[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <72aa8863-a534-b8df-6b9e-f69cf4dd5c4d@i-love.sakura.ne.jp>
Date: Sat, 26 Jan 2019 15:10:38 +0900
From: Tetsuo Handa <penguin-kernel@...ove.sakura.ne.jp>
To: Arkadiusz Miśkiewicz <a.miskiewicz@...il.com>
Cc: Tejun Heo <tj@...nel.org>, cgroups@...r.kernel.org,
Aleksa Sarai <asarai@...e.de>, Jay Kamat <jgkamat@...com>,
Roman Gushchin <guro@...com>, Michal Hocko <mhocko@...e.com>,
Johannes Weiner <hannes@...xchg.org>,
linux-kernel@...r.kernel.org,
Linus Torvalds <torvalds@...ux-foundation.org>
Subject: Re: pids.current with invalid value for hours [5.0.0 rc3 git]
On 2019/01/26 11:41, Arkadiusz Miśkiewicz wrote:
> dmesg:
> http://ixion.pld-linux.org/~arekm/cgroup-oom-2.txt
What is wrong with this output? It seems to me that the OOM killer is killing
processes as soon as python 3 fork()s one. Although "potentially unexpected
fatal signal 7." messages due to handling SIGBUS at get_signal() after the
OOM reaper reclaimed the OOM victim's memory is a bit noisy, I don't think
that something is wrong.
Jan 26 03:33:49 xps kernel: python3 invoked oom-killer: gfp_mask=0x6000c0(GFP_KERNEL), order=0, oom_score_adj=0
Jan 26 03:33:52 xps kernel: potentially unexpected fatal signal 7.
You seem to be monitoring cgroup statistic values for only 1 minute, but
since the fork bombs might take longer than 1 minute, I'm not sure that
the statistic values after terminating cg.py and waited for a while (e.g.
1 minute) is still showing unexpected values.
Powered by blists - more mailing lists