Machine hangs with syslog-ng after a while
Posted: 2018/09/18 06:22:40
Hello,
I am facing a strange issue with syslog-ng v3.5.2 (also faced same issue with v3.9) running on centos 6.10.
After a few hours of running syslog-ng the machine becomes unresponsive, we cannot pass the ssh login screen, it accepts username but there's no password prompt. I cannot run any commands (crontab etc).
The only way to get into the machine, is to always keep an extra ssh session and login into that to kill syslog-ng. After killing syslog-ng machine becomes responsive again.
This machine accepts on average a couple of thousants packets per second from 300-400 hosts.
$ ss -s
Total: 330 (kernel 1024)
TCP: 1124 (estab 136, closed 958, orphaned 10, synrecv 0, timewait 948/0), ports 988
Transport Total IP IPv6
* 1024 - -
RAW 0 0 0
UDP 30 18 12
TCP 166 97 69
INET 196 115 81
FRAG 0 0 0
Anybody has faced a similar behaviour before?
Thanks
I am facing a strange issue with syslog-ng v3.5.2 (also faced same issue with v3.9) running on centos 6.10.
After a few hours of running syslog-ng the machine becomes unresponsive, we cannot pass the ssh login screen, it accepts username but there's no password prompt. I cannot run any commands (crontab etc).
The only way to get into the machine, is to always keep an extra ssh session and login into that to kill syslog-ng. After killing syslog-ng machine becomes responsive again.
This machine accepts on average a couple of thousants packets per second from 300-400 hosts.
$ ss -s
Total: 330 (kernel 1024)
TCP: 1124 (estab 136, closed 958, orphaned 10, synrecv 0, timewait 948/0), ports 988
Transport Total IP IPv6
* 1024 - -
RAW 0 0 0
UDP 30 18 12
TCP 166 97 69
INET 196 115 81
FRAG 0 0 0
Anybody has faced a similar behaviour before?
Thanks