Massive apache slowdown after high load period

[Steven Hartland]

We've just had a box which was subjected to a
large amount of small apache requests once this
subsided the response of apache was still very
slow. Many seconds between requests showing
in tcpdump and response being returned.

Upon investigation it seemed the number of
sockets in teardown / mbufs allocated might
be the cause has anyone else experienced that?

netstat -m
4294880568 mbufs in use
29885/76800 mbuf clusters in use (current/max)
0/5/6656 sfbufs in use (current/peak/max)
38088 KBytes allocated to network
0 requests for sfbufs denied
0 requests for sfbufs delayed
0 requests for I/O initiated by sendfile
687 calls to protocol drain routines

netstat showed
   1 CLOSE_WAIT
   6 LAST_ACK
  41 FIN_WAIT_1
 129 FIN_WAIT_2
 295 ESTABLISHED
1083 TIME_WAIT

A restart of apache made no difference a reboot
of the machine was required to return normal
performance.

N.B. 5.4-RELEASE-p8 is in use on the machine in
question.

    Steve


================================================
This e.mail is private and confidential between Multiplay (UK) Ltd. and the person or entity to whom it is addressed. In the event of misdirection, the recipient is prohibited from using, copying, printing or otherwise disseminating it or any information contained in it. 

In the event of misdirection, illegible or incomplete transmission please telephone +44 845 868 1337
or return the E.mail to postmaster at multiplay.co.uk.