Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 22 Oct 2024 18:01:03 +0000
From:      bugzilla-noreply@freebsd.org
To:        bugs@FreeBSD.org
Subject:   [Bug 280846] Low memory freezes / OOM: a thread waited too long to allocate a page
Message-ID:  <bug-280846-227-j86Nxv3Dm8@https.bugs.freebsd.org/bugzilla/>
In-Reply-To: <bug-280846-227@https.bugs.freebsd.org/bugzilla/>
References:  <bug-280846-227@https.bugs.freebsd.org/bugzilla/>

next in thread | previous in thread | raw e-mail | index | archive | help
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=3D280846

--- Comment #53 from Mark Millard <marklmi26-fbsd@yahoo.com> ---
(In reply to Henrich Hartzer from comment #50)

"failed to reclaim memory" is from multiple attempts
to increase the free RAM to not be below a target
threshold. There is a parameter for controlling the
number of attempts before the related OOM kills start.
For example in /boot/loader.conf I have:

#
# Delay when persistent low free RAM leads to
# Out Of Memory killing of processes:
vm.pageout_oom_seq=3D120

(You might well want bigger than 120. The default
is 12 --last I knew anyway. You might want 1200 or
12000 for experimenting. I expect too large a
number might end up with overflow problems but
have not checked the details.)

The setting does not ever disable the "failed to
reclaim memory" OOM activity. It just makes it
take longer to happen. It is more useful for
spanning temporary heavy RAM use, such as can
happen during buildworld, for example. It is not a
fix for permanent heavy RAM use. But it can help
allow exploring a context by giving more time
before the OOM activity happens.

--=20
You are receiving this mail because:
You are the assignee for the bug.=



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?bug-280846-227-j86Nxv3Dm8>