Hi All Unix Fellows...
Any encounter high ping timeout in AIX Server ? I scheduled the job to ping all the same segments servers all the time for 2 days without terminate. Found sometimes there are time the ping timeout is quite high ie. 10000ms
but most of time is 0ms. what cause this ? is this common ?
Thanks.
No, that is of course not normal. Try to narrow it down, checking if your network interfaces on the AIX server have any problems using commands like netstat -i , netstat -v , entstat or traceroute etc.
Also check if other servers in that network have similar problems.
If it is only the AIX server, check if it has maybe any performance problems CPU, Memory or I/O wise using vmstat to get a first glance.
Try to see if there is some kind of pattern, maybe related to workload, starting jobs, etc.
Good advice by zaxxon. Of course it could be a myriad of other reasons: network congestion, switch(router,hub, ....) bottleneck, broadcast storms, or even the temporary drop of bandwidth in a line.
Many years ago, i had a Laser-link as part of my network infrastructure, which mysteriously dropped in bandwidth every day at around 10:30 only to get back to speed at about 14:30 again. Turned out there was a cantina beneath the line of sight, which started to cook at this time and their exhaust air along with the steam from the cooking disturbed the connection, which throttled the bandwidth. It took about 2 months to find this one out, because everybody was busy looking at their network analyzers and nobody looked out the window.
In problems like this the only way is to meticulously describe the "chain of events" (or participating parts) and start to falsify every part of this until one hits the problematic one. And - see above - sometimes it helps to get some fresh air. ;-))
I hope this helps.
bakunin