Re: Segfaults using PEAR
| From: | Vinai Kopp | Date: | Tue, 25 Sep 2001 15:51:39 +0000 |
| Subject: | Re: Segfaults using PEAR | ||
| References: | 1 | Groups: | php.pear.dev |
| Request: | Send a blank email to pear-dev+get-2039@lists.php.net to get a copy of this message | ||
Sorry I can't give more detailed information, but if I had found a
reliable clue to what caused the problem I would have posted it.
Maybe it's a hardware fault, since this doesn't happen on other machines
(that's what I told the management).
This happend about 2 Months ago, here is what I recall:
The Segfault in XML/Parser seemed to happen when xml_set_object() was
called. But as I said, I couldn't nail it, because even when I removed the
call in some test code, the segfault happend when xml_parser_create() was
called (earlier in the script). I only could reproduce the segfaults using
the xml functions. The other segfaults seemed to happen somewhat randomly.
I used Expat bundeled with apache 1.3.19, I hadn't changed that.
Simple testing scripts ran without a complaint, but we use a komplex
structure of objects, and that seemed to facilitate the segfaults.
The more I looked the more confused I got about it.
I feel like a lamer writing this, since I can't supply any usefull
information, so I'll stop here.
Still interested to hear what others find out,
Vinai
On Tue, 25 Sep 2001, Martin Jansen wrote:
> On Tue, 25 Sep 2001 16:46:29 +0200 (CEST), Vinai Kopp wrote:
>
> >We got segfaults on one of our dev machines, too. They seemed to be
> >connected to XML/Parser.php (but not only),
>
> What does 'not only' mean? What other PEAR scripts were involved? What
> version of Expat did you use (perhaps it was somewhat buggy)?
--
A Microsoft Haiku: First snow, then silence.
This thousand-dollar screen dies
So beautifully.