httpd-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Aram Mirzadeh <...@qosina.com>
Subject Re: WWW Form Bug Report: "core dump reported almost every time
Date Sun, 26 May 1996 17:37:56 GMT
At 01:22 PM 5/26/96 -0400, you wrote:
>So let me see if I have this right:
>
>Old Linux = no core dump
>New Linux + libc 5.3.12 = core dump

make that no core dump.  When they say it core dumps they mean that
it got a signal and exited.  You really have to core on linux to get to dump
core.

>IRIX 5.3 + gcc 2.7.2 = no core dump
>All other OS so far = no core dump
>
>Am I wrong, or does the new Linux malloc() look broken?

Same libc for all of them... that's where malloc is isn't it?

>
>Is Apache the only app having this problem on Linux?

Nope...  I've heard of a couple... mostly scientific packages. 

>
>Why not try GNU malloc() on Linux, to help narrow down the possibilities?
>I assume the one from libg++ would be OK to use.

Yes.  I've gotten one person to compile it via g++ and it seems to have 
worked. 

>
>If it isn't this, it could be the shared memory functions, shm*() or sbrk().

I don't think so.  In fact I don't think it's the malloc that's broke.  I
think it's
our small log-memory leak that causing the problem.  malloc is probebly
now more sensative.

--
Aram W. Mirzadeh, MIS Manager, Qosina Corporation
http://www.qosina.com/~awm/, awm@qosina.com
Apache httpd server team http://www.apache.org



Mime
View raw message