Title: ntop child processes - CLOSE_WAIT state

I am running ntop 3.1 on Fedora Core 3 on an x86 with 512 MB RAM.
Processing about 5 million packets per hour.
After quite a few packets (number unknown), ntop will fork child processes (that do not end) and those processes seem to be acting strangely.

1) Most of the time the processes will show up (via netstat -tp) as having a single packet in their Recv-Q and a status of CLOSE_WAIT.  This status seems to never clear (i.e. the packet never gets processed).

2) Other child processes will be forked that have no sockets open at all (via 'lsof -c ntop -a -i' or via a netstat) but are running as real processes (status is sleeping).

None of these processes show up as being zombies.

Has anyone seen this before?
Is this a bug?
Is there a fix?


Arjen Dragt

_______________________________________________
Ntop mailing list
[email protected]
http://listgateway.unipi.it/mailman/listinfo/ntop

Reply via email to