sdyates has asked for the wisdom of the Perl Monks concerning the following question:
Well folks, this one has been puzzling me for weeks. I have read RFCs and many documents, but it appears that no one can give me a definate.
Background
Trying to determine the level of close-wait states on an Internet0based application. Netstat -na is very good for getting the basic info. Now I can monitor the states with a script. However, while I cannot terminate the sockets, I can restart the application, causing all sockets to resolve themselves. However, before I start monkeyong around with this, I need to get definitive answers as the the cause of close-wait states and instead of writing a script to restart an application after a number of close-wait state appear, perhaps a script to prevent them would be better?
Below is a simple diagram taken from the RFC on TCP based connections:
Can anyone provide me with a few guidlines on which way to go? Or, has something already been created to do the job.
By the way, I also tried tunning the tcp values of the OS which has made little to no difference.
|
|---|
| Replies are listed 'Best First'. | |
|---|---|
|
Re: Smartly dealing with CLOSE-WAIT states
by fglock (Vicar) on Sep 27, 2002 at 17:15 UTC | |
|
Re: Smartly dealing with CLOSE-WAIT states
by fglock (Vicar) on Sep 28, 2002 at 06:10 UTC | |
by sdyates (Scribe) on Oct 01, 2002 at 21:11 UTC |