1 INFO-VAX	Mon, 24 Apr 2000	Volume 2000 : Issue 229       Contents:/ "Reverse" Telnet as AlphaServer serial console? 3 Re: "Reverse" Telnet as AlphaServer serial console? 3 RE: "Reverse" Telnet as AlphaServer serial console? 3 Re: "Reverse" Telnet as AlphaServer serial console?  Another VAX to Alpha question $ Re: Backup Strategies for Enterprise0 Re: Catastrophic EVE/TPU documentation error :-)0 Re: Catastrophic EVE/TPU documentation error :-) Re: CXML and SETI   For Stephen Hoffman: USB support$ Re: For Stephen Hoffman: USB support. how do i kwno how much memory is on my system?2 Re: how do i kwno how much memory is on my system?2 RE: how do i kwno how much memory is on my system?2 Re: how do i kwno how much memory is on my system?, Re: Infoserver CD-R (was: Verify of Backups)& RE: LK46W-A2 & "DO" key in Reflections Re: Mozilla M15's out... Re: Mozilla M15's out... Re: Mozilla M15's out... Re: NT Breaks Again  Re: NT Breaks Again  OpenVMS marketing  Re: OpenVMS marketing  Re: OpenVMS marketing / Problem with upgrading an Alpha Server to 7.2-1 , Re[2]: Calculating Discordian dates from DCL& Re: The Movie "Breaking Point" + PDP11& Re: The Movie "Breaking Point" + PDP113 Re: UCX "SET NOHOST/NOCONFIRM" Wipes Out Host Table  UCX SMTP Mail Headers  Re: UCX SMTP Mail Headers ! Re: UCX TCP/IP Packet Window Size  Re: Verify of Backups  RE: Verify of Backups  Re: Verify of Backups  RE: Verify of Backups  Re: Verify of Backups   F ----------------------------------------------------------------------   Date: 24 Apr 2000 13:46:02 GMT0 From: "Dale A. Dellutri" <ddellutr@enteract.com>8 Subject: "Reverse" Telnet as AlphaServer serial console?+ Message-ID: <8e1j6q$ar$1@news.enteract.com>   > I will soon have 7 separate ES40s, each of which need a system> console.  When I just 3, I used a VT520 with three sessions to> have three serial consoles with one screen.  Now, with 7, that won't be practical.   > I'm considering doing some sort of "reverse" telnet setup: use? a simple terminal server with 8 serial ports and a TCP/IP port, = attach 7 of the serial ports to each of the ES40s, then use a @ cross-over cable to attach the TCP/IP port to a cheap PC running= Win98, start 7 telnet sessions on the PC, and connect each of 7 them to one of the serial ports on the terminal server.   < I assume that I can find a terminal server that will do this "reverse" telnet.   : Is anyone doing this?  What terminal server are you using? Any issues I should consider?   ? I'm trying to do this without buying any special software (like  Polycenter Console Manager).   --  & Dale Dellutri -- ddellutr@enteract.com   ------------------------------  + Date: Mon, 24 Apr 2000 08:16:04 -0600 (MDT) ) From: John Nebel <nebel@athena.csdco.com> < Subject: Re: "Reverse" Telnet as AlphaServer serial console?F Message-ID: <Pine.OSF.4.21.0004240802280.7129-100000@athena.csdco.com>   Dale,   G A DECserver 700 should work fine.  Reverse LAT is easier to set up, but E reverse telnet works too.  The cable requires an RJ45 on one end with D an offset-modular on the other.  I make my own with 6-conductor flatI cable.  Needle nose pliers are helpful as the center two positions (if I  . remember correctly) of the RJ45 need a jumper.  D PT525 (part of Pathworks) will connect over a LAN to the DS 700 withE either telnet, LAT, or the serial port.  The advantage of using a LAN G connection is that you can connect to all the console ports from any of 
 the hosts.  
 John Nebel    ' On 24 Apr 2000, Dale A. Dellutri wrote:   @ > I will soon have 7 separate ES40s, each of which need a system@ > console.  When I just 3, I used a VT520 with three sessions to@ > have three serial consoles with one screen.  Now, with 7, that > won't be practical.  > @ > I'm considering doing some sort of "reverse" telnet setup: useA > a simple terminal server with 8 serial ports and a TCP/IP port, ? > attach 7 of the serial ports to each of the ES40s, then use a B > cross-over cable to attach the TCP/IP port to a cheap PC running? > Win98, start 7 telnet sessions on the PC, and connect each of 9 > them to one of the serial ports on the terminal server.  > > > I assume that I can find a terminal server that will do this > "reverse" telnet.  > < > Is anyone doing this?  What terminal server are you using? > Any issues I should consider?  > A > I'm trying to do this without buying any special software (like  > Polycenter Console Manager). >  > --  ( > Dale Dellutri -- ddellutr@enteract.com >    ------------------------------  % Date: Mon, 24 Apr 2000 11:16:36 -0400 # From: John Vottero <John@MVPSI.com> < Subject: RE: "Reverse" Telnet as AlphaServer serial console?D Message-ID: <C15945A9D9EFCF11BA8B08002BBF1CCC0CD6C9@berry.mvpsi.com>  L I'm doing this with Reflections and a DECserver 90TL  I'm not sure what typeI of console connector an ES40 has but, if it's a PC 9 pin serial connector L then you  can use an H8585-AA to go from 9 pin to RJ45 and then use standardK twisted pair Ethernet cables to connect to the 90TL (which has RJ45 not MMJ  connectors).   > -----Original Message-----7 > From: Dale A. Dellutri [mailto:ddellutr@enteract.com] & > Sent: Monday, April 24, 2000 9:46 AM > To: Info-VAX@Mvb.Saic.Com : > Subject: "Reverse" Telnet as AlphaServer serial console? >  > @ > I will soon have 7 separate ES40s, each of which need a system@ > console.  When I just 3, I used a VT520 with three sessions to@ > have three serial consoles with one screen.  Now, with 7, that > won't be practical.  > @ > I'm considering doing some sort of "reverse" telnet setup: useA > a simple terminal server with 8 serial ports and a TCP/IP port, ? > attach 7 of the serial ports to each of the ES40s, then use a B > cross-over cable to attach the TCP/IP port to a cheap PC running? > Win98, start 7 telnet sessions on the PC, and connect each of 9 > them to one of the serial ports on the terminal server.  > > > I assume that I can find a terminal server that will do this > "reverse" telnet.  > < > Is anyone doing this?  What terminal server are you using? > Any issues I should consider?  > A > I'm trying to do this without buying any special software (like  > Polycenter Console Manager). >  > --  ( > Dale Dellutri -- ddellutr@enteract.com >    ------------------------------  % Date: Mon, 24 Apr 2000 11:00:32 -0500 ) From: "John E. Malmberg" <wb8tyw@qsl.net> < Subject: Re: "Reverse" Telnet as AlphaServer serial console?7 Message-ID: <012e01bfae06$3c43db20$020a0a0a@xile.realm>   , Dale A. Dellutri <ddellutr@enteract.company>  @ > I will soon have 7 separate ES40s, each of which need a system@ > console.  When I just 3, I used a VT520 with three sessions to@ > have three serial consoles with one screen.  Now, with 7, that > won't be practical.  > @ > I'm considering doing some sort of "reverse" telnet setup: useA > a simple terminal server with 8 serial ports and a TCP/IP port, ? > attach 7 of the serial ports to each of the ES40s, then use a B > cross-over cable to attach the TCP/IP port to a cheap PC running? > Win98, start 7 telnet sessions on the PC, and connect each of 9 > them to one of the serial ports on the terminal server.  > < > Is anyone doing this?  What terminal server are you using? > Any issues I should consider?   K I prefer to use separate terminals.  This could be a space problem for you, / if you do not have room for another two VT520s.   H The only time that you need access to the consoles is when you are doingD some software upgrades, standalone backup, and hardware diagnostics.  J For all other operations, I use a VMS workstation with DECWindows-Motif toL access the systems.  The instructions for setting up DECNET objects to allow" easy launching of remote DECterms.    E There are two big drawbacks for using a terminal server on a Console.   G 1. If the Terminal Server is down, you  have no access to your systems.   F Your systems may not even boot if the terminal server is down.  If theJ terminal server does not have it's own flash ROM, it requires at least oneJ load host to boot it.  Just think, a full computer room power failure, andH you have no way of recovering.  Have you heard all the stories of people0 accidentally hitting the emergency power button?  K 2. With the VMS systems I am familiar, you have to use a hardware switch to K enable halting.  In this case, a reboot of the terminal server can halt the J machine.  If you are planning to be able to halt and do console operationsJ in a lights out computer room remotely, you have introduced a failure mode into your system.    -John  wb8tyw@qsl.network   ------------------------------  # Date: Mon, 24 Apr 2000 13:52:43 GMT / From: "John Nixon" <jorlnixon@worldnet.att.net> & Subject: Another VAX to Alpha questionG Message-ID: <LkYM4.30851$fV.1865372@bgtnsc05-news.ops.worldnet.att.net>   L I appreciate the comments I have received on VAX to Alpha conversion. I haveL been asked another question about this effort (and I am sure I will be asked a lot more).  H One of the major warnings of areas to be concerned about have to do withG boundary allingments.  The question is whether these problems will show L themself during Compile or Link, or are we likely to see those problems whenF the program is run.  Many of the problems we had with the VAXC to DECCI Conversion did not show up until after the executables were built and run 4 and then failed (frequently with access violations).  K I look forward to one day  riding the wave, rather than constantly paddling  to catch up.   ------------------------------  # Date: Mon, 24 Apr 2000 15:58:08 GMT  From: jgessling@yahoo.com - Subject: Re: Backup Strategies for Enterprise ) Message-ID: <8e1qu6$o9d$1@nnrp1.deja.com>   H In article <EA5FE99F5DEAD311A6CB00805F199992BCA805@misnts1.dalsemi.com>,.   Don Rogstad <Don.Rogstad@dalsemi.com> wrote:D > Our current environment has lots of Sun boxes, Digital Unix boxes,G Windows NT boxes and of course, OpenVMS VAX and Alpha machines.  We are D now looking for ways to consolidate the backup process from multiple machines / platforms.    Don,  G I've never understood this need.  When will you ever restore a VMS file @ to an NT server?  or Unix to VMS etc.?  Given the differing fileD systems and special needs, like image backups on VMS, registry filesD on NT and such, there doesn't exist one good solution.  I personallyF think that if the solution is good enough for the particular platform,D like being mostly automatic, then having multiple solutions is not a	 big deal.   G Others have mentioned ABS as being good for VMS, I have one BIG problem F with it.  To lookup a file to restore, you must know what disk it cameE from!  Sorry folks, even my development system has more that 16 disks D and files move from time to time so locating one can be a real pain.D According to Cpaq this is the way it's designed so tough luck.  Look@ in dsnlink in the storage database for this article if you don't believe me.  search for:  0 "ABS Can't Find Data in Catalog After Dir Moved"  D This references version 2.2 from december 1998, but it's not changed in 3.0   Jim     & Sent via Deja.com http://www.deja.com/ Before you buy.    ------------------------------  % Date: Mon, 24 Apr 2000 01:30:21 -0500 , From: Howard S Shubs <hshubs@mindspring.com>9 Subject: Re: Catastrophic EVE/TPU documentation error :-) > Message-ID: <hshubs-3F4D0C.01302124042000@news.mindspring.com>  9 In article <3903D1FB.64A2D223@vl.videotron.ca>, JF Mezei  ( <jfmezei.spamnot@vl.videotron.ca> wrote:  M >Shouldn't that be a variable which automatically dials into Compaq's systems F >to find out what name Compaq's version of UNIX is this week ? :-) :-)  6 And it should be labeled as "OpenVMS" these days.  :-/ --  ; Howard S Shubs      hshubs@mindspring.com    hshubs@bix.com ? The Denim Adept     Which is better, Maryann or pickled Ginger? > SPAM: uce@ftc.gov   postmaster@[127.0.0.1]   abuse@[127.0.0.1]   ------------------------------  # Date: Mon, 24 Apr 2000 12:49:35 GMT , From: koehler@eisner.decus.org (Bob Koehler)9 Subject: Re: Catastrophic EVE/TPU documentation error :-) ' Message-ID: <2000Apr24.084935.1@eisner>   b In article <3903D1FB.64A2D223@vl.videotron.ca>, JF Mezei <jfmezei.spamnot@vl.videotron.ca> writes: > VMS VAX, 7.2 > 
 > EDIT/TPU >  > Then HELP EVE  > ' > "EVE Runs on VMS or ULTRIX" :-) :-)    > N > Shouldn't that be a variable which automatically dials into Compaq's systemsG > to find out what name Compaq's version of UNIX is this week ? :-) :-)   F The statement is true and unchanged.  OSF1/DU/Tru64 is not ULTRIX.  On6 the other hande TPU didn't ship with ULTRIX very long.  F ----------------------------------------------------------------------? Bob Koehler                     | Computer Sciences Corporation = Hubble Space Telescope Payload  | Federal Sector, Civil Group E  Flight Software Team           | please remove ".aspm" when replying    ------------------------------  % Date: Mon, 24 Apr 2000 10:38:13 -0400 # From: Jim Agnew <agnew@hsc.vcu.edu>  Subject: Re: CXML and SETI+ Message-ID: <39045C55.406E3DC2@hsc.vcu.edu>   G If you all are doing a peer-reviewed paper, you MUST be able to account H for boundary conditions, zerodivides, etc...  the SETI folks really needH to keep their code base locked down tight, or else the research will notH be based on rock-solid math...  computational math is really difficult..   Jim    Steve Lionel wrote:  > 4 > On Sun, 16 Apr 2000 02:22:57 -0400, David A Froble > <davef@tsoft-inc.com> wrote: > R > >It would probably be counter-productive, but I'd like to see Compaq release theQ > >ultra fast SETI client.  It would then force the SETI people to decide whether = > >they wanted results, or are just another political entity.  > E > The SETI project does not want us to release the version of the 1.3 C > client that uses CXML, and the rules are that 2.x clients are not H > allowed to use vendor-supplied FFT routines - you have to use the code? > as they supply it.  The SETI folks claim that the "science is B > compromised" if different FFT code is used.  Personally, I don'tB > understand that, but it's their project and they get to make the > rules. > / > Steve Lionel (mailto:Steve.Lionel@compaq.com)  > Fortran Engineering ( > Compaq Computer Corporation, Nashua NH > 8 > Compaq Fortran web site: http://www.compaq.com/fortran   ------------------------------  # Date: Mon, 24 Apr 2000 10:27:23 GMT  From: Dirk Munk <munk@home.nl>) Subject: For Stephen Hoffman: USB support ' Message-ID: <39042151.6D88EA00@home.nl>   C If I'm not mistaken, the next version of VMS will have USB support. F The local PC shops over here have several PCI-USB adapter boards based on the CMD chip.H Can we expect that these boards will be supported in VMS (without asking for guarantees of course) ?    ------------------------------   Date: 24 Apr 2000 14:58:56 GMT2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)- Subject: Re: For Stephen Hoffman: USB support 6 Message-ID: <8e1nfg$qlb$1@mailint03.im.hou.compaq.com>  H In article <39042151.6D88EA00@home.nl>, Dirk Munk <munk@home.nl> writes:D :If I'm not mistaken, the next version of VMS will have USB support.  G   I would not make that assumption -- we are incorporating USB into an  I   upcoming OpenVMS release, but (at least at present) I would not assume  H   that the next big OpenVMS release will be the one to (initially) have    USB support.  G :The local PC shops over here have several PCI-USB adapter boards based  :on the CMD chip.   I   We did at least some of the initial development and testing of the USB  D   support on OpenVMS with a Symbios 6800 PCI-USB card, but the only I   supported platform for USB will be (at least initially) the integrated  H   USB interface -- this once all of the hardware, firmware, and OpenVMS H   issues are worked out, and also for a limited set of USB devices.  We I   did definitely find some problems with at least one other (widget name  I   has been expurgated as it was a pre-release version) PCI-USB widget we  I   tried...  Donno if these problems  got fixed with the released version  K   of the (expurgated) PCI-USB widget, nor if the Symbios PCI-USB card will     continue to operate.  I :Can we expect that these boards will be supported in VMS (without asking  :for guarantees of course) ?  G   Initially, I would tend to doubt it -- it might work, it might not...SJ   Given my experience with I/O widgets and with the general compatibility H   found in the "commodity" I/O hardware space, I would initially assume F   that any random I/O widget will not work, until proven otherwise... J   Further, these PCI-USB widgets will likely not autoconfigure, but -- if I   you want to try using one -- then see about copying the USB entry from  E   SYS$CONFIG.DAT over to SYS$USER_CONFIG.DAT, and modifying the entrytE   appropriately (based on the ID string retrieved via ANALYZE/SYSTEM  J   CLUE CONFIG).  (You will also need the "early adopter" USB kit that was J   around for a while, pending the incorporation of the USB support into a    main-line OpenVMS release.)   N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------  # Date: Mon, 24 Apr 2000 17:01:09 GMT  From: joecarlos@yahoo.com (Joe) 7 Subject: how do i kwno how much memory is on my system? 5 Message-ID: <8F20856F0joeyahoocom@news.supernews.com>k  B is there a commnad that tells me the physical memory of my system?   ------------------------------  # Date: Mon, 24 Apr 2000 17:11:24 GMTc) From: healyzh@aracnet.com (Zane H. Healy)E; Subject: Re: how do i kwno how much memory is on my system?a4 Message-ID: <0f%M4.376$pK4.8050@typhoon.aracnet.com>  W In article <8F20856F0joeyahoocom@news.supernews.com>, joecarlos@yahoo.com (Joe) writes: E |> is there a commnad that tells me the physical memory of my system?s |> a    I The first part of the SHOW MEMORY command will tell you how much there issJ if you've already go the system up.  On an Alpha at the SRM, and IIRC, on % most VAXen you can also do a SHO MEM.    $sho mem@               System Memory Resources on 24-APR-2000 10:14:04.34  L Physical Memory Usage (pages):     Total        Free      In Use    ModifiedL   Main Memory (320.00Mb)           40960       27650       12127        1183   <snip a lot of other info>   				Zane   ------------------------------  % Date: Mon, 24 Apr 2000 13:31:43 -0400i. From: "Crowley, George M." <gmcrowley@tva.gov>; Subject: RE: how do i kwno how much memory is on my system? N Message-ID: <E43037234EFCCF11B99B0000F8014A40034935E1@knxwbnois1b.wbn.tva.gov>   show memory    > ----------& > From: 	Joe[SMTP:joecarlos@YAHOO.COM]' > Sent: 	Monday, April 24, 2000 1:01 PM   > To: 	INFO-VAX@LISTSERV.UGA.EDU: > Subject: 	how do i kwno how much memory is on my system? > D > is there a commnad that tells me the physical memory of my system? >    ------------------------------  % Date: Mon, 24 Apr 2000 10:51:25 -0700t' From: "George Shouse" <GHS@Shouses.com>T; Subject: Re: how do i kwno how much memory is on my system?.2 Message-ID: <sK%M4.317$kd6.802@newsfeed.slurp.net>  * Joe <joecarlos@yahoo.com> wrote in message/ news:8F20856F0joeyahoocom@news.supernews.com...tD > is there a commnad that tells me the physical memory of my system?   $ HELP SHOWh $ HELP SHOW MEMORY     --
 George Shousea
 -------------s Always a Laker Fan   ------------------------------  % Date: Mon, 24 Apr 2000 13:44:56 -0400a0 From: Glenn and Mary Everhart <Everhart@gce.com>5 Subject: Re: Infoserver CD-R (was: Verify of Backups)t' Message-ID: <39048818.62AEF47D@gce.com>   G Just sent you the Fall 1999 sigtape cd. Sent to several others as well.o Glenn Everhart     Tim Shoppa wrote:  >  > Richard L. Dyson wrote:eQ > >         I make OpenVMS (i.e., ODS-2) CD-R discs all the time.  I too used the  > > above geocitesJ > > references, amoung others.  The first of two methods I found that work > > best (for me) are toK > > master an ISO 'image file' on an OpenVMS box with a free program from ae > > guy from Hungry thatL > > works on VAX and Alpha.  Then transfer the ISO file to my Win95 box with > > EasyCDPro and burn aJ > > CD-R disc.  Note, I also have all the versions of EasyCDPro and EasyCD > > Creator and nowaF > > CD Creator Deluxe and none of them will accept that the ISO file I? >                                                           ^^^a > > provide. > < > Small nitpick (but it is, IMHO, an important distinction): > I > An ODS-2 disk image is *not* an "ISO file"!  In common usage "ISO file"dG > seems to mean "CD-ROM image", but not all CD-ROM images are ISO-9660,k* > and an ODS-2 disk image certainly isn't. > I > Of course, maybe you're just copying Adaptec's terminology, and they're-D > no hero when it comes to proper usage.  After all, they sell theirB > SCSI host adapters as "SCSI controllers", when a SCSI controller > it is *not*. >  > Tim.   ------------------------------  % Date: Mon, 24 Apr 2000 08:26:35 -0300l1 From: "Boyle, Darren" <boyledj@bankofbermuda.com>e/ Subject: RE: LK46W-A2 & "DO" key in Reflections K Message-ID: <F150836441C5D311A11700508B6FF01A8356E9@bdant024.bda.bobda.com>   
 Thanks, Marc,   ? 	I have been through the Keyboard map and have followed all thehH obvious steps (click on the keyboard map, click the LK46W DO Button, mapF them together), but still it wont take.  I can get around it buy usingF various other key sequences or even by using the mouse on a map of theK keyboard, but I like to use the DO button, I've been doing it for years and0L it's easy, don't get me wrong I'm not against  change, hey I like change butH I also like to know how to fix those irritating little things like this. Thanks,t Darren   > ----------. > From: 	Marc Lippmann[SMTP:marc@lippmann.com]) > Sent: 	Thursday, April 20, 2000 4:33 PM  > To: 	Info-VAX@Mvb.Saic.Com2 > Subject: 	Re: LK46W-A2 & "DO" key in Reflections > G > heh heh heh. Hey, Norm.  Darren, on the current version of ReflectiontG > for UNIX and Digital (V7) you'll find it under the SETUP/Keyboard MapnH > menu option. On older versions, the same functionality is there, but IF > don't remember if the menu location was identical. Then you can save; > the profile which will make the mapping change permanent.s > G > On Tue, 18 Apr 2000 13:48:46 -0400, norm.raphael@jamesbury.com wrote:o >  > >h > >u > >a > >e > >Hi, > > G > >     Can anyone tell me how to map the "DO" key in reflections so iteK > >works under a TPU OpenVMS 7.1 session, also the F17/18/19/20 keys do not K > >appear to send the appropriate escape sequence.  I have been through the:G > >Keyboard map in reflections and mapped the PC keyboard to the US DEC0 > LK4500$ > >keyboard but it still won't work. > >TIA > >- DarrencL > >------------------------------------------------------------------------- > ---sE > >------------------------------------------------------------------e> > >Darren James Boyle,              Telephone : (441) 299-6425 > >Lead Analyst Programmer,d= > >ISD VMS Technical Services,  Fax          : (441) 299-6502e > >Bank of Bermuda3 > >Par-la-Ville Branch,                EMail      :h > BoyleDJ@BankofBermuda.com  > >Hamilton Bermuda.L > >------------------------------------------------------------------------- > ---lE > >------------------------------------------------------------------a > >t< > >I always just use F16=DO by <Shift-F6> on my PC keyboard. > >n > >r > >r >     F **********************************************************************C This message and any files transmitted with it are confidential andoJ may be privileged and/or subject to the provisions of privacy legislation.M They are intended solely for the use of the individual or entity to whom theywL are addressed. If the reader of this message is not the intended recipient, B please notify the sender immediately and then delete this message.I You are notified that reliance on, disclosure of, distribution or copyingr of this message is prohibited.   Bank of BermudaaF **********************************************************************   ------------------------------  # Date: Mon, 24 Apr 2000 10:38:28 GMTs' From: Colin Blake <colin@theblakes.com>h! Subject: Re: Mozilla M15's out... - Message-ID: <39042423.C1CF008E@theblakes.com>,   Brian Wheeler wrote:  . > Dumb question:  can it be built for the VAX?  H No. OpenVMS for VAX doesn't contain the required threading capabilities.   ------------------------------  # Date: Mon, 24 Apr 2000 10:42:57 GMTa' From: Colin Blake <colin@theblakes.com> ! Subject: Re: Mozilla M15's out... , Message-ID: <39042530.DA9CD35@theblakes.com>   Brian Wheeler wrote:  . > Dumb question:  can it be built for the VAX?  d By the way, this question and more are answered in the newly expanded FAQ section in the Mozilla for OpenVMS Release Notes:  M http://www.openvms.digital.com/openvms/products/ips/mozilla_relnotes.html#faqo   ------------------------------  # Date: Mon, 24 Apr 2000 10:45:27 GMTY' From: Colin Blake <colin@theblakes.com>l! Subject: Re: Mozilla M15's out...c- Message-ID: <390425C5.DC57796D@theblakes.com>i   Wayne Sewell wrote:a  L > That is not a factor with a self-extracting zip file, which can be done asL > easily with zip as with dcx.   The user does not need to have zip or unzipP > installed on his system.   What, the run command is a tool not in the standardM > openvms distribution?  If you run a program and get a file, what difference P > does it make to the user what algorithm is used?   Other than the much smaller" > size of the zip file, of course.    M16 should ship as a ZIPSFX kit.   ------------------------------  % Date: Mon, 24 Apr 2000 10:44:01 -0400 # From: Jim Agnew <agnew@hsc.vcu.edu>  Subject: Re: NT Breaks Again+ Message-ID: <39045DB1.23F09D15@hsc.vcu.edu>1  > hahahahahaahaha.......   shades of Ringworld, by Larry Niven..   ever read it??   j   ( Andrew Harrison SUNUK Consultancy wrote: >  > Art Rice wrote:D > # > > Ya just gotta love NT security:e > >m5 > > http://news.cnet.com/news/0-1003-200-1707928.htmlu > >r > > -- > > Art Rice   *# ' > > Special Data Processing CorporationT* > > --------------------------------------* > > All opinions expressed are mine and do) > > not reflect the views of my employer.e > @ > On a simular vein this is a bug report from Microsoft that may > amuse you. > J > check out http://support.microsoft.com/support/kb/articles/Q131/1/09.asp > / > It will the the Microsoft Flat Earth next :-)k > 	 > Regardsi > Andrew Harrisonn > Enterprise IT Architect    ------------------------------  % Date: Mon, 24 Apr 2000 10:50:09 -0700 ! From: Shane.F.Smith@healthnet.comb Subject: Re: NT Breaks Again8 Message-ID: <882568CB.0061FC6C.00@WHDOM99.HEALTHNET.COM>  P It was only the first edition of Ringworld where the ring rotated the wrong way.J The second editon and onwards have that and a few other technical glitchesN fixed. It's not until "The Ringworld Engineers" that he gets the attitude jetsP in place, that one was found by a bunch of physics students who were working out the numbers as an exercise.d  ? Shane (who actually reads the author's intro section sometimes)v                7 Jim Agnew <agnew@hsc.vcu.edu> on 04/24/2000 07:44:01 AM P                                                                                 P                                                                                 P                                                                                     >                                                               >                                                               >                                                               >  To:      Info-VAX@Mvb.Saic.Com                               >                                                               >  cc:      (bcc: Shane F Smith/WH/HN/FHS)                      >                                                               >                                                               >                                                               >  Subject: Re: NT Breaks Again                                 >                                                                             > hahahahahaahaha.......   shades of Ringworld, by Larry Niven..   ever read it??   jV  ( Andrew Harrison SUNUK Consultancy wrote: >0 > Art Rice wrote:n > # > > Ya just gotta love NT security:v > >r5 > > http://news.cnet.com/news/0-1003-200-1707928.htmle > >  > > -- > > Art Rice   *#U' > > Special Data Processing Corporationn* > > --------------------------------------* > > All opinions expressed are mine and do) > > not reflect the views of my employer.n > @ > On a simular vein this is a bug report from Microsoft that may > amuse you. >nJ > check out http://support.microsoft.com/support/kb/articles/Q131/1/09.asp >T/ > It will the the Microsoft Flat Earth next :-)- >-	 > Regards- > Andrew Harrison- > Enterprise IT Architect    ------------------------------  # Date: Mon, 24 Apr 2000 12:55:04 GMT ) From: "Neil Rieck" <n.rieck@sympatico.ca>t Subject: OpenVMS marketing; Message-ID: <IuXM4.70168$2D6.1806433@news20.bellglobal.com>-   Folks,  H The problem with OpenVMS marketing is not Compaq, it's the people in the OpenVMS marketing group.  C Check this out: if you're interested in getting involved with Tru64pL (previously known as DUNIX) then all you have to do it plunk down US$99. SeeL http://www.unix.digital.com/noncommercial-unix/ for more details. IMHO, thisD isn't much different than US$75 for a Solaris-8 CD-ROM (which can be@ downloaded and run for free on machines with fewer than 8 CPUs).  G Now compare this to getting an OpenVMS Hobbyist license. First you mustvI become a DECUS member (which is difficult if not living in the US) and isuF not free (C$50 in Canada unless you're a full time student). After you2 finally get your member id, you then need to go toJ http://www.montagar.com/hobbyist/media.html to order a media kit for US$30 and free OpenVMS license.o  J The OpenVMS marketing group should just produce a no-hassle US$99 kit like the Tru64 marketing people did.e  
 Neil Rieck* Kitchener(New Berlin?)/Waterloo/Cambridge, Ontario, Canada.! http://www3.sympatico.ca/n.rieck/a6 http://www3.sympatico.ca/n.rieck/docs/alpha_diary.html   ------------------------------  # Date: Mon, 24 Apr 2000 13:51:00 GMT * From: young_r@eisner.decus.org (Rob Young) Subject: Re: OpenVMS marketing' Message-ID: <2000Apr24.095100.1@eisner>o  g In article <IuXM4.70168$2D6.1806433@news20.bellglobal.com>, "Neil Rieck" <n.rieck@sympatico.ca> writes:- > Folks, > J > The problem with OpenVMS marketing is not Compaq, it's the people in the > OpenVMS marketing group. > E > Check this out: if you're interested in getting involved with Tru64nN > (previously known as DUNIX) then all you have to do it plunk down US$99. SeeN > http://www.unix.digital.com/noncommercial-unix/ for more details. IMHO, thisF > isn't much different than US$75 for a Solaris-8 CD-ROM (which can beB > downloaded and run for free on machines with fewer than 8 CPUs). > I > Now compare this to getting an OpenVMS Hobbyist license. First you mustjK > become a DECUS member (which is difficult if not living in the US) and is0H > not free (C$50 in Canada unless you're a full time student). After you4 > finally get your member id, you then need to go toL > http://www.montagar.com/hobbyist/media.html to order a media kit for US$30 > and free OpenVMS license.t >   A 	Maybe you would have more impact if you got your facts straight. > 	Sustaining membership costs $40 for a year.  Basic membership	 	is FREE.)  * http://www.decus.org/Membership/join.shtml  B 	So by your own analysis, with FREE membership, the DECUS hobbyist 	offer is better!e  C 	Of course, "after you finally get your member id."  Oh the pain of % 	waiting so patiently by the mailbox.c   				Robe   ------------------------------  # Date: Mon, 24 Apr 2000 14:54:06 GMTe0 From: "Terry C. Shannon" <shannon@world.std.com> Subject: Re: OpenVMS marketing& Message-ID: <FtIzKG.Ey2@world.std.com>  7 "Rob Young" <young_r@eisner.decus.org> wrote in messageo! news:2000Apr24.095100.1@eisner... J > In article <IuXM4.70168$2D6.1806433@news20.bellglobal.com>, "Neil Rieck" <n.rieck@sympatico.ca> writes:
 > > Folks, > >mL > > The problem with OpenVMS marketing is not Compaq, it's the people in the > > OpenVMS marketing group. > >i <snip> > >nK > > Now compare this to getting an OpenVMS Hobbyist license. First you mustwJ > > become a DECUS member (which is difficult if not living in the US) and isJ > > not free (C$50 in Canada unless you're a full time student). After you6 > > finally get your member id, you then need to go toH > > http://www.montagar.com/hobbyist/media.html to order a media kit for US$30  > > and free OpenVMS license.v > >h >iB > Maybe you would have more impact if you got your facts straight.? > Sustaining membership costs $40 for a year.  Basic membershipy
 > is FREE. >r, > http://www.decus.org/Membership/join.shtml >GC > So by your own analysis, with FREE membership, the DECUS hobbyistn > offer is better! >/  J Rob, the gentleman **does** have his facts straight... basic membership isG free in the DECUS US Chapter; some international chapters--the new CACUt= organization in Canada, f'rinstance--assess a membership fee.-  E Still, it's significant to note that the OpenVMS Hobbyist License washF available long before the Tru64 Technology Enthusiast License. And theL Dallas-Fort Worth Compaq Users Group is largely responsible for the birth of the OpenVNS Hobbyist License.h   ------------------------------  # Date: Mon, 24 Apr 2000 13:20:16 GMTy0 From: sfm1115@bjcmail.carenet.org (Shawn Malone)8 Subject: Problem with upgrading an Alpha Server to 7.2-11 Message-ID: <390443c8.842282257@news.starnet.net>1  F I began upgrading a DEC 3000 Alpha Server on Thursday from OpenVms 7.1@ to 7.2-1.  The server was running UCX 4.2 and DecNet Phase IV.    F I ran the upgrade and the OS portion loaded without a problem.  I then; upgraded UCX to TCPIP050A, again no problem, I have TCP/IP.r  : I applied the VMS721_UPDATE-V0100 patch, again no problem.  A I then set out to upgrade my DecNet to DECNET-PLUS-V7_2.  This is B where the problem began.  The first thing it did was complain that= GBLSECTIONS, GBLPAGES, GBLPAGFIL and KSTACKPAGES were to low.   B I was able to modify the first 3, but it is not letting me set theB KSTACKPAGES.  I have tried editing the ModParams.dat file and haveC included both MIN_KSTACKPAGES=2 and KSTACKPAGES=2.  Then reboot the6C system.  When I look at tha params after the reboot it goes back to  one.  ? DecNet is giving me the error Net$Startup-F-NOREQSYSGEN, SYSGENi" parameters not at required values.  : I have also installed the DNVOSI072ECO01 patch for Decnet.  > When installing this file the firs thing that happened was the following error:  7 The following product will be installed to destination::F     DEC AXPVMS DNVOSIECO01 V7.2            DISK$ALPHASYS:[VMS$COMMON.]B %PCSI-I-RETAIN, file [SYSLIB]DTSS$SHR.EXE was not replaced because) file from kit has lower generation numbert  C Can someone please give me some ideas of what I need to do next and-0 what I need to do to get the KSTACKPAGE to hold.   Thanks   Shawn    ------------------------------  % Date: Mon, 24 Apr 2000 09:59:40 -0500n* From: WILLIAM WEBB <WWEBB1@email.usps.gov>5 Subject: Re[2]: Calculating Discordian dates from DCL - Message-ID: <0033000021341576000002L062*@MHS>r  H =0AIn article <39007ab7.524144494f47414741@radiogaga.harz.de>, Martin V=	 orlaender  wrote:4 > From: martin@RADIOGAGA.HARZ.DE (Martin Vorlaender) > Newsgroups: comp.os.vmsn4 > Subject: Re: Calculating Discordian dates from DCL' > Date: Fri, 21 Apr 2000 17:58:47 +0200a >E' > Paul Sture (sture.ch@sture.ch) wrote:I > : Mpatt644 wrote:VH > : >  Working in the finance sector I keep getting asked for the curre= nt datea > : > in Discordian format, ...a > :e > : ???? > :lH > : I may be missing the point here, but what has the Discordian format=  to do > : with the finance sector? > 
      SNIP:  D      I thought Discordians were those Church of the Subgenius folks?  )      You know, J.R. "Bob" Dobbs and such?o        WWWebb=   ------------------------------  # Date: Mon, 24 Apr 2000 15:29:38 GMTe* From: "David Cressey" <David@DCressey.com>/ Subject: Re: The Movie "Breaking Point" + PDP11O5 Message-ID: <CLZM4.162$u4.19518@petpeeve.ziplink.net>   < I remember Stevie Wonder with his DEC Rainbow synthesizer...    ! David Turner wrote in message ...CK >Anyone ever noticed the title in the beginning of the movie Breaking Point  ?  > C >In the "music by" section it gives some guy's name with "his PDP11 	 computer"t   ------------------------------  # Date: Mon, 24 Apr 2000 17:24:48 GMTo) From: healyzh@aracnet.com (Zane H. Healy)e/ Subject: Re: The Movie "Breaking Point" + PDP11D4 Message-ID: <Ar%M4.377$pK4.8050@typhoon.aracnet.com>  ] In article <sg7b57c5cdo172@corp.supernews.com>, "David Turner" <d_b_turner@yahoo.com> writes: O |> Anyone ever noticed the title in the beginning of the movie Breaking Point ?  |> eO |> In the "music by" section it gives some guy's name with "his PDP11 computer"- |> - |> Wow !  I So was he using the PDP-11 as a syth or what?  ISTR, that there was some  K sound hardware available, but don't remember the specifics.  I'd love to be ' able to hook my /73 up to my stereo :^)6  1 |> Shame they never pushed Alpha's that way huh ?  |> I |> How about this: |> yF |> "Indepedence day  - rendering by Digital Alpha Technology  Personal |> Workstation 433a6, |> Call 1 800 Digital for sales information" |>  N |> Instead it was Shhhhhhh... don't let them know they were using Linux on ourH |> Alpha's - in fact, don't let them know they were using Alpha's at all |> e |> Can we say Pathetic ?????  M Doubly so, I didn't even know they'd used them on "Independence Day", though o I did know about Titanic.o   			Zanen   ------------------------------  # Date: Mon, 24 Apr 2000 07:08:53 GMTr4 From: LESLIE@209-16-45-102.insync.net (Jerry Leslie)< Subject: Re: UCX "SET NOHOST/NOCONFIRM" Wipes Out Host Table% Message-ID: <9qSM4.83$w6.2011@insync>t  * Jerry Leslie (leslie@clio.rice.edu) wrote:; : The following UCX command wipes out the local host table:( : ! :     $ ucx set nohost /noconfirmo : G : This seems like a bug since no host was specified, per the UCX "help"bI : on "set host", included below. The command should require an "*"; e.g.:e : # :     $ ucx set nohost /noconfirm *n : I : I haven't tried this yet on a V5.0A DIGITAL TCP/IP Services for OpenVMSm	 : system.w :   / I have now, and V5.0A has the same bug/feature:      $ ucx show version  ;     DIGITAL TCP/IP Services for OpenVMS Alpha Version V5.0AU=     on a Digital Personal WorkStation  running OpenVMS V7.2-1p     $ tcpip show host /local          LOCAL databasee     Host address    Host name   &   127.0.0.1       LOCALHOST, localhost    138.55.251.38   sccvxf, SCCVXF   134.163.248.61  titanr     $ ucx set nohost /noconfirm-     $ tcpip show host /local1   %TCPIP-E-HOSTERROR, cannot process host request-*   -TCPIP-W-NORECORD, information not found   -RMS-E-RNF, record not found    E Jerry Leslie  leslie@clio.rice.edu  (my opinions are strictly my own)g   ------------------------------  # Date: Mon, 24 Apr 2000 06:28:04 GMT ' From: Roose Chua <roose_chua@yahoo.com>h Subject: UCX SMTP Mail Headers) Message-ID: <8e0phc$ld5$1@nnrp1.deja.com>e   Hi!   G Would just like to know if it is possible to remove the RFC 822 headers-G that we receive whenever an SMTP mail is received from a non-VMS server-G (i.e. MS Exchange, internet)? Also, how is it that on our two clusters,.C the location of these headers are different: one is before the mail*E message itself while the other cluster has it after the mail message?t( Any advise would be greatly appreciated!   Sample mail:  /     # From                 Date         Subjecta  ,     1 SMTP%"roose_chua@yah 24-APR-2000  TEst   MAIL> 1t           #1          24-APR-2000 4 13:57:05.63                                  NEWMAIL From:   SMTP%"me@somewhere.com"a To:     system@bogus.com CC:g Subj:   TEst   This is a test  5 ================== RFC 822 Headers ==================e Return-Path: me@somewhere.com & Received: from nowhere.com ([1.1.2.6])7          by bogus.com (UCX V4.2-21C, OpenVMS V7.1 VAX);-'         Mon, 24 Apr 2000 13:57:02 -08000	 Received:M ...r    & Sent via Deja.com http://www.deja.com/ Before you buy.M   ------------------------------  % Date: Mon, 24 Apr 2000 11:48:33 -0400e- From: Jonathan Boswell <jsb@ost.cdrh.fda.gov>o" Subject: Re: UCX SMTP Mail Headers0 Message-ID: <39046CD1.493450CE@ost.cdrh.fda.gov>   Roose Chua wrote:-I > Would just like to know if it is possible to remove the RFC 822 headerslI > that we receive whenever an SMTP mail is received from a non-VMS server3  I Please avail yourself of the excellent HELP facility within UCX/TCPIP viab thel  ) ucx> help set configuration smtp /optionsn  P command.  You will find you can place the headers at the top, bottom, or get rid of them entirely.p   ------------------------------  # Date: Mon, 24 Apr 2000 07:28:06 GMTo From: Dirk Munk <munk@home.nl>* Subject: Re: UCX TCP/IP Packet Window Size' Message-ID: <3903F74E.E268B3A6@home.nl>   
 Hi Arturo,  = Seems to me as if you are confusing two different parameters.mG The MaxMTU size is the maximum packet size you want to use on a certainn
 interface.F It is limited to the maximum size that is possible on the architectureD used on that interface, and that is 1500 byte for ethernet, which is also the default.n1 For FDDI for instance it is a bit more than 4 kB.   H The 8760 byte parameter is the Receive Window Size, and it specifies how8 many packets the sender can transmit without receiving a acknowledgement.  G A TCPIP ethernet packet without headers has a size of 1460 byte, so the-- window size in NT has a default of 6 packets. E When a connection is set up the window size is negotiated between thee sender and the receiver.D Increasing the window size can give you a much better performance inG certain cases, I run my windooz 98 box with a window size of 32 packets  (= 32 x 1460 byte).e   regards,   Dirk         arturo saavedra wrote: > 
 > Hi all.. > C > I have been digging around the UCX documentation but I'm a little E > confused as to the correct way to define the size of packet through- > UCX. > E > I know that the Ethernet standard size is 1500 bytes.  If I were to G > want to match Windows's NT default size of 8760 bytes/packet, is this C > done through $ UCX set prot tcp/quota=(send:8760,receive:8760) oroF > through $ UCX set service/socket=(send:8760,receive:8760) XXX  where0 > XXX is whichever service I'm trying to modify? > F > To confuse me even more, there is also a mentioning of TCP qualifierA > that either enables or disables MTU (Maximum Transmission Unit)e	 > sizing.j >  > Thanks in advance! > C > p.s. reason for the attempt to match NT configuration was that ofhF > troubleshooting a pesky emulation software to server network related
 > problem. > C > UCX version 4.2 eco 2 running on Alpha Server 8400, OpenVMS 7.1-2o   ------------------------------  # Date: Mon, 24 Apr 2000 12:55:17 GMTe From: briggs@eisner.decus.org  Subject: Re: Verify of Backups' Message-ID: <2000Apr24.085517.1@eisner>   b In article <B5265C3A-EA642@165.247.45.143>, "Robert Deininger" <rdeininger@mindspring.com> writes:? > On Fri, Apr 21, 2000 12:19 PM, briggs@eisner.decus.org wrote: & >>The driver can command the drive to: >> >>	Read forward  >>	Read reverse  >>	Skip forward n records  >>	Skip forward n tape marks >>	Skip backward n records >>	Skip backward n tape marks  >>	Write a record  >>	Write a tape mark? >>	(Ignoring a few more esoteric commands of limited relevance)  >>E >>Not surprisingly, these are the same operations that a user program  >>can request of the driver. > G > At or around 7.1, MKDRIVER has the fast-skip modifier which works for  > (some) newer tape drives.   E And if you read the documentation, you'll see that the IO$M_ALLOWFAST A modifier allows the driver to implement an IO$_SKIPFILE operation F by commanding the drive to skip by tape marks.  By default, the driver& commands the drive to skip by records.  7 Both of those commands appear in the list I gave above.   J >>On drives with accelerated seek capabilities (like the DLT drives have),L >>the drive knows where various landmarks on the tape reside.  For instance,J >>"the start of serpentine track 34 is 17 tape marks plus 435 records in".+ >>The driver does not share this knowledge.  > B > How sure are you of this?  It doesn't match my recollection from > scanning the DLT 4000 manual.   D No, I'm not certain.  But it's the only implementation that fits the facts.  @ > Does the drive's extra information last only until the tape isB > dismounted?  I don't recall anything about the drive saving thisD > info on the tape (for example in a special area near the beginning > of tape).p  4 I recall that the drive does save this info on tape.  : > How many "shortcuts" can it store?  What happens when it > runs out of room?o  @ Beats me.  If I were implementing it, I'd store one landmark forF each end of the tape at each possible head position.  A couple hundred landmarks total.  G >>When the driver issues a "skip 5 tape marks" command, the drive takescG >>its current position in tape marks and records (it keeps track), addstF >>the seek offset and tries to find a match in the tape directory.  ItG >>it can then quickly skip to a landmark near the final destination andhH >>do an unaccelerated seek from there.  I've never seen this documented. >>But it seems obvious to me.d > G > I recall nothing in the SCSI command set that matches this.  Maybe ithI > is completely transparent.  But I don't have the manual in front of me..G > (I'm speaking specifically about DLTs, I've no experience with DATs.)s  C Since it's the drive doing it, it is indeed completely transparent.a  @ > What I do recall seeing are two SCSI commands that are roughlyH > Tell me the "internal block number" at the current tape position, and @ > Skip to a given "internal block number".  I believe the latterE > command takes the shortest path, rather than the logically-linear, rE > physically-serpentine simple path.  The trick, I think, is teachingeD > your application to know which "internal block number" to ask for.H > I haven't found an MKDRIVER function code or modifier that correspondsB > to these functions, which is why I doubt VMS supports them.  ButE > there could be an undocumented mechanism for VMS-internal use only.u  F Interesting.  I agree that this would be useless for VMS with existing& application software and $QIO support.  & 	John Briggs			briggs@eisner.decus.org   ------------------------------  # Date: Mon, 24 Apr 2000 14:52:48 GMTg From: briggs@eisner.decus.org  Subject: RE: Verify of Backups' Message-ID: <2000Apr24.105248.1@eisner>e  b In article <B52658D0-DD8D7@165.247.45.143>, "Robert Deininger" <rdeininger@mindspring.com> writes:> > On Fri, Apr 21, 2000 7:45 AM, briggs@eisner.decus.org wrote:B >>In article <B524D10D-2398D@165.247.26.226>, "Robert Deininger" <$ >>rdeininger@mindspring.com> writes: > G >>> I do my backups with /verify, with multiple save sets per DLT tape.oH >>> Backup does NOT go back to the beginning of the tape for each verifyJ >>> on our systems. It just backs up to the beginning of the last save set >>> and compares it to disk. >> >>How do you know? > B > When one of my co-workers asked me this question, I sent him the1 > following e-mail (I never throw anything away):r > ? >     Here is a little summary of tonight's incremental backup.u	 [snipped]    And here's my test.y  I OpenVMS/Alpha 7.1  Backup link date 25-NOV-1996, image ID X6C7-0040069227b $1$MUA9 is a TLZ06 on an HSJ-40d  2 Test 1:  Append large save set (6000 blocks or so)  J $ backup sys$Login:b*.* $1$Mua20:test.bck /save /verify /ignore=label /log   Start time 			8:59:38t/ Files start hitting the tape    9:38:00 (or so)a! Backup pass done		9:38:26 (or so)e/ Verification pass starts       10:17:00 (or so)2/ Verification pass done         10:17:30 (or so)"  > That's 38 minutes to find EOT and 39 minutes to find it again.E During the pause before the verification pass, Control-T consistently>G showed a delta in I/O count of a multiple of 3 plus 2.  Since Control-TMH has an overhead of 2 I/O per use, I inferred that BACKUP was hitting theH device driver with three I/O operations per saveset skipped over.  OtherC evidence ruled out the possibility of a multiple of 3 I/Os per save E set.  It is conceivable that the device driver is emulating a massiveiF skip file operation from BACKUP and racking up the I/O count that way.  2 Test 2:  Append a small saveset (couple of blocks)?          Note: Tape still positioned at EOT from previous test.   N $ backup sys$Login:login.com $1$mua9:test2.bck /ignore=label /verify /norewind  $ Start time:		       10:18:00 (or so)/ Files start hitting the tape   10:20:00 (or so)a& Backup pass done	       within seconds- Verification pass starts       within seconds - Verification pass done         within secondsF  C Control-T showed significant I/O count during the two minute delay.r0 There was no observed multiple of three pattern.  D Based on this test the 1000 block limit would appear to be in force.6 This does fly in the face of the results you reported.  A One plausible theory to reconcile the two sets of results is thatiD BACKUP implements the "skip to beginning of this save set" for large@ save sets as a rewind followed by a massive forward IO$_SKIPFILE@ operation.  On my hardware, the device driver emulates this as aA series of skip record operations and counts one I/O per tape mark E found (three tape marks per save set is standard).  On your hardware, C the device driver ships the skip file request straight to the driver  which then accelerates the seek.  & 	John Briggs			briggs@eisner.decus.org   ------------------------------  # Date: Mon, 24 Apr 2000 16:01:53 GMT  From: briggs@eisner.decus.orga Subject: Re: Verify of Backups' Message-ID: <2000Apr24.120153.1@eisner>P  b In article <B5265C3A-EA642@165.247.45.143>, "Robert Deininger" <rdeininger@mindspring.com> writes:? > On Fri, Apr 21, 2000 12:19 PM, briggs@eisner.decus.org wrote:7J >>On drives with accelerated seek capabilities (like the DLT drives have),L >>the drive knows where various landmarks on the tape reside.  For instance,J >>"the start of serpentine track 34 is 17 tape marks plus 435 records in".+ >>The driver does not share this knowledge.t > B > How sure are you of this?  It doesn't match my recollection from > scanning the DLT 4000 manual.  > @ > Does the drive's extra information last only until the tape isB > dismounted?  I don't recall anything about the drive saving thisD > info on the tape (for example in a special area near the beginningE > of tape).  How many "shortcuts" can it store?  What happens when itN > runs out of room?r  ! I found a confirming reference at   / http://www.dlttape.com/About/university/u07.aspo  E On the tape prologue (prior to BOT) there are three zones.  A scratchXB area, a calibration area and a directory area.  The directory area is used for fast seek.  G What I've never seen is documentation on exactly how the directory areamI works.  What it stores, how it is indexed, how big it is, how it is used,-E when it is updated, how it is cached, whether it can ever be accessed B by an application and how errors are handled.  I've already postedI some of my conjectures in that regard.  I won't bore you with them again.t  & 	John Briggs			briggs@eisner.decus.org   ------------------------------    Date: 24 Apr 2000 12:19:30 -04004 From: "Robert Deininger" <rdeininger@mindspring.com> Subject: RE: Verify of Backups+ Message-ID: <B529EC54-4A459@165.247.46.172>   = On Mon, Apr 24, 2000 10:52 AM, briggs@eisner.decus.org wrote:L >And here's my test. >0J >OpenVMS/Alpha 7.1  Backup link date 25-NOV-1996, image ID X6C7-0040069227  >$1$MUA9 is a TLZ06 on an HSJ-40  E I'm not up on the DEC part numbers.  What kind of beastie is a TLZ06?o Some kind of 4mm DAT?e  3 >Test 1:  Append large save set (6000 blocks or so)b >wK >$ backup sys$Login:b*.* $1$Mua20:test.bck /save /verify /ignore=label /log  >I >Start time 			8:59:380 >Files start hitting the tape    9:38:00 (or so)" >Backup pass done		9:38:26 (or so)0 >Verification pass starts       10:17:00 (or so)0 >Verification pass done         10:17:30 (or so) >r? >That's 38 minutes to find EOT and 39 minutes to find it again."  . Ok, it looks like it went back to the the BOT.  F >During the pause before the verification pass, Control-T consistentlyH >showed a delta in I/O count of a multiple of 3 plus 2.  Since Control-TI >has an overhead of 2 I/O per use, I inferred that BACKUP was hitting theII >device driver with three I/O operations per saveset skipped over.  OtherrD >evidence ruled out the possibility of a multiple of 3 I/Os per saveF >set.  It is conceivable that the device driver is emulating a massiveG >skip file operation from BACKUP and racking up the I/O count that way.c >.3 >Test 2:  Append a small saveset (couple of blocks)>@ >         Note: Tape still positioned at EOT from previous test. >cE >$ backup sys$Login:login.com $1$mua9:test2.bck /ignore=label /verifyI	 /norewindc >"% >Start time:		       10:18:00 (or so)"0 >Files start hitting the tape   10:20:00 (or so)' >Backup pass done	       within seconds,. >Verification pass starts       within seconds. >Verification pass done         within seconds > D >Control-T showed significant I/O count during the two minute delay.1 >There was no observed multiple of three pattern.: >>E >Based on this test the 1000 block limit would appear to be in force.n7 >This does fly in the face of the results you reported.   C I don't understand the 2 minute delay.  That wasn't enough time to h> rewind and skip forward to the end.  But it seems too long forG the system to decide "yup, I'm at EOT".  (Unless a TLZ06 is an Exabyte; @ I don't remember them being quite *that* slow, but I've tried to
 forget them.)   A Certainly you saw different behaviour in the two tests before the- verify pass.  B >One plausible theory to reconcile the two sets of results is thatE >BACKUP implements the "skip to beginning of this save set" for largebA >save sets as a rewind followed by a massive forward IO$_SKIPFILEcA >operation.  On my hardware, the device driver emulates this as aaB >series of skip record operations and counts one I/O per tape markF >found (three tape marks per save set is standard).  On your hardware,D >the device driver ships the skip file request straight to the drive! >which then accelerates the seek.r  C I'm getting more confused.  When I used MKSET to force fast skip on < and off, I wasn't able to see any difference in Backup.  ButB SET MAGTAPE/SKIP=FILE was faster with fast skip.  I'll have to dig0 up those results and see if they make any sense.     ---------------------------i Robert Deininger rdeininger@mindspring.comc   ------------------------------    Date: 24 Apr 2000 12:40:07 -04004 From: "Robert Deininger" <rdeininger@mindspring.com> Subject: Re: Verify of Backups+ Message-ID: <B529F129-5C6EE@165.247.46.172>   = On Mon, Apr 24, 2000 12:01 PM, briggs@eisner.decus.org wrote:s >e" >I found a confirming reference at >n0 >http://www.dlttape.com/About/university/u07.asp   Thanks. I'll check it out.     ---------------------------o Robert Deininger rdeininger@mindspring.come   ------------------------------   End of INFO-VAX 2000.229 ************************