ext3 filesystem corruption - more info (in text)

Martial Herbaut martial at server101.com
Thu Apr 13 22:30:19 UTC 2006



> >But we didn't actually lose power on the raid or hosts
> >just the connecting switches, so we lost all communication.
> >Presumably, in this situation  the controller cache should have been 
> >emptied Is my reasoning correct here ?
> 
> Correct.  If your RAID has w/b cache enabled, but is battery backed, you
> should be OK.
> 
> Beyond this, I'm not sure what else you can look at.
> 

don't mean to barge in, however I have seen similar corruption happen in 
the past where the fabric went away momentarily, like unplugging and 
replugging a fibre cable on a non-dualpath/failover setup but the host 
was not killed/rebooted. From memory the corruption was not immediately 
apparent and became so later. 

I think the best thing to do in that case scenario is force a reboot of 
the host and then force fsck as opposed to continuing on and hope for the 
best.

Martial Herbaut
---------------
Server101.com




More information about the Ext3-users mailing list