Chapter 3. Getting Started Troubleshooting - IBM Storwize V7000 Unified Series Problem Determination Manual

Chapter 3. Getting started troubleshooting

This topic is an entry point to troubleshooting your system. The content provides
help in correctly identifying which of the recovery procedures must be run to
recover a Storwize V7000 Unified system from a problem.
Important: After you successfully fix a problem by using the instructions that
follow, use the "Health status and recovery" on page 22 procedure to set the health
status back to green.
If you are here because you installed a new system and cannot initialize it by
using the USB key, go to "Installation troubleshooting" on page 10.
If one of the file modules does not boot up and join the GPFS
hardware problem by using the light-path diagnostics LEDs. See "File node
hardware indicators" on page 34. If you suspect that the boot software is
corrupted, call IBM support.
If any orange fault LEDs are illuminated on the control enclosure, front or rear, see
"Resolving a problem" on page 189.
If you are having problems accessing the management GUI or the CLI, see "GUI
access issues" on page 21. For information about accessing the management GUI,
see "Accessing the Storwize V7000 Unified management GUI" on page 49.
If the health status indicator in the lower right corner of management GUI is not
green, hover over the icon on the left side of the indicator to see the type of error
that is causing the poor health status. Select an error type, and you are shown the
critical errors in the event log. First try to fix the critical errors under the Block tab
of the Monitoring > Events page before trying to fix the critical errors under the
File tab of the Monitoring > Events page.
Log into the CLI interface and run the CLI command, lslog. Review the results for
problems that may need to be resolved.
If users or applications are having trouble accessing data that is held on the
Storwize V7000 Unified system, or if the management GUI is not accessible or is
running slowly, the Storwize V7000 control enclosure might have a problem.
If you cannot ping the management IP address for the Storwize V7000 control
enclosure, try to access the control enclosure service assistant. Use the service IP
address of the node canisters in the control enclosure to resolve any reported node
errors. See "Procedure: Fixing node errors" on page 204.
Note: Use the access information that you previously recorded for the service IP
address of the node canisters in the control enclosure. See "Record access
information" on page 3. If you do not know the service IP addresses for the node
canisters in the control enclosure, see "Problem: Unknown service address of a
node canister" on page 192.
If all nodes show either node error 550 or node error 578, you might need to
perform a system recovery. See "Recover system procedure" on page 240 for more
details.
© Copyright IBM Corp. 2011, 2012
™
cluster, look for a
9