4.
Click
Okay
.
To collect logs and command results using the CLI, use the following command:
ibrix_collect -c -n NAME
NOTE:
•
Only one manual collection of data is allowed at a time.
•
When a node restores from a system crash, the
vmcore
under
/var/crash/<timestamp>
directory is processed. Once processed, the directory will be renamed
/var/crash/
<timestamp>_PROCESSED
. HP Support may request that you send this information to assist
in resolving the system crash.
•
HP recommends that you maintain your crash dumps in the
/var/crash
directory. Ibrix
Collect processes the core dumps present in the
/var/crash
directory (linked to
/local/
platform/crash
) only. HP also recommends that you monitor this directory and remove
unnecessary processed crashes.
Downloading the archive file
When data is collected, a compressed archive file is created and stored in a compressed archive
file (
.tgz
) under
/local/ibrixcollect/archive
directory. To download the collected data
to your desktop, select the collection and click
Download
from the Fusion Manager.
NOTE:
•
Only one collection can be downloaded at a time.
•
The average size of the archive file depends on the size of the logs present on individual
nodes in the cluster.
•
You may later be asked to email this final
.tgz
file to HP Support.
Deleting the archive file
You can delete a specific data collection or all collections simultaneously in the GUI and the CLI.
To delete a specific data collection using the GUI, select the collection to be deleted, and click
Delete
. The
tgz
file stored locally will be deleted from each node.
To delete all of the collections, click
Delete All
.
To delete a specific data collection using the CLI, use the following command:
ibrix_collect -d -n NAME
144 Troubleshooting
