The following describes known problems with the NFS Toolkit
and workarounds for them. However, this is subject to change without
notice. For the most current information contact your HP support representative.
More recent information on known problems and workarounds
may be available on the Hewlett Packard IT Resource Center:
JAGad91624:
NFS remote mount table discrepancy on package fail over |
 |
What is the problem? The counter number
of remote mount entries in the /var/lib/nfs/rmtab may or may not be correct after running NFS packages
for a long period. This problem does not disrupt any client access
of NFS files.
Each entry in the /var/lib/nfs/rmtab file contains three fields. These fields are: Client IP address, Exported file system , and Counter number of the exported file system mounted by a client.
An rmtab file, for example, could contains following entries:
128.12.148.2:/expo/pkg1fs:0x00000001
128.12.148.3:/expo/pkg1fs:0x00000001
128.12.148.2:/expo/pkg2fs:0x00000001
|
The counter field, in normal case, contains 0x00000001 (e.g. the counter can be 2 if the same client mounted the same file system
twice). The counter may or may not be updated correctly when a client
un-mounts the file system.
What is the workaround? The system
administrator needs to be aware of the counter number that may not
be consistent with the number of clients currently mounting the
file systems. The administrator can choose to ignore the counter
number; however if the file, /var/lib/nfs/rmtab, or the counter number becomes too large, following
steps can be taken in order to maintain the file.
Verify that clients are not currently
mounting the file systems.
Verify that all NFS packages are running on all
nodes in the cluster.
On each node, perform:
# cat /dev/null > /var/lib/nfs/rmtab
Resume the normal client mounting process.
Printable version
