We recently experienced a problem on a SQL Server 2005 (Active/Passive config), which brought down our SQL Cluster several times during a 2 day period.
Disabling the SAV services on both nodes have resolved all our issues (including some minor issues encountered on a daily basis).
in the first instance the SQL Agent service restarted during the night for no identifiable reason on the Active node.
Later the next day it did so again, but this time it was because the SQL Service was issued a restart command from the OS, as the cluster service had determined it had encountered a problem (no clear cause, but seems it lost the hearbeat). Within minutes of the restart, the process repeated itself (this occured several times).
We initiated a manual fail-over to the Passive node (which now became the active node). After about 10 mins of stable operation it presented with the same issues as the initial active node.
The initial active node DBA (now passive) was bounced. SAV services were disabled. When it was back up we failed back to it from the current active node DBB which was now restarting the SQL Service as well randomly every few minutes.
DBB was then shut down completely. DBA continued working 100% for several hours (only node now, SAV disabled). SAV was then enabled again on the server. About 60 - 90 minutes later the SQL cluster starting going down with the same issues as before. The SAV services were once again disabled and the server bounced.
It (DBA / The cluster) has been stable with SAV disabled for almost 2 weeks now. DBB was added back onto the Cluster (also with SAV disabled) a week after the initial issues were resolved.
This cluster had been running for nearly a year with SAV without problems, and no server or sql config changes were applied prior to the sudden instability.
The other minor issues which now appear to be fixed:
-
We do manual log shipping using .bat files, which would fail a few times each day with "Access denied" errors or shell commands suddenly not being recognised (even though they run fine 99% of the time). We've had 0 failures since SAV has been disabled.
-
We use a 3rd party app to do db backup encryption/compression. This too would fail inexplicably once or so every day or two, usually with the "Access Denied" message. We've also had 0 failures since we disabled SAV for this.
The only identifiable cause seems to be SAV, throughout our investigations as to the cause.
We've tried to log support requests with the support tool, but we're getting blank drop downs on the first step, and they are required selections, so we cannot get past that point. This has been logged 3 x with support (web and customer), but we have not received any feedback in more than a week. Appaling customer service/support... not sure why we bothered paying for it then.
Is someone able to assist from symantec to either fix the support tool so we can log the issue, or open a case from the back-end, or possibly suggest some way we can safely continue using SAV (possibly a manual definition update or something?).
Thanks
=============================== Server / SAV info ==================================================================================
SQLCluster (Windows 2003 R2 32 bit / MSSQL2005 SP2)
1 Active Node: DBA
1 Passive Node: DBB
SAV:
Program Version: 10.1.8.8000
Scan engine: 91.2.1.10
Virus Definitions: 5/7/2010 rev.16
Exclusions:
Extension:
BAK
bak
CHK
chk
EDB
edb
GZ
gz
LDF
ldf
LOG
MDF
mdf
ndf
SDB
sdb
TGZ
tgz
TRC
trc
TRN
trn
Folders:
C:\Documents and Settings\clustersql1\Temp
C:\Windows\Cluster
Quorum(Q:)