Re: process running under Cgroup2 control is OOM'ed if its stdout goes to a file at at tmpfs filesystem

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

To: Michal Koutný <mkoutny@xxxxxxxx>
Subject: Re: process running under Cgroup2 control is OOM'ed if its stdout goes to a file at at tmpfs filesystem
From: Toralf Förster <toralf.foerster@xxxxxx>
Date: Sat, 16 Nov 2024 17:04:06 +0100
Cc: cgroups@xxxxxxxxxxxxxxx
In-reply-to: <tuvclkyjpsulysyz6hjxgpyrlku5zuov6gyyhjzvadrqt4qpse@bwmb7ddutwzj>
References: <e0dccc65-3446-4563-8a0d-1ebda4bd7b81@gmx.de> <tuvclkyjpsulysyz6hjxgpyrlku5zuov6gyyhjzvadrqt4qpse@bwmb7ddutwzj>
Ui-outboundreport: notjunk:1;M01:P0:6tySPlg3z2o=;ec9+zKdjfseIt8bdzkO995uam2N 6p0u2ifULh/Yeo19tX9E8Eh+gufRzdaWhCcc4VnS/fxD/rMDxLSkZX+plKx3lMpc1isubXjfY NBZioKFis9H+D2dEDFbRI2LIqgBqgYTN7h7GqhWJTrvcDRETPf1exvcoG/+SU/Ohys2SYxRIz NhP/W8BKoCPNy+V/FoPYh4DpuNNuWO2jGlPVA08/JqN4Fs2I/Uaj7735LbD/KKPmu6cFrS396 53QZ1zdQLjIwHhijAibAPus6Jqsjk20iVGcsGftkmkj0fesUIMFmh9CEksagx96JO0Ejuv9z+ SdBq/DmTg5yehmXrj5b2O5Mpg95dpFsDv1k//7Fb8vxisjj/gnQcw32PoySroRKoJPvOkHZOR /lBoILqMfusmZGoWkETAB/UZXXfX/5hmFyYstZeYF6aMdIvvX392SLh2GHUK6xI5Sbf2QPXLu Vwzwkq/NkLT7J1ErpZmBTEiDAziG2kCKCKZ7Tc7FldE0pVz9FEcHG3Ap5R0/i+2Mmovpu7+m7 Ej2zX7NilBviT4HfkEKhLPq3YWGWx6Ki4c3akFbEMv+KqMjTCtS0TrI4S3UUVZpeZzyibRIhl mHH4F4Uyz7VyssH3g5f38ptfm2qZBc+IccYUKB+wn9EwQhUXA3NMtI7mBzHbalWVhUltF0hCE mGRQdjMcIRULUBV37YqvSNpB8CA7UNfGkFoFliIbfBmBUbj83AImjWC44KLH/5TVY5RRK8m8m DBlr8gSnGwA7eErHpEZPkcjFlgvoDPbn2nVwRWc9/mpz2uQbZiZ/9h1Mj6OO17occYwtAgm6r ne5neoD249Uu+1kJBZbhyEL4i0jp8G+eWYx2mToHHJsVaCaPgCWUmFxHppi5r/T5CkshE1mlh Rp8GChG/MrGcBDso6EQv2rrA5WF2xq8NmNa8gu7W0//V8+XfvmZ8mllwe
User-agent: Mozilla Thunderbird

On 11/15/24 13:52, Michal Koutný wrote:

Your reproducer disables swap, so there's no option to write out the
anonymous memory. OTOH, regular page cache can be written out to the
backing persistent filesystem and free up RAM.


Ah, thx!

I removed any limitation for memory.swap.max and have set memory.max to
the RAM which is needed for the fuzzer.
That should make it, right?

--
Toralf

Follow-Ups:
- Re: process running under Cgroup2 control is OOM'ed if its stdout goes to a file at at tmpfs filesystem
  - From: Michal Koutný

References:
- process running under Cgroup2 control is OOM'ed if its stdout goes to a file at at tmpfs filesystem
  - From: Toralf Förster
- Re: process running under Cgroup2 control is OOM'ed if its stdout goes to a file at at tmpfs filesystem
  - From: Michal Koutný

Prev by Date: Re: [PATCH v2 1/2] sched/isolation: Exclude dynamically isolated CPUs from housekeeping masks
Next by Date: [RFCv1 0/6] Page Detective
Previous by thread: Re: process running under Cgroup2 control is OOM'ed if its stdout goes to a file at at tmpfs filesystem
Next by thread: Re: process running under Cgroup2 control is OOM'ed if its stdout goes to a file at at tmpfs filesystem
Index(es):
- Date
- Thread

[Index of Archives] [Linux ARM Kernel] [Linux ARM] [Linux Omap] [Fedora ARM] [IETF Annouce] [Security] [Bugtraq] [Linux OMAP] [Linux MIPS] [eCos] [Asterisk Internet PBX] [Linux API] [Monitors]