1 INFO-VAX	Mon, 30 Oct 2000	Volume 2000 : Issue 606       Contents: BACKUP problem Re: BACKUP problem Backup tape  Re: Backup tape  Backup tape question! boot hangs after upgrade of 7.2-1 % Re: boot hangs after upgrade of 7.2-1 6 Re: Cheap Alpha - What config is good and WHAT PRICE ?6 Re: Cheap Alpha - What config is good and WHAT PRICE ? Re: DECdtm QUESTIONS  Decwindows TPU/EVE: Window title% Re: Ethernet failure on MicroVAX 3400 % Re: Ethernet failure on MicroVAX 3400 % Re: Ethernet failure on MicroVAX 3400 % Re: Ethernet failure on MicroVAX 3400 $ Re: FTP from VMS to the demon's seed$ Re: FTP from VMS to the demon's seed Re: Galaxy doubt Re: Help with VMS 5_5_2 and GCC  Re: Microsoft gets hacked 4 Re: Narrow compatible SCSI controller for OpenVMS...+ Re: OpenVMS/VAX Installation over a Network % Re: Print drivers for OKI printers??? % RE: Print drivers for OKI printers??? % Re: Print drivers for OKI printers???  Re: Q: Why not (2^n)-bit?  RE: Rdb on VMS reference?  Re: Record not current RE: Sun "uptime" belly laugher Re: Sun "uptime" belly laugher System Crash Dump , TCPIP services: SMTP programming interface ?0 Re: TCPIP services: SMTP programming interface ?0 Re: TCPIP services: SMTP programming interface ?0 Re: TCPIP services: SMTP programming interface ?0 Re: TCPIP services: SMTP programming interface ?0 Re: TCPIP services: SMTP programming interface ? Urgent Help Needed.....   F ----------------------------------------------------------------------  % Date: Mon, 30 Oct 2000 13:02:11 +1300 " From: "kevinf" <kevinf@gns.cri.nz> Subject: BACKUP problem - Message-ID: <8tiedi$2qa8$1@news.comnet.co.nz>   , This is a multi-part message in MIME format.  + ------=_NextPart_000_0080_01C04271.9DD8E970  Content-Type: text/plain;  	charset="iso-8859-1" + Content-Transfer-Encoding: quoted-printable   F I have been thwarted by an apparent bug in AXP/VMS version 7.1 BACKUP.  I If I use the keywords /RECORD /SINCE=3DBACKUP, I expect the backup save = F set to contain all files modified since the last BACKUP /RECORD data =1 pass, or those which have not been backed up yet.   I This works correctly on our older disks, but not the two large capacity = H disks (> 20 Gbytes). They always backup the entire directory, ignoring =I the recorded backup date. They are both (apparently) Files 11 Structure = G Level 2 volumes, and the backup record date is correctly held against = E each file. One of them is the internal disk supplied with our Alpha = / server, so I'd expect that to behave correctly.   4 Here is the SH DEV /FULL for the internal hard disk:  A Disk A2G$DKA100:, device type DEC RZ1EF-BB, is online, mounted, = 
 file-oriented I     device, shareable, served to cluster via MSCP Server, error logging =  is     enabled.  H     Error count                    0    Operations completed           = 16895252H     Owner process                 ""    Owner UIC                      = [SYSTEM]=     Owner process ID        00000000    Dev Prot            =  S:RWPL,O:RWPL,G:R,W J     Reference count                4    Default buffer size              =    512J     Total blocks            35565080    Sectors per track                =    254J     Total cylinders             7001    Tracks per cylinder              =     20  J     Volume label           "SEISMO2"    Relative volume number           =      0J     Cluster size                  35    Transaction count                =      3J     Free blocks             14162155    Maximum files allowed            = 400000J     Extend quantity               72    Mount count                      =      2<     Mount status              System    Cache name         = "_A2G$DKA0:XQPCACHE"J     Extent cache size             64    Maximum blocks in extent cache   = 708107J     File ID cache size            64    Blocks currently in extent cache = 504105J     Quota cache size               6    Maximum buffers in FCP cache     =    446B     Min ret. period    7-00:00:00.00    Max ret. period          = 14-00:00:00.005     Volume owner UIC           [1,1]    Vol Prot    =  S:RWCD,O:RWCD,G:RWCD,W:RWCD   B   Volume Status:  subject to mount verification, file high-water = marking, write-        back caching enabled.     Volume is also mounted on M5G.    I Is this a known problem to the group? Is there more information I could =  supply?    Many thanks,   Kevin Fenaughty * Institute of Geological & Nuclear Sciences Wellington, New Zealand     + ------=_NextPart_000_0080_01C04271.9DD8E970  Content-Type: text/html; 	charset="iso-8859-1" + Content-Transfer-Encoding: quoted-printable   0 <!DOCTYPE HTML PUBLIC "-//W3C//DTD W3 HTML//EN"> <HTML> <HEAD>  0 <META content=3Dtext/html;charset=3Diso-8859-1 = http-equiv=3DContent-Type>8 <META content=3D'"MSHTML 4.72.3110.7"' name=3DGENERATOR> </HEAD>  <BODY>F <DIV>I have been thwarted by an apparent bug in AXP/VMS version 7.1=20
 BACKUP.</DIV>  <DIV>&nbsp;</DIV> I <DIV>If I use the keywords /RECORD /SINCE=3DBACKUP, I expect the backup =  save set=20 H to contain all files modified since the last BACKUP /RECORD data pass, = or those=20 ( which have not been backed up yet.</DIV> <DIV>&nbsp;</DIV> E <DIV>This works correctly on our older disks, but not the two large =  capacity=20 B disks (&gt; 20 Gbytes). They always backup the entire directory, = ignoring the=20 E recorded backup date. They are both (apparently) Files 11 Structure = 
 Level 2=20J volumes, and the backup record date is correctly held against each file. =	 One of=20 I them is the internal disk supplied with our Alpha server, so I'd expect = 
 that to=20 behave correctly.</DIV>  <DIV>&nbsp;</DIV> ? <DIV>Here is the SH DEV /FULL for the internal hard disk:</DIV> J <DIV><BR><FONT color=3D#000000 face=3DCourier size=3D3>Disk A2G$DKA100:, = device type=20G DEC RZ1EF-BB, is online, mounted, file-oriented<BR>&nbsp;&nbsp;&nbsp; = 
 device,=20> shareable, served to cluster via MSCP Server, error logging=20. is<BR>&nbsp;&nbsp;&nbsp; enabled.</FONT></DIV>F <DIV><FONT color=3D#000000 face=3DCourier size=3D3></FONT>&nbsp;</DIV>G <DIV><FONT color=3D#000000 face=3DCourier size=3D3>&nbsp;&nbsp;&nbsp; =  Error=20J count&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&n=1 bsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;=20 ! 0&nbsp;&nbsp;&nbsp; Operations=20 H completed&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;=20' 16895252<BR>&nbsp;&nbsp;&nbsp; Owner=20 J process&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;=! &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;=20 ' &quot;&quot;&nbsp;&nbsp;&nbsp; Owner=20 J UIC&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbs=; p;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;=20 / [SYSTEM]<BR>&nbsp;&nbsp;&nbsp; Owner process=20 I ID&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; 00000000&nbsp;&nbsp;&nbsp; =  Dev=20I Prot&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;=20 6 S:RWPL,O:RWPL,G:R,W<BR>&nbsp;&nbsp;&nbsp; Reference=20J count&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&n= bsp;&nbsp;&nbsp;&nbsp;=20 % 4&nbsp;&nbsp;&nbsp; Default buffer=20 J size&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nb= sp;&nbsp;&nbsp;&nbsp;&nbsp;=20" 512<BR>&nbsp;&nbsp;&nbsp; Total=20J blocks&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; =  ) 35565080&nbsp;&nbsp;&nbsp; Sectors per=20 J track&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&n=+ bsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;=20 " 254<BR>&nbsp;&nbsp;&nbsp; Total=20J cylinders&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbs= p;&nbsp;=20 $ 7001&nbsp;&nbsp;&nbsp; Tracks per=20J cylinder&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp=( ;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;=20 20</FONT></DIV> F <DIV><FONT color=3D#000000 face=3DCourier size=3D3></FONT>&nbsp;</DIV>G <DIV><FONT color=3D#000000 face=3DCourier size=3D3>&nbsp;&nbsp;&nbsp; = 	 Volume=20 D label&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;=208 &quot;SEISMO2&quot;&nbsp;&nbsp;&nbsp; Relative volume=20J number&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&= nbsp;&nbsp;&nbsp;&nbsp;=20" 0<BR>&nbsp;&nbsp;&nbsp; Cluster=20J size&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nb=$ sp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;=20# 35&nbsp;&nbsp;&nbsp; Transaction=20 J count&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&n=7 bsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;=20  3<BR>&nbsp;&nbsp;&nbsp; Free=20 J blocks&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&= nbsp;=20+ 14162155&nbsp;&nbsp;&nbsp; Maximum files=20 J allowed&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;= =20 & 400000<BR>&nbsp;&nbsp;&nbsp; Extend=20J quantity&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp= ;&nbsp;&nbsp;&nbsp;=20 72&nbsp;&nbsp;&nbsp; Mount=20 J count&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&n=J bsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nb= sp;&nbsp;&nbsp;=20  2<BR>&nbsp;&nbsp;&nbsp; Mount=20J status&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&= nbsp;&nbsp;=20! System&nbsp;&nbsp;&nbsp; Cache=20 7 name&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;=20 D &quot;_A2G$DKA0:XQPCACHE&quot;<BR>&nbsp;&nbsp;&nbsp; Extent cache=20J size&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nb= sp;=20B 64&nbsp;&nbsp;&nbsp; Maximum blocks in extent cache&nbsp;&nbsp;=20- 708107<BR>&nbsp;&nbsp;&nbsp; File ID cache=20 I size&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;=20 8 64&nbsp;&nbsp;&nbsp; Blocks currently in extent cache=20+ 504105<BR>&nbsp;&nbsp;&nbsp; Quota cache=20 J size&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nb= sp;&nbsp;&nbsp;=20- 6&nbsp;&nbsp;&nbsp; Maximum buffers in FCP=20 1 cache&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; =   446<BR>&nbsp;&nbsp;&nbsp; Min=20I ret. period&nbsp;&nbsp;&nbsp; 7-00:00:00.00&nbsp;&nbsp;&nbsp; Max ret.=20 ? period&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;=20 4 14-00:00:00.00<BR>&nbsp;&nbsp;&nbsp; Volume owner=20B UIC&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;=205 [1,1]&nbsp;&nbsp;&nbsp; Vol Prot&nbsp;&nbsp;&nbsp;=20 ( S:RWCD,O:RWCD,G:RWCD,W:RWCD</FONT></DIV>F <DIV><FONT color=3D#000000 face=3DCourier size=3D3></FONT>&nbsp;</DIV>B <DIV><FONT color=3D#000000 face=3DCourier size=3D3>&nbsp; Volume = Status:&nbsp; subject=202 to mount verification, file high-water marking,=20J write-<BR>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; back caching enabled.<BR>&nbsp; =	 Volume=20 $ is also mounted on M5G.</FONT></DIV> <DIV>&nbsp;</DIV>  <DIV>&nbsp;</DIV> E <DIV><FONT color=3D#000000>Is this a known problem to the group? Is = 
 there more=20 ( information I could supply?</FONT></DIV>. <DIV><FONT color=3D#000000></FONT>&nbsp;</DIV>4 <DIV><FONT color=3D#000000>Many thanks,</FONT></DIV>. <DIV><FONT color=3D#000000></FONT>&nbsp;</DIV>7 <DIV><FONT color=3D#000000>Kevin Fenaughty</FONT></DIV> C <DIV><FONT color=3D#000000>Institute of Geological &amp; Nuclear=20  Sciences</FONT></DIV> ? <DIV><FONT color=3D#000000>Wellington, New Zealand</FONT></DIV> < <DIV><FONT color=3D#000000></FONT>&nbsp;</DIV></BODY></HTML>  - ------=_NextPart_000_0080_01C04271.9DD8E970--    ------------------------------  # Date: Mon, 30 Oct 2000 01:10:17 GMT + From: John Santos <john.santos@verizon.net>  Subject: Re: BACKUP problem > Message-ID: <MPG.146693aad49d8797989689@news.bellatlantic.net>  G In article <8tiedi$2qa8$1@news.comnet.co.nz>, kevinf@gns.cri.nz says... H > I have been thwarted by an apparent bug in AXP/VMS version 7.1 BACKUP. > O > If I use the keywords /RECORD /SINCE=BACKUP, I expect the backup save set to  G contain all files modified since the last BACKUP /RECORD data pass, or  ( those which have not been backed up yet. > O > This works correctly on our older disks, but not the two large capacity disks E (> 20 Gbytes). They always backup the entire directory, ignoring the  J recorded backup date. They are both (apparently) Files 11 Structure Level E 2 volumes, and the backup record date is correctly held against each  J file. One of them is the internal disk supplied with our Alpha server, so $ I'd expect that to behave correctly.	 --------- J (Please put line breaks in your posts or use a competent news reader that I does this automatically.  It is hard to read and reply to posts with very  long lines.)  I This probably has nothing to do with disk size or age.  This has been the I behaviour of BACKUP for a long time (V6.2 or V7.0).  Have you ever done a G backup with /IMAGE and /RECORD on the new disks?  If not then the stuff B below applies.  If so, check the MODIFIED and BACKUP dates on the E directories in the path, since if and directory has been modified, it G still applies.  (Creating or deleting files doesn't change the MODIFIED E date of a directory.  The MODIFIED date is changed by such things as  > changing the protection code of the directory or renaming it.)  H If any *DIRECTORY* in the path to a sub-directory hasn't had its backup D date recorded (including the MFD, [0,0]) then all files in that sub-F directory and any of its descendant directories will be included in a H backup.  This is to prevent missed files when directories get renamed.   There are 3 solutions:  I 1) Do a proper (I.e. /IMAGE) backup of the disk with /RECORD.  This will  I set the backup dates on all directories including [000000].  This is the  H first step if you follow the procedures for doing incremental backups in the Fine Manuals.   F 2) Record the backup dates on all directories by using a dummy backup,C such as "$ backup/record disk:[000000...]*.dir nl:foo.bck/saveset". ) (Add /nocrc/group=0 to speed it up some.)   E 3) Define a logical name (V7.1 or later) that tells BACKUP to use the 
 old behavior.    RTFF (Read The Fine FAQ)   --   John Santos    ------------------------------  % Date: Sun, 29 Oct 2000 20:45:45 -0500 ) From: Steven Xie <r33300@MailAndNews.com>  Subject: Backup tape& Message-ID: <3A350448@MailAndNews.com>   Hello there,  I I have a question to ask regarding to the backup tape. I have a dlt tape   drive M attacthed to my alpha server (OS version is V7.2-1). When I do the backup, I  L do  backup/norewind... so every time when the back up get executed, it will J continue with last time's backup. My quesiton is, how do I know my backup  tape  H is getting close to the end and is there a way to determine that I need  change  	 the tape?    Many thanks, Steven  < ------------------------------------------------------------8  Get your FREE web-based e-mail and newsgroup access at:&                 http://MailAndNews.com  7  Create a new mailbox, or access your existing IMAP4 or 4  POP3 mailbox from anywhere with just a web browser.< ------------------------------------------------------------   ------------------------------  # Date: Mon, 30 Oct 2000 02:49:09 GMT * From: Alan E. Feldman <alan48@my-deja.com> Subject: Re: Backup tape) Message-ID: <8tinj4$ief$1@nnrp1.deja.com>   & In article <3A350448@MailAndNews.com>,,   Steven Xie <r33300@MailAndNews.com> wrote: > Hello there, > E > I have a question to ask regarding to the backup tape. I have a dlt  tape > drive D > attacthed to my alpha server (OS version is V7.2-1). When I do the	 backup, I E > do  backup/norewind... so every time when the back up get executed,  it will D > continue with last time's backup. My quesiton is, how do I know my backup > tapeD > is getting close to the end and is there a way to determine that I need > change > the tape?   G During the backup, if the end of tape is reached, BACKUP will ask for a G new tape. If you used the /NOASSIST qualifier, the request will be sent G to your terminal. Otherwise, the request will be sent to the console or G anyone logged in with REPLY/ENABLE=TAPES (other variations of REPLY may G apply, I'm not sure which). In the former case, you either press return E or enter yes and press return after the new tape is loaded and ready. G In the latter case, you need to issue an appropriate REPLY command. See 2 HELP REPLY or look up REPLY in the DCL dictionary.  A If you just want to know how full the tape is before starting the0E backup, well, that's more difficult. BACKUP/LIST on all the save setspC on the tape will show how large they are in blocks. If you know thenG capacity of the tape, you can then guess. I believe by default there ishG approx. 10% overhead due to redundancy groups. Someone else may be ablec to answer this in more detail.   -- Disclaimer: JMHO Alan E. Feldman  alan48  &-)  dellnet.comf    & Sent via Deja.com http://www.deja.com/ Before you buy.0   ------------------------------  % Date: Mon, 30 Oct 2000 09:58:45 +0800X+ From: Steven Xie <r33300@email.sps.mot.com>  Subject: Backup tape question 1 Message-ID: <39FCD5D5.7145CBCF@email.sps.mot.com>    Hello there,  H I have a question to ask regarding to the backup tape. I have a dlt tape drive/A attached to my alpha server (OS version is V7.2-1). When I do thes	 backup, IrF do  backup/norewind... so every time when the back up get executed, it willB continue with last time's backup. My question is, how do I know my backup tapeuG is getting close to the end and is there a way to determine that I needm change	 the tape?    Many thanks, Steven   ------------------------------  # Date: Sun, 29 Oct 2000 22:05:06 GMTs2 From: "Chris Rodriguez" <crodrigu@midsouth.rr.com>* Subject: boot hangs after upgrade of 7.2-1: Message-ID: <ma1L5.23376$Fi.66337@typhoon.midsouth.rr.com>  F I was wondering if anyone has seen this scenerio on a VMS 7.1 to 7.2-1 upgrade before?-   2 node cluster 4100's-  E After the upgrade process, the system boots it mounts the quorum diskg3 remotely the waits to join Cluster and hangs there.i  K We did the same thing in our other cluster of 2 4100's and had no problems.   Our procedure has been the same.  H Main difference is that the one we are trying to upgrade now has 6gig vs; 2gig mem and larger global sections allocated on each node.    Any ideas would be appreciated.7     Chris    ------------------------------  # Date: Mon, 30 Oct 2000 01:24:06 GMT + From: John Santos <john.santos@verizon.net>h. Subject: Re: boot hangs after upgrade of 7.2-1> Message-ID: <MPG.14669707ca6b199498968a@news.bellatlantic.net>  ; In article <ma1L5.23376$Fi.66337@typhoon.midsouth.rr.com>, t  crodrigu@midsouth.rr.com says...H > I was wondering if anyone has seen this scenerio on a VMS 7.1 to 7.2-1 > upgrade before?r >  > 2 node cluster 4100'se > G > After the upgrade process, the system boots it mounts the quorum diskh5 > remotely the waits to join Cluster and hangs there.I > M > We did the same thing in our other cluster of 2 4100's and had no problems.a" > Our procedure has been the same. > J > Main difference is that the one we are trying to upgrade now has 6gig vs= > 2gig mem and larger global sections allocated on each node.  > ! > Any ideas would be appreciated.  >  >  > Christ  H I've not had any problems upgrading clustered systems, but the behaviourH you describe is consistent with having incorrect VOTE-related parametersG in MODPARAMS.DAT.  (Maybe EXPECTED_VOTES is too high, or VOTES is 0 foroC one of the systems, or you have a QUORUM disk that isn't defined in C MODPARAMS.)  Upgrading runs AUTOGEN, which reads MODPARAMS.  If you H adjusted these parameters manually with SYSGEN or SYSMAN, you could haveC been running for ages with incorrect MODPARAMS values that wouldn'ti) cause any problems until you ran AUTOGEN.l  F There is a method to get the system unstuck from QUORUM-wait.  I'm notB sure if it works on all systems and it is different for VAXes and E Alpha's.  Basically, you interrupt the system from the console, force H a special kernel-mode interrupt which runs a special kernel-mode programE that re-computes QUORUM based on the current number of cluster votes. G The procedure is documented in the VMS Cluster Manual (or maybe the 2ndiF volume of the System Management Manual), and maybe in the FAQ.  Wish IB had a copy of the VMS doc set at home - I could tell you for sure.  F Or you could re-boot with conversational boot (definitely in the FAQ),C adjust VOTES, EXPECTED_VOTES, QDVOTES, etc. as needed) which should E un-stick you.  I'm not sure if you need to do this on both systems orvC if fixing one will get the other one running again.  Once it is up,e4 make sure MODPARAMS.DAT is correct for both systems.  D Sorry to be vague here, but it sounds like you're stuck and a littleC advice to get you going in the right direction *now* is more useful  than exact information later.e   -- n John Santos-   ------------------------------  % Date: Sun, 29 Oct 2000 20:48:37 -0500"- From: "Island Computers" <sales@islandco.com>-? Subject: Re: Cheap Alpha - What config is good and WHAT PRICE ?H/ Message-ID: <svpkfmndf4c655@corp.supernews.com>=   David9  H The fact that VMS systems need certain devices such as SCSI disk, cd rom( andController is what bumps the price up    7 I can sell a 128mb system in a box with floppy for $5890  J But then once you add all the SCSI bits n' pieces which are as we all know: overpriced (same basic disk  technology in the IDE drives)  J If the systems were (VMS Bootable with IDE ) then I could get one in under $1000 with a 15" NEW monitor   Well approximately... ;0)    Davidi     ----- Original Message -----7 From: "David J. Dachtera" <djesys.nospam@earthlink.net>0 Newsgroups: comp.os.vmsk& Sent: Sunday, October 29, 2000 1:19 PM? Subject: Re: Cheap Alpha - What config is good and WHAT PRICE ?t     > Island Computers wrote:, > >t* > > Please take a look at www.islandco.com > >a: > > Click on low cost alpha and let me know what you think > % > Well, close but no cigar. Try this:t >=? > Alpha Workstation / Server  PC164LX for $650 + shipping with:e >  > 128mb . > 4.3 GB SCSI disk (RZ29 or equiv., or better)
 > DE500-BA
 > Keyboard > mouse  >eJ > That's still more than $150 too high, but in my mind at least that wouldG > be the closest thing to an "ideal" hobbyist machine that anyone wouldc > dare ask for.e >r > Wanna REALLY clinch it?n >;: > Throw in a reburb'd 15-inch monitor and free shipping toB "David J. Dachtera" <djesys.nospam@earthlink.net> wrote in message' news:39FC6A35.597F965F@earthlink.net...p | Island Computers wrote:& | >&* | > Please take a look at www.islandco.com | >b: | > Click on low cost alpha and let me know what you think | % | Well, close but no cigar. Try this:o |e? | Alpha Workstation / Server  PC164LX for $650 + shipping with:0 |6 | 128mb>. | 4.3 GB SCSI disk (RZ29 or equiv., or better)
 | DE500-BA
 | Keyboard | mouseb |&J | That's still more than $150 too high, but in my mind at least that wouldG | be the closest thing to an "ideal" hobbyist machine that anyone would& | dare ask for.& |= | Wanna REALLY clinch it?; |sJ | Throw in a reburb'd 15-inch monitor and free shipping to the continental | US.D |s	 | F.Y.I.:s |nG | A friend of mine recently became adminstrator of a private school fornJ | the learning disabled. She needs some new computers but has only a $3000I | budget to cover at least five(5) new machines including software. AfterbG | being deeply disappointed by Compaq, I discovered that the local Best&I | Buy was selling eMachines computers - complete - for less than $500 US.sC | Each will need to be upgraded, but I can handle that in my sleep.  |< | -- | David J. Dachtera= | dba DJE Systemss | http://www.djesys.com/ |s< | Unofficial Affordable OpenVMS Home Page and Message Board:! | http://www.djesys.com/vms/soho/p |bH | This *IS* an OpenVMS-related newsgroup. So, a certain bias in postings | is to be expected. |<B | Feel free to exercise your rights of free speech and expression. |&H | However, attacks against individual posters, or groups of posters, are | strongly discouraged.s   ------------------------------  % Date: Sun, 29 Oct 2000 21:05:08 -0500p% From: JM <vmswiz@geonospamcities.com><? Subject: Re: Cheap Alpha - What config is good and WHAT PRICE ?>O Message-ID: <C1485D99D473E35E.B3429B83D3CB66A9.E77F954C04D419ED@lp.airnews.net>p   > Island Computers wrote:& > >&* > > Please take a look at www.islandco.com > >0: > > Click on low cost alpha and let me know what you think  ) Great! That's getting where I want to be.s  G $849 for my home dream machine (533mhz, 128mb mem, no disk) is not bad.&E That's close to the price I paid for the Alphastation 200 three yearsn< ago through Onsale.com. Gives me a number to save towards...  H ... and I can start hinting to the wife that I really NEED to spend that$ much money to be a happier person :)  4 Ever notice computer widows have no sense of humour?  B Any chance of lowering the price by dropping the processor speed?   @ 266mhz or 333mhz would be ideal starter systems if significantlyH cheaper, or are the old chips just unavailable or more expensive? I know= they probably aren't typical "commodity" items. What can this; motherboard handle?n  ; I'd also want to make the Cdrom reader optional (I've got a=G lightning-fast Plextor 12plex already). You might want to use IDE cdrompE readers on this box. The XP-900 at the office seems to work just finen" and plenty fast using IDE for CDs.  E The narrow scsi adapter is useful and relatively necessary in all VMSo situations.s  D How about a good (333mhz), better (433mhz), best (533mhz) selection?  C At the low end, even a $50 reduction is a big deal. We want as many;B C-code porting college students buying these things as possible :)  F Links directly to the vendors web documents on each of the items wouldE be very useful. Keep your margins up by not having to answer too manyx
 questions.   Thanks!a   			*JM*b   ------------------------------  % Date: Sun, 29 Oct 2000 20:25:05 -0000s- From: "Alan Potter" <alan_potter@hotmail.com>s Subject: Re: DECdtm QUESTIONSo/ Message-ID: <8ti13v$76p$1@newsg2.svr.pol.co.uk>;  - fabio_compaq@ep-bc.petrobras.com.br wrote ...sK > Do the DECdtm works over SCS  in a Cluster Configuration ? What means, if& Ip > begin to use SMCI (Shared H > Memory Cluster Interconnect) , it will  improve the speed of my DECdtm > transactions ???  G From memory, the amount of CPU time and network traffic taken up by the2G DECdtm part of most database transactions was miniscule compared to the0I amount of work being done by the database applications themselves.  HencepL the benefit to be gained from speeding up DECdtm traffic, as compared to anyA other parts of the transaction, would be likely to be very small.<  < Are you sure that your problem lies in the DECdtm subsystem?   regards, /alan3   ------------------------------  % Date: Sun, 29 Oct 2000 21:32:05 -0400 - From: JF Mezei <jfmezei.spamnot@videotron.ca>=) Subject: Decwindows TPU/EVE: Window title;, Message-ID: <39FCCF91.29E51728@videotron.ca>  % This one has gotten me a bit stumped.D  N The title that appears at the top of a EVE window (on a workstation), seems to, be taken from the file "SYS$LIBRARY:EVE.DAT"% which contains, amongst other things:   > Tpu.Tpu$MainWindow.ApplicationTitle:                       EVE    0 And inside of TPU, I can access that value with D TPU COPY_TEXT(GET_DEFAULT("Tpu.Tpu$MainWindow.ApplicationTitle",""))  E But how can I set that value ? I can't seem to find documentation for;I GET_DEFAULT on the documenmtation (except for the TPU built-in help), and,M there doesn't seem to be any reference to the file EVE.DAT that is being readY nor for its format.e  U I want each window to show the filename being edited as the title. (get buffer name).c  * Does anyone know how this can be changed ?   ------------------------------  % Date: Sun, 29 Oct 2000 16:16:35 -0600a) From: "John E. Malmberg" <wb8tyw@qsl.net>n. Subject: Re: Ethernet failure on MicroVAX 3400. Message-ID: <svp8b1m5r4f4f@corp.supernews.com>  4 "Chris Chiesa" <cchiesa1@rochester.rr.company> wrote= in message news:TUEK5.4145$_n.1063304@typhoon.nyroc.rr.com...p  ? > First, John E. Malmberg <wb8tyw@qsl.network> wrote in messageE+ > news:svckm9kuvsm293@corp.supernews.com...e >a; > > Has someone verified that the ethernet segment does notb2 > > have any problems and somewhat conforms to the > > specifications?h >n= > If I remember my Ethernet topological jargon, this may be ab? > non-issue as there is only ONE "segment."  The entire networkcB > is a single linear run of coaxial cables joined by T connectors.E > No routers, gateways, etc. Is this the same definition of "segment"aD > you're using, John?  If so, I should mention that there are twentyB > or so other workstations on this network / segment, all of which> > continue to function and "inernetwork" perfectly on the SAME$ > backbone as the VAX-that-does-not.  J Generally a segment can be considered any section of a network that can be easily isolated.  K Others consider a segment to consist of  a "collision domain" for ethernet.e  L Because of the nature of how ethernet works, you can have a network cable onJ the brink of total failure and most if not all systems on it can appear to	 be happy..  I Only with network monitoring tools or by watching the network counters onrG the various devices that you might have a clue that there is a problem.t  L Just because only one device is having a problem, does not totally exonerate the network configuration.  K But I agree with you that it is not likely that the network is the problem.-I It is just after replacing the VAX processor and bulkhead did not fix theaJ problem, I recommend verifying that the network, or it's connection to the VAX is not the problem.D  ? > For what it's worth, though, no, I don't think anybody's EVERl7 > been particularly concerned with whether this segment0B > "conforms to the specifications;"  I did my best, ten years ago,@ > to whip up awareness of such things as minimal-curving-radius,4 > minimum/maximum cable lengths, and maximum segment= > lengths, but not too many people paid much attention.  Moree > on this below.  K Considering that for a typical segment of thin-net, the average utilizationaL of a larger network shows at 2.5 % of bandwidth, quite a lot of problems can
 be hidden.  2 The bottom line though is if it gets the job done.   >07 > > The first thing it to verify if the problem is withc/ > > network hardware, the VAX ethernet port, or@  > > the software running on VMS. > 1 > Exactly what distinction are you making between.1 > "network hardware" and "the VAX ethernet port?" * > To me, the latter is PART OF the former.  J The network hardware is any part of the physcal network from the bulk headH of the VAX outward.  It includes cables, tranceivers, and anything else.  9 > I agree with you, but unfortunately I have no tools fore8 > this task at this time.  Best we can do is observe the9 > activity LED(s) on the transceiver(s) as we try to Ping- > to/from the VAX. >-7 > > Create a small compliant thin-net network with onlya9 > > the VAX and a network monitor on it.  [...] There are 4 > > are supposed to be downloadable network monitors2 > > for x86 PCs running LINUX, or a commercial one$ > > available for Windows NT Server. >-B > (Forgive me for quoting some of your remarks out-of-sequence...) >0? > Sounds like a great idea, if I can pull it off.  I don't have:6 > any Linux PCs or NT Server machines at my own hands,= > and I doubt that my client does either, but I'll definitely3) > look into it.  Thanks for the pointers.,  B Through an MSDN license, I have a laptop that is running a NetworkJ Monitoring program that I can easily check things with.  If I was using myJ own money for it though I would be using a LINUX based laptop.  None of my9 VMS systems are portable enough for that type of purpose.m  > > It's not entirely clear to me what you mean by a "compliant"8 > network.  Just properly-terminated, observing minimum-< > curving-radius and proper length on all coax cables, etc.?3 > Or is there something more?  I must admit we weree< > always pretty lax on this, except on the rare occasions it6 > arose to bite us and we were FORCED to deal with it.  G For a two node thin-net network, compliance should be easy, and it also L should not be as picky as a multinode network.  I do not quite remember what= exactly all of the specifications are off the top of my head.e  : > > With a network monitor working, you can try booting to< > > the internal ethernet card and watch the requests on the
 > > ethernet.D >d= > I assume by "booting to..." you mean what I'd call "bootingo@ > from..." :-)  Is it as simple as getting to the console prompt9 > and issuing a 'b esa0' (or whatever the Ethernet devicey2 > name is)?  If not, please tell me the procedure.  I B esa0: should cause the VAX 3400 to start transmitting.  Unless you havedJ something that speaks MOP on the other end, you will not be able to test a two way transmission.r  9 > > Then boot the VAX normally and use the TCP/IP programt< > > and the network monitor using the PING command.  If thatA > > works, you know the problem is in the other network hardware.  >m5 > What "other network hardware?"  Seems to me that ifu7 > "that works," we've successfully verified ALL networks5 > hardware having to do with the VAX.  Are you sayingr5 > some OTHER network hardware could be at fault here?   > Anything is possible.  We covered the other semantics earlier.  7 > (Hmm. The VAX is at ONE END of a linear backbone.  If ; > my client had recently added a workstation to the networkk6 > and exceeded the maximum cabling length,  I'd expect< > problems -- but could they be limited to "just the VAX" as4 > we are seeing?  Or would OTHER hosts have problems1 > TOO?  Currently the 20+ other hosts on the same , > backbone/segment are NOT having problems.)  9 As you say, they are NOT *appearing* to have problems :-)f  I Basically, divide and conquer.  A simple two node network can tell you ifo= you are wasting your time replacing VAX 3400 processor cards.   6 > > A time-domain reflectometer can be used to quickly$ > > find physical defects in cables. > 4 > We don't have a TDR.  I've only ever even SEEN one9 > ONCE, and that many years ago.  I believe it was rentedt7 > for an exorbitant fee, and I have no idea from where.e > Suggestions?  E I do not know about the fees, where I am now, the thin-net cables are L scanned with a TDR before they are put into service, and there is a scheduleE of retests.  But the size of the network is much larger than what youe	 describe.c  F With the proliferation of network hardware, the price of purchasing orK renting a TDR could have come down.  I would not get one just to solve this9 problem though.n  K If you can prove that the VMS system runs on a small test network, then theiL TDR and possibly a network troubleshooting service to run it could save time in finding the cause.   9 > > The things that can cause an ethernet segment to stop = > > working reliably can be endless.  An intermittent circuith4 > > that sometimes comes back to life (or dies) when8 > > something is touched is a common problem in thin-net > > ethernet circuits. >a= > Hmm.  That's interesting.  I was hoping our nine-plus yearst3 > of successful operation were a sign of stability.   D They are until someone pulls a cable too hard, and then jams it backL together.  As long as it is not bumped, it can run for a very long time. :-)  4 > > The TEE connectors can go bad this way, rare but > > it happens.t >t5 > They've been swapped, several times among a handfulS: > of connectors, so I think we've probably eliminated this; > problem.  I'll consider buying one or two brand new ones,f > though, just to be safe.  J I have had terminators fall off in my hands after someone told me they hadI checked a network thorougly and could not find out why it was not workingi
 correctly.   > ------H > Next, Robert Deininger rdeininger@mindspring.company> wrote in messageD news:rdeininger2510000008590001@user2ive7ej.dialup.mindspring.com...D  (I've edited out duplicate points already made by John; no offense, Robert.) >u9 > > > I assume "current" TCP/IP packages probably requiree; > > > newer VMS versions than V5.4-2, and that such "newer" 9 > > > VMS versions are unsupported on hardware as ancientt; > > > as a MicroVAX 3400-- although they probably WILL RUN,n > > > bless 'em... > >oL > > The SPD still lists this system as supported in current versions of VMS: > > $ > > o  MicroVAX 3400, VAXserver 3400 >uF > I am VERY glad to hear that!  Dare I ask the embarrassingly-ignorantH > question, "what version of VMS is now current?"  I'm REALLY out of the
 loop these > days.u  F OpenVMS VAX 7.2, OpenVMS ALPHA 7.2-1, with 7.3 for both in field test.  H The problem is that the version of third-party TCP/IP stack may have hadG proprietary calls in it.  If the application is expecting that brand of 5 TCP/IP, it would not run with any other TCP/IP stack.   9 > > But a VMS upgrade will require that the custom devicer= > > driver is re-assembled and re-linked.  If the source code-( > > isn't available, there will be pain.  : > > Also, I think there were some changes to device driver8 > > architecture since 5.4, so source code changes might& > > be needed.  I don't know for sure. >s< > The source code is effectively unavailable except at great9 > effort and possibly expense.  Ditto someone to make any ; > needed source changes.  It's very unlikely that my clientp? > can afford to pay for either.  However, I'll dutifully remind ; > him that "given sufficiently large amounts of money, veryl > little is impossible."  L You are not alone in this.  I suspect that your q-bus drivers may still workI as far as OpenVMS 5.5-2, but with out you testing the application, you do 	 not know.s  ; > > Vax 4000 series machines are nice, and they have Q-bus.n >0: > Now THAT is a VERY GOOD IDEA.  This product used to run,; > as-is (no driver changes!) on 4000 series machines.  I'llA > definitely suggest that.  H A VAX 4000-200, VAX 4000-300, VAX 4000-500 may be able to work at 5.4-2,& some of them may need a manditory ECO.  K Also some of the VAX 4000 processors will require minimum revisions of sometL of the Q-BUS boards.  I am assuming that the faster VAX processors uncoveredH some timing irregularities.  How this will affect your third party Q-BUS cards can not be predicted.o  6 What you need to find is the S.P.D. for VAX/VMS 5.4-2.  L If you are not dependent on it having an imbedded ethernet controller, a useK VAX 3500 or 3800 processor, memory, and either a DELQA, or a DESQA(sp?) mayr also be substituted.  J It may be cheaper for you to find a used VAX with all of the above than to just purchase the parts :-)s   -Johnu wb8tyw@qsl.network   ------------------------------  # Date: Mon, 30 Oct 2000 01:20:31 GMTl0 From: "Chris Chiesa" <cchiesa1@rochester.rr.com>. Subject: Re: Ethernet failure on MicroVAX 3400< Message-ID: <z14L5.102377$JS3.15712825@typhoon.nyroc.rr.com>  
 Hello all,  H Somehow I missed responding to Marcin Szczecinski <marcin@lodz.tpsa.pl>, who on 20 October wrote:  / > >>  Now THAT has died, and he's called ME in. ) > >What THAT? Transceiver? AUI interface?   D My client worked around his original thinwire-connector failure mode
 (inabilityJ to establish or accept an IP connection) by switching over to (use of) theH thickwire connector (with transceiver).  Sometime thereafter, the systemB became unable to establish or accept an IP connection through this
 thickwire-0 connector-and-transceiver configuration as well.  I > >You can choose interface (BNC or AUI) but there is only _one_ ethernetyH > >port in microVAX 3400, not _pair_. You can not use BNC and AUI at the
 > >same time.e  E Ah. Thank you for correcting my jargon.  I didn't distinguish betweenh "inter-tH face" (connector) and "port" (internal hardware).  I've tried to be more carefule in my followup postings.  L > >>    2) ALL 3400 CPUs (KA-630? 640?  I don't remember) "expire" vis-a-visE > >> their on board Ethernet capabilities, around this point in theiri
 lifetimes. > >tK > >I dont think so. My microVAXen are working without "ethernet expiration"pB > >for several years (and will work for another several - I hope).  I Thanks.  I think we've pretty much eliminated this far-out possibility --- but Ie needed to ask. :-)  E > I think the front panel problem might be mechanical. As long as thep9 > ethernet connection isn't disturbed everything is fine.   J Hmm... I'll have to ask my client exactly when he relocated from his "old"J office suite to his current/"new" one, relative to when his VAX networking prob-l lems began.s  H > If you are at the other extreme and have a 3400 where the door doesn't shutI > properly because the connection sticks out or the cable is likely to bed trippeds- > over then the problem might be more likely.   = I will definitely look into this next time I visit my client.v  H > I think Hoff posted that the CPU and front panel are always changed atC > the same time - unless I misunderstood him - but when we had thispF > problem last field service replaced the CPU board and only when that' > didn't work replaced the front panel.m >oH > Basically what I'm saying in all these posts is if the front panel wasC > replaced along with the cpu then it's not that. If it wasn't thenl > replace it iimmediately.  L If by "front panel" you mean what I would also call "bulkhead," then in this caseJ it WAS replaced along with the CPU.  However, the replacement CPU was alsoH a USED one, and might well suffer from the SAME PROBLEM if, as it seems,1 this is a "common" problem with this type of CPU.D   Thanks,a       Chris Chiesa   ------------------------------  # Date: Mon, 30 Oct 2000 01:30:08 GMT 0 From: "Chris Chiesa" <cchiesa1@rochester.rr.com>. Subject: Re: Ethernet failure on MicroVAX 3400< Message-ID: <Aa4L5.102385$JS3.15720308@typhoon.nyroc.rr.com>  8 JF Mezei <jfmezei.spamnot@videotron.ca> wrote in message& news:39FB5FCE.1C970EC4@videotron.ca...  B > Have you tried ETHERMON on the VAX ? It is capable of becoming a promiscuous L > voyeur that will tell you if your vax is seeing traffic from other devices on
 > the lan.  H No, I haven't had the opportunity to try much of anything beyond a quick
 look-see withIJ essentially NO tools.  It would be ... tricky, to say the least, to try to get any externalI software onto this machine at this time.  I would have to download onto aa	 WinTel or&K *nix machine, and attempt to put the ETHERMON package onto magnetic tape in  ar VAX/VMS-readable format...  H > There was a way to patch MONITOR to enable the MONITOR ETHERNET.  Does= > anyone know where that documentation is hidden these days ?a  D Hm.  Just knowing it's POSSIBLE, I should be able to reconstruct the technique. :-)  L > If that VAX has DECNET, you might wish to turn on a ethernet link and then usee; > NCP to show the counters to see if there are many errors.b  E It does.  I shall.  Except, what exactly do you mean by "turn on a[n]s ethernet link?"  I'm a little rusty...-   Thanks,-      Chris Chiesa2   ------------------------------  % Date: Sun, 29 Oct 2000 22:37:57 -0400 - From: JF Mezei <jfmezei.spamnot@videotron.ca>s. Subject: Re: Ethernet failure on MicroVAX 3400, Message-ID: <39FCDEFB.9A0B3CA6@videotron.ca>   Chris Chiesa wrote:oJ > No, I haven't had the opportunity to try much of anything beyond a quick > look-see with L > essentially NO tools.  It would be ... tricky, to say the least, to try to > get any external* > software onto this machine at this time.  H You could probably obtain ethermon on a TK cartridge or something from a@ friendly neibourhood DEUS member. What media does the VAX have ?  : ethermon.exe is 115 blocks and the help file is 53 blocks.  8 Are there no PCs with a serial connection to that VAX ?   N When I supported a bank's SWIFT software, I devised a good deal: Since I had aN TK50 at home, and they had TK70, they agreed to allow me to bring tapes to theR site because they knew I could not steal any data and read it on my home machines.  N This allowed me to get Digital to download patches onto my home vax during theN day (via dialup), and at night when I got back home, I would cut a TK50 tape I4 would bring to the bank the next day with the patch.  G > It does.  I shall.  Except, what exactly do you mean by "turn on a[n]r > ethernet link?"  I'm > a little rusty...@   $MC NCP> NCP> SHOW KNOWN LINES CHAR  B This should tell you if you have one line defined on the ethernet.   then SHOW KNOWN LINE COUNTER  M will give you statistics on the traffic, collisions, errors etc on that line.    ------------------------------  % Date: Sun, 29 Oct 2000 15:00:30 -0400s- From: JF Mezei <jfmezei.spamnot@videotron.ca>r- Subject: Re: FTP from VMS to the demon's seedj, Message-ID: <39FC73CD.905C3E5D@videotron.ca>  & "Brian Schenkenberger, VAXman-" wrote:K > every file by name.  The WRQ, regardless of the *$FTP_NO_VERSIONS, seemed @ > to want to write that version number as part of the extension.   it is  TCPIP$FTP_NO_VERSION  5 there is no S at the end. And the value should be "1"h  N If WRQ wanted to write the version numbers is it because it was being suppliedH with the version numbers. The goal of *$FTP_NO_VERSION is to prevent theK *server* from giving version numbers, and when set, the client doesn't knowkN that there are multiple versions of a file. If the client gets a list of files3 without version numbers then it can't request them.t  N Note that the doc says that the logical affect the mget *.* command as well as the ls command.g  J If your client builds the list of files to request differently, it may not
 work for you.   J But I know that for me, (MAC FTP client), the difference was quite visible without the versions.t   ------------------------------  # Date: Mon, 30 Oct 2000 01:22:00 GMTe= From: system@SendSpamHere.ORG (Brian Schenkenberger, VAXman-)i- Subject: Re: FTP from VMS to the demon's seed 0 Message-ID: <009F256E.5B137E7B@SendSpamHere.ORG>  \ In article <39FC73CD.905C3E5D@videotron.ca>, JF Mezei <jfmezei.spamnot@videotron.ca> writes:' >"Brian Schenkenberger, VAXman-" wrote: L >> every file by name.  The WRQ, regardless of the *$FTP_NO_VERSIONS, seemedA >> to want to write that version number as part of the extension.g >t >it is h >TCPIP$FTP_NO_VERSIONp >p6 >there is no S at the end. And the value should be "1"  I I should have lowercased the S.  *$FTP_NO_VERSIONs.  I have TCPIP and UCXr6 on machines here and it doesn't seem to matter to WRQ.     --O VAXman- OpenVMS APE certification number: AAA-0001     VAXman(at)TMESIS(dot)COM             nO city, n., 1. a place where trees are cut down and streets are named after them.r   ------------------------------  # Date: Mon, 30 Oct 2000 02:57:23 GMTa* From: Alan E. Feldman <alan48@my-deja.com> Subject: Re: Galaxy doubt?) Message-ID: <8tio2h$is5$1@nnrp1.deja.com>e  / In article <sv9b369vdqnsdf@news.supernews.com>,a0   wspencer@ap.nospam.org (Warren Spencer) wrote:= > alan48@my-deja.com (Alan E. Feldman) wrote in <8t1vtv$crg$1e >i > -- snip -- >tG > >THERE'S NOTHING WRONG WITH SPLITTING AN INFINITIVE. Check the FAQ ats > >www.grammarlady.com, e.g. > >t >t > -- snip -- >w9 > This is off-topic banter up with which I shall not put.c >l   About what was it?  :-)  &-)     -- Disclaimer: JMHO Alan E. Feldman  alan48  &-)r dellnet.comu    & Sent via Deja.com http://www.deja.com/ Before you buy.u   ------------------------------  % Date: Sun, 29 Oct 2000 16:23:07 -0600 ) From: "John E. Malmberg" <wb8tyw@qsl.net>r( Subject: Re: Help with VMS 5_5_2 and GCC/ Message-ID: <svp8n76pll5u15@corp.supernews.com>:  4 "Martin Vorlaender" <martin#radiogaga.harz.de> wrote@ in message news:39f9a9d6.524144494f47414741@radiogaga.harz.de..." > sms@antinode.organization wrote: > :    ditt7@my-deja.com:iA > : >   I'm working on an old vaxstation with OS VMS 5.52.  [...]m > : B > :    I don't know what your license situation is, but DEC C V6.0< > : (distributed on the Hobbyist CD-ROM) works on (I quote):G > :       o  OpenVMS VAX Operating System, Version 5.2-2 to Version 7.2t >lG > With the big BUT in that it needs the backport library for VMS pre-V6e4 > (which must then be distributed with the program).  F In addition to the BACKPORT library, the DEC C compiler comes with theL AACRTL060 kit that can be distributed as a shared image with the application- to be installed on OpenVMS VAX 5.5-2 systems.   9 Programs built this way do not need the backport library.d  J Should you have the AACRTL060 library installed on a system, GCC VAX 2.8.1 can also use it.   -Johnc wb8tyw@qsl.network   ------------------------------  % Date: Sun, 29 Oct 2000 16:45:33 -0500g, From: "Glenn C. Everhart" <Everhart@GCE.com>" Subject: Re: Microsoft gets hacked' Message-ID: <39FC542D.2A5716D4@GCE.com>b  B There have been reports (on places like slashdot.org) that someone= has been offering NT source code, and also cautions to anyonee@ developing free software NOT to look. It appears the breakers inA want to sell, but it is claimed they don't have Windows 2000, nora( office source code, nor even windows ME.  > This incident could be used by Microsoft to sue people who getB their functionality too close to what Microsoft has, on the theory@ that any such close duplication "must" be a result of having the? code, and with no evidence needed that such is indeed the case.   ? It would become moot if one of two things happened, as far as Im
 can guess:A 1. The perps get caught and it is seen who has in fact had accesse due to the breakin, orC 2. The perps or someone else releases the entire code suite in news < so that it basically flood fills every news server on Earth.@ That would achieve such wide distribution that the secrets wouldC be forever lost. It would also mean no money for copies, for anyone @ ever. If a distribution THAT wide occurred, trying to claim someB idea that was in it was Microsoft's own would become pretty nearlyB unprovable.  Did person X have the idea independently and happened? to see the broadcast or did he get the idea from the broadcast?   > It does not appear that 2 will occur though; dumping that much> code into a news server would be risky...unless of course they? have an "in" in some Microsoft shop and can post the stuff fromc  there unbeknownst to the owners.  = The Gnu and the Samba people, who have been concerned to keepe; their code base squeaky clean because they have had attacks @ based on stolen code before, are evidently worried that baseless< lawsuits could occur and are taking extra precautions to try* to prevent anyone from contaminating them.   Aunty MS wrote:n > ? > In article <01JVVS5EYMQQ005WKR@tgmail.tg.nsw.gov.au>,  wrote:a) > > From: paddy.o'brien@zzz.tg.nsw.gov.aue > > Newsgroups: comp.os.vmsl& > > Subject: Re: Microsoft gets hacked) > > Date: Sat, 28 Oct 2000 19:49:13 +0010  > >  > > >h5 > > >I'm surprised it hasn't happened before now. :-)n > > >o > > > : > > >http://cnnfn.cnn.com/2000/10/27/technology/microsoft/ > > S > > Hmmm, in the first sentence it talks about the world's most powerful technologyo9 > > company.  Isn't Microsoft and technology an oxymoron?e > > T > > This was also on the evening news here in Sydney about an hour ago.  The gist ofS > > the news item was similar in that they only gained access to "future" products. N > > [If anything can be called a "product" from Microsoft, it must be future.] > >.Q > > They also said that the hacking was done by an email loophole, with mail to auK > > Microsoft employee giving them access when he clicked on an attachment.e > >hQ > > With the past n virus alerts being caused by attachments to Microsoft mail, IaT > > would have thought that anyone internal who had such access that could be passedN > > on would be very sensitive to his/her own mail system, and particularly to > > attachments. > > S > I have to include this, because it's the first time I've seen such a public blastf  > against MS from such a source. > U > Note this comes from the City Editor, not just some hack, in today's (London) Daily  > Telegraph. > U > -----------------------------------------------------------------------------------r > Unlocked Windows > City Editor's commenta > / > Microsoft humiliated as hackers crack Windowsp > B > JANET RENO must be scratching her head. After nearly a decade ofK > massive US government expense involving dozens of investigators traipsingaA > through Microsoft's systems and Seattle headquarters, America'sn? > attorney-general is still years away from nailing Bill Gates.t > J > Maybe she should visit Russia. Whoever hacked their way into Microsoft'sI > Redmond campus has taken just three months to scare the life out of theoH > world's richest man. It is the FBI, which ironically is part of Reno'sN > Department of Justice, which is crawling all over the place. Whether it is aN > lamentable act of software espionage or a prank by another bored 15-year-oldL > computer geek remains to be seen, but it will cause shockwaves through theJ > industry, even though the stock market shrugged off the issue yesterday. > M > If Microsoft can have its firewalls breached, then everyone else must be attO > risk, not least from corrupted versions of the Windows software on which they ' > all rely infecting their own systems.d > O > Yet Microsoft is notorious in the computer industry for security flaws. There M > are enough holes and bugs in the company's software to give mischief-makers5G > months of fun. A new bug seems to emerge almost daily and critics say Q > Microsoft's attitude is partly responsible for the Melissa and Love Bug virusesi > that caused so much damage.m > M > The industry is full of Microsoft-haters, who are unwittingly helped by thedE > open architecture net culture the company works so hard to promote. I > Microsoft receives thousands of emails a day, many with attachments and J > downloadable programmes, and one company executive recently boasted that* > he had 800 unread messages in his queue. > K > A blanket ban would remove the flow of ideas and technology that has made,N > Microsoft mighty. Scares like yesterday's are the price it has to pay, alongO > with the millions of dollars to computer security consultants who have joinedyM > the gravy train. In her quieter moments, Mrs Reno may wish that she was onek
 > of them. > U > ----------------------------------------------------------------------------------- 8 > Gasp - /boasting/ that you have 800 unread messages!!! > R > PS The full article contains the author's email address, which I won't reproduceQ > to save him from the spambots we get here. You'll need to subscribe to read thefQ > article however. You can find the link off today's (Sat 28-Oct-2000) front pageb4 > underneath the lead article as "See City Comment".   ------------------------------  % Date: Sun, 29 Oct 2000 14:35:12 -0500e2 From: rdeininger@mindspring.com (Robert Deininger)= Subject: Re: Narrow compatible SCSI controller for OpenVMS...JL Message-ID: <rdeininger-2910001435120001@user-2ivea5v.dialup.mindspring.com>  \ In article <v04210124b620aa0dbcb7@[140.186.88.130]>, Dick Munroe <munroe@csworks.com> wrote:  C > Anybody have an recommendations?  I still have a bunch of narrow nF > devices that I don't want to throw away.  Are there any PCI OpenVMS H > compatible controllers out there for OpenVMS?  Preferably not costing  > an arm and a leg?a   We recently bought a KZPAA-AA SCSI2 Card.  We use it in a DPW 600au with VMS, talking to a narrow DLT 4000 tape drive.  No problems.  E I don't think Compaq still sells it, but there are still some hoards.oA Contact David Turner at Island Computers:   dbturner@islandco.comr   -- y Robert Deininger rdeininger@mindspring.com    ------------------------------  % Date: Sun, 29 Oct 2000 15:15:39 -0600t7 From: "David J. Dachtera" <djesys.nospam@earthlink.net>b4 Subject: Re: OpenVMS/VAX Installation over a Network- Message-ID: <39FC937B.CEFF7839@earthlink.net>t   Christos Ricudis wrote:  [snip]  K > 2) The system has a really weird external SCSI connector I have no cablesn@ > for, and as such, I cannot find an easy way to bootstrap VMS.   : What constitutes "a really weird external SCSI connector"?  G I take it you mean something other than an LD50 ("Centronics" style) ore an HD50 or HD68 connector...   -- i David J. Dachtera. dba DJE Systems  http://www.djesys.com/  : Unofficial Affordable OpenVMS Home Page and Message Board: http://www.djesys.com/vms/soho/e  F This *IS* an OpenVMS-related newsgroup. So, a certain bias in postings is to be expected.  @ Feel free to exercise your rights of free speech and expression.  F However, attacks against individual posters, or groups of posters, are strongly discouraged.m   ------------------------------  % Date: Sun, 29 Oct 2000 15:36:04 -0600f7 From: "David J. Dachtera" <djesys.nospam@earthlink.net> . Subject: Re: Print drivers for OKI printers???- Message-ID: <39FC9844.62812C3E@earthlink.net>    "Main, Kerry" wrote: > 
 > Ray, David:t > : > These folks might be able to assist or provide pointers:M > www.nls.com  (just tried accessing their site and got timeout, but it might  > be issue at my end)m  	 Well, ...   ( DJAS01::DDACHTERA$ mult nslo www.nls.com Server:  LANmodem.djesys.com Address:  169.254.1.1s   Non-authoritative answer:e Name:    nls.com Address:  209.218.60.194 Aliases:  WWW.NLS.COM   1 DJAS01::DDACHTERA$ mult nslo www.nls.com/type=anyp Server:  LANmodem.djesys.com Address:  169.254.1.1p   Non-authoritative answer:s( WWW.NLS.COM     canonical name = nls.com  ( Authoritative answers can be found from:! nls.com nameserver = NS2.HOME.NET ! nls.com nameserver = NS1.HOME.NETl, NS2.HOME.NET    internet address = 24.2.0.27, NS1.HOME.NET    internet address = 24.0.0.27  * ...but Netscape failed to access anything.   -- e David J. Dachteras dba DJE Systemsi http://www.djesys.com/  : Unofficial Affordable OpenVMS Home Page and Message Board: http://www.djesys.com/vms/soho/   F This *IS* an OpenVMS-related newsgroup. So, a certain bias in postings is to be expected.  @ Feel free to exercise your rights of free speech and expression.  F However, attacks against individual posters, or groups of posters, are strongly discouraged.F   ------------------------------  % Date: Sun, 29 Oct 2000 19:21:40 -0600 + From: "Main, Kerry" <Kerry.Main@COMPAQ.com>e. Subject: RE: Print drivers for OKI printers???N Message-ID: <910612C07BCAD1119AF40000F86AF0D805284946@kaoexc3.kao.cpqcorp.net>   David,  L Their site must have been down. I just retried www.nls.com with no problems.  
 Kerry Main Senior Consultant  Compaq Canada Inc. Professional Servicesc Voice: 613-592-4660  Fax  :  819-772-7036 Email: Kerry.Main@Compaq.com     -----Original Message-----< From: David J. Dachtera [mailto:djesys.nospam@earthlink.net] Sent: October 29, 2000 4:36 PM To: Info-VAX@Mvb.Saic.ComM. Subject: Re: Print drivers for OKI printers???     "Main, Kerry" wrote: > 
 > Ray, David:  > : > These folks might be able to assist or provide pointers:G > www.nls.com  (just tried accessing their site and got timeout, but it  mightt > be issue at my end)   	 Well, ...h  ( DJAS01::DDACHTERA$ mult nslo www.nls.com Server:  LANmodem.djesys.com Address:  169.254.1.1e   Non-authoritative answer:t Name:    nls.com Address:  209.218.60.194 Aliases:  WWW.NLS.COM   1 DJAS01::DDACHTERA$ mult nslo www.nls.com/type=any0 Server:  LANmodem.djesys.com Address:  169.254.1.1i   Non-authoritative answer:h( WWW.NLS.COM     canonical name = nls.com  ( Authoritative answers can be found from:! nls.com nameserver = NS2.HOME.NET ! nls.com nameserver = NS1.HOME.NETh, NS2.HOME.NET    internet address = 24.2.0.27, NS1.HOME.NET    internet address = 24.0.0.27  * ...but Netscape failed to access anything.   -- e David J. Dachteraa dba DJE Systemso http://www.djesys.com/  : Unofficial Affordable OpenVMS Home Page and Message Board: http://www.djesys.com/vms/soho/e  F This *IS* an OpenVMS-related newsgroup. So, a certain bias in postings is to be expected.  @ Feel free to exercise your rights of free speech and expression.  F However, attacks against individual posters, or groups of posters, are strongly discouraged.    ------------------------------  % Date: Sun, 29 Oct 2000 20:22:12 -0600t7 From: "David J. Dachtera" <djesys.nospam@earthlink.net>a. Subject: Re: Print drivers for OKI printers???- Message-ID: <39FCDB54.28604E12@earthlink.net>    "Main, Kerry" wrote: >  > David, > N > Their site must have been down. I just retried www.nls.com with no problems.  9 Ah, yes - Northlake software. I didn't recognize the URL.n  F Should stand out in my mind because the town where I grew up is calledD "NorthLake", named after the two closest major roads, North Ave. and	 Lake St. e   -- h David J. DachteraC dba DJE Systemsv http://www.djesys.com/  : Unofficial Affordable OpenVMS Home Page and Message Board: http://www.djesys.com/vms/soho/2  F This *IS* an OpenVMS-related newsgroup. So, a certain bias in postings is to be expected.  @ Feel free to exercise your rights of free speech and expression.  F However, attacks against individual posters, or groups of posters, are strongly discouraged.h   ------------------------------   Date: 29 Oct 2000 20:43:06 GMT  From: inwap@best.com (Joe Smith)" Subject: Re: Q: Why not (2^n)-bit?, Message-ID: <8ti24q$rks$1@nntp1.ba.best.com>  2 In article <39f94c2c.1535925093@news.newsguy.com>,& Alan Greig <A.Greig@virgin.net> wrote:B >On 26 Oct 2000 14:32:00 GMT, fdc@watsun.cc.columbia.edu (Frank da
 >Cruz) wrote:g >r >>H >>I remember this quite well.  We had just installed our first DEC-20 inI >>1977 when our DEC salesperson brought a brand-new VT100 to show it off.bL >>We connected it to a front-end port at 9600 bps, enabled smooth scrolling,J >>TYPE'd a file, and the front end crashed right away because of the floodI >>of Xoffs and Xons.  As soon as it came back, it crashed again, etc etc.  >>H >>A similar incident had occurred previously when one of our programmersK >>fell asleep on the keyboard of his (non-VT100) terminal, causing the keys0K >>to autorepeat.  When we complained to DEC about this vulnerability, their-G >>answer was "don't do that".  Similarly for terminals that had a "page.H >>transmit" function.  But when their own VT100 became the culprit, they# >>tightened up RSX20F pretty quick.c >>M >>Nevertheless, the front end was very definitely designed with only people'sRF >>fingers in mind on the input side.  Even after the fixes, RSX20F wasJ >>markedly intolerant of large bursts of input.  Which is why the original; >>Kermit protocol enforced a 94-byte maximum packet length." > E >I think this partly came down to the size of a TOPS-20 buffer called G >something like ttbuf. There was some logiic that upped the buffer sizemG >depending on line speed. I remember for certain that I patched this toO? >use larger buffers on all lines but it's highly unlikely I didn( >anything similar to the front end code. >-E >Just before I first saw Kermit (around 1981?) I'd  written some codenE >to transfer between a 10 and a 20 and got away with 256 byte buffersDA >but to do this I needed to disable XON/XOFF handling and have it.A >handled by the code itself. This sounds counter-intuitive but it D >allowed me to use longer buffers than if I left flow control to theB >front end. When the TOPS-10 system XOFFed it still had room for aD >number of characters in its own input buffer so the fact that a fewC >more were sent  before the program had time to detect the XOFF and @ >stop the output didn't matter. Of course if you were sending toD >something like an Apple II then any chars transmitted after an XOFF >just got dropped. >yD >It's all slowly coming back now as I also remember having a programE >that scanned the tty input buffer counts looking for lines where the:G >input character rate approximated to the input line speed and then setc@ >the speed to zero or something like that. I think we wrote that) >locally based on an idea from elsewhere.i >g >>	 >>- Frank  >@ >--o >Alan Greigo  C Around that time I came up with a hack to the TOPS-10 Monitor to dosB both local and remote XON/XOFF processing.  If the terminal was inC packed-image mode, and XON/XOFF was enabled in the LDB, then when aiE Control-S came in, I had it mark the line as being XOFFed and put theeD Control-S character in the input buffer.  (Previously, it was one orC the other, not both.)  With the hack in, the smooth-scrolling VT100iI could tell the local system to stop sending, and KERMIT (in connect mode)c: could tell the remote TOPS-20 system to also stop sending.  I I also had some help from Dawn Banks (who was working at DEC at the time)29 on tweeks that could be applied to the RSX20F load image.o 	-Joea --8 See http://www.inwap.com/ for PDP-10 and "ReBoot" pages.   ------------------------------  % Date: Sun, 29 Oct 2000 15:25:20 -0600-+ From: "Main, Kerry" <Kerry.Main@compaq.com>c" Subject: RE: Rdb on VMS reference?N Message-ID: <910612C07BCAD1119AF40000F86AF0D805284940@kaoexc3.kao.cpqcorp.net>   Marco,  K The Lillian Hobbs book is a great start. As I recall, the Rdb documentationh is also very good.  I Btw - in case you are interested in a few examples of who is using Oraclea Rdb, check out:t  J http://www.openvms.compaq.com/openvms/brochures/ISE/ (New OpenVMS Customer: that just went live in May and who has very high IO loads)L http://www.iseoptions.com/about/technology_compaq.html (statement of support for previous pointer) G http://www.compaq.com/services/success/stories/ss_directv.html (Big Rdb  shop)o8 http://www.compaq.com/alphaserver/gs/quotes/etrade.html    and many others ..   :-)u   Regards,  
 Kerry Main Senior Consultant  Compaq Canada Inc. Professional Servicesc Voice: 613-592-4660w Fax  :  819-772-7036 Email: Kerry.Main@Compaq.com     -----Original Message-----+ From: Marco Shaw [mailto:marco@nbnet.nb.ca]  Sent: October 27, 2000 6:43 AM To: Info-VAX@Mvb.Saic.Comw Subject: Rdb on VMS reference?    L Any good references for Oracle Rdb on VMS?  I've seen a few books, but don't/ know if they address the OpenVMS environment...    Thanks,r Marco    ------------------------------  % Date: Mon, 30 Oct 2000 11:30:27 +1300r" From: "kevinf" <kevinf@gns.cri.nz> Subject: Re: Record not currentd- Message-ID: <8ti91i$2ned$1@news.comnet.co.nz>r  H Thanks for your suggestions, and I appreciate that in my haste I did notC include VMS version or even the programming language. My apologies.h  I I was getting Fortran 77 error 53, NO_CURREC, running under AXP/VMS v7.1.n  F I have now grasped the concept that I have been reading sequentially aL secondary key, but a second process has been inserting records concurrently,I sometimes into the middle of my sequential read. Hence a loss of currencys% when the NEXT pointer went undefined.u  G I have re-written my code to re-establish currency whenever this occursn (which is rare).   Many thanks, Kevin Fenaughtyw   ------------------------------  % Date: Sun, 29 Oct 2000 12:59:22 -0600u+ From: "Main, Kerry" <Kerry.Main@compaq.com>g' Subject: RE: Sun "uptime" belly laugher:N Message-ID: <910612C07BCAD1119AF40000F86AF0D80528493E@kaoexc3.kao.cpqcorp.net>   Paul,o  0 >>> Kerry, can we please nail this one shut. <<<  F Paul - let me explain in more detail, as your reference to node faili= ng4 scenarios is not related to what I am talking about.  F I am not talking about the reactive side of availability ie. node cra= shing ' type stuff. That is another discussion.s  F I am talking about the proactive side of availability ie. planned shu= tdowns.dF How does one tell all of those Internet / client connections not to c= onnectD to one system, but still allow current connections on that system to continue until they are done?   F You are correct that this will not work if you have folks that stay l= ogged F in for days. There is also a issue with multi-day batch jobs for obvi= ous  reasons.=20   F However, how many people stay logged into a single system for days at=  a time?=20  F Many sites (and I agree) view inactive users staying logged in as a s= ecurity.F risk and use Hitman type programs in the off hours to disconnect them= . Yes,F I know you can set a workstation to lock, but many people either forg= et toC  set it to lock automatically.=20  F I remember walking around a site with a Customer and moving the mouse=  on-6 many workstations to find numerous sessions wide open.  D Anyway, these are a very small minority and for those few exceptionsF (expecially system manager types), you simply ask them to log out tha= t F session and log back in again using the cluster alias. They will now = be ongF another system. I prefer Hitman or equivalent programs, but thats ano= ther discussion as well.n  F We have Customers today that do this and shut systems down in prime t= imetF (say 07:30 AM - start of shift) to save OT costs and give Operators a=  betters' life style. No user failovers - period.h  F So, Operations gets to reboot systems for planned shutdowns when they=  wanteF to with ZERO impact on business applications (with exception of long = time- logged in users and multi-day batch jobs).=20   F Question : How many Customers are running old versions of OS SW or ha= ve notD applied patches they want to because they could not get the downtimeF approved by theor business groups? In a B2B, eBusiness, or any type o= fr7 external web application, this is becoming a big issue.   F With OpenVMS clusters, flaky HW, tuning reboots, OS patches and with = twocF system disks, OS upgrades etc can all be done with no user failover a= nd nos# impact on application availability.   F For those that work in high availability environments, this is is a h= uge. huge advantage.=20   Regards,  
 Kerry Main Senior Consultant  Compaq Canada Inc. Professional Servicesb Voice: 613-592-4660d Fax  :  819-772-7036 Email: Kerry.Main@Compaq.com   -----Original Message-----) =46rom: Paul Sture [mailto:paul@sture.ch]d Sent: October 28, 2000 1:49 PM To: Info-VAX@Mvb.Saic.Com ' Subject: Re: Sun "uptime" belly laughers       ___In article=20F <910612C07BCAD1119AF40000F86AF0D805284930@kaoexc3.kao.cpqcorp.net>, K= erry Main=20c wrote:  - =46rom: "Main, Kerry" <Kerry.Main@compaq.com>u > Newsgroups: comp.os.vmso) > Subject: RE: Sun "uptime" belly laugher ' > Date: Sat, 28 Oct 2000 10:53:34 -0500s >=20	 > Andrew,  >=20! > Combining two replies in one ..h >=20? > >>> Solaris 8 allows you to patch it while it is running. <<<e >=20F > So does OpenVMS. OpenVMS has static parameters and dynamic paramete= rs. F > Changing the static ones requires a reboot. Changing dynamic ones d= oes- not. >=20F > So does Solaris 8. Please correct me if I am mistaken, but ALL OS's=  on ALLTF > platforms have some key kernel mode parameters or cache type sizes = that ifo: > it is desired to change them, then a reboot is required. >=20F > >>> Kerry this is just marketing. You cannot do upgrades to a runni= ng systemF > with no impact to the users because users have to be disconnected f= rom the-, > running node that needs to be upgraded.>>> >=20F > Not with OpenVMS clusters. We have Customers that today do this all=  the > time.t >=20F > One great feature of OpenVMS clusters is that this can be done with=  ZEROiF > application availability impact. ie. no fail-over of processes, res= ources,]C > no restarting of applications, no disconnecting active users etc.m >=20F > As you have heard me explain many times before - (you must be tired=  of me > explaining this by now)M >=20F > Using the shared disk architecture (which allows users to log into = anyaF > system for full read-write access to any files on any disks at the = sameF > time) and the TCPIP V5 load brokers load balancing (standard DNS on= ly does'F > round robin), you simply disable logins on the system that is plann= ed to  beF > shutdown. This allows current connections / users to continue worki= ng, butiF > all NEW connections are directed at other systems in the cluster. A=  number1F > of hours later (next day?), when all db/user connections are all on=  otherF > systems in the cluster, then that system which needs maint (tuning = staticF > parameters, OS upgrades using 2 system disk rolling upgrade, OS pat= ches,l HWF > replace ect) can be shut down with ZERO application availability im= pact.=20 >=20F Kerry, can we please nail this one shut. (I'm getting tired of the sa= me old! argument between you and Andrew).i  F Andrew is totally correct when he says that _I_ lose my sessions when=  a given=20F node is closed. As a system manager I do not log out at the end of a = dayr (With=20F an average of 80 sessions, would you?), although obviously my worksta= tion isn   well and truly locked.  F But then, all those sessions are sitting at the command line prompt, = whichr is=20 " not the normal end user interface.  F When it comes to clients us=EEng OPS, or an assortment of middleware = talkingo to=20mF something else, it is really not a problem. One node bombs due to a h= ardwarem  F problem (yes, we _do_ admit to the occasional one Andrew), the other = node(s)r  7 simply take over. The end user sees no loss of service.n  F > HW is sized such that as an example - in a three node cluster, any = twol9 > systems can carry peak loads for short periods of time.d >=20F > Great feature of this is that no user warnings are required, theref= ore no  > business approval is required. >=20F > Key is to separate SYSTEM availability from APPLICATION availabilit= y. WithvF > OpenVMS clusters, systems can be shutdown and the applications rema= in 100%  > available. >=20
 > Cool eh? >=20 > :-)= >=20F > One of the next big issues for vendors in the future is going to be=  thatr ourWF > age old "well of course, our 99.999999x numbers do not included sch= eduled> > downtime - no one does" is no longer going to be acceptable. >=20F > With the exponentially increasing importance of availability, this = will bee a=E > key consideration for those looking at e*Whatever and other mission  criticalF > applications that can no longer be simply scheduled down for planne= dM maint. >=20F > For Customers who have vendors in front of them quoting 99.999 (wha= tever)F > levels of availability, ask them if that includes scheduled system = down! > times. Then watch them squirm..e >=20   __
 Paul Sture Switzerlands   ------------------------------  # Date: Mon, 30 Oct 2000 00:18:41 GMTg* From: "Nikita V. Belenki" <kit@nospam.net>' Subject: Re: Sun "uptime" belly laugherm6 Message-ID: <B73L5.11$A26.183@sjc-read.news.verio.net>  = "andrew harrison" <andrew.nospam@uk.sun.com> wrote in messageF# news:39F95F75.F71394E@uk.sun.com...c  K > > OpenVMS clusters can do this by allowing current users to complete workm onI > > their active system, but force all new user / db connections to othersF > > systems in the cluster. When all current connections are completed (usersJ > > went home etc), then that system can be shutdown with ZERO applicationI > > availability impact. Cluster HW is designed such that in a three nodeMD > > cluster, any two nodes can carry peak loads for short durations.C > Kerry this is just marketing. You cannot do upgrades to a running J > system with no impact to the users because users have to be disconnected2 > from the running node that needs to be upgraded.F > Of course you can schedule this, get the users to stop what they are > doingoF > connect to another node, do the upgrade, reboot the node and you are > back.s  I Let's consider eBay as an example. Where would you see the "impact to them users" in this case?   Kit. kit # kits.net   ------------------------------  + Date: Sun, 29 Oct 2000 16:43:15 -0600 (CST) 0 From: Michael Angello <techwebsite@netscape.net> Subject: System Crash Dump? Message-ID: <20001029224315.707.qmail@www0u.netaddress.usa.net>h   Hello,G I am using OVMS VAX V6.2 on MicroVax 3100-80, and when I reboot the OS,2F the system appears to work well, but almost at the ending point of theJ SYS$MANAGER:SYSTARTUP_VMS.COM the system fails, and a dump action occurs.=   In9 have already tried to purge all of the logs I can, due to/, I have not enough free space on system disk.J I have just accessed the system shell once, and run the AUTOGEN to analyz= e H the system parameters values with the ones displayed on the OPA0: at theJ booting process, which were now modified to the AUTOGEN suggessted values= =2E-J After I made those changes, the system was started up completely and work= ingh7 well for a shor time, and then continue the crash loop.dJ The system does not have SYS$SYSTEM:SYSDUMP.DMP, so the SAVEDUMP is equal=  toi8 1, but It still fail at the last of the booting process.J I have tried to use the SDA with the SYS$SYSTEM:PAGEFILE.SYS, but It says=  ( that the referred file has no dump info.  , What do I have to do to correct the problem?  % Any comments are appreciated, thanks.m   Mike.t      D ____________________________________________________________________J Get your own FREE, personal Netscape WebMail account today at http://home= =2Enetscape.com/webmailg   ------------------------------  % Date: Sun, 29 Oct 2000 18:06:51 -0400n- From: JF Mezei <jfmezei.spamnot@videotron.ca> 5 Subject: TCPIP services: SMTP programming interface ?e, Message-ID: <39FC9F6D.B40F0AF6@videotron.ca>  M Is there a programming interface to have programs deal directly with the SMTP-J server on VMS (TCPIP Services V5.x) totally bypassing VMSmail's restricted
 features ?  N For instance, could an application create and submit a message directly to the; SMTP gateway to be processed and sent by the SMTO gateway ?,  N And more importantly, is it possible to have the SMTP gateway deliver messages= to an application instead of delivering messages to VMSmail ?r  N In other words, does the SMTP software in TCPIP services have the potential ofG becoming a replacement for Message Router and PMDF if one hooks various B gateways to it ? Or is it tied solely and exclusively to VMSmail ?   ------------------------------  % Date: Sun, 29 Oct 2000 19:02:42 -0500f# From: sol gongola <sol@adldata.com>,9 Subject: Re: TCPIP services: SMTP programming interface ?o' Message-ID: <39FCBAA2.1814@adldata.com>m  $ You can use the VMS version of pine.   sol gongolat   JF Mezei wrote:e > O > Is there a programming interface to have programs deal directly with the SMTP L > server on VMS (TCPIP Services V5.x) totally bypassing VMSmail's restricted > features ? > P > For instance, could an application create and submit a message directly to the= > SMTP gateway to be processed and sent by the SMTO gateway ?e > P > And more importantly, is it possible to have the SMTP gateway deliver messages? > to an application instead of delivering messages to VMSmail ?i > P > In other words, does the SMTP software in TCPIP services have the potential ofI > becoming a replacement for Message Router and PMDF if one hooks various D > gateways to it ? Or is it tied solely and exclusively to VMSmail ?   ------------------------------  % Date: Mon, 30 Oct 2000 09:48:29 +0800O4 From: Dave Sneddon - bigpond <dbsneddon@bigpond.com>9 Subject: Re: TCPIP services: SMTP programming interface ? + Message-ID: <39FCD36D.76787EAF@bigpond.com>)   JF Mezei wrote:c > O > Is there a programming interface to have programs deal directly with the SMTP L > server on VMS (TCPIP Services V5.x) totally bypassing VMSmail's restricted > features ?  : There is a currently undocumented/unsupported interface... see Ask the Wizard item 4492I I had this requirement recently and have put together a DCL interface forf use by our inhouse application.s  P > For instance, could an application create and submit a message directly to the= > SMTP gateway to be processed and sent by the SMTO gateway ?c  " The DCL basically has two options;  ' 	$ imail send filespec "him@over.there"t  5 will create the necessary headers then send the file,y   	$ imail deliver filespecm  C assumes that the file contains all the appropriate headers and justn passes it on to SMTP.   P > And more importantly, is it possible to have the SMTP gateway deliver messages? > to an application instead of delivering messages to VMSmail ?b  3 Don't know about this one - haven't had to do this.@  P > In other words, does the SMTP software in TCPIP services have the potential ofI > becoming a replacement for Message Router and PMDF if one hooks variousnD > gateways to it ? Or is it tied solely and exclusively to VMSmail ?     -- - Regards, Dave.iI -------------------------------------------------------------------------pI David B Sneddon (dbs)  OpenVMS Systems Programmer   dbsneddon@bigpond.comkI DBS software at ...   http://www.users.bigpond.com/dbsneddon/software.htmnI "Life is what happens to you while you're busy making other plans" Lennonl   ------------------------------  # Date: Mon, 30 Oct 2000 01:55:30 GMT + From: John Santos <john.santos@verizon.net> 9 Subject: Re: TCPIP services: SMTP programming interface ?w> Message-ID: <MPG.14669e5466bf774798968b@news.bellatlantic.net>  ? In article <39FCBAA2.1814@adldata.com>, sol@adldata.com says...i& > You can use the VMS version of pine.  D JF's asking for a programming interface.  Pine works fine as a humanE interface (I use it on my Alpha at work), but I don't think it has a o$ callable interface for sending mail.   > 
 > sol gongolas >  > JF Mezei wrote:  > > Q > > Is there a programming interface to have programs deal directly with the SMTP N > > server on VMS (TCPIP Services V5.x) totally bypassing VMSmail's restricted > > features ? > > R > > For instance, could an application create and submit a message directly to the? > > SMTP gateway to be processed and sent by the SMTO gateway ?c > > R > > And more importantly, is it possible to have the SMTP gateway deliver messagesA > > to an application instead of delivering messages to VMSmail ?  > > R > > In other words, does the SMTP software in TCPIP services have the potential ofK > > becoming a replacement for Message Router and PMDF if one hooks various F > > gateways to it ? Or is it tied solely and exclusively to VMSmail ?  F I haven't tried this, but I think it is a fairly trivial use of TCP/IPH programming to use the SOCKET routines to open a connection to socket 25F on host 127.0.0.1 (local host) and send and receive the SMTP protocol,I which is also pretty trivial.  You send HELO, etc. messages to get logged H in, then send FROM and TO messages to set up for sending a message, thenH DATA followed by the message itself.  The only tricky bit is in the DATAH section: Each line of text that begins with a period (".") must have theG period doubled.  After the last line of text, send a line consisting ofa@ "." by itself.  You can then send another message or logout and I disconnect.  (Each command and each line of text in a message ends with ay= line delimiter.  I can't remember if it is <CR> or <CR><LF>.)r  F Most books discussing TCP/IP protocols include a complete description.: They often use SMTP as an example because it is so simple.   -- e John Santosy   ------------------------------  % Date: Sun, 29 Oct 2000 22:43:34 -0400a- From: JF Mezei <jfmezei.spamnot@videotron.ca>-9 Subject: Re: TCPIP services: SMTP programming interface ?r, Message-ID: <39FCE04C.3B6D8C35@videotron.ca>   Dave Sneddon - bigpond wrote:g< > There is a currently undocumented/unsupported interface... > see Ask the Wizard item 4492   Thanks for the pointer.e  O PIPE AnA/IMAGE SYS$LIBRARY:TCPIP$SMTP_MAILSHR.EXE | SEARCH SYS$INPUT symbol:      L yields very interesting results. But it is is unclear on whether this allowsD just sending or sending and receiving message from the smtp gateway.   ------------------------------  % Date: Sun, 29 Oct 2000 22:48:20 -0400o- From: JF Mezei <jfmezei.spamnot@videotron.ca>d9 Subject: Re: TCPIP services: SMTP programming interface ?n, Message-ID: <39FCE16A.20F48C7D@videotron.ca>   John Santos wrote:H > I haven't tried this, but I think it is a fairly trivial use of TCP/IPJ > programming to use the SOCKET routines to open a connection to socket 25H > on host 127.0.0.1 (local host) and send and receive the SMTP protocol, > which is also pretty trivial.e    L Well, I already have that part done in a message-router to SMTP gateway (andM one from POP to message-router), but eventually, since I will have to get rideJ of message router, I would like to see if the SMTP gateway that comes withG TCPIP services could be made into an MTS that can route message betweene+ various gateways (including external SMTP).w  I For instance, if I could migrate my gateways for fax and sms from messageaL router to SMTP, then I could then have access to those from any mail packageK that interfaces to the smtp gateway. But this require that the SMTP productsS route some message to my gateway instead of trying to deliver them to the internet.l  ' PMDF can do that. But it isn't free :-)m   ------------------------------  % Date: Mon, 30 Oct 2000 12:16:57 +0530a/ From: Suresh Kumar-T <sureshkumar@mascon.co.in>a  Subject: Urgent Help Needed.....< Message-ID: <6DB7DDFF3075D41191F3001083FC5FB81803E0@LCSRVR1>   Hi ,  K We are using VMS operating system and oracle 8.0.We have new problem in ourtF VMS operating system. That is I want use operating system username andK Password to Oracle also. So we use following command and successfully solve  te problem.t  / <create user ops$suresh identified  externally>m  H and then we simply type sqlplus /. we enter the sqlplus use of Operating system username and password.e  J Now the   problem is suppose we have 100 users. We want to create each and; every user use  ops$USERS command. It will waste of  time .g  G So I want to create some common user ID for Eg. suresh. All other users K should go this ID only. So I want to grant this(suresh)id to all users. Goto it.=  I (eg)some x enter <sqlplus />.  It should invoke ops$suresh and then enter  SQlplus using Suresh ID.  J I want to grant this(suresh)id to all users. In oracle we simply use grant	 command.    6 In oprating system <VMS>level How can I get around it.   This is very urgent.   Waiting for your reply.p   Bye    Suresh.T sureshkumar@mascon.co.in Mascon global Ltdp India.   ------------------------------   End of INFO-VAX 2000.606 ************************