If they are not mounted on either file module, then reboot each file module.
6. Use the lsnode command to determine when the file modules are back up and
when GPFS and CTDB are both active.
The file systems might take several minutes to get mounted after GPFS
becomes active on both file modules. More than one reboot might be required
to bring the file system back up. Allow time between reboots because the file
system might take some time before it comes back up after a reboot.
7. Ensure that the CTDB status is now shown as active on both file modules, as
described in “Checking CTDB health” on page 187.
8. If a GPFS file system fails to mount, complete the following steps:
a. Check the output log of the lslog command and look for the latest
messages about mounting the file system.
If you find input/output errors and messages about a failure to read the
super block, the problem is with the DMAPI clients of the TSM/HSM
system.
Check for disk-related problems, such as errors reading from a disk or
errors showing a non-existent disk. For these errors, check whether the
path to the storage system is working. If it is, verify that the system
itself is in working order.
b. For additional information, refer to the “Diagnostics: Troubleshooting
tables” information in “Troubleshooting the System x3650” in the IBM
Storwize V7000 Unified Information Center.
c. If file systems remain unmounted, contact IBM support.
Resolving stale NFS file systems
You can resolve problems with stale NFS file systems on Storwize V7000 Unified
file modules. A file module might have the file system mounted, but the file
system remains inaccessible due to a stale NFS file handle.
About this task
Procedure
To identify and resolve stale file handle problems, complete this procedure:
1. To identify stale NFS file handle problems, log in to the CLI as privileged user
and enter the sc onnode all df | grep NFS command:
# onnode all df | grep NFS
df: `/ibm/gpfs0’: Stale NFS file handle
2. If the command reports a stale NFS file handle on a particular file system, see
“Working with file modules that report a stale NFS file handle” on page 403 for
instructions on file system recovery.
Checking user and server authentication issues
Due the authentication failure, the users cannot log in to the system or a
connection cannot be established between the servers.
About this task
For login issues, if you are sure that you have used the correct user ID and
password, your user account might have been deleted or corrupted.
Chapter 4. File module 191
To Next Page
Loading...
Need help?
General
