ipfs1 102592 0 (unused)
If either
grep
command returns empty, contact HP Support.
6.
From the active management console node, verify that the new version of StoreAll software
FS/IAS is installed on the file serving nodes:
<ibrixhome>/bin/ibrix_version -l -S
Completing the upgrade
1.
Remount the StoreAll file systems:
<ibrixhome>/bin/ibrix_mount -f <fsname> -m </mountpoint>
2.
From the node hosting the active management console, turn automated failover back on:
<ibrixhome>/bin/ibrix_server -m
3.
Confirm that automated failover is enabled:
<ibrixhome>/bin/ibrix_server -l
In the output,
HA
should display
on
.
4.
From the node hosting the active management console, perform a manual backup of the
upgraded configuration:
<ibrixhome>/bin/ibrix_fm -B
5.
Verify that all version indicators match for file serving nodes. Run the following command from
the active management console:
<ibrixhome>/bin/ibrix_version -l
If there is a version mismatch, run the
/ibrix/ibrixupgrade -f
script again on the
affected node, and then recheck the versions. The installation is successful when all version
indicators match. If you followed all instructions and the version indicators do not match,
contact HP Support.
6.
Verify the health of the cluster:
<ibrixhome>/bin/ibrix_health -l
The output should show
Passed / on
.
Troubleshooting upgrade issues
Automatic upgrade fails
Check the
upgrade.log
file to determine the source of the failure. (The log file is located in the
installer directory.) If it is not possible to perform the automatic upgrade, continue with the manual
upgrade procedure.
ibrixupgrade hangs
The installation can hang because the RPM database is corrupted. This is caused by inconsistencies
in the Red Hat Package Manager.
Rebuild the RPM database using the following commands and then attempt the installation again.
Note that
rm
is followed by a space and then two underscores, and
rpm
is followed by a space
and then two dashes:
cd /var/lib/rpm
rm __*
rpm --rebuilddb
On the management console,
ibrixupgrade
may also hang if the NFS mount points are stale.
In this case, clean up the mount points, reboot the management console, and run the upgrade
procedure again.
200 Cascading Upgrades