Hi all,
The latest version has seen this problem go away. Still not too sure what
the underlying cause was, but am glad that it's working now.
Thanks,
Mark
On Tue, Jul 17, 2012 at 9:11 PM, Mark Slade <markandrewslade[@]gmail.com>wrote:
> Hi all,
>
> I'm new to Xdebug and trying to get started. I it successfully installed
> and configured; I'm able to start debugging from my remote IDE and step
> through code. But I keep hitting segfaults and I'm not sure why, nor am I
> sure the best direction investigate. Guidance would be much appreciated.
>
> xdebug 2.2.0
> apache 2.2.15
> PHP 5.4.3
> NetBeans 7.1.2
>
> The script runs fine if I run it without debugging (by removing xdebug
> from php.ini or just not starting a session). It doesn't generate
> segfaults, exceptions, errors, warnings, or notices, and it doesn't take
> abnormally long.
>
> If I start debugging and step through, it reaches a certain line where
> NetBeans disconnects indicating a "Socket Exception", and I see a segfault
> in Apache's error_log. This always happens on a line before the line that
> produces output. None the less, I do end up seeing output and it's
> identical to when this doesn't happen, as if the script is completing after
> xdebug dies and then the process is segfaulting. The remote log (
> http://pastebin.com/9VUY2AcJ) seems like xdebug is just crashing. That
> is the end of the log after the line that causes problems. The
> conversation is stopping right after context_get, note the lack of a
> closing "-->" to the command, and then the log re-opens.
>
> If I load the Apache coredump into gdb and backtrace (
> http://pastebin.com/w21zXnH8) it shows xdebug_var_export_xml_node at the
> top of the stack. I'm not a gdb pro so I'm not sure where to go from here.
> I can't provide the code that's triggering this because it's a sizeable
> codebase and proprietary.
>
> I tried running the script itself in gdb with debugclient on the other
> end, but I'm apparently not doing something right. When I first connect
> with debugclient I run "step_over -i XXXX" and it seems to just run the
> entire script and produce output. In this case, no segfault / error. The
> raw response I get from debugclient includes status="stopping" reason="ok".
>
> Any advice? Can I provide more information that might be useful?
>
> Mark
>
Received on Sat Jul 21 2012 - 02:37:50 BST
This archive was generated by hypermail 2.2.0 : Mon Jun 25 2018 - 06:00:04 BST