From: Theodore Ts'o <tytso@mit.edu>
To: "Toralf Förster" <toralf.foerster@gmx.de>
Cc: linux-ext4@vger.kernel.org
Subject: Re: fuzz testing a KVM machine (x86) with ext4fs forces much more often the oom killer than btrfs
Date: Mon, 24 Nov 2014 22:15:12 -0500 [thread overview]
Message-ID: <20141125031512.GE31339@thunk.org> (raw)
In-Reply-To: <547390CC.2050306@gmx.de>
No idea; can you grab /proc/slabinfo and /proc/meminfo every minute or
so until it oom kills trinity? Maybe that will give us a hint of at
least where the memory is getting consumed.
It could be the extent status tree cache, for which the need for a
better shrinker is a known bug that we're working on, but that's just
a wild guess.
Thanks,
- Ted
next prev parent reply other threads:[~2014-11-25 3:15 UTC|newest]
Thread overview: 3+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-11-24 20:10 fuzz testing a KVM machine (x86) with ext4fs forces much more often the oom killer than btrfs Toralf Förster
2014-11-25 3:15 ` Theodore Ts'o [this message]
2014-11-25 21:31 ` Toralf Förster
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20141125031512.GE31339@thunk.org \
--to=tytso@mit.edu \
--cc=linux-ext4@vger.kernel.org \
--cc=toralf.foerster@gmx.de \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).