[SOLVED] RHEL 5.3 Hangs when yum, rpm or df -h is run

protos78 · 12-09-2009, 03:38 PM

We have Red Hat Enterprise Linux Server release 5.3. For the last two weeks, the server started not to respond "df -h", rpm and yum requests. Here is an example of yum package installation.

[root@helios tmp]# yum install sysstat
Loaded plugins: rhnplugin, security
Setting up Install Process
Parsing package install arguments
Resolving Dependencies
--> Running transaction check
---> Package sysstat.x86_64 0:7.0.2-3.el5 set to be updated
--> Finished Dependency Resolution

Dependencies Resolved

================================================================================ Package Arch Version Repository Size
================================================================================Installing:
sysstat x86_64 7.0.2-3.el5 rhel-x86_64-server-5 174 k

Transaction Summary
================================================================================Install 1 Package(s)
Update 0 Package(s)
Remove 0 Package(s)

Total download size: 174 k
Is this ok [y/N]: y
Downloading Packages:
sysstat-7.0.2-3.el5.x86_64.rpm | 174 kB 00:00
Running rpm_check_debug
Running Transaction Test
Killed
(I killed the process. I waited for more than2 hours. This should be done instantly. )

If I run, df -h, it shows some of the mounted drives and hangs. Waited for more than 12 hours, nothing happened. I checked the system log, and there is no error message. It would be great if you can tell me how I can diagnose the issue?

Thanks,

protos78 · 12-09-2009, 04:56 PM

I solved the issue. THe problem was stale nfs mounts. I first checked /etc/mtab and removed the stale nfs mounts. Then, commented stale nfs mounts at /etc/fstab. Restarted nfs client. Mounted drives again. Rebuild rpm databases. Bingo, it works!!!

stormtracknole · 12-09-2009, 06:17 PM

Ah, thanks for sharing that. As a matter of fact, I had the same problem, and I didn't realized that having a stale NFS mount would do that.

DrLove73 · 12-10-2009, 04:32 AM

When you have stale NFS mounts in fstab, or no connection to NFS server, even booting will take forever. Just a week ago I have disabled NFS from fstab and set it via autofs.

The trick I used to be able to see autofs mounts from krusader (ls and others open them without a problem but file managers can not see them) is to mount them all in some other folder then desired one, and then (while mounted via ls) create a softlink (symlink) in the location where you want it to be. That way you can double click on the softlink and have autofs mounted NFS folders from file manager.

Procedure:

My server will be caled vmaster, with export named "extra". Example uses nfs4 because of the firewall friendly caracteristics. Mount point will be /vmaster/extra.

1. I created folder /autofsmounts and within that folder a subfolder called "vmaster".
2. I added following line in /etc/auto.master:

Code:

/autofsmounts/vmaster  /etc/auto.vmaster --timeout=90

3. I created a file called /etc/auto.vmaster with following content:

Code:

#
# $Id: auto.misc,v 1.2 2003/09/29 08:22:35 raven Exp $
#
# This is an automounter map and it has the following format
# key [ -mount-options-separated-by-comma ] location
# Details may be found in the autofs(5) manpage

extra	-fstype=nfs4,rw,proto=tcp	vmaster:/extra

4.Then I executed following command from terminal:

Code:

ls /autofsmounts/vmaster/extra

.

5. After that, I created "/vmaster" folder and created symlink of /autofsmounts/vmaster/extra folder in "/vmaster" folder (/vmaster/extra).

Now I have NFS mount point visible all the time from file managers like krusader, but mounted only when they are needed. On boot this mount is not checked hence there is no freezing if mount is unaccessible.