June 23, 2012 at 6:01 am
A client had a an old Dell T300 with a RAID 5, and ECC Memory.
They had corruption in the Active Directory DB and in a SQL DB. Both from two different times. We quickly replaced their server with a newer HP in fear of not knowing why the disk corruption was taking place.
The built in Dell diagnostics detect no issues. And while the server is "stable", the disk writes are not.
I suspect its memory or maybe the RAID controller doesn't know the BBU is broken etc, or SMART on a disk is not reporting correctly.
I wanted to test for the BBU problem by running a long running SQL query and unplugging the power.
Any suggestions for a query that is likely to then after inspection after the restart, to have caused corruption if it happened mid transaction?
EDIT: I am not looking for the queries that inspect the DB, I am looking for advice on what I query I can run that would write a lot of data that would also not deal well with an ungraceful shutdown.
We would like to figure out the source of the issue, so we can make it a replica destination for Veeam (basically a backup ready to run).
Thanks!
June 23, 2012 at 7:32 am
DBCC CheckDB is what you use to detect corruption.
Gail Shaw
Microsoft Certified Master: SQL Server, MVP, M.Sc (Comp Sci)
SQL In The Wild: Discussions on DB performance with occasional diversions into recoverability
June 24, 2012 at 6:07 am
Hi,
I edited my post as I'm not sure it was clear exactly what I was looking for.
June 24, 2012 at 6:22 am
Update a large table.
However if you want to test the stability of the IO subsystem, shut SQL Server down and use SQLIOSim
Gail Shaw
Microsoft Certified Master: SQL Server, MVP, M.Sc (Comp Sci)
SQL In The Wild: Discussions on DB performance with occasional diversions into recoverability
Viewing 4 posts - 1 through 3 (of 3 total)
You must be logged in to reply to this topic. Login to reply