware address         = 00-00-F8-21-E3-F4" Load file                = APB.EXE2 Load Assist Agent        = SYS$SHARE:NISCS_LAA.EXE/ Load Assist Parameter    = POALF2$DKA0:<SYS12.>    poalf4 is the new node.    Jutta   c lewis@spyder.mitre.org (Keith A. Lewis) wrote in message news:<c8g5m8$34u$2@newslocal.mitre.org>...  > jutta.reichel@gmx.de (jutta) writes in article <ec7e473.0405190622.77145268@posting.google.com> dated 19 May 2004 07:22:55 -0700: 7 > >We have a OpenVMS (Alpha) Cluster with 4 Satellites. F > >One of them is defect. We took another Client with Operating System? > >DEC UNIX. We changed the Hardware Address on the Server with  > >Cluster_config.) > >The boot parameter on the new Client:   > >>>> os_type OpenVMS > >>>> ew0_protocols MOP > >>>> boot_dev ewa0.0.0.14.0  > >>>> boot_osflags 0,0  > >>>> ewa0_inet_init  bootp > >>>> pal VMS PALcode V1.20-14  > >>>> version V6-9-7  > > 8 > >We try to boot the new Client and we got the message: > >>>> b > >boot ewa0.0.0.3.0 -flags 0,0  > >Trying MOP boot > >..... bootstrap failure > A > That's the message I get if the server is down, or ignoring me.  > G > Are you running DECnet IV or V?  I know the commands for phase IV, to $ > double-check the hardware address: >   > $ MC NCP SHOW NODE (node) CHAR > H > Also, you can narrow down the problem by doing a "REPLY/ENABLE" on theJ > server, then trying to boot again.  If the server gets the request there* > will be an OPCOM message to that effect. > E > Check ewa0_mode and make sure it's compatible with your hub/switch.  > 2 > --Keith Lewis              klewis {at} mitre.org@ > The above may not (yet) represent the opinions of my employer.   ------------------------------    Date: 24 May 2004 00:39:02 -0700" From: jutta.reichel@gmx.de (jutta) Subject: Re: OpenVMs Cluster; Message-ID: <ec7e473.0405232339.1c5a524@posting.google.com>    Thanks for the answer.   I Think we have DECnet* $ mc ncp list known node characteristics :   Remote node =   2.32 (POALF4) , Hardware address         = 00-00-F8-75-6C-7F" Load file                = APB.EXE2 Load Assist Agent        = SYS$SHARE:NISCS_LAA.EXE/ Load Assist Parameter    = POALF2$DKA0:<SYS11.>    Remote node =   2.33 (POALF5) , Hardware address         = 00-00-F8-21-E3-F4" Load file                = APB.EXE2 Load Assist Agent        = SYS$SHARE:NISCS_LAA.EXE/ Load Assist Parameter    = POALF2$DKA0:<SYS12.>    poalf4 is the new node.    $ mc lancp sho client * E %LANCP-E-NORSPACP, No response from LANACP, LANACP process apparently  not running * %LANCP-E-CMDERROR, Error executing command  F The new Client is a digital personal workstation 433au . Perhaps it is# not the right hardware for OpenVMS?    Jutta   ^ "Hans Vlems" <hvlems.dotweg@zonnet.nl> wrote in message news:<2h1q9oF853dlU1@uni-berlin.de>...3 > "jutta" <jutta.reichel@gmx.de> schreef in bericht 8 > news:ec7e473.0405190622.77145268@posting.google.com...8 > > We have a OpenVMS (Alpha) Cluster with 4 Satellites.G > > One of them is defect. We took another Client with Operating System @ > > DEC UNIX. We changed the Hardware Address on the Server with > > Cluster_config. ) > > The boot parameter on the new Client:  > > >>> os_type OpenVMS  > > >>> ew0_protocols MOP  > > >>> boot_dev ewa0.0.0.14.0 > > >>> boot_osflags 0,0 > > >>> ewa0_inet_init  bootp   > > >>> pal VMS PALcode V1.20-14 > > >>> version V6-9-7 > > 9 > > We try to boot the new Client and we got the message: 	 > > >>> b   > > boot ewa0.0.0.3.0 -flags 0,0 > > Trying MOP boot  > > .... bootstrap failure > >  > >  > > one of the other clients: ! > >  >>> pal  VMS PALcode V5.65.2  > >  >>> version V6.9-4  > > 4 > > Do we need to do somethings more on the server ?" > > Is the firmware the wrong one? > >  > > Who can help us? > > 
 > >  Jutta	 > > Jutta  > : > There are two ways to boot a satellite, DECnet or LANCP.L > The MAC address of the satellite is used to figure out what SYS$SYSROOT is8 > used to boot from. The MAC address is either in LANCP:, > Have a look at the output of this command: >  > $ mc lancp sho client *  > 5 > or in a DECnet database, for phase IV that would be  >  > $ mc ncp list known node char    ------------------------------  # Date: Mon, 24 May 2004 14:14:13 GMT $ From: "Mikey" <mikey117@hotmail.com> Subject: Paypal Scams...4 Message-ID: <Vqnsc.33940$KO2.31300@fe2.texas.rr.com>  N I am so sick & tired of reading bullshit on how you can get rich with Paypal!  ***News Flash***- PAYPAL DOESN"T GIVE THE $5 REFERRALS ANYMORE!  If you don't believe me, go to  S https://www.paypal.com/cgi-bin/webscr?cmd=_help-ext&leafid=54253&answer_id=20106782  and read all about it!  U All these people are doing is trying to get your $5 & you will get NOTHING in return!  You have been advised...   ------------------------------    Date: 24 May 2004 08:51:50 -07002 From: mcaccavone@manfinancial.com (Mark Caccavone)9 Subject: Previous Post - Upgrading Cluster Communications = Message-ID: <c667babc.0405240751.5f4c1e4c@posting.google.com>   D It seems that VMS is far smarter than me , I have just checked SCACPC and I see the Gigabit Cards that were installed this weekend and it C would suggest they are already enabled for SCS traffic - without me  having to run CLUSTER_CONFIG.   C Therefore the cluster will always exist when changing communication C channels so disks will not drop out of the Shadow Set when changing C from FDDI to GIGABIT. I take it that to force whether or not to use D FDDI or GIGABIT I just set the PRIORITY on the LAN DEVICE via SCACP.     Mark Caccavone MAN Financial IT VMS Support    ------------------------------  % Date: Mon, 24 May 2004 17:01:58 +0100 - From: Roy Omond <Roy.Omond@BlueBubble.UK.Com> = Subject: Re: Previous Post - Upgrading Cluster Communications * Message-ID: <2hekl0Fb0t1eU1@uni-berlin.de>   Mark Caccavone wrote:   F > It seems that VMS is far smarter than me , I have just checked SCACPE > and I see the Gigabit Cards that were installed this weekend and it E > would suggest they are already enabled for SCS traffic - without me  > having to run CLUSTER_CONFIG.  > E > Therefore the cluster will always exist when changing communication E > channels so disks will not drop out of the Shadow Set when changing E > from FDDI to GIGABIT. I take it that to force whether or not to use F > FDDI or GIGABIT I just set the PRIORITY on the LAN DEVICE via SCACP.  6 Mark, to re-iterate - "you don't need to do anything".  A If your FDDI rings are physically being decommissioned, then once @ you've got the Gigabit Ethernet hooked up (and verified that SCS@ traffic is flowing successfully over it), then simply disconnectA the FDDI cables.  VMS is smart enough to notice, and in any case, @ would have started to use the Gigabit path in preference to FDDI anyway.   @ If the FDDI rings are going to be physically left in place, then@ leave it all connected.  This way, you'll still have redundancy.  2 So once more - "you don't need to do anything" :-)  	 Roy Omond  Blue Bubble Ltd.B "VMS ... when only much better than the next best is good enough".   ------------------------------  % Date: Mon, 24 May 2004 11:26:03 -0400 ( From: "Hein" <hein.nomail@hp.nomail.com>6 Subject: Re: Puzzling LIB$SPAWN problem with DEC BASIC, Message-ID: <40b2152f$1@usenet01.boi.hp.com>  K Specifically this TRMMBX problem appears due to BROADCAST TRAPPING (through  SMG).   H Mind you, in the time you have worried about this problem you could haveF written ( or googled ) an example of calling the MAIL$ utility routineL directory and integrate that with your applicaiton. Much better performance,L better error control, for a little bit more work. (50 lines of code versus 5 lines + command file)    Cheers,  Hein.   E "Stephen Eickhoff" <operagost@operagost.example.com> wrote in message . news:6uLrc.5352$ZQ.285@nwrddc03.gnilink.net... > G > "Stephen Eickhoff" <operagost@operagost.example.com> wrote in message 2 > news:XfLrc.3284$o97.3264@nwrddc01.gnilink.net... > > > > > "David B Sneddon" <dbsneddon@bigpond.com> wrote in message( > > news:40AF00BC.1080300@bigpond.com..., > > > Stephen Eickhoff was overheard to say: > > > > K > > > > I got the following output when I fed the value of X into lib$stop:  > > > >  > > > > 000EEF37 000EEF37 E > > > > ----- above condition handler called with exception 000388F4: L > > > > %CLI-F-TRMMBX, terminal has associated mailbox - terminate image and > > then
 > > > > SPAWN  > > > > K > > > > Well, I don't use any mailboxes in this program - at least not yet.  > I'm / > > > > wondering where this mailbox came from! L > > > > What creates temporary mailboxes, other than an explicit sys$crembx? > > > G > > > Is the code using SMG routines?  The "mailbox" may be a result of  > > > using SMG. > > K > > Bingo! This is Galactic Trader - tons of SMG routines. Now the question  > is - > > what do I do about it? > K > I get to answer my own question! Question 9544 on the OpenVMS Wizard says L > that using the null device "NL:" for command input will avoid the problem. I = > don't need any command input, so I put in NL: and it works.  >  > Thanks everyone! >  >    ------------------------------  # Date: Mon, 24 May 2004 16:01:31 GMT # From: hoff@hp.nospam (Hoff Hoffman) 6 Subject: Re: Puzzling LIB$SPAWN problem with DEC BASIC1 Message-ID: <v%osc.2142$_e.1218@news.cpqcorp.net>   s In article <K1yrc.2142$No1.1723@nwrddc02.gnilink.net>, "Stephen Eickhoff" <operagost@operagost.example.com> writes: G :I'm trying to do something very simple - spawn a process from a BASIC   :program (VAX BASIC 3.8).   G   You are using asynchronous coding techniques now, meaning you must be @   rather more careful around error handling and synchornization.  F   But first, I'd tend to use callable mail API and not a spawn of the F   MAIL command.  The MAIL API is slightly more work to use, but it hasF   both increased flexibility and reduced system overhead when comparedJ   with repeatedly spawning subprocesses and the ensuing image activations.  D   If you want to stick with the current approach, you can reduce theD   entire operation to a one-line MAIL send command; no need to @ the   DCL command procedure.    > :close #8%                ! It works this far - I get the file9 :x=lib$spawn("@gal-email.com")    ! I never get the email  ..  F   If the program exits before the mail is sent, the subprocess can getF   deleted.  (Use WAIT on the lib$spawn, or synchronize the parent exit5   with subprocess completion.  Or use callable mail.)   E   Oh, and I'll assume there can only ever be one GAL-EMAIL.TMP active G   at a time in this directory, as otherwise you'll collide with an open F   version of the file and/or you'll find cases where the file containsF   the wrong data.  Again, you're coding asynchronously and you'll needF   to look at the data synchronization and the call lifetimes involved.     :I've tried expanding it to M :X=LIB$SPAWN("@GAL-EMAIL.COM",,"dka300:[galtest]gal-email.log",NOWAIT,"GALALE  :RT")   H   For common coding mistakes -- and a failure to check the return statusG   from a call such as lib$spawn is an example of an error and I'm here  H   assuming you would have shown the check if it existed -- if the returnI   status check does exist, ignore this -- please see Ask The Wizard topic    1661.* and related topics.  N  ---------------------------- #include <rtfaq.h> -----------------------------K     For additional, please see the OpenVMS FAQ -- www.hp.com/go/openvms/faq N  --------------------------- pure personal opinion ---------------------------E         Hoff (Stephen) Hoffman   OpenVMS Engineering   hoff[at]hp.com    ------------------------------  # Date: Mon, 24 May 2004 16:14:07 GMT # From: hoff@hp.nospam (Hoff Hoffman) 6 Subject: Re: Puzzling LIB$SPAWN problem with DEC BASIC0 Message-ID: <jbpsc.2144$_e.734@news.cpqcorp.net>  s In article <XfLrc.3284$o97.3264@nwrddc01.gnilink.net>, "Stephen Eickhoff" <operagost@operagost.example.com> writes: L :> > Well, I don't use any mailboxes in this program - at least not yet. I'm, :> > wondering where this mailbox came from!I :> > What creates temporary mailboxes, other than an explicit sys$crembx?  :>D :> Is the code using SMG routines?  The "mailbox" may be a result of
 :> using SMG.  : M :Bingo! This is Galactic Trader - tons of SMG routines. Now the question is -  :what do I do about it?     I   There are SMG calls which relinquish the terminal and to later reassert I   the SMG display, and these are commonly used to bracket operations that K   might want to control the terminal display: the smg$save_physical_screen  *   and smg$restore_physical_screen calls.    C   There are also process-control calls within the SMG library, too.   N  ---------------------------- #include <rtfaq.h> -----------------------------K     For additional, please see the OpenVMS FAQ -- www.hp.com/go/openvms/faq N  --------------------------- pure personal opinion ---------------------------E         Hoff (Stephen) Hoffman   OpenVMS Engineering   hoff[at]hp.com    ------------------------------  # Date: Mon, 24 May 2004 15:40:56 GMT - From: JF Mezei <jfmezei.spamnot@teksavvy.com> ! Subject: Re: Request for new SMTP @ Message-ID: <c9599849cf8c5dfbdd7ae14e2438fd83@news.teranews.com>   Michael Austin wrote: g > system) and bounces the message to the sender in the  "MAIL FROM:"   Because this address is **also** q > bogus/forged, the receiving system bounces the message back to "postmaster" or equivelent, in my case, the user  > TCPIP$SMTP gets the message.  E YOu should properly define your postmaster alias so that non-delivery N notifications are signed by "postmaster@chocolate.com" instead of "TCPIP$SMTP"  1 define/system TCPIP$POSTMASTER_ALIAS "postmaster"   N (make sure that you have a VMS mail profile entry for "postmaster" that points to a valid username).   @ Many implementations will not return messages to a "postmaster".   ------------------------------  % Date: Mon, 24 May 2004 11:56:08 +0200 ( From: "Rudolf Wingert" <win@fom.fgan.de># Subject: SDLT 320 Tape Drive Issues 3 Message-ID: <009401c44175$5648d7d0$994614ac@wat153>    Hello,   Davies did wrote:    >>> C It seems that after a tape is used for a backup - the drive will no F longer accept the tape.  If you put it in - it spits it right back out after a few seconds. <<<   C We do have SDLT320 for our OpenVMS backup. We do use it without any C problem. I did see your behavior only with a bad media. Did you try 2 other media? Did you a visual check of your media?   Best regards R. Wingert    ------------------------------  % Date: Mon, 24 May 2004 13:02:29 +0100 , From: Adrian Birkett <aaa@notreallyhere.com> Subject: Sortwork file query' Message-ID: <c8socf$3uc$1@lore.csc.com>    Hi,   H One of out applications sorts a file. The logical names of the sortwork H files are defined as DSA1:[NODIR]. This directory does not exist on the H device and the files are showing in SDA as DSA1:[]. Is there any way of I viewing these files to get an idea of how big they are as the seem to be  - eating disk space at  quite an alarming rate?    Regards,   Ade    ------------------------------  % Date: Mon, 24 May 2004 13:51:52 +0100 - From: Roy Omond <Roy.Omond@BlueBubble.UK.Com>   Subject: Re: Sortwork file query* Message-ID: <2he9giFc3av2U1@uni-berlin.de>   Adrian Birkett wrote:   J > One of out applications sorts a file. The logical names of the sortwork J > files are defined as DSA1:[NODIR]. This directory does not exist on the J > device and the files are showing in SDA as DSA1:[]. Is there any way of K > viewing these files to get an idea of how big they are as the seem to be  / > eating disk space at  quite an alarming rate?   C Use DFU.  If your files really have the name ".", then a simple DFU & search will tell you how big they are.  ! E.g.	$ mcr dfu search dsa1/file=.   6 Always remember, DFU is your (very special) friend :-)  	 Roy Omond  Blue Bubble Ltd.   ------------------------------  % Date: Mon, 24 May 2004 11:18:51 -0400 ( From: "Hein" <hein.nomail@hp.nomail.com>  Subject: Re: Sortwork file query* Message-ID: <40b21380@usenet01.boi.hp.com>  J If sort was not allowed to use enough memory to sort a file in memory then it has to go to the workfile. I That workfile will eb roughly as big as the input file. If there is a big J difference, then this might have been caused bythe "Longest Record Length"K (LRL) being out of whack with the actuall file contents. Make sure that LRL L matches the true LRL (use SET FILE/ATTR=LRL if need be) and is not just some
 large number. K If your file is variable length and truly has a wide range on record sizes,  then you my be SOL.  - Check out TAG sort+ - Read the SORT/STAT output very carefully.    hth, Hein.             9 "Adrian Birkett" <aaa@notreallyhere.com> wrote in message ! news:c8socf$3uc$1@lore.csc.com...  > Hi,  > I > One of out applications sorts a file. The logical names of the sortwork I > files are defined as DSA1:[NODIR]. This directory does not exist on the I > device and the files are showing in SDA as DSA1:[]. Is there any way of J > viewing these files to get an idea of how big they are as the seem to be/ > eating disk space at  quite an alarming rate?  > 
 > Regards, >  > Ade  >    ------------------------------  % Date: Tue, 25 May 2004 00:10:19 +0800 , From: Paul Repacholi <prep@prep.synonet.com>  Subject: Re: Sortwork file query- Message-ID: <87fz9p6c2c.fsf@prep.synonet.com>   . Adrian Birkett <aaa@notreallyhere.com> writes:  @ > One of out applications sorts a file. The logical names of theE > sortwork files are defined as DSA1:[NODIR]. This directory does not E > exist on the device and the files are showing in SDA as DSA1:[]. Is E > there any way of viewing these files to get an idea of how big they D > are as the seem to be eating disk space at quite an alarming rate?  B Use ANAL/DISK to list the header, and then with the file IDs, dump the header for them.   --  < Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda. @                                              West Australia 6076* comp.os.vms,- The Older, Grumpier Slashdot. Raw, Cooked or Well-done, it's all half baked.F EPIC, The Architecture of the future, always has been, always will be.   ------------------------------    Date: 24 May 2004 08:10:03 -07002 From: mcaccavone@manfinancial.com (Mark Caccavone)8 Subject: Upgrading Cluster from FDDI to Gigabit Ethernet= Message-ID: <c667babc.0405240710.1391af7b@posting.google.com>   B We have a split site VMS cluster with Enterprise Virtual Arrays at@ either end with VMS controlling the Volume Shadowing between theE EVA's. Currently the SCS traffic is all going via FDDI , which is due E to be phased out in the UK next year. Temporarily we are going to use C an existing Gigabit circuit to control the cluster traffic , before  going to Fibre Channel.   D Question is this , when I run CLUSTER_CONFIG to change SCS from FDDIE to GigE , will that node lose quorum temporarily? If so does that not F mean that the DISK's that are controlled by that node will drop out ofF the Shadow Set and spend several hours rebuilding? If thats the case I@ can Dismount the disks using /POLICY=MINICOPY prior to executing Cluster_Config.      Mark Caccavone MAN Financial  VMS Support    ------------------------------  % Date: Mon, 24 May 2004 16:19:46 +0100 - From: Roy Omond <Roy.Omond@BlueBubble.UK.Com> < Subject: Re: Upgrading Cluster from FDDI to Gigabit Ethernet* Message-ID: <2hei5sFbrr6eU1@uni-berlin.de>   Mark Caccavone wrote:   D > We have a split site VMS cluster with Enterprise Virtual Arrays atB > either end with VMS controlling the Volume Shadowing between theG > EVA's. Currently the SCS traffic is all going via FDDI , which is due G > to be phased out in the UK next year. Temporarily we are going to use E > an existing Gigabit circuit to control the cluster traffic , before  > going to Fibre Channel.    Good.  Sounds nice.   F > Question is this , when I run CLUSTER_CONFIG to change SCS from FDDIG > to GigE , will that node lose quorum temporarily? If so does that not H > mean that the DISK's that are controlled by that node will drop out ofH > the Shadow Set and spend several hours rebuilding? If thats the case IB > can Dismount the disks using /POLICY=MINICOPY prior to executing > Cluster_Config.   G You don't *need* to do *anything* (and you most certainly don't need to B run CLUSTER_CONFIG).  This is VMS, not some other inferior system.C If you've already got the Gigabit controllers in the machines, then E they will be candidates for SCS traffic by default (easily verifiable + in SDA, or SCACP on later versions of VMS).   C And because you don't need to do anything, your disks will not drop  out of the shadowsets :-)   	 Roy Omond  Blue Bubble Ltd.+ "VMS ... when only the best is good enough"    ------------------------------  % Date: Mon, 24 May 2004 11:31:06 -0400  From: norm.raphael@metso.com< Subject: Re: Upgrading Cluster from FDDI to Gigabit EthernetQ Message-ID: <OF2814ACEC.2C947931-ON85256E9E.0055277A-85256E9E.0055285F@metso.com>     > before going to Fibre Channel.  # Is SCS over FC announced/supported?   I mcaccavone@manfinancial.com (Mark Caccavone) wrote on 05/24/2004 11:10:03  AM:   D > We have a split site VMS cluster with Enterprise Virtual Arrays atB > either end with VMS controlling the Volume Shadowing between theG > EVA's. Currently the SCS traffic is all going via FDDI , which is due G > to be phased out in the UK next year. Temporarily we are going to use D > an existing Gigabit circuit to control the cluster traffic, before > going to Fibre Channel.  > F > Question is this , when I run CLUSTER_CONFIG to change SCS from FDDIG > to GigE , will that node lose quorum temporarily? If so does that not H > mean that the DISK's that are controlled by that node will drop out ofH > the Shadow Set and spend several hours rebuilding? If thats the case IB > can Dismount the disks using /POLICY=MINICOPY prior to executing > Cluster_Config.  >  >  > Mark Caccavone > MAN Financial 
 > VMS Support    ------------------------------  # Date: Mon, 24 May 2004 16:15:55 GMT / From: brooks@cuebid.zko.dec.nospam (Rob Brooks) < Subject: Re: Upgrading Cluster from FDDI to Gigabit Ethernet- Message-ID: <feiC2VPM2CfK@cuebid.zko.dec.com>    norm.raphael@metso.com writes:  % > Is SCS over FC announced/supported?    No.    --    M Rob Brooks    VMS Engineering -- I/O Exec Group     brooks!cuebid.zko.dec.com    ------------------------------  % Date: Mon, 24 May 2004 12:09:44 -0400  From: norm.raphael@metso.com< Subject: Re: Upgrading Cluster from FDDI to Gigabit EthernetQ Message-ID: <OFCC202629.F499FBA0-ON85256E9E.0058AB19-85256E9E.0058B1C8@metso.com>   I Thanks, Rob.  That was my understanding.  So I wonder why he's driving at  GigE speed on a * road with a "Bridge Out Ahead" sign on it.  J brooks@cuebid.zko.dec.nospam (Rob Brooks) wrote on 05/24/2004 12:15:55 PM:    > norm.raphael@metso.com writes: > ' > > Is SCS over FC announced/supported?  >  > No.  >  > -- > 1 > Rob Brooks    VMS Engineering -- I/O Exec Group  brooks!cuebid.zko.dec.com    ------------------------------  % Date: Mon, 24 May 2004 11:38:37 -0500 / From: Chris Scheers <chris@applied-synergy.com> . Subject: Re: Vaxstation 3100 and graphics card3 Message-ID: <40B2250D.1B882B01@applied-synergy.com>    JF Mezei wrote:  > L > So, I finally installed a WS01X (SPX) graphics card in my microvax 3100-30 >  > Alas, no success.  >  > A test 50 shows: >   > ?? SPGFX       0001.2112  V1.3A >           00FFFF04 00000000 39B00014 00000002 00000000 00000000  > G > I know that the ?? means trouble. But what can I tell from the rest ?  > ) > a TEST 4 yiels an error code of 84 FAIL  > C > test 80 doesn't complain (but I have to use break key to end it).  > N > Now, this is all from the serial console. I built my own adaptor from the 15N > pin vax monitor plug to the 15 pin d-sub VGA plug. However, all of the aboveK > happen even if the adaptor is not connected. The monitor does not see any ; > video signal (it worked fine with the monochrome output).  >  > Questions;M > How do I find documenmtation on the above error codes and byte displays for O > that device ? Shouldn't HP make all VAX technical documentation available for / > download now that they are retired products ?  > J > Are there any dip/jumpers that need to be changed on the VAX motherboardO > before adding the WS01X board ? Or does one just add the board, and power the 1 > unit back on and it should magically all work ?  > I > If there is no monotor connected, should the varous tests still succeed  > without any complaints ?    H I don't remember for sure, but I would not be surprised to find out thatG the tests will fail if a monitor is not correctly connected to at least  the green lead.   , Make sure that you cable is wired correctly.  G ----------------------------------------------------------------------- $ Chris Scheers, Applied Synergy, Inc.  C Voice: 817-237-3360            Internet: chris@applied-synergy.com     Fax: 817-237-3074    ------------------------------  % Date: Mon, 24 May 2004 11:59:39 -0500 / From: Chris Scheers <chris@applied-synergy.com> . Subject: Re: Vaxstation 3100 and graphics card3 Message-ID: <40B229FB.99C580AC@applied-synergy.com>    JF Mezei wrote:  > K > If one installs the SPX (WS01X) graphics card on a vaxstation 3100-30, is N > there a need to rerun  autogen ? Or does the graphics card come with all its4 > own resources (such as screen bitmap memory) etc ? > O > Also, is there a way through software/sysgen to disable the graphics card and E > have the built-in monochrome card take over (either at boot time or W > dynamically) ? (for istance to test how x software performs on a monochrome display).  > O > Also, something I don't quite understand. What does "8 plane" actually mean ?  > N > Also, what sort of services does the graphics card provide to VMS ? does VMSK > just use the graphic card's memory to just dump bitmaps, or does the card P > provide smart services such as functions to draw lines, boxes, fill areas with > a colour pattern etc ? > P > Does the graphic card use specific drivers at the VMS level ? (if so, which ?)  E The SPX graphics card has its own bitmap memory.  But DECwindows will F need some additional memory to track the additional resources provided* by the card, so an AUTOGEN is a good idea.  C When you install an SPX card in a VS3100, both the SPX card and the * VS3100's onboard mono graphics are active.  H Take a look at SYS$MANAGER:DECW$PRIVATE_SERVER_SETUP.TEMPLATE for how toG control these resources.  In particular, take a look at the section for E a dual head VS3100 system.  With the appropriate definitions, you can D have DECwindows use the mono card, the color card, or both.  You canH also have it generate a mono signal on the green lead of the color card.  E The color and mono cards use different pins on the vido connector, so H you need to make sure that you have the correct cabling to bring out the signal(s) you want.   C "8 plane" refers to the color capabilities of the card.  Each plane E provides one bit of color information, so an 8 plane card can support 5 2**8 or 256 colors.  (A mono card is a 1 plane card.)   E The SPX card includes a graphics processor which provides much of the > capability of the card.  The SPX card is optimized for the X11H primitives, as opposed to the GPX card which was optimized (???) for the VWS primitives.   
 Good luck!  G ----------------------------------------------------------------------- $ Chris Scheers, Applied Synergy, Inc.  C Voice: 817-237-3360            Internet: chris@applied-synergy.com     Fax: 817-237-3074    ------------------------------  % Date: Mon, 24 May 2004 11:34:16 +0100 O From: Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com> 2 Subject: Re: You'll never guess what HP advertised0 Message-ID: <c8sj39$mhh$1@new-usenet.uk.sun.com>   John Smith wrote: M > "David J. Dachtera" <djesys.nospam@NeOaSrPtAhMlNiOnWk.net> wrote in message 2 > news:40AEBAC5.39082ED0@NeOaSrPtAhMlNiOnWk.net... >  >>John Smith wrote:  >> >>>Dave, >>>  >>>It's tougher than that....  >>> ( >>>No Exchange compatible server on VMS, >>A >>Really? Then how was my ISP (DLS and FSI before them) doing it?  >> >>) >>>now no more AV software (EOL announced  >>>for Sophos),  >>D >>VMS doesn't need anti-virus software. Virus scanners are purely toJ >>accomodate inferior systems that may be using a VMS-based system as mailB >>server, to scan e-mail for malicious code targetting he inferior >>systems, not VMS.  >  >  > 
 > Exactly. > K > However in the SMB (small/medium business space, the (dare I say)  common I > practice is to use but one os on the server side, hence the need for AV J > software, Exchange compatible mail/calendaring gateway, etc.... preciousI > little of which VMS has available, and when such 3rd party products are M > available, I wouldn't call having the choice of only one package in a given  > category a thriving market.  >   B No one really thinks that UNIX systems require anti-virus software@ either,  but we have large numbers of systems hosting anti-virus= packages such as Sophos and Trend Micro because these are the > fileserver and mail servers for Windows boxes and removing the? viruses from their directories or their attachments before they ! get to the PC's is a requirement.   D We also have systems being used as SMTP virus gateways as an adjunct to being a FireWall.   Intrusion detection ditto  Authentication ditto   regards  Andrew Harrison N > HP claims its year-over-year VMS revenues are up. Whether that's from higherG > services-related revenue, extra attributed storage sales, or real VMS N > licencing revenue (no unit volumes given either), and the number of customerM > that HP targets directly is small (750-1000 named accounts in the USA), the F > SMB space offers a huge opportunity for unit sales growth for VMS --. > providing the apps and advertising is there. >  >  >  >  >    ------------------------------    Date: 24 May 2004 07:17:05 -0700( From: bob@instantwhip.com (Bob Ceculski)2 Subject: Re: You'll never guess what HP advertised= Message-ID: <d7791aa1.0405240617.600ca386@posting.google.com>    Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com> wrote in message news:<c8sj39$mhh$1@new-usenet.uk.sun.com>...  > D > No one really thinks that UNIX systems require anti-virus softwareB > either,  but we have large numbers of systems hosting anti-virus? > packages such as Sophos and Trend Micro because these are the @ > fileserver and mail servers for Windows boxes and removing theA > viruses from their directories or their attachments before they # > get to the PC's is a requirement.  > F > We also have systems being used as SMTP virus gateways as an adjunct > to being a FireWall. >  > Intrusion detection ditto  > Authentication ditto > 	 > regards  > Andrew Harrison   : we run sophos on vms and do all the above also, so what is your point? :)   ------------------------------   End of INFO-VAX 2004.287 ************************