Hi all, and sorry for using this kind of bandwidth :)
In my mail yesterday I indicated that RAID-5 was finally working well
for me, with 2.0.35 pre15.
Before taking the RAID in production, I wanted to stress-test it. So, I
ran a tar and two bonnie jobs.
The system locked up to some extent. I could still ping it, but the fs
layer seemed to have died.
The syslog said:
Oct 20 02:02:09 malthe kernel: raid5: lock contention, rw == 0, sh->cmd == 1
Oct 20 02:02:14 malthe kernel: raid5: lock contention, rw == 1, sh->cmd == 2
Oct 20 02:02:55 malthe kernel: raid5: lock contention, rw == 0, sh->cmd == 1
Oct 20 02:02:57 malthe kernel: raid5: lock contention, rw == 1, sh->cmd == 2
Oct 20 02:02:57 malthe kernel: raid5: bug, nr_write == 1, nr_read == 1, sh->cmd == 2
Oct 20 02:02:57 malthe kernel: raid5: bug, nr_write == 1, nr_read == 1, sh->cmd == 2
Oct 20 02:02:57 malthe kernel: raid5: bug, nr_write == 2, nr_read == 1, sh->cmd == 2
I got a lot of the lock contention messages. But as far as I could see from the raid5.c
block driver, this is normal. The bug message seems worrying though.
What can I do to help clear up this issue ?
Thanks,
................................................................
: [EMAIL PROTECTED] : And I see the elder races, :
:.........................: putrid forms of man :
: Jakob Østergaard : See him rise and claim the earth, :
: OZ9ABN : his downfall is at hand. :
:.........................:............{Konkhra}...............: