Internal Traffic

[TechLineX]

Hello,

i have the problem, that my cluster nodes are randomly didn't see one specific host. I cannot ping it.

When I try to ping from the not reachable server, I get ping: sendmsg: Operation not permitted

After ~ 20-40 Seconds it work again for about 3 Minutes.

Help me.. :-(

#Found this in kern.log

 

[TechLineX]

Randomly multiple lxc containers are not reachable. And I can't get out why.

Pings out of the container looks like this:

ping 8.8.8.8
PING 8.8.8.8 (8.8.8.8) 56(84) bytes of data.
64 bytes from 8.8.8.8: icmp_seq=1 ttl=58 time=553 ms
64 bytes from 8.8.8.8: icmp_seq=10 ttl=58 time=1167 ms
64 bytes from 8.8.8.8: icmp_seq=11 ttl=58 time=143 ms
64 bytes from 8.8.8.8: icmp_seq=19 ttl=58 time=0.554 ms
From 212.48.117.147 icmp_seq=20 Destination Host Unreachable
From 212.48.117.147 icmp_seq=21 Destination Host Unreachable
From 212.48.117.147 icmp_seq=22 Destination Host Unreachable
64 bytes from 8.8.8.8: icmp_seq=24 ttl=58 time=2.46 ms
64 bytes from 8.8.8.8: icmp_seq=35 ttl=58 time=1.06 ms
^C
--- 8.8.8.8 ping statistics ---
36 packets transmitted, 6 received, +3 errors, 83% packet loss, time 35702ms
rtt min/avg/max/mdev = 0.554/311.445/1167.797/429.891 ms, pipe 3

 

[Stoiko Ivanov]

Found this in kern.log

Seems you've run into the default setting for the connection-tracking buckets - you need to increase them via sysctl
see e.g.
* https://security.stackexchange.com/questions/43205/nf-conntrack-table-full-dropping-packet
* https://morganwu277.github.io/2018/...e-of-nf-conntrack-table-full-dropping-packet/
* https://www.kernel.org/doc/Documentation/networking/nf_conntrack-sysctl.txt

Hope this helps!