Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 23 Nov 2021 15:17:18 -0500
From:      Chris Ross <cross+freebsd@distal.com>
To:        Ronald Klop <ronald-lists@klop.ws>
Cc:        Andriy Gapon <avg@freebsd.org>, Mark Johnston <markj@freebsd.org>, freebsd-fs <freebsd-fs@freebsd.org>
Subject:   Re: swap_pager: cannot allocate bio
Message-ID:  <42CC8B5C-1E08-4EE4-8CCE-4EC959C60BD1@distal.com>
In-Reply-To: <op.1da5pwr2kndu52@joepie>
References:  <9FE99EEF-37C5-43D1-AC9D-17F3EDA19606@distal.com> <09989390-FED9-45A6-A866-4605D3766DFE@distal.com> <op.1cpimpsmkndu52@joepie> <4E5511DF-B163-4928-9CC3-22755683999E@distal.com> <YY7KSgGZY9ehdjzu@nuc> <19A3AAF6-149B-4A3C-8C27-4CFF22382014@distal.com> <6DA63618-F0E9-48EC-AB57-3C3C102BC0C0@distal.com> <35c14795-3b1c-9315-8e9b-a8dfad575a04@FreeBSD.org> <YZJzy%2ByI40wXFYjd@nuc> <b2121d25-0782-5cc3-2b55-33ba11c41995@FreeBSD.org> <471B80F4-B8F4-4D5A-9DEB-3F1E00F42A68@distal.com> <A945F841-3EFD-46EE-91B2-38287D2F2F8D@distal.com> <op.1da5pwr2kndu52@joepie>

next in thread | previous in thread | raw e-mail | index | archive | help


> On Nov 23, 2021, at 06:18, Ronald Klop via freebsd-fs =
<freebsd-fs@freebsd.org> wrote:
>=20
> Just a wild guess. Would it help if you set a limit in the =
vfs.zfs.arc_max variable?
> Maybe that will help lower the memory pressure and gain some =
stability.
> You can use the zfs-stats package to see the current ARC size.
>=20
> My RPI4 gives:
> # zfs-stats -A
> ...
> ARC Size:                               28.19%  1.93    GiB
>        Target Size: (Adaptive)         30.47%  2.08    GiB
>        Min Size (Hard Limit):          3.58%   250.80  MiB
>        Max Size (High Water):          27:1    6.84    GiB
> ...
>=20
> You can use your stats to tune it to not use too much memory for ARC =
and leave more for the running applications so swapping might also be =
reduced.
>=20
> You can check zfs-stats -E to see if the ARC cache hit ratio is still =
ok with limited ARC.


Thanks for that detail regarding zfs-stats.  Someone else had suggested =
limiting
arc_max, but the concern was that it might just mask the real problem.  =
It was
my fallback plan, though, if we couldn=E2=80=99t make progress towards =
avoiding the
lock-up.

                  - Chris=




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?42CC8B5C-1E08-4EE4-8CCE-4EC959C60BD1>