HP StorageWorks Scalable File Share - Manual (Page 250)

Troubleshooting 9–26
NOTE: When a new file system is created or an existing file system is modified, you can run the
repair-lfsck
script so that a shell script will be available if the file system later needs to be repaired. If
you want to repair a file system that has been changed since a shell script was last generated for the file
system (for example, new OST services have since been added to the file system), you must run the
repair-lfsck
script again to generate a new shell script for the file system.
A script generated by the
repair-lfsck
script performs the following tasks for the file system that is to be
examined and/or repaired:
• Creates a
config
file that specifies the servers serving the MDS and OST information and specifies
the services used on each server.
• Launches the MPI Lustre repair utility.
• Verifies that the information on the
config
file is consistent with the current file system configuration.
• Launches the
e2fsck-lfsck
command in parallel on all MDS and OST services used by the file
system.
• Traverses the file system information of the MDS and OST services and reports/repairs as necessary.
To run the
repair-lfsck
script, perform the following steps:
1. Unmount the file system on all client nodes. The
/proc/fs/lustre/mds/*/num_export
s
counter on the MDS server must be 0 (zero). This counter is set internally when client nodes mount
and unmount file systems.
2. Stop the file system and ensure that the file system devices are not being used. Check the
/proc/
mounts
file and the
/proc/fs/lustre/obdfilter/mntdev
(or
/proc/fs/lustre/mds/
mntdev
) files on each server to ensure that the file system devices are not mounted on any mount
point.
3. Ensure that the administration server can connect to all of the other servers used by the file systems
using the
ssh
utility without a password.
4. Enter the command to run the script, as follows. If you want to generate a script for a specific file
system, specify the file system, as shown in the following example. If you do not specify a file system,
a script will be generated for each file system:
# repair-lfsck -f scratch
Section 9.26.5.1.1 provides instructions for running the shell scripts generated by the
repair-lfsck
script.
9.26.5.1.1 Running a generated file system repair script
The shell scripts generated by the
repair-lfsck
script are named in the following format:
do_lfsck-filesystem_name.sh
The generated scripts do not take any arguments, and must be run as
root
user on the administration
server.
HP recommends that you capture the output of file system repair sessions in log files so that it will be
available for analysis.
A file system repair session can take several hours to complete. The length of time taken depends on the size
of the file system, the number of files it contains, and the number of services over which the files may be
distributed.
To run a shell script generated by the
repair-lfsck
script, perform the following steps:
1. Unmount the file system on all client nodes. The
/proc/fs/lustre/mds/*/num_export
s
counter on the MDS server must be 0 (zero). This counter is set internally when client nodes mount
and unmount file systems.