OPEN VMS, SCSI Card and now always reboot, why_??

[Sebastian Perez]

We are working with ES-40 Alpha Server, 850MHz.
We added a SCSI card to conect a Backup driver <100-300GB>.We work on OPEN VMS.
The problem is when we make a backup, always the system reboot and we don't know why?
Any idea about this??
Thnx.

[Martin P.J. Zinser]

Do you have a crash dump? If yes what does SDA give you? If not, enable dump writing before your next backup ;-)

[Veli Körkkö]

You don't tell us your VMS version but it must at least V7.1-2 of course. System does write memory dumps by default and on on reboot it usually produces SYS$ERRORLOG:CLUE$sysnam_...LIS file. Contents of those CLUE$....LIS would be extremely useful.

Typical things to do:

a) upgrade ES40 firmware to recent level
b) apply all relevant, preferably ALL VMS patches
c) upgrade the firmware on your tape drive
(Certain tape drivels models have had issues on their firmware causing havoc on VMS host)

d) contact your CS organixation for a-c if needed

_veli

[Antoniov.]

Until you have solved, disbale reboot after bug;
$ SYSMAN PARAM SET BUGREBOOT 0
After you solved, restore orignal value
$ SYSMAN PARAM SET BUGREBOOT 1

Bye
Antoniov

[Antoniov.]

Oops, I've forgotten MCR
$ MCR SYSMAN
PARAM SET BUGREBOOT 0
PARAM WRITE CURRENT
EXIT

Excuse me
Antoniov

[Willem Grooters]

Some hints and questions...
* Seen the size, I assume you use a DLT-type of drive. Is that correct? If this is SDLT, I've been told there are some issues in combination with VMS.
* When does the reboot occur: when DLT is inserted, at start of write operation, or during this operation, change of tape, end (during or after final processing)? This might shine a light.
* If you can load a tape, can you mount it?
* What does console tell you: does the new SCSI show up in SHOW CONFIG and SHOW DEVICE? * Same for the backup device
* What happens if you INIT ?
* What happens if you MOUNT?
* Is the unit number of the unit not conflicting with another device on that SCSI controller?

For the rest: indeed check crashdump.

HTH

Willem

[Jeff Chisholm]

Yes, the output from...
$ ana/cras sys$system
SDA> clue crash

...would be really useful if you could post it here please. There's a couple of open elevations with certain scsi cards that use the sys$pkwdriver, backups to SDLT tape fail or hang. Make sure you've got modern firmware and the latest fibre_scsi kit for your version... which was what?

If you need assistance setting up a dumpfile, let me know and I'll post the article. /jeff

[Sebastian Perez]

I have some information that can be useful to help us.
It´s attached in txt format.
Thank

[Jeff Chisholm]

Exception Frame:
----------------
R2 = FFFFFFFF.800A2D54 MNTVER$PACKACK_VOL_C+00134
R3 = FFFFFFFF.A68FFC78
R4 = FFFFFFFF.85CC95C0 EXE$MNTVERSIO
R5 = FFFFFFFF.8105F040
R6 = FFFFFFFF.800A5448 MNTVER$GET_CKSUM_BUF_C+00D18
R7 = FFFFFFFF.8105F740
PC = FFFFFFFF.8105F74C
PS = 38000000.00000804

Note the lack of symbols for the PC. Note that R5, R7 and the PC are all in the same neighborhood.

Note that this is V7.1-2, I wonder what sort of tapedrive this is. The system seems to be patched up to rev pretty well.

Note the PACKACK. Has the bus been correctly terminated? Do we have errors in the errorlog just prior to the crash. A device error from the tapedrive and a scsi bus reset?

I'm going to guess that the tape drive is on the PKB SCSI bus, Symbios 895, using the sys$pkwdriver. Let's move it to one of the other adapters and see what the behavior is.

Log a support call, we'll look at this.
/jeff

[Sebastian Perez]

Here is the tape drive especifications.
And what We have here:

Magtape AXPPRO$MKB500:, device type COMPAQ SuperDLT1, is online, file-oriented
device, error logging is enabled.

Error count 0 Operations completed 0
Owner process "" Owner UIC [SYSTEM]
Owner process ID 00000000 Dev Prot S:RWPL,O:RWPL,G:R,W
Reference count 0 Default buffer size 2048
Density unknown Format Normal-11

Volume status: no-unload on dismount, position lost, odd parity.

110/220 GB SDLT
Series 3306
Serial: RBD0400650.