IBM Storwize V7000, Problem Determination Manual

Page: 285 / 314

Note: This procedure involves analyzing various logs depending on the errors
displayed by the initial SCM error log.
1.
Open the CNSCM log located at
/var/log/cnlog/cnscm
for the file module that
reported the error.
2. Review the error entries around the listed time stamp and then check the log
for issues that seem related that occurred before the listed time stamp. For
example, you might find GPFS-related issues appearing earlier and later, too.
3. Based on the log entries, check the appropriate corresponding log. If the issue
appeared to be related to GPFS, for example, you could look for the root cause
in
/var/adm/ras/mmfs.log
.
4.
Continue to evaluate and analyze the logs until you resolve the error.
Recovering from an httpd service error
Use this procedure to recover from an httpd service error when the service is
reported as unhealthy or off.
To fix the httpd error, perform the following steps:
1. Attempt to start the http service manually.
a. Log in as root.
b. Issue the service http start command.
2. When you complete the service action, refer to “Health status and recovery” on
page 22.
Recovering from an sshd_data service error
Use this procedure to recover from an sshd_data service error.
This recovery procedure starts the sshd_data when it is down.
1. Log in as
root
.
2. Issue the service
sshd_data start
command.
3. If the problem persists, restart the node.
4. If the restart action does not resolve the issue, contact the next level of support.
Recovering from an sshd_int service error
Use this procedure to recover from an sshd_int service error.
This recovery procedure starts the sshd_int when it is down.
1. Log in as
root
.
2. Issue the service
sshd_int start
command.
3. If the problem persists, restart the node.
4. If the restart action does not resolve the issue, contact the next level of support.
Recovering from an sshd_mgmt service error
Use this procedure to recover from an sshd_mgmt service error.
This recovery procedure starts the sshd_mgmt when it is down.
1.
Log in as
root
.
2. Issue the service
sshd_mgmt start
command.
3.
If the problem persists, restart the node.
4. If the restart action does not resolve the issue, contact the next level of support.
Chapter 7. Recovery procedures
261