November 15, 2012 at 2:58 pm
Hi All,
I know that this is not the first thread related to this issue and i am fully aware of what this error means, but i am having trouble working out exactly what the problem in in my particular case.
Our production server is connected to an MD3000 SAS disk array. Within our array we have three virtual desks configured, one (DISK0) is used to store data files, another (DISK1)is used to store log files and the third (DISK2) is used to store the data and log files for the tempDB database.
The array has 2 built in RAID controllers; DISK0 is mapped to one of the controllers while disk 1 and 2 are mapped to the other controller and it has worked fine for the past with no signs of the error detailed in the subject line.
This week the first controller experienced an "issue" and a failover occurred; meaning that all three virtual disks were mapped to a single controller and we experienced massive performance issues.
We have since verified that the controller is fine and mapped the disks back to their original configuration. Dell (who supplied and support the MD3000) had a look at the hardware for us and they are convinced it is still in optimal condition.
However, we are seeing occasional occurences of the warning detailed in the subject line; the culprit is always files that reside on either DISK1 or DISK2. These errors always appear after a log backup operation has completed but also pop up randomly as well.
Now the common denomenator here is the RAID controller because both DISK1 and DISK2 are mapped to the same controller and we did have this failover event which occurred for an "unidentifiable" reason according to DELL.
So that is the state of affairs currently, i wondered if anyone had come across a similar scenario and may be able to advise how i could suggest any reasons why we have seen the sudden drop in performance.
Any help that you could provide would be greatfully received.
Thanks in advance.
November 15, 2012 at 3:12 pm
what are the status' of the arrays, check for dead disks, etc first. This may degrade RAID performance
-----------------------------------------------------------------------------------------------------------
"Ya can't make an omelette without breaking just a few eggs" 😉
November 15, 2012 at 3:34 pm
Thanks very much for the reply. All disks are working fine, we have sent performance information exported from the array to the manufacturers support team and they are convinced that the array (including the disks) is working perfectly.
Viewing 3 posts - 1 through 2 (of 2 total)
You must be logged in to reply to this topic. Login to reply