Re: (ipt_log_packet, sb_add) 3.6.0-rc2 kernel panic - not syncing; Fatal exception in interrupt

On Sun, Sep 02, 2012 at 15:28:34 +0200, Florian Westphal wrote:

Sami Farin [off-list ref] wrote:

quoted

I get this panic every 1-2 days.
Also with 7a611e69b26069a511d9d5251c6a28af6c521121 (commit before 3.6.0-rc4).

Could you please post iptables-save output?
My guess is you're using NFLOG in INPUT?

I don't have NFLOG but I have some like this
-j LOG --log-tcp-options --log-ip-options --log-uid --log-macdecode
I also have CONFIG_IP_ROUTE_CLASSID=y , using also htb + fq_codel.

using these ipt* modules
ipt_ECN                 1849  1 
ipt_REJECT              2661  12 
iptable_filter          1650  1 
iptable_mangle          1631  1 
iptable_raw             1422  1 
ip_tables              17482  3 iptable_filter,iptable_mangle,iptable_raw
x_tables               20149  26 ip6table_filter,arp_tables,ip6table_mangle,xt_length,ip_tables,xt_tcpudp,arptable_filter,xt_limit,xt_owner,xt_state,xt_connlimit,xt_conntrack,xt_NOTRACK,xt_LOG,xt_set,xt_hashlimit,xt_multiport,iptable_filter,xt_CLASSIFY,ip6t_rt,xt_connmark,ipt_REJECT,iptable_mangle,ipt_ECN,ip6_tables,iptable_raw

If so, I bet its caused by the tcp early demux stuff.
Does that crash go away with ip_early_demux sysctl off?

I try that, thanks!
 

My guess is its assigning skb->sk with TIMEWAIT sockets, which
would explain the crash.

-- 
Do what you love because life is too short for anything else.