CentOS 6.4: problems after update

6 posts / 0 new
Last post
#1 Wed, 09/18/2013 - 11:14

CentOS 6.4: problems after update


I recently ran an update using the standard "yum update" in CentOS 6.4. The LAMP stack on the server is still working. However, system calls and certain command line programs are hanging. Virtualmin/Webmin at times will not load the interface. As of right now, I cannot even SSH and the Virtualmin system info does not load at all.

Kernel is currently: Linux x2vps.com 2.6.32-358.18.1.el6.x86_64 #1 SMP Wed Aug 28 17:19:38 UTC 2013 x86_64 GNU/Linux

Has anyone else run into these problems? Any suggestions/tips for diagnosis?

Wed, 09/18/2013 - 13:23


What you're describing is very unusual, I hadn't heard of that occurring before... except perhaps when certain VPS restrictions are being reached.

What kind of VPS are you using? Is it by chance OpenVZ? If so, can you paste in the contents of your /proc/user_beancounters file?

Also, is your system using any third party repositories? To determine that, can you paste in the output of this command:

ls /etc/yum.repos.d/

Wed, 09/18/2013 - 13:52 (Reply to #2)

Hello, and thanks for the very quick response.

This is a bare metal machine; the VPSs are all Apache virtual hosts / system users managed by Virtualmin. This is Virtualmin GPL, so I'm not sure it has the ability to manage actual distinct virtual machines.

It does has EPEL, because there are a few PHP modules that we need that aren't available in the default repositories, so yes there are third-party repositories. I can't get any verbatim output from the command line right now (because the server is non-responsive).

Wed, 09/18/2013 - 16:20

Yeah, the server being unresponsive is a big issue -- it sounds like there's something really weird going on now, though it's difficult to say what might be causing that.

It sounds like you're saying that this is a dedicated server -- if you look at the console, do you see any errors there?

You may want to try rebooting though, if your system is otherwise unresponsive.

The EPEL repository shouldn't be a problem though, BTW. And it's okay that you're hosting VPS's -- some folks are actually running within a VPS though, that's what I was wondering (sounds like a dedicated server though).


Wed, 09/18/2013 - 16:48 (Reply to #4)

I really appreciate your time and tireless effort on this great software.

After some investigation, I found something odd had been going on with the system's hostname. I sorted it out and the problem (so far) appears to have been solved.

Long story:

About a year ago, when I was working on getting the Virtualmin installer to run, I had to change the hostname to a fully-qualified domain name. However, I was then unaware of how CentOS has yet a third place, apart from /etc/hosts and /etc/hostname, where the hostname is defined, and where I'd have to change it: in /etc/sysconfig/network. This particular file never really changed, and the problem only surfaced when I restarted the server and CentOS set the hostname as it booted up. This is probably why SSH/Sudo/Virtualmin were stuck.

Does that sound accurate? Could the problem have been caused by just that?

Wed, 09/18/2013 - 23:54

Well, the symptoms you were describing seem a bit harsh for an incorrectly set hostname... so while it's good that you have the hostname straightened out, I'm not sure why things were working that poorly for you.

It's possible that something else was solved during the reboot... but that's tough to say :-)

However, I'm glad it's all working now!


Topic locked