mismatch_cnt on various raid's

[Kevin Fenzi]

So, we have been getting weekly mismatch_cnt emails from a number of
machines with raid setups. 

http://fedoraproject.org/wiki/Infrastructure/SOP/Raid_Mismatch_Count

Describes a process to run repair and check on them, but in these cases
the mismatches have come back right away. 

Why is that you might ask? 

It's because these are all raid-1 setups. There are cases where raid1
can and does have a mismatched sector setup and it's perfectly fine. 
It's still good to run the weekly check on them in case there is a
marginal drive that drops out or otherwise shows lower layer problems. 

See: https://bugzilla.redhat.com/show_bug.cgi?id=566828

So, moving forward I think I will try and do a "hotfix" for 
/etc/cron.weekly/99-raid-check and make it not mail
for raid1 mismatches. (see patch in above bug)

Longer term, perhaps there will be a mdadm update for rhel5, or we will
get everything moved to rhel6 (which has the fix already). 

kevin
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
Url : http://lists.fedoraproject.org/pipermail/infrastructure/attachments/20110315/4cf88801/attachment.bin