[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Condor-users] Condor Daemons Fail to run on node

On Thu, Dec 2, 2010 at 12:30 PM, Xenia Fave <xfave2008@xxxxxxxxxx> wrote:
The file system is not full nor is the file system mounted in read-only

[xfave@compute-1-0 condor]$ df -h
Filesystem            Size  Used Avail Use% Mounted on
/dev/sda1              19G  5.5G   13G  31% /
/dev/sda3             193G   13G  171G   7% /scratch
tmpfs                 7.9G     0  7.9G   0% /dev/shm
nas-0-0:/nas0         9.6T  6.8T  2.8T  72% /mnt/nas0

Are quotas in use on the filesystem (note: it doesn't look like it based on you fstab output, but check just in case)? Maybe the user hit it's quota?

Can you touch a new file, as the user Condor writes its log files, as on that file system?

sudo -u <user> touch /scratch/condor/log/test.txt

Can you put data in the file?

sudo -u <user> echo "hello world" >> /scratch/condor/log/test.txt

The relevant mount output looks the same for both machines, so the filesystem is mounted the same on working and non-working nodes (I trimmed the output so it's easier to see):
[xfave@compute-1-0 condor]$ mount
/dev/sda3 on /scratch type ext3 (rw)

However, the results of the mount command on one of the working nodes is slightly different, I'm not sure if that's relevant.

[xfave@compute-1-1 ~]$ mount
/dev/sda3 on /scratch type ext3 (rw)

Have you tried unmounting, remounting the /scratch share on a bad machine?

- Ian