IBM Storwize V7000 Unified Problem Determination Manual page 285

6. Verify that the state of each affected file module is healthy after you resume the
Recovering the GPFS
1. From the root, enter lsnode -r. This displays GPFS and CTDB status.
2. Check for the GPFS that are in stale mount by running lsmount or mmlsmount
3. Enter onnode all df. This displays df: 'ibm/gpfsX': Stale NFS handle where
To recover gpfs, follow the procedure below:
1. Enter lsnode -r
2. Enter ssh<node>
3. Enter initnode -r - n <node> to reboot the node.
4. Enter exit.
Ping the node or check uptime or wait for the node to come up.
1. lsnode -r
2. resumenode <node>
Perform the above procedure for all the nodes in the cluster.
File module-related issues
This section covers the recovery procedures related to file module issues.
Restoring System x firmware (BIOS) settings
During critical repair actions such as the replacement of a system planar in an IBM
Storwize V7000 Unified file module, you might have to reset the System x
firmware.
Before you begin
The firmware and software code package for the Storwize V7000 Unified
microcode can automatically configure the default settings for the System x
firmware to the required Storwize V7000 Unified settings. However, to enable the
automatic configuration, you must reset the System x firmware from its current
state to the default configuration.
About this task
Use the following procedure to set the System x firmware to the default state and
start the automatic Storwize V7000 Unified configuration.
Procedure
file module and that no new "Stale NFS file handle" CIMs appear in the alert
log.
gpfs1 -L.
X is the gpfs number such as gpfs0.
1. Attach the USB and VGA cable from the KVM switch to the video and USB
port of the Storwize V7000 Unified file module that is to be reset.
2. Open the KVM unit in the base rack.
3. Turn on the KVM unit if it is off.
4. Press PrtSc to display the KVM selector screen.
5. Scroll down to select the USB cable, then press Enter.
Chapter 7. Recovery procedures
265