503 Error: Litespeed Web Server Enterprise v4.2.4

[Legba]

I have tried for days trying to overcome an issue with Litespeed. I have read every post I could find and have done everything they all said to no avail. I have a PHP script that needs to run for at least 20 minutes. I initially set the max executuion time to 3600 thinking that would do it. However, no matter what I do after about 300 seconds I get the following in the error log:

2013-11-03 09:05:01.085 [STDERR] Killing runaway process PID: 27214 with SIGTERM

and the web page displayed is:

503 Service Unavailable
The server is temporarily busy, try again later!

I have set every remotely relevant timeout I could find in the Litespeed config to 100 times its initial value and still I get the same issue after about 300 seconds.

Please could someone help?

Below are my settings, and they are confirmed using phpinfo.php

Tuning:
Max Connections 2000
Max SSL Connections 500
Connection Timeout (secs) 36000
Max Keep-Alive Requests 32000
Smart Keep-Alive (I have tried with this value set to No and Yes)
Keep-Alive Timeout (secs) 60
PHP5:
Initial Request Timeout (secs) 36000
Max Idle Time 1000

 

[NiteWave]

admin console->Server->External App->Environment,add
LSAPI_MAX_PROCESS_TIME=1800
default is 300 seconds(5 minutes)

please refer:
https://www.litespeedtech.com/support/wiki/doku.php/litespeed_wiki:php:lsapi-environment-variables

 

[bobykus]

Mine is dying too on

Killing runaway process PID: 463 with SIGTERM
2014-03-04 09:27:07.874 [STDERR] Child process with pid: 463 was killed by signal: 15, core dump: 0

which is super suxxx. It happens 2-3 times a day on servers with high i/o.
And it is not just php dying, whole web server stop respongind.
Why? Any way to avoid it?

 

[NiteWave]

which is super suxxx. It happens 2-3 times a day on servers with high i/o.

WHM/cPanel installed ?
check the setting of log rotate in WHM, it may cause restart

 

[bobykus]

No, not WHM/cPanel. What it have todo with log rotate?