Oopses in Glibc... [message #12332] |
Wed, 25 April 2007 09:21 |
m.s.tremer
Messages: 9 Registered: April 2007
|
Junior Member |
|
|
Hello and a good morning to everybody!
I also asked for support in the german area but got no answers ( http://forum.openvz.org/index.php?t=tree&th=2387&sta rt=0&). Now, I am going to try it out in the international area of your forum
gettext [ 0.14.5 ] [ 95 ] [ DONE ]
kbd [ 1.12 ] [ 8 ] [ DONE ]
popt [ 1.10.4 ] [ 31 ] [ DONE ]
sysvinit [ 2.86 ] [ 2 ] [ DONE ]
misc-progs [ ipfire ] [ 1 ] [ DONE ]
e2fsprogs [ 1.39 ] [ 71 ] [ DONE ]
libaal [ 1.0.5 ] [ 27 ] [ DONE ]
reiser4progs [ 1.0.5 ] [ 188 ] [ DONE ]
reiserfsprogs [ 3.6.19 ] [ 38 ] [ DONE ]
sysfsutils [ 1.3.0 ] [ 19 ] [ DONE ]
util-linux [ 2.12r ] [ 12 ] [ DONE ]
pciutils [ 2.2.3 ] [ 4 ] [ DONE ]
zlib [ 1.2.3 ] [ 8 ] [ DONE ]
wget [ 1.10.2 ] [ 29 ] [ DONE ]
libxml2 [ 2.6.26 ]
*** glibc detected *** double free or corruption (!prev): 0x08de97f8 ***
[ 174 ] [ FAIL ]
ERROR: Building libxml2
Check /serv/ipfire/trunk/log/_build.installer.log for errors if applicable
[ FAIL ]
***Sending error message*** glibc detected *** double free or corruption (!prev): 0x08de97f8 ***
I'm going to build a linux distribution from source in a VE.
I have chosen OpenVZ because I can manage the resources of the node.
But during the build process the compiler sometimes says "Aborted" and dies with the message over there.
After that I installed a fresh Centos5 on the node (before: FC5) and the error was the same.
In the VE there were a FC5 and an Ubuntu but they all produce the same errors.
Today I tried a build on the hardware node itself and hey --> Same error.
Is there a chance to change the kernel to a devel version? (Linux hsclix17.intra 2.6.18-8.el5.028stab027.1 #1 SMP Thu Apr 5 20:23:39 MSD 2007 i686 i686 i386 GNU/Linux)
Or do you think what causes the error?
Thanks in advance for help!
Michael
[Updated on: Wed, 25 April 2007 09:21] Report message to a moderator
|
|
|
|
Re: Oopses in Glibc... [message #12356 is a reply to message #12338] |
Wed, 25 April 2007 16:11 |
m.s.tremer
Messages: 9 Registered: April 2007
|
Junior Member |
|
|
1) There is really nothing in the logs... Neither /var/log/messages nor any other (dmesg...). GCC simply dies with SegFault.
2) The environment is a bunch of makefiles that run "./configure .. . && make && make install"
3) The same build process runs on any other kernel I know...
(Fedora Core 5/6, SuSE 9.3/10.0/10.2, Debian 3.1)
Now, it is running on the real Centos kernel... I'm still waiting... No errors yet...
Hardware was completely tested by memtest & cpuburn.
Michael
[Updated on: Wed, 25 April 2007 16:12] Report message to a moderator
|
|
|
|
|
|
|
|