Messages in this thread | | | Date | Thu, 24 Jul 2008 00:49:28 +0100 | From | Alan Cox <> | Subject | Re: libata timeout, MCE and kernel panic |
| |
> recovery was required. It seems the journal recovery combined with the > md device reconstruction was enough to cause an ata command > timeout. After that I saw several MCEs and the kernel paniced.
An MCE is amachine check, and indicates the hardware detected a problem. Given the sequence of heavy I/O, followed by a timeout and the processors detecting inconsistencies - is your PSU sufficient ?
| |