2009/4/7 Deepan Chakravarthy <codeshepherd@gmail.com>:
> Hi,
>   I am using nginx with fast-cgi .  When I run
> $netstat -np | grep 127.0.0.1:9000
> I find lot of connections in TIME_WAIT state. Is this because of high
> keepalive_timeout value ?   When lot of people use (5 requests per second)
>  nginx takes more  time to respond. System load goes more than 10 during
> peak hours.

This is because of how TCP works.


> debian:~# netstat -np | grep 127.0.0.1:9000
> tcp        0      0 127.0.0.1:9000          127.0.0.1:45603
> TIME_WAIT  -
> tcp        0      0 127.0.0.1:9000          127.0.0.1:45601
> TIME_WAIT  -


If you were on FreeBSD, you could disable TIME_WAIT on loopback
completely by setting:

sysctl net.inet.tcp.nolocaltimewait=1

Maybe debian has similar /proc tuning or you can try to decrease msl
(maximum segment lifetime) or beter don't use inet sockets on loopback
- use unix sockets.





--
regards,
Artis Caune

<----. CCNA | BSDA
<----|====================
<----' didii FreeBSD