Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 14 Jul 2015 10:26:15 +0100
From:      Matthew Seaman <matthew@freebsd.org>
To:        freebsd-fs@freebsd.org
Subject:   Re: FreeBSD 10.1 Memory Exhaustion
Message-ID:  <55A4D5B7.2030603@freebsd.org>
In-Reply-To: <55A3A800.5060904@denninger.net>
References:  <CAB2_NwCngPqFH4q-YZk00RO_aVF9JraeSsVX3xS0z5EV3YGa1Q@mail.gmail.com> <55A3A800.5060904@denninger.net>

next in thread | previous in thread | raw e-mail | index | archive | help
This is an OpenPGP/MIME signed message (RFC 4880 and 3156)
--CnuH8DIVhwl4sPakXbhMRW56gHDDcql24
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: quoted-printable

On 07/13/15 12:58, Karl Denninger wrote:
> Put this on your box and see if the problem goes away.... :-)
>=20
> https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=3D187594
>=20
> The 2015-02-10 refactor will apply against 10.1-STABLE and 10.2-PRE (th=
e
> latter will give you a 10-line fuzz in one block but applies and works.=
)
>=20
> I've been unable to provoke misbehavior with this patch in and I run a
> cron job that does auto-snapshotting.  There are others that have run
> this patch with similarly positive results.

I know that you, Karl, and a number of others have been advocating to
get this patch set committed.  Having now personally run into the sort
of problems that this addresses I can say that I would very much like to
see this go in.  Conditional of course on this actually solving the
problems I and others have been experiencing without introducing
significant regressions elsewhere. It's only had a day's testing from me
so far, but it's looking good.  If it survives a week without the system
locking up, I'll be convinced.

I also know that some of the most experienced ZFS developers are
cautious about applying this patchset, and I do not doubt that they have
good reason to think so.

What I wonder is could it be possible to generate some sort of
regression tests where

   * The memory exhaustion effect or equivalent memory pressures can be
     triggered at will
   * The test doesn't require unfeasibly large resources to run
   * The behaviour provides a good model for real-world deployments

Maybe these tests would be too large-scale to run every day in Jenkins,
but having them available as part of, say, the release process, seems
like a no-brainer to me.

ZFS is one of FreeBSD's big selling points, and not having it work
reliably at scale is an embarrassment, to say the least.

	Cheers,

	Matthew




--CnuH8DIVhwl4sPakXbhMRW56gHDDcql24
Content-Type: application/pgp-signature; name="signature.asc"
Content-Description: OpenPGP digital signature
Content-Disposition: attachment; filename="signature.asc"

-----BEGIN PGP SIGNATURE-----
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=3wyn
-----END PGP SIGNATURE-----

--CnuH8DIVhwl4sPakXbhMRW56gHDDcql24--



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?55A4D5B7.2030603>