Skip site navigation (1)Skip section navigation (2)
Date:      Fri, 4 Dec 2009 13:07:49 -0800
From:      "K. Macy" <kmacy@freebsd.org>
To:        Mike Tancsa <mike@sentex.net>
Cc:        Zaphod Beeblebrox <zbeeble@gmail.com>, FreeBSD Stable <freebsd-stable@freebsd.org>
Subject:   Re: Quggaa locking hard.
Message-ID:  <82c4140e0912041307m5f764f00vee1a7f37d4782a5b@mail.gmail.com>
In-Reply-To: <200912041456.nB4Eu8oN017070@lava.sentex.ca>
References:  <5f67a8c40912031946v1ab3efua5e43bd7d767361a@mail.gmail.com> <200912041456.nB4Eu8oN017070@lava.sentex.ca>

next in thread | previous in thread | raw e-mail | index | archive | help
If you have a large number of routes then you will want to disable the
flowtable. The default maximum number of cacheable flows is fairly
small, raising it can help on the low-end, but fundamentally its an
optimization for systems that have fewer than a few thousand
simultaneous peers - the common case.

I do have longer term plans for moving to lock-free L3 and L2 so that
applications with large numbers of prefixes will also no longer be
hampered by high locking overhead.


-Kip



On Fri, Dec 4, 2009 at 6:56 AM, Mike Tancsa <mike@sentex.net> wrote:
> At 10:46 PM 12/3/2009, Zaphod Beeblebrox wrote:
>>
>> I'm still investigating this, but my quagga is locking hard on FreeBSD 8=
.0
>> and not locking hard on 7.2. =A0It seems (at this early point in the
>> investigation) that both bgpd and zebra are wedging and zebra is listed =
as
>> being in the "RUN" state.
>>
>> curiously, the load is also 4.0 (exactly the number of cores in the
>> machine)
>> even though the machine also reads 100% idle.
>
>
> I think I am seeing something similar on a test box. =A0I was loading up =
the
> box with 200k routes to do testing with. =A0Kernel is default, save for a=
 few
> unused drivers removed. If I take out
> options =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0FLOWTABLE =A0 =A0 =A0 =A0 =A0 =A0 =
=A0 # per-cpu routing cache
> from the kernel, load avg is back to normal. =A0This issue only seems to =
have
> come up in the past week or so as the previous kernel from ~8 days ago wa=
s
> OK.
>
> last pid: =A06229; =A0load averages: =A02.00, =A02.00, =A02.00 =A0 =A0 =
=A0 =A0 =A0 =A0 =A0 up
> 1+17:33:02 =A009:39:31
> 141 processes: 7 running, 106 sleeping, 28 waiting
> CPU: =A00.0% user, =A00.0% nice, 22.2% system, =A00.0% interrupt, 77.8% i=
dle
> Mem: 98M Active, 2233M Inact, 187M Wired, 36K Cache, 112M Buf, 979M Free
> Swap: 8192M Total, 8192M Free
>
> =A0PID USERNAME PRI NICE =A0 SIZE =A0 =A0RES STATE =A0 C =A0 TIME =A0 WCP=
U COMMAND
> =A0 22 root =A0 =A0 =A076 =A0 =A0- =A0 =A0 0K =A0 =A0 8K CPU3 =A0 =A03 =
=A041.5H 100.00% flowcleaner
> =A0 11 root =A0 =A0 171 ki31 =A0 =A0 0K =A0 =A032K CPU2 =A0 =A02 =A041.5H=
 100.00% {idle: cpu2}
> =A0 11 root =A0 =A0 171 ki31 =A0 =A0 0K =A0 =A032K CPU1 =A0 =A01 =A041.5H=
 100.00% {idle: cpu1}
> =A0 11 root =A0 =A0 171 ki31 =A0 =A0 0K =A0 =A032K RUN =A0 =A0 0 =A041.4H=
 100.00% {idle: cpu0}
> =A0869 root =A0 =A0 =A0 4 =A0 =A00 64860K 64488K select =A00 =A0 4:12 =A0=
0.00% bgpd
> =A0 11 root =A0 =A0 171 ki31 =A0 =A0 0K =A0 =A032K RUN =A0 =A0 3 =A0 2:09=
 =A00.00% {idle: cpu3}
> =A0 20 root =A0 =A0 =A044 =A0 =A0- =A0 =A0 0K =A0 =A0 8K syncer =A00 =A0 =
1:00 =A00.00% syncer
> =A0 12 root =A0 =A0 -32 =A0 =A0- =A0 =A0 0K =A0 224K WAIT =A0 =A01 =A0 0:=
47 =A00.00% {swi4: clock}
> =A0 =A00 root =A0 =A0 -68 =A0 =A00 =A0 =A0 0K =A0 =A080K - =A0 =A0 =A0 2 =
=A0 0:03 =A00.00% {fw0_taskq}
> =A01230 root =A0 =A0 =A076 =A0 =A00 =A03348K =A01160K ttyin =A0 2 =A0 0:0=
2 =A00.00% getty
> =A0863 root =A0 =A0 =A096 =A0 =A00 24640K 24232K RUN =A0 =A0 2 =A0 0:02 =
=A00.00% zebra
> =A0 12 root =A0 =A0 -32 =A0 =A0- =A0 =A0 0K =A0 224K WAIT =A0 =A02 =A0 0:=
01 =A00.00% {swi4: clock}
> =A0 14 root =A0 =A0 -16 =A0 =A0- =A0 =A0 0K =A0 =A0 8K - =A0 =A0 =A0 0 =
=A0 0:01 =A00.00% yarrow
>
>> _______________________________________________
>> freebsd-stable@freebsd.org mailing list
>> http://lists.freebsd.org/mailman/listinfo/freebsd-stable
>> To unsubscribe, send any mail to "freebsd-stable-unsubscribe@freebsd.org=
"
>
> --------------------------------------------------------------------
> Mike Tancsa, =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =
=A0 =A0 =A0 =A0tel +1 519 651 3400
> Sentex Communications, =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =
=A0 =A0mike@sentex.net
> Providing Internet since 1994 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0www.=
sentex.net
> Cambridge, Ontario Canada =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0=
 www.sentex.net/mike
>
>



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?82c4140e0912041307m5f764f00vee1a7f37d4782a5b>