[FOLLOW-UP] :- SWXCR RAID 200 problems on an AS2100

From: Smith, Gavin <SmithG_at_capitagroup.co.uk>
Date: Wed, 30 Sep 1998 16:48:13 +0100

Hi,

Apologies for the delay in sending this follow up to the list, I didn't
get a
chance to 'deal' with the machine as soon as I'd have liked...

We're still getting errors from the SWXCR at startup. According to our
DEC
engineer we have the incorrect type of terminator for our RAID array.
The type
we have doesn't have a 'Shelf OK' jumper, and is causing the errors. He
doesn't belive this will cause us any problems though (apparently the
person
responsible for the machine when it was running VMS complained of
similar
errors. Ho-Hum)

The intermittant crahses on the machine were due to spurious entries
manually added to /etc/sysconfigtab by our customer. The parameter
SSM-THRESHOLD was being set to its maximum value, an impossibly huge
18446744073709551615.

Having fixed /etc/sysconfigtab and patched the OS, the system has been
up
for nearly a month now. If we ever get round to replacing the terminator
(our
users are not keen on anymore downtime at the moment) and it 'fixes' the
error messages I'll post a summary.

In the meantime thanks to :-

Saar Picker (saarp_at_socrates.berkeley.edu)
Neil Dyce (Neil.Dyce_at_bristol.ac.uk)
Paul Yahnig (Yahnig_at_GroupWise.Kean.edu)

For their suggestions.

Gavin Smith

--
Original Message :-
We've an Alphaserver 2100 4/200 with a DEC RAID 200 array (SWXCR) that
we've installed Digital Unix V4.0D on. The disks in the SWXCR are just 
configured as seven JBOD's.
Previously the Alphaserver was running OpenVMS V6.1 and the RAID array 
was configured with two JBOD's and a five disk RAID 5 set.
When we boot we see the follwoing error :-
Initializing xcr0.  Please wait.....
vmunix: xcr0 at eisa0
vmunix: xcr_logger: XCR_ERROR packet
vmunix: xcr_logger: cntrl 0 unit 2
vmunix: re_getdrive
vmunix: Cmd should always return good status
vmunix: Hard Error Detected
vmunix: Active XCR_COM at time of error
vmunix: xcr_logger: XCR_ERROR packet
vmunix: xcr_logger: cntrl 0 unit 2
vmunix: re_getdrive
vmunix: Cmd memory lost
vmunix: Possible Software Problem - Impossible Cond Detected
vmunix: Active XCR_COM at time of error
vmunix: xcr_logger: XCR_ERROR packet
vmunix: xcr_logger: cntrl 0 unit 0
vmunix: re_getdrive
vmunix: Cmd should always return good status
vmunix: Hard Error Detected
vmunix: Active XCR_COM at time of error
vmunix: xcr_logger: XCR_ERROR packet
vmunix: xcr_logger: cntrl 0 unit 0
vmunix: re_getdrive
vmunix: Cmd memory lost
vmunix: Possible Software Problem - Impossible Cond Detected
vmunix: Active XCR_COM at time of error
vmunix: re1 at xcr0 unit 1 (unit status = ONLINE, raid level = JBOD)
vmunix: Created FRU table binary error log packet
I installed the swxcrmon / swxcrmgr 2.3 software. Swxcrmon reports the
following error :-
10-Aug-1998 12:39:08 Shelf failure on channel 0
In addition to the errors during the boot we're experiencing
intermittant crashes
but there's no crash dump or indications (other than a reboot entry) of
failure
in the binary errorlog. The crash does not produce an error message of
any
sort, the machine simply halts and sits at the console (>>>) prompt.
Attempts
to 'crash' it from here have been unsuccessful. 
Our resident DEC engineer believes the errors during the boot are a red
herring
and that they may be caused by an incorrect jumper setting. Just to
muddy the
waters further the machine was moved (not by DEC) to a remote site, and
it 
hasn't been patched (yet).
Anything else? The customer is trying to run Informix 7.3 and ODBC 4.5.
They
also use a piece of software called Workbench-Engine on their PC's to
access
the database.
Has anybody come across these errors before? Am I looking at a problem
with
a single drive or the power supplies / fans on the shelf? I'll know more
when I
visit the site this Friday to patch the OS, but any suggestions as to 
what to try in the meantime would be appreciated!
TIA
Gavin Smith
END
**********************************************************************
This email and any files transmitted with it are confidential and 
intended solely for the use of the individual or entity to whom they   
are addressed. If you have received this email in error please notify 
the system manager.
This footnote also confirms that this email message has been swept by 
MIMEsweeper for the presence of computer viruses.
www.mimesweeper.com
**********************************************************************

Received on Wed Sep 30 1998 - 15:46:44 NZST