Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 15 Jan 2013 16:16:40 -0800
From:      Artem Belevich <art@freebsd.org>
To:        Nicolas Rachinsky <fbsd-mas-0@ml.turing-complete.org>
Cc:        freebsd-fs <freebsd-fs@freebsd.org>
Subject:   Re: slowdown of zfs (tx->tx)
Message-ID:  <CAFqOu6jJnWdbikPmE1-UML5i_x7meF%2BiyY=9WBRyv2j7AeOaSg@mail.gmail.com>
In-Reply-To: <20130115224556.GA41774@mid.pc5.i.0x5.de>
References:  <CAFqOu6jrng=v8eVyhqV-PBqJM_dYy%2BU7X4%2B=ahBeoxvK4mxcSA@mail.gmail.com> <20130110193949.GA10023@mid.pc5.i.0x5.de> <20130111073417.GA95100@mid.pc5.i.0x5.de> <CAFqOu6gWpMsWN0pTBiv10WfwyGWMfO9GzMLWTtcVxHixr-_i3Q@mail.gmail.com> <20130114094010.GA75529@mid.pc5.i.0x5.de> <CAFqOu6hxfGt_M6Jo9qWeifDz9YnNc_Bd9H-GEe4RYtutaPvH5w@mail.gmail.com> <20130114195148.GA20540@mid.pc5.i.0x5.de> <CAFqOu6jwJ4qhbOovN_NhzusdQJvrbvUC3g93sziR=Uw99SGenw@mail.gmail.com> <20130114214652.GA76779@mid.pc5.i.0x5.de> <CAFqOu6jKX-Ks6C1RK5GwZ51ZVUSnGSe7S99_EfK%2BfwLPjAFFYw@mail.gmail.com> <20130115224556.GA41774@mid.pc5.i.0x5.de>

next in thread | previous in thread | raw e-mail | index | archive | help
On Tue, Jan 15, 2013 at 2:45 PM, Nicolas Rachinsky
<fbsd-mas-0@ml.turing-complete.org> wrote:
>  147     0 100098 kernel           zio_write_issue_ mi_switch+0x176 sleep=
q_wait+0x42 _cv_wait+0x129 space_map_load_wait+0x20 metaslab_activate+0x73 =
metaslab_alloc+0x7b2 zio_dva_allocate+0x9a zio_execute+0xc3 taskqueue_run_l=
ocked+0x85 taskqueue_thread_loop+0x4e fork_exit+0x11f fork_trampoline+0xe

It appears that lots of threads are stuck in
metaslab_activate->space_map_load_wait path. This sounds like CR#
6876962 in Solaris: "degraded write performance with threads held up
by space_map_load_wait(). This bug is fixed in patch 147440-05, -06 or
-07, which is current and contains the fix." Alas, I could not find
specifics on how the issue got fixed and whether the same fix is
present in illumos and FreeBSD.

You may want to update your system to very recent FreeBSD as quite a
few fixes were recently imported from illumos. Hopefully it will deal
with the issue. I'm out of ideas otherwise. Sorry.

--Artem



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?CAFqOu6jJnWdbikPmE1-UML5i_x7meF%2BiyY=9WBRyv2j7AeOaSg>