Enhancement: Logging of when process/thread limit is reached

[GoogleCodeExporter]

From what I can tell, when you're running in daemon mode and the
processes/threads are all busy dealing with requests, any other requests
received will be queued and handled once a process/thread becomes available.

It would be useful to be able to determine whether a daemon was hitting
these limits. This would allow you to manually increase the number of
processes/threads available in order to provide more capacity.

I've not thought in detail how this would be achieved, but perhaps an extra
option on the WSGIDaemonProcess directive that would result in a "notice"
or "warning" level message being written to the Apache error log should the
process/thread limit be reached.

Original issue reported on code.google.com by fras...@gmail.com on 28 Jan 2009 at 9:29

[GoogleCodeExporter]

This can be done within a specific daemon process using a WSGI middleware 
wrapper around the application. 
You will just need to tell the middleware how many threads are in process 
manually.

This has been discussed before on the mailing list as part of a more 
comprehensive middleware component that 
could be developed to analyse thread utilisation and determine capacity of a 
specific configuration.

Original comment by Graham.Dumpleton@gmail.com on 29 Jan 2009 at 12:51

• Added labels: Type-Other
• Removed labels: Type-Defect

[GoogleCodeExporter]

Not going to make any code changes related to this, although possibly need to 
document how this could be done 
as WSGI middleware.

Original comment by Graham.Dumpleton@gmail.com on 6 Mar 2009 at 4:44

• Changed state: WontFix