lkml.org 
[lkml]   [2017]   [May]   [23]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
SubjectRe: [PATCH] mm/oom_kill: count global and memory cgroup oom kills
From
Date


On 23.05.2017 10:49, David Rientjes wrote:
> On Mon, 22 May 2017, Konstantin Khlebnikov wrote:
>
>> Nope, they are different. I think we should rephase documentation somehow
>>
>> low - count of reclaims below low level
>> high - count of post-allocation reclaims above high level
>> max - count of direct reclaims
>> oom - count of failed direct reclaims
>> oom_kill - count of oom killer invocations and killed processes
>>
>
> In our kernel, we've maintained counts of oom kills per memcg for years as
> part of memory.oom_control for memcg v1, but we've also found it helpful
> to complement that with another count that specifies the number of
> processes oom killed that were attached to that exact memcg.
>
> In your patch, oom_kill in memory.oom_control specifies that number of oom
> events that resulted in an oom kill of a process from that hierarchy, but
> not the number of processes killed from a specific memcg (the difference
> between oc->memcg and mem_cgroup_from_task(victim)). Not sure if you
> would also find it helpful.
>

This is worth addition. Let's call it "oom_victim" for short.

It allows to locate leaky part if they are spread over sub-containers within common limit.
But doesn't tell which limit caused this kill. For hierarchical limits this might be not so easy.

I think oom_kill better suits for automatic actions - restart affected hierarchy, increase limits, e.t.c.
But oom_victim allows to determine container affected by global oom killer.

So, probably it's worth to merge them together and increment oom_kill by global killer for victim memcg:

if (!is_memcg_oom(oc)) {
count_vm_event(OOM_KILL);
mem_cgroup_count_vm_event(mm, OOM_KILL);
} else
mem_cgroup_event(oc->memcg, OOM_KILL);

\
 
 \ /
  Last update: 2017-05-23 12:33    [W:0.181 / U:0.284 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site