On Sat, 23 Jun 2012, Rudy Zijlstra wrote:
On 22-06-12 21:38, Henrique de Moraes Holschuh wrote:
On Fri, 22 Jun 2012, Rudy Zijlstra wrote:
let system run with IPv4& IPv6 routing for about 1 month
IPv6 routing will start to fail
IPv4 routing becomes slow and unpredictable
no obvious causes visible in the system. top and friends do not show a cpu hog
a reboot will bring the system back to normal behaviour.
Please use (as root) "ip neigh show", and "ip route list cache" to try to
track down any weird differences between the box when it is behaving
normally, and the box when wedged. You may want to compare it to a healthy
box on the same network segment.
You can also try to see if "ip route flush cache" and "ip neigh flush" can
unwedge the system. After a flush, "ip neigh show" and "ip route list
cache" should return very few, if any, entries.
Thanks, i've stored the current output of these commands, including
the IPv6 version, so i can compare when trouble hits again in some
weeks.
You probably want to store their output once a day. If it is a
neighbour/route cache leak or malfunction of some sort (e.g. routes getting
stuck in the presence of ICMP redirects), you should be able to notice that
old crap is accumulating over time.
If possible, do the same in a box that does not show the same problem
(ideally in the same network segment), so that you have a baseline to
compare to.
Note that it could be something else entirely, don't rule out hardware
malfunction (sometimes cleared if you down the interfaces and then bring
them up again), or driver issues (sometimes cleared if you rmmod + modprobe
the buggy driver). And make sure the box is running the latest firmware
(BIOS/UEFI, NIC firmware...).