>> Hi,
>>
>> Does anybody know what this means?
>>
>> Mar 9 16:32:12 localhost kernel: audit(1110403917.4294966169:0):
initialized
>> Mar 9 16:32:59 localhost kernel: I/O error in filesystem ("sda")
>> meta-data dev sda block 0x1e8ed3bff ("xfs_read_buf") error 5 buf
> >count 512
>>
>
>It means the device driver returned an error when XFS
>attempted a metadata read. Looks like one of the AG
>header structures in an AG fairly far into the FS, based
>on the size and block# there. Probably a hardware issue.
>
>cheers.
>
>--
>Nathan
Hi Nathan,
Thanks for your previous replies. But could you be a little more specific?
1) What do you mean by "probably a hardware issue"?
2) Is there danger that the xfs filesystem on this RAID has somehow lost its
integrity or has become corrupted? Or was the metadata eventually read?
(I've seen the same error 3 times in the logs -- over the course of 2 days).
3) Should I be running xfs_repair? Or, is there anything else I that I CAN
and/or SHOULD do to:
-- a) investigate the error, or
-- b) correct the error
In other words, should I just ignore the logged event, or take some action?
Thanks,
Andy Liebman
|