[Info-vax] SCSI errors
Phillip Helbig---undress to reply
helbig at astro.multiCLOTHESvax.de
Wed Oct 19 15:17:35 EDT 2011
Consider the output after my questions below.
This is the output from $ SHOW ERROR. The 1 and 2 errors (except for
that on DSA510:) show up regularly shortly after booting and don't
increase. The ones now at 32 and 33 increase slowly (few per day). All
disks are on DKB, not DKA. Most or all of the DQ devices are not real.
DSA510: has one member on each node. DSA110: has both members on node
JANDER. Each node has two SCSI controllers, but only one (the external
one) is in use. All are "Qlogic ISP1020" except PKB0: on JANDER which
is "SYM53C895 LVD SCSI" (operating in SE mode, though; all disks are
single-ended). I have the shortest cables I have (longer ones gave more
errors) and I'm pretty sure I am within the allowed cable length.
Connected to each of the external controllers is a BA356 with 7 disks,
essentially the same types on each node.
Considering the fact that the error count increases only slowly (ca.
3--4 per day) a) what could be causing this and b) is it something to
worry about?
---------8<-------------------------------------------------------------------
%SYSMAN-I-ENV, current command environment:
Clusterwide on local cluster
Username FOOBAR will be used on nonlocal nodes
%SYSMAN-I-OUTPUT, command execution on node MINNIM
Device Error Count
DSA510: 1
PEA0: 1
$170$DQA0: (MINNIM) 1
$170$DQA1: (MINNIM) 1
$170$DQB1: (MINNIM) 1
%SYSMAN-I-OUTPUT, command execution on node JANDER
Device Error Count
DSA110: 32
PKB0: 2
$110$DKB0: (JANDER) 1
$110$DKB200: (JANDER) 1
$110$DKB300: (JANDER) 1
$110$DKB500: (JANDER) 2
$110$DKB600: (JANDER) 1
PEA0: 33
$110$DQA1: (JANDER) 1
$110$DQB0: (JANDER) 1
$110$DQB1: (JANDER) 1
%SYSMAN-I-OUTPUT, command execution on node LEEBIG
Device Error Count
$150$DQA0: (LEEBIG) 1
$150$DQA1: (LEEBIG) 1
$150$DQB1: (LEEBIG) 1
More information about the Info-vax
mailing list