Re: Advice for nf_conntrack: table full: increase netfilter.ip_conntrack or descrease timeouts?

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



Are you sure nf_conntrack_max is the problem? do you see "hashtable
full drop connection" on dmesg?

as a general advise , run conntrack -L (or cat /proc/xxx session
table) when problem happens and analyse where the session comes from
first.

Cheers.

On Thu, Jun 14, 2012 at 5:52 PM, José Pablo Pérez
<josepablo@xxxxxxxxxxxx> wrote:
> Hello again guys...
>
> I have a webserver who should be able to handle 10,000 established
> requests, right now its been tested with 1000 but it fails because of
> to many connections on TIME_WAIT
>
> I have been reading prior to posting and it seems I have two options:
>
> Increse these two:
> sysctl -a | grep conntrack | grep net.nf_conntrack_max
> net.nf_conntrack_max = 31772
>
> sysctl -w net.netfilter.nf_conntrack_max=131072
> echo 32768 > /sys/module/nf_conntrack/parameters/hashsize
>
> Or decrease the timeouts.. which right now I have...
>
> Right now I have:
>
> sysctl -a | grep conntrack | grep timeout
> net.netfilter.nf_conntrack_generic_timeout = 600
> net.netfilter.nf_conntrack_tcp_timeout_syn_sent = 120
> net.netfilter.nf_conntrack_tcp_timeout_syn_recv = 60
> net.netfilter.nf_conntrack_tcp_timeout_established = 432000
> net.netfilter.nf_conntrack_tcp_timeout_fin_wait = 120
> net.netfilter.nf_conntrack_tcp_timeout_close_wait = 60
> net.netfilter.nf_conntrack_tcp_timeout_last_ack = 30
> net.netfilter.nf_conntrack_tcp_timeout_time_wait = 120
> net.netfilter.nf_conntrack_tcp_timeout_close = 10
> net.netfilter.nf_conntrack_tcp_timeout_max_retrans = 300
> net.netfilter.nf_conntrack_tcp_timeout_unacknowledged = 300
> net.netfilter.nf_conntrack_udp_timeout = 30
> net.netfilter.nf_conntrack_udp_timeout_stream = 180
> net.netfilter.nf_conntrack_icmp_timeout = 30
> net.netfilter.nf_conntrack_events_retry_timeout = 15
>
>
>
> Iam more keen to the second because high
> net.ipv4.netfilter.ip_conntrack_max according to what I read can lead
> to system freeze.... so can anybody offer me a reference to tune and
> lower these paratemers? I tried lowering the
> net.ipv4.netfilter.ip_conntrack_tcp_timeout_time_wait  to 5 ... but
> that didnt change anything much.
>
> Thanks..!
>
>
>
> --
> To unsubscribe from this list: send the line "unsubscribe netfilter" in
> the body of a message to majordomo@xxxxxxxxxxxxxxx
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
--
To unsubscribe from this list: send the line "unsubscribe netfilter" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html


[Index of Archives]     [Linux Netfilter Development]     [Linux Kernel Networking Development]     [Netem]     [Berkeley Packet Filter]     [Linux Kernel Development]     [Advanced Routing & Traffice Control]     [Bugtraq]

  Powered by Linux