April 15, 2015 at 10:35 am
Hi
I'm just looking for a little feedback from out there in the wilderness, here's my situation.
5 months ago added locally attached SSD storage
•741148-B21 - HP 200GB 12G SAS High Endurance SFF 2.5in Enterprise Performance Solid State Drive
in order to alleviate severe pressure on the i/o subsystem.
I have 7 SQL instances (2012) running on a rather meaty machine with shared storage
trying to be clever I migrated tempdb for all SQL instances onto the new SSD
Wow what a difference!
However 3 weeks ago I experienced the first issue, although the drive concerned was view-able and indeed navigable in Windows Explorer All SQL instances shut themselves down complaining that it could no longer find tempdb.
This issue then occurred for the second time on Monday and once again today. I have a long night ahead of me moving database files around in order to mitigate the effects should this fault re-occur, but I'd like to throw this open to the community to see if anyone else has experienced similar issues.
Any feedback greatly appreciated.
Paul
April 15, 2015 at 11:53 am
paul.connors 70350 (4/15/2015)
HiI'm just looking for a little feedback from out there in the wilderness, here's my situation.
5 months ago added locally attached SSD storage
•741148-B21 - HP 200GB 12G SAS High Endurance SFF 2.5in Enterprise Performance Solid State Drive
in order to alleviate severe pressure on the i/o subsystem.
I have 7 SQL instances (2012) running on a rather meaty machine with shared storage
trying to be clever I migrated tempdb for all SQL instances onto the new SSD
Wow what a difference!
However 3 weeks ago I experienced the first issue, although the drive concerned was view-able and indeed navigable in Windows Explorer All SQL instances shut themselves down complaining that it could no longer find tempdb.
This issue then occurred for the second time on Monday and once again today. I have a long night ahead of me moving database files around in order to mitigate the effects should this fault re-occur, but I'd like to throw this open to the community to see if anyone else has experienced similar issues.
Any feedback greatly appreciated.
Paul
Had a similar issue which only occourred once on a dev server, iirc the firmware was upgraded and all fine since. Quick question, any clues in the logs?
😎
April 15, 2015 at 12:39 pm
Thanks for the swift reply, I'm thinking along the same lines myself, unfortunately the cursory scan of the event logs etc revealed not much detail at all and then as a mere DBA I had to hand over to our Internal Support guys who passed it straight onto the hosted solution provider, so sadly no, not much info from any logs currently although a Root Cause Analysis has been requested!
Thanks again for the information, I will request that all firmware and driver revisions are up to date!
regards
paul
April 15, 2015 at 1:13 pm
paul.connors 70350 (4/15/2015)
Thanks for the swift reply, I'm thinking along the same lines myself, unfortunately the cursory scan of the event logs etc revealed not much detail at all and then as a mere DBA I had to hand over to our Internal Support guys who passed it straight onto the hosted solution provider, so sadly no, not much info from any logs currently although a Root Cause Analysis has been requested!Thanks again for the information, I will request that all firmware and driver revisions are up to date!
regards
paul
Dug a little into my notes and found that it was indeed a firmware issue, make certain that the firmware and drivers are up to date across the board, not just the SSDs. Check also the I/O subsystem's logs, the O/S logs and all other applicable logs. There is bound to be a clue in there.
😎
Further on the dev-box, I upgraded to Win 2012 R2 and 2014 recently, no issues and about 20% increase in overall performance.;-)
May 29, 2015 at 4:25 am
Thanks very much for the reply and sorry I didn't respond sooner. The firmware was 'old' and has since been upgraded. This didn't completely resolve the issue, although we do have a little more stability we've experienced 2 outages since. Not getting a lot of luck with the IT support guys and their cronies out in the cloud, so I assume there's nothing obvious in any logs. I've asked if an additional controller can be installed, maybe I'm just asking too much of a single controller..?
Thanks again for the additional info.
Paul
Viewing 5 posts - 1 through 4 (of 4 total)
You must be logged in to reply to this topic. Login to reply