Herbert Xu wrote: > Tobias Diedrich <ranma+kernel@xxxxxxxxxxxx> wrote: > > > > Meanwhile I added a slab statistic rrd script. Nothing obvious to > > see on ari or yumi yet, but on oni (which after all is the most > > affected by this) I can see 'size_2048' and 'TCPv6' growing > > steadily along with the route cache size (Presumably 'ip_dst_cache', > > which is also growing larger at a more or less constant rate). > > OK that almost certainly means that we have a leak of TCPv6 > objects which would then cause a leak of dst objects. > > So you need to track down why TCPv6 isn't getting cleaned up. > First check the obvious, do you have any process(es) that are > hoarding TCPv6 sockets? Hmm, how do I look for that, if netstat doesn't look suspicous? > If not do you see a growing list of dead sockets in netstat? No, I don't think so. oni:~# netstat -an | wc -l 139 oni:~# netstat -tun Active Internet connections (w/o servers) Proto Recv-Q Send-Q Local Address Foreign Address State tcp 0 0 217.172.187.75:53668 xx.xxx.xx.x:6543 ESTABLISHED tcp 0 0 217.172.187.75:40180 yyy.yyy.yyy.yy:6871 TIME_WAIT tcp 0 0 217.172.187.75:58480 xx.xxx.xx.x:6543 ESTABLISHED tcp 0 0 217.172.187.75:49213 xx.xxx.xx.x:6543 ESTABLISHED tcp 0 0 217.172.187.75:34828 xx.xxx.xx.x:6543 ESTABLISHED tcp 0 0 217.172.187.75:44859 xx.xxx.xx.x:6543 ESTABLISHED tcp 0 0 217.172.187.75:41863 xx.xxx.xx.x:6543 ESTABLISHED tcp 0 0 217.172.187.75:38063 xx.xxx.xx.x:6543 ESTABLISHED tcp6 0 0 ::ffff:217.172.187.7:80 ::ffff:zz.zzz.zz.z:3729 TIME_WAIT tcp6 0 0 ::ffff:217.172.187.7:80 ::ffff:zz.zzz.zz.z:3726 TIME_WAIT tcp6 0 0 ::ffff:217.172.187.7:80 ::ffff:aa.aa.aaa.:59396 ESTABLISHED tcp6 0 992 ::ffff:217.172.187.7:22 ::ffff:bb.bbb.bb.:57870 ESTABLISHED HTH, -- Tobias PGP: http://9ac7e0bc.uguu.de このメールは十割再利用されたビットで作られています。 - To unsubscribe from this list: send the line "unsubscribe linux-net" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
