when this happen my server is unworkable but after some time it back to normal operation my iptables configuration: *mangle -A POSTROUTING -j CONNMARK --restore-mark -A POSTROUTING -m layer7 --l7proto applejuice -j MARK --set-mark 10 -A POSTROUTING -m layer7 --l7proto ares -j MARK --set-mark 10 -A POSTROUTING -m layer7 --l7proto bittorrent -j MARK --set-mark 10 -A POSTROUTING -m layer7 --l7proto directconnect -j MARK --set-mark 10 -A POSTROUTING -m layer7 --l7proto edonkey -j MARK --set-mark 10 -A POSTROUTING -m layer7 --l7proto fasttrack -j MARK --set-mark 10 -A POSTROUTING -m layer7 --l7proto gnutella -j MARK --set-mark 10 -A POSTROUTING -m layer7 --l7proto imesh -j MARK --set-mark 10 -A POSTROUTING -m layer7 --l7proto kugoo -j MARK --set-mark 10 -A POSTROUTING -m layer7 --l7proto mute -j MARK --set-mark 10 -A POSTROUTING -m layer7 --l7proto napster -j MARK --set-mark 10 -A POSTROUTING -m layer7 --l7proto openft -j MARK --set-mark 10 -A POSTROUTING -m layer7 --l7proto poco -j MARK --set-mark 10 -A POSTROUTING -m layer7 --l7proto soulseek -j MARK --set-mark 10 -A POSTROUTING -m layer7 --l7proto xunlei -j MARK --set-mark 10 -A POSTROUTING -j CONNMARK --save-mark COMMIT
Created attachment 102699 [details] log from kernel
Well, if it gets back alive after some time, then it doesn't hang. Insufficient info to debug the problem.
what more info i should post?
ok,more info: # uname -a Linux STALOWA 2.6.17-gentoo-r8 #1 Fri Nov 24 20:11:15 CET 2006 i686 AMD Athlon(tm) Processor AuthenticAMD GNU/Linux net-misc/l7-protocols-2006.10.18 net-misc/l7-filter-2.6 net-firewall/iptables-1.3.6-r1
For starters, stuff like kernel version and .config?
Created attachment 102700 [details] kernel config
use flags for iptables: l7filter -extensions -imq -ipv6 -static
(In reply to comment #2) > Well, if it gets back alive after some time, then it doesn't hang. so how should i call this situation?
when I hash some those line in iptables server stoped soft locking up: #-A POSTROUTING -m layer7 --l7proto imesh -j MARK --set-mark 10 #-A POSTROUTING -m layer7 --l7proto kugoo -j MARK --set-mark 10 #-A POSTROUTING -m layer7 --l7proto mute -j MARK --set-mark 10 #-A POSTROUTING -m layer7 --l7proto openft -j MARK --set-mark 10 #-A POSTROUTING -m layer7 --l7proto poco -j MARK --set-mark 10 #-A POSTROUTING -m layer7 --l7proto soulseek -j MARK --set-mark 10 #-A POSTROUTING -m layer7 --l7proto xunlei -j MARK --set-mark 10 so one of those rules may be the problem
I tried with your setup, I was not able to reproduce it. net-misc/l7-protocols-2006.10.18 net-misc/l7-filter-2.6 net-firewall/iptables-1.3.6-r1 uname -a Linux pluto 2.6.18.5 #1 PREEMPT Mon Dec 4 10:05:51 CET 2006 i686 VIA Samuel 2 CentaurHauls GNU/Linux
(In reply to comment #10) > I tried with your setup, I was not able to reproduce it. > > net-misc/l7-protocols-2006.10.18 > net-misc/l7-filter-2.6 > net-firewall/iptables-1.3.6-r1 > > uname -a > Linux pluto 2.6.18.5 #1 PREEMPT Mon Dec 4 10:05:51 CET 2006 i686 VIA Samuel 2 > CentaurHauls GNU/Linux > > > did you try to use those protocols I listed? I mean use them to download/upload something via p2p
p.s. I have LAN with 130 computers and with those configuration a mentioned above my main server soft lockup average once a day. Sometimes it soft lockups over and over for an hour or two, especially in the evening.
Is this just the imesh? Does it work without imesh? http://sourceforge.net/tracker/index.php?func=detail&aid=1602960&group_id=80085&atid=558668
reported upstream. I'm sure i'm not willing to debug dual cpu faults without the hardware. Sorry. Hope upstream finds it soon.