HPUX TCP TIME_WAIT increasing in version 11.23

[Mohandass]

Dear All,

Attached is the output of (netstat -a|grep TIME_WAIT|grep 4327. As you can see in the file, the TIME_WAIT connection increasing a lot.

Please advice why we are having this symptoms and anyway to resolve it.

Regards
Mohan.

[johnsonpk]

Hi Mohan,

What application running on port 4327 ?

Are you facing any performance issue ?

The TIME_WAIT is defined by an ndd parameter tcp_time_wait_interval and most of the time the default value should be ok(60 seconds) unless you application vendor ask to change it


The HP-UX TCP stack can track literally millions of TIME_WAIT
connections with no particular decrease in performance and only a
slight cost in terms of memory. So, it should almost never be the case
that you need to decrease this value from its default of 60 seconds.

The above statement is from the url ftp://ftp.cup.hp.com/dist/networking/briefs/annotated_ndd.txt


regards!
Johnson

[rick jones]

Adding - basically, is suggests that whatever application is associated with port 4327, it seems to like to "churn" through TCP connections.

So, what application is listening at port 4327 on your system?

[Mohandass]

Thanks for your prompt reply, it is billing application running with real time diameter traffcis. Pump from the other client to this server. Apparently, I guess large number of connections was the root cause of the tcp TIME_WAIT problem.

We guess its application bugs. But we need some sharing from you as well to understand on the TCP connections.

[Raj D.]

To find more, Find out the process running on 4327 port, and do tusc on that pid ( # tusc pid ). See the progress. What ever you see , you can suggest to the app team, It seems to be it is waiting on the connection to be estblish ,

Hth,
Raj.