Re: Strange "flush" process bahaviour

[Piotr Szymaniak <szarpaj-TbOm9Ca2r9GrDJvtcaxF/A@public.gmane.org>]

[-- Attachment #1: Type: text/plain, Size: 1449 bytes --]

On Sun, Mar 31, 2013 at 03:14:00PM +0400, Vyacheslav Dubeyko wrote:
> Your sysrq-triger output is really strange. The sysrq-triger takes
> momentary system state snapshot. So, this output makes sense only
> inside issue occurrence. And I feel that this output was taken after
> issue ending. Or, maybe, it needs to make several trying of getting
> sysrq-triger output during issue reproducibility.
> 
> I think so because flusher doesn't make something NILFS2 related and, moreover, it simply sleeps:
> *snip*
> The nilfs_cleanerd daemons sleep also:
> *snip*
> The segctor thread sleeps too:
> 
> Maybe I misunderstand something but your sysrq-triger output doesn't
> contain info about any NILFS2 related activity. So, I can't see any
> details about the issue with kernel flusher thread in shared
> sysrq-triger output.

Hrm, maybe this is btrfs realated then? But I suppose I don't have any
btrfs on the second machine with flush issues.

I had a system freeze yesterday and right after reboot flush is back.
So it looks like I can reproduce it (something makes the flush thing
while booting. Maybe I can catch some debug info there?). Right now I'm
going back to 3.4.x series kernel.


Piotr Szymaniak.
-- 
Siedziałem z (...) takim, co włamał się do budki z piwem i całą noc
wiadrami nosił piwo do domu. Miał pełną wannę i akwarium, kiedy rano
go złapali.
  -- Jacek Hugo-Bader, "Egzamin na świra"

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 836 bytes --]