Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 06 May 2008 20:47:58 +0300
From:      Oleksandr Samoylyk <oleksandr@samoylyk.sumy.ua>
To:        net@freebsd.org
Subject:   Re: Problems with netgraph
Message-ID:  <482099CE.5050802@samoylyk.sumy.ua>
In-Reply-To: <4820801A.90504@samoylyk.sumy.ua>
References:  <48207C8B.4020509@samoylyk.sumy.ua> <4820801A.90504@samoylyk.sumy.ua>

next in thread | previous in thread | raw e-mail | index | archive | help
Oleksandr Samoylyk пишет:
> Oleksandr Samoylyk wrote:
>> Dear developers,
>>
>> Please read this thread:
>> http://lists.freebsd.org/pipermail/freebsd-net/2008-May/017891.html
>>
>> I'm using no encryption and no compression in mpd, so netgraph should 
>> fly.
>> It seems to get bad after 500 sessions...
>>
>> In FreeBSD 6 - it's swi1: net - 100% CPU
>> In FreeBSD 7 - it's em0 taskq - 100% CPU
>>
>> After playing with it I can make guess that's a netgraph problem.
>>
>> Maybe it's a poor design in netgraph the way it handle its tables and 
>> it's is probably not designed to add so many interfaces.
>>
>> Might be there something in the source code I can edit to improve the 
>> table lookups or hash table or whatever netgraph use to store and 
>> process the node information and do the ppp/gre work.
>>
> 
> I can also offer a shell access to server in order to investigate the 
> problem.
> 

# ngctl list
ngctl: send msg: No buffer space available

However:
kern.ipc.maxpipekva="400000000"
net.graph.maxalloc="8192"
net.graph.maxdgram="256000"
net.graph.recvspace="256000"

-- 
  Oleksandr Samoylyk
  OVS-RIPE



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?482099CE.5050802>