The almighty .plan

Wed, 09 Apr 2008 00:00:00 +1000

Apr  6 06:20:04 clarence kernel:
   ad4: TIMEOUT - READ_DMA48 retrying (1 retry left) LBA=470360994
Apr  6 06:20:04 clarence kernel:
   ad4: FAILURE - READ_DMA48 status=51<READY,DSC,ERROR> error=4<ABORTED>
                                                           LBA=470360994
Apr  6 06:20:04 clarence kernel:
   g_vfs_done():ad4s2f[READ(offset=57219432448, length=16384)]error = 5


ad4 is the drive on clarence holding swap, root, /var, /usr and
/archive. ad4s2f is /usr.

I only managed to take a look at it last night. All FS are consistent
(props to UFS2) although there were lots of Incorrect Block Counts and
Soft Update Inconsitencies. It doesn't look like anything important
was lost, although I haven't done a full comparison with the last
known good dumps.

The proximate causes were probably (a) power issues due to the storms
(b) ourmon. Ourmon is an insanely overengineered network monitoring
system I was using, and it can put a lot of strain on the system;
nearly all the errors were in its logs and rrd files (which occupied
about 70% of the space on /usr too).

I wasn't too worried anyway - I do a full fs dump every week, with
daily incremental ones. All are sent to another local machine, and the
weeklies are copied offsite.

PS: for those who like their evidence anecdotal, the drive is a Maxtor.