| < Day Day Up > |
|
When contacting the IBM Support Center, you will be required to have the following information:
Description of the problem.
Any syslog entries related to the problem, which can be found with the following command:
# grep "mmfs:" /var/log/messages
mmfsadm dumps all output on at least the failing node and the file system manager's node. Some errors may require that you have this information from all nodes.
A master mmfs log file that is merged and chronologically sorted for the date of the failure. This can be done by copying the logs of all nodes for the date of the failure and using the cat and sort commands to merge them, as in Example 8-22 on page 242.
The actual application output of the failure.
Traces on affected nodes, if available. In some cases, this may mean all the nodes.
Any internal dumps written to /tmp/mmfs relating to the time of the failure.
On the failing node, the output of:
# rpm -aq
For all failing file systems, the output of:
# mmlsdisk <filesystem>
Example 8-23: Gathering and sorting log data
# cp /var/adm/ras/mmfs.log.latest node001 # scp node002:/var/adm/ras/mmfs.log.latest node002 mmfs.log.latest 100% |*********************************| 44912 00:00 # scp node003:/var/adm/ras/mmfs.log.latest node003 mmfs.log.latest 100% |*********************************| 82790 00:00 # ls -la total 24 drwxr-xr-x 2 root root 4096 Nov 29 21:49 . drwxr-x--- 13 root root 4096 Nov 29 00:26 .. -rw-r--r-- 1 root root 2440 Nov 29 00:29 node001 -rw-r--r-- 1 root root 4912 Nov 29 00:29 node002 -rw-r--r-- 1 root root 2790 Nov 29 00:30 node003 # cat node* | sort -k 3,4 > merged.log # ls -la total 36 drwxr-xr-x 2 root root 4096 Nov 29 21:49 . drwxr-x--- 13 root root 4096 Nov 29 00:26 .. -rw-r--r-- 1 root root 10142 Nov 29 21:49 merged.log -rw-r--r-- 1 root root 2440 Nov 29 00:29 node001 -rw-r--r-- 1 root root 4912 Nov 29 00:29 node002 -rw-r--r-- 1 root root 2790 Nov 29 00:30 node003
| < Day Day Up > |
|