2009/2/24 Rob Kampen rkampen@kampensonline.com:
Alain Terriault wrote:
I experienced 3 problems that i am 99% sure related to kernel-2.6.18-92.1.22.el5.centos.plus.x86_64.rpm
You cannot be sure 99% until you can reproduce the problems...
#1 on a nfs server problem : constantly appearing on all my machines "lockd: server 192.168.10.2 not responding, timed out"
solution : move to kernel-2.6.18-92.1.22.el5, never saw that message again.
I run a CentOS 2.6.18-92.1.22.el5 #1 SMP Tue Dec 16 11:57:43 EST 2008 x86_64 x86_64 x86_64 GNU/Linux kernel serving NFS with no issues. I have a number of CentOS 2.6.18-92.1.22.el5.centos.plus #1 SMP Wed Dec 17 10:49:19 EST 2008 x86_64 x86_64 x86_64 GNU/Linux workstations that have no issues as NFS clients. No experience of plus kernel used on servers.
I have been running 2.6.18-92.1.22.el5.centos.plus both on my own workstations and as VMware guests. The real (host) machine works as an nfs server as well. I have never seen any issue there.
#2 on the same nfs server problem : it is a important server for our department, so i rarely do a "shutdown". when i moved to kernel-2.6.18-92.1.22.el5 the "/sbin/shutdown -h now" did not succeed, i had to press the power button to shut it down.
I have not seen this behavior either on any of the machines running a centosplus kernel.
#3 on a vmware virtual machine problem : Feb 15 18:55:41 www kernel: nscd[5631]: segfault at 00002b89938405c0 rip 00002b891f30f7c5 rsp 00000000411256d0 error 4
(snip)
I did not take that many notes or investigate that much, so i can not tell you much more. but i will do my best to reply to direct emails. I am also not in a position to reproduce those problems. I sent this post because i am concern they may be serious bugs in the centos.plus kernels.
Unless this can be reproduced by you or other users, it will be very difficult to even identify it as a bug. Troubleshooting is nearly impossible. The centosplus kernel has several drivers and other options enabled, but otherwise its behavior should be the same as the distro kernel. Yes, some of those options might be involved, but nothing is immediately obvious as to the cause of the issues you experienced.
Akemi