1 INFO-VAX	Sun, 30 Apr 2000	Volume 2000 : Issue 241       Contents:% Re: Disk subsystems for OpenVMS Alpha $ Re: DLT Drive Software Write Locking Help with RSH timeout  Test - please ignore Re: TPU for unix6 Re: VMS Fibre channel connectivity with 3rd party disk Re: Where is Operator.log ?  Re: Where is Operator.log ?   F ----------------------------------------------------------------------  % Date: Sun, 30 Apr 2000 16:54:29 +0930 % From: Jeremy Begg <jeremy@vsm.com.au> . Subject: Re: Disk subsystems for OpenVMS Alpha* Message-ID: <390BDFAD.3E13CE1E@vsm.com.au>   Hi Terry  K >   You'll still need a single-ended SCSI adapter (usually the Q-logic one) L > to support the internal CD-ROM (the KZPAC "notices" CD-ROMs but won't makeI > them available to the system), so you can hang your TZ tape off that as H > well. There's a bogus restriction on mixing internal and external SCSIL > devices on the same controller, but it works fine (and the controller man-K > ufacturers don't know why Digipaq is saying this). I suspect it's because : > it makes it easier to goof when setting the termination.  I Last year I had to set up an AlphaServer 1000A where we wanted to have an M external DLT drive on the same bus as the internal shelf & CD-ROM.  It proved K to be very unstable (bus lockups) even though the cable length restrictions L etc were observed.  In the end we installed a second SCSI controller for the* tape drive and all our problems went away.  K >   A later note shows you've been confused by the literature on the KZPAC. I > The total amount of disk storage it can serve to the host is 256GB, but M > it can have over twice that connected - it can serve up to 8 logical drives M > (which appear as DRAn:) and each of those logical drives can be up to 32GB. M > Extra disks in RAID 1 mirror sets don't count, nor do hot spare drives. The K > limit is entirely in the amount of data served to the host. I have mostly M > RAID 5 sets, and one controller has 87GB attached - 7 9GB drives internally ? > on 2 of the buses and 6 4GB drives externally on the 3rd bus.    Thanks for the clarification.   L BTW, what is UltraLVD SCSI?  I notice you referred to it (and UltraHVD) in aH later note and I've seen the term around but I don't know what it means.   Thanks,    	Jeremy Begg   ------------------------------  % Date: Sun, 30 Apr 2000 05:47:57 +0200 $ From: Paul Sture <sture.ch@sture.ch>- Subject: Re: DLT Drive Software Write Locking + Message-ID: <VA.00000025.04b120dd@sture.ch>   J In article <20000428123218.14595.00000548@ng-ch1.aol.com>, BrianNFO wrote:# > From: briannfo@aol.com (BrianNFO)  > Newsgroups: comp.os.vms + > Subject: DLT Drive Software Write Locking   > Date: 28 Apr 2000 16:32:18 GMT > N > I have a pair of TZ887's connected to an HSJ52.  All of a sudden, the drivesQ > are write locking whenever you load a tape.  From a hardware standpoint, things M > look fine.  The tapes are NOT write protected, and when the tape loads, the L > drive "write protect" light is NOT on.  When you mount the tape to VMS, itP > indicates the device is write locked...and a show dev/full indicates the driveQ > is software write locked.  (This happens on multiple tapes...both brand new and  > used.) > L > This is happening on 2, TZ887's that before now we're perfectly fine.  Any1 > thoughts on what is causing this?  Thanks much.  >  > Brian  >  > P.S.  VMS 6.2  > R Have you tried deleting the tape units from the HSJ and recreating them? I've had N disks, tapes and robots "go wrong" from time to time. One fairly reproducible M example I've seen of a tape "going wrong" is when backup tries to write to a  J cleaning tape (e.g. an abnormally large backup needs another tape and the : operators load the net set of tapes, cleaning tape first).  L Not at work at the moment, so I can't give a sample session, but at the HSJ O controller prompt, you determine which unit references the tape and delete it,  O then delete the tape. Follow this with RUN CONFIG and the HSJ will rebuild its  R tables, in the process adding the tape back in. You then add the unit to the tape  and the job is done.  O Something like (WARNING - this is from memory, so I cannot be held responsible   for inaccuracies):   hsj00> show tape  .  unit t2 shows that it is connected to tape650  3 hsj00> del t2  ! This is the unit entry for $1$MUA2  hsj00> del tape650 hsj00> run config   M the HSJ tells you it is rebuilding its tables, and logs the fact that it has  + discovered tape650. Then add the unit back:    hsj00> add unit t2 tape650  M If you are not familiar with or happy with entering HSJ commands, I strongly  @ recommend that you call CSC and ask them to talk you through it. ___ 
 Paul Sture Switzerland    ------------------------------  # Date: Sun, 30 Apr 2000 14:34:24 GMT , From: Stanley Hippler <shippler@my-deja.com> Subject: Help with RSH timeout) Message-ID: <8ehg9g$epj$1@nnrp1.deja.com>   A I am using rsh from OpenVMS 7.2-1, TCPIP service 5.0A, to execute G commands on an IBM VM platform.  Works well, except for cases where the - IBM command takes over 5 minutes to complete.   2   $ rsh/user=myuser/pass=mypass ibmhost ibmcommand    ibmcommand: Cannot be executed#   Timer expired:  Connection closed   D Is there a logical or config parameter for the rsh client that I can# use to increase this timeout value?    Thanks.    Stanley Hippler     & Sent via Deja.com http://www.deja.com/ Before you buy.    ------------------------------  # Date: Sun, 30 Apr 2000 09:20:46 GMT - From: Dave Pickles <davep@nugate.demon.co.uk>  Subject: Test - please ignore * Message-ID: <8egtp8$1lf$1@cyw.uklinux.net>  , My ISP seems to have stopped taking c.o.v...   ------------------------------   Date: 30 Apr 2000 08:00:40 GMT- From: djweath@attglobal.net (Dave Weatherall)  Subject: Re: TPU for unix 5 Message-ID: <DTiotGxQ0bj6-pn2-VUDWhKEtqJWF@localhost>   C On Sun, 29 Apr 3900 14:51:44, "Ray T." <lists@aik.tec.sc.us> wrote:   C > Not what you are asking for, but my favorite is a better than EDT 
 > editor, D > called SEDT.  It is available for PC, VMS, and Unix as binaries or
 > source.  > 3 > http://www.ultranet.com/~anker/sedt/sedt_main.htm    I second that.   Cheers - Dave.   ------------------------------  # Date: Sun, 30 Apr 2000 17:23:42 GMT % From: Uwe Zessin <zessin@my-deja.com> ? Subject: Re: VMS Fibre channel connectivity with 3rd party disk ) Message-ID: <8ehq6h$otg$1@nnrp1.deja.com>   ' In article <2000Apr26.150156.1@eisner>, /   kaplow_r@eisner.decus.org (Bob Kaplow) wrote: 1 > In article <Pine.BSI.4.00.10004251940080.14953- E 100000@hometown.idirect.com>, "Michael Kwan, GCN" <mkwan@idirect.com>  writes: H > > I am not sure if this has been asked before. I am working to connect0 > > some non-Compaq (non-DEC) disk array to VMS. > > G > > I have the KGPSA-BY Fibre channel HBA on my AS8400, and connect the  > > disk array via a switch .   C Hm. -BY is the part number of just the board. The full order number 8 (incl. manual) for the 32-bit adapter is called KGPSA-BC- (or 380574-001 in the idiotic Compaq system).   E > > The strange thing is, I can see these drives at the ALpha console > > > (v5.6) mode using SHOW DEVICE, ( $1$DRA250m $1$DRA300 etc)   More strange things:+ 1) FC-connected disks are named "$1$DGAu:". C 2) You don't see FC-connected disks on the console level unless you A    explicitly configure them via 'wwidmgr' for boot/dump support.   G > > but after I booted up VMS (v7.2-1 with latest FC tima kit v0200), I % > > can't see any of these FC drives.   $ OpenVMS requires that on the HSG-80:% 1) the command console lun is enabled   2) SCSI-VERSION is set to SCSI-3% 3) each LUN is assigned an identifier   F > > I looked through quite a bit of Compaq/DEC documentation, there isC > > no mention of the need to do any configuration on the VMS side, F > > however, the DEC docs always refer to the HSG80, which in my case,: > > I don't have, since I am attaching to a non-DEC array. > > D > > Am I missing something? Any insight would be deeply appreciated. > D > Both at his DECUS presentation, and at a recent local FibrechannelB > presentation late last year, Richie Lary made it VERY CLEAR thatG > interoperability just does not exist in todays Fibrechannel products, A > especially in switches. Are you at least using a Compaq switch?   E Is that really a Fabric-switch? Gadzoox, f.ex. produces a 'switch' to B connect arbitrated-loop ports together (so-called switched loops).2 They just some days ago announced fabric services.  H In recent Alpha console revisions you can switch the KGPSA between FC_AL and fabric at console level.  
 > 	Bob Kaplow    --
 Uwe Zessin    & Sent via Deja.com http://www.deja.com/ Before you buy.    ------------------------------  % Date: Sun, 30 Apr 2000 05:44:59 +0200 $ From: Paul Sture <sture.ch@sture.ch>$ Subject: Re: Where is Operator.log ?+ Message-ID: <VA.00000024.04ae66dc@sture.ch>   ? In article <390BC1CC.4B13F394@vl.videotron.ca>, JF Mezei wrote: 2 > From: JF Mezei <jfmezei.spamnot@vl.videotron.ca> > Newsgroups: comp.os.vms " > Subject: Where is Operator.log ?' > Date: Sun, 30 Apr 2000 01:17:10 -0400  > ( > This may seem like a silly question... > L > I have a cluster at home. There was a short power outage today requiring aU > reboot (damm, I was just starting to get a respectable uptime of over 150 days :-(.  > K > I was not there at the time, so I decided to investigate a bit. I came to P > realise that my workstation has not been keeping operator.log since it was put > into a cluster.  > , > VELO is an all-mighty teenage Microvax II.U > BIKE is a  Vaxstation 3100 with decwindows. Both have separate (local) system disks  > P > VELO seems to be the keeper of the OPERATOR.LOG with BIKE no longer creating a: > new one at boot as it did before the two were clustered. >  > Which begs the question: > L > What determines which of the nodes will be the keeper of OPERATOR.LOG in a' > cluster with separate systems disks ?  > H > Does BIKE decide it need not keep its own operator.log because it is a > workstation ?  > N > What if BIKE were to boot first into the cluster, would it become the keeper > of OPERATOR.LOG ?  > T Firstly, you didn't mention what version of VMS you are running. IIRC the behaviour * has changed in this area over the years...  Q Have a look in syslogicals.com (or sylogicals.template if yours doesn't have the  T info) for a description of the OPC$* logicals. You'll probably see code relating to N workstations in there too (it's a long time since I've had my hands on a VAX).  9 To create an operator log on the fly, issue the following   D $ define/user sys$command opa0 ! not necessary if entered from OPA0: $ reply /log  E This can also be put in your startup files as a brute force approach.  ___ 
 Paul Sture Switzerland    ------------------------------  % Date: Sun, 30 Apr 2000 02:14:10 -0400 0 From: JF Mezei <jfmezei.spamnot@vl.videotron.ca>$ Subject: Re: Where is Operator.log ?/ Message-ID: <390BCF24.E0C3D8DC@vl.videotron.ca>    Paul Sture wrote: R > Have a look in syslogicals.com (or sylogicals.template if yours doesn't have theU > info) for a description of the OPC$* logicals. You'll probably see code relating to P > workstations in there too (it's a long time since I've had my hands on a VAX).  K Thanks. I never thought of looking there. It is on fact documented that for @ workstations in a clsuyer,. operator.log is not enabled locally.   ------------------------------   End of INFO-VAX 2000.241 ************************