A fcgid timeout was the cause of the strange behavior.
I just set the time-limit to 0 (unlimited) and everything works as
suspected.
Maybe a hint in the documentation should be placed, that debugging without
this limit set to 0 is cumbersome and strange behavior is to be expected?
Best regards
Andreas
2011/6/8 Andreas Paeffgen <oegeeks[@]googlemail.com>
> Hi all,
>
> using xdebug.so 2.1.1 on centos 5.6 and php 5.3.6 (xdebug Compiled from the
> sources) in
> various client-debug-environments i always get socket errors, internal
> server errors and strange behaviors.
>
> All debugging is remote on a Mac with only one breakpoint.
>
> 1. Eclipse: Debug-Session starts correctly and shows some Variables.
> Meanwhile the website tries to load. After 30 sec, or so an internal server
>
> error occurs. No further debugging possible. Restart of the debug-session
> is
> necessary. Then the same happens again.
> 2. Netbeans: Configuring the Watch and Balloning in the options dialog
> gives a warning about unstable behavior, so i suspected a somewhat weird
> behavior. Debug-Session starts, but
> watch variables only show null values. Working with the debugger leads
> to internal server errors in the debug-session or to a socket error.
> Restart of netbeans is necessary after a socket error occurs to
> be able to start debugging again.
> 3. MacGDB (The more or less stable solution): Jumping to the
> breakpoint (a few very quick clicks are necessary) and the site still works
>
> and variables / functions are shown. Clicking not fast enough, a
> "Internal Server Error" occurs before reaching the breakpoint. The
> same happens with Step-Debugging up to the breakpoint. I suspect a
> timeout.
>
> Is this normal behavior? Any hints how to go further are appreciated.
> It is a symfony-framework based application. Firewall is disabled for
> testing purposes on the linux machine.
>
> Apache log shows a lot of the following lines with different pid, but
> other-
> wise the info is not very helpfull:
>
> [warn] mod_fcgid: process 10935 graceful kill fail, sending SIGKILL
>
>
>
> What kind of logs, etc should i look for to find out the cause of the
> problem?
>
> Best regards
> Andreas
>
>
Received on Fri Jun 10 2011 - 14:22:05 BST
This archive was generated by hypermail 2.2.0 : Mon Jun 25 2018 - 06:00:04 BST