> You can't blame me for making a mess of the package system.
I can. You should not use Debian unstable and experimental if you
don't understand packaging system well. Also you didn't include
relevant information (reportbug template is minimum) - please read
http://www.debian.org/Bugs/Reporting
>
> "OS" == Ondřej Surý writes:
OS> Could you please install libmysqlclient15off from s-p-u?
OS> There are random segfaults caused by regression in upstream mysql
OS> library which was fixed in 5.0.51a-24+lenny2+spu1
I don't know what "s-p-u" is.
All I know is I have this obsolete package ins
Could you please install libmysqlclient15off from s-p-u?
There are random segfaults caused by regression in upstream mysql
library which was fixed in 5.0.51a-24+lenny2+spu1
Thanks,
Ondrej
On Sun, Jan 10, 2010 at 20:04, wrote:
> # strace php /dev/null 2>&1
> Segmentation fault
> # dmesg |tail -
# strace php /dev/null 2>&1
Segmentation fault
# dmesg |tail -n 1
[ 9964.678720] php[3698]: segfault at b68e7dc0 ip b68e7dc0 sp b684e39c error 4
in libXdmcp.so.6.0.0[b6ad2000+4000]
So there's your answer, libxdmcp apparently. Note that this bug only
occurs on my Thinkpad R50e.
It even happens in
Is mysql or mysqli loaded? Could you try to not load them and repeat
those tests?
Ondrej
On Fri, Jan 8, 2010 at 15:11, wrote:
> OS> Does the first machine (on which bug occurs) have nvidia hardware or
> OS> module loaded?
> It's just a standard IBM Thinkpad R50e.
> Maybe next time I'm on it I'l
OS> Does the first machine (on which bug occurs) have nvidia hardware or
OS> module loaded?
It's just a standard IBM Thinkpad R50e.
Maybe next time I'm on it I'll run
$ lshw|grep -i nvidia
to try and find out. I do know for sure that none of the
# aptitude search ~nnvidia|wc -l
48
packages are inst
Does the first machine (on which bug occurs) have nvidia hardware or
module loaded?
Ondrej
On Fri, Jan 8, 2010 at 14:31, wrote:
> Odd. With the same setup, on an Asus EEEPC 702 8G, the bug does not occur!
>
> *-cpu
> product: Intel(R) Celeron(R) M processor 900MHz
>
Hi,
I did some additional testing; I wrote a small program that loads each
module separately and then runs php 100 times (on an empty file,
/dev/null) to see if it crashes.
The full code and results is available at
http://pastie.org/private/yg18udvpnpmmkyg2hht1kq
To summarize the results:
On Thursday 13 August 2009 14:58:58 Bas Verhoeven wrote:
[...]
>
> Although this information doesn't seem to have become more usable, so it
> could be that I'm doing something wrong here.
>
> > On what architecture(s) are you experiencing the problem?
> >
> > How many cores does the CPU have?
>
>
9 matches
Mail list logo
