What's your RAID configuration? Are the OS LUN and data LUN in the same storage pool? What's the brand/model of your HDDs? It's highly recommended to use HDDs in the Tested hardware and operating system list (THOL).
thanks for your quick reply.
System and storage pools are separated.
In the data pool we have 7 LUNs and only two have these problems.
The have started to change the hard drives from Toshiba to Hitachi after 4 have been marked as defect in a very short time. This was also recommended by our german support company Thomas Krenn (the also delived the new hard discs) - we are also in direct contact with them, but we still have no solution.
Last night we shut down all of the servers and disconnectde the Storage Controller. After reinserting it the controller the system came up again, but with the same errors as before.
Tonight we are planning to delete the LUNs and recreate them. (Of course we don't want to try that during wokring hours.)
Another strange effect we had last night:
I formated one of the LUNs with NTFS,
chkdsk did not show any errors until we restored the data from tape ( data files only .. not sectors or so) chkdsk showed errors on the drive again.
(Ok, we thought that might have been the problem ... but !!!)
Than I copied the data from the LUN to a SAN. Chkdsk on SAN = no errors
reformattnig defective LUN & copy back data from SAN
reboot of server using these drives showed automated chkdsk again with the same errors as before
result ... data broken. Same error as before !
And like I mentioned before. The Modular Server shows all hard discs as fully functional (including two spares)
If you have any ideas .. please let me know.
Have you tried to upodate the storage driver on your compute modules? Is it a single or dual storage controller configuration?