Hi,

Last Friday we got bitten by a rather vicious raid controller
firmware/codeset that corrupted some of our raidsets when a disk
failed in the array and the controller started a rebuild. Seeing that
I quickly disabled some of DLEs involved by either commenting them out
or inserting a `strategy skip' stanza in the dumptypes while we
researched/solved the issue. 

This is with amanda-2.6.1p2 both on clients and servers.

Once the firmware issue was sorted out and the affected raidsets
rebuilt I tried to restore them as they were prior to the failure
just to see that the index files for those DLEs had disappeared.
I had to retrieve them from the backups, but that made me a little
nervous that something else was at work. Of course I could have used
amfetchdump rather than amrecover, but that begs the question:

What is the `proper' way to deal with a situation like this:
commenting away the DLEs or use 'strategy skip'?

I'm quite please to say: I restored of 3TB of data with no glitch
whatsoever once the index files were put back in place. 
Amanda is quite a nice piece of software!

thanks,
jf
-- 
<° >< Jean-François Malouin          McConnell Brain Imaging Centre        
Systems/Network Administrator       Montréal Neurological Institute
3801 Rue University, Suite WB219          Montréal, Québec, H3A 2B4
Phone: 514-398-8924                               Fax: 514-398-8948

Reply via email to