Kyle Yencer wrote:
List,
Here is the message I have gotten from my latest crash, it is finally a full
one, usually, there is no trace left in the log:
Jul 11 09:07:28 vps2 kernel: Unable to handle kernel NULL pointer dereference at virtual address 00000004
[cut]
What is it with you people? A user-mode program should in NO WAY WHAT SO EVER crash a kernel, even if it is a vserver patched linux. The problem has to reside inside some part of the kernel. It can be a race condition that only happens when you have alot of processes, memory, io, ... or something totaly different. It might or might not be vserver related.
Okay, this came out overly hard, but still i wonder why some of you keep on talking about java. This log clearly says
"Unable to handle kernel NULL pointer dereference at ..."
It might happen while running a java program, it might not. It is my opinion that there is something in the current vserver code that causes the kernel to dereference a NULL pointer. Either directly or because of a conflict.
My own machine runs mostly a user mode NFS server in a vserver, but also a mysql, apache, ezbounce, and of course several ssh clients. I have a similar (software) machine at work, with the exception that it doesnt run with the vserver patches. The rest, LVM, loop-aes, kernel 2.4.21 is the same. Granted the hardware is not the same, but i used to run my home server without vserver, and that didnt cause trouble. The server at work has a much higher load than my home server, and more memory, but there is still plenty of free memory
17:47:05 up 1 day, 16:36, 1 user, load average: 0.00, 0.00, 0.00 36 processes: 35 sleeping, 1 running, 0 zombie, 0 stopped CPU states: 0.8% user, 1.2% system, 0.0% nice, 98.0% idle Mem: 191864K total, 167148K used, 24716K free, 48464K buffers Swap: 511992K total, 1536K used, 510456K free, 39252K cached
this is from the "root" server, and not inside a vserver, but it looks similar inside a vserver.
So all in all, i dont think vserver is even close to come into 2.6 yet.
JonB
