Skip site navigation (1)Skip section navigation (2)
Date:      Mon, 11 Jul 2011 14:02:22 +0200
From:      Vlad Galu <dudu@dudu.ro>
To:        Eugene Grosbein <egrosbein@rdtc.ru>
Cc:        "net@freebsd.org" <net@freebsd.org>
Subject:   Re: Repeating kernel panic within dummynet
Message-ID:  <30BDDC28-275C-4DAA-9AC8-621F36A19D3B@dudu.ro>
In-Reply-To: <4E1AE3C8.8030708@rdtc.ru>
References:  <4E1AE1AD.4020907@rdtc.ru> <833EECF8-245B-4085-B853-AC753DBE0D19@dudu.ro> <4E1AE3C8.8030708@rdtc.ru>

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

On Jul 11, 2011, at 1:51 PM, Eugene Grosbein wrote:

> 11.07.2011 18:45, Vlad Galu =D0=BF=D0=B8=D1=88=D0=B5=D1=82:
>>=20
>> On Jul 11, 2011, at 1:42 PM, Eugene Grosbein wrote:
>>=20
>>> Hi!
>>>=20
>>> My FreeBSD 8.2/amd64 routers use dummynet heavily
>>> and keep panic with the *same* KDB backtrace:
>>>=20
>>> dummynet: bad switch -256!
>=20
> Forgot to mention that I use io_fast dummynet mode
> and have increased pipe lengths:
>=20
> net.inet.ip.dummynet.pipe_slot_limit=3D1000
> net.inet.ip.dummynet.io_fast=3D1
>=20
> Distinct pipes do really use long lengths.
>=20
>>> Sadly, I cannot add options INVARIANTS to the kernel because it =
makes my mpd-based
>>> routers to panic very often (every 2-3 hours) due to famous =
'dangling pointer'
>>> problem - PPPoE user disconnects, its ngXXX interface got removed, =
then its traffic
>>> goes out various system queues (netisr, dummynet etc.) and another =
kind of panic
>>> occurs due to INVARIANTS' references to non-existent ifp.
>>=20
>> Hi Eugene,
>> If your ISR threads aren't already bound to CPUs, you can bind them =
and try using INVARIANTS.
>=20
> Please explain how to bind them. I have 4-core boxes with 4 NICs =
grouped to 2 laggs,
> one lagg(4) for uplink and another one for downlink.
>=20

net.isr.bindthreads=3D1

I'm not sure how and if that would help your particular setup, but it =
did so in Adrian Minta's recent netgraph/mpd experiments. According to =
an off-list chat I had with him, the machine would panic unless the ISRs =
were bound.

> Eugene Grosbein




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?30BDDC28-275C-4DAA-9AC8-621F36A19D3B>