IBM Storwize V7000 Unified Problem Determination Manual page 81
Hide thumbs
Also See for Storwize V7000 Unified:
- Introduction and implementation manual (670 pages) ,
- Problem determination manual (474 pages) ,
- Troubleshooting and maintenance manual (189 pages)
Table of Contents
Advertisement
6. After replacing the failing part and replacing the file module cover, replace the
file module in the rack, if necessary, and reconnect the power cords.
After reconnecting the power cords, the power indicator LED on the front of
the file module begins to flash quickly.
7. Push the file module back into the rack.
8. After the power indicator LED on the front of the file module begins to flash
slowly, press the power switch that surrounds the indicator light to turn on the
file module.
As the file module reboots, the Storwize V7000 Unified system reintegrates it
back into the cluster.
What to do next
You can now perform maintenance on another file module that requires it. After
completing all maintenance, perform the system checkout procedure and return the
system to the customer.
Removing a file module without disconnecting power
You can work on an IBM Storwize V7000 Unified file module to perform a
maintenance action that does not require you to remove its power cords.
About this task
Perform the following procedure to remove and replace a hot swappable field
replaceable unit (FRU) in a file module when you do not have to remove the file
module from the rack to work on it.
Procedure
1. Access and log in to the Storwize V7000 Unified system from the
command-line interface.
2. Issue the suspendnode command to remove the file module from the system so
that you can work on it.
To remove the mgmt001st001 file module from the system, for example, issue
the following command:
# suspendnode mgmt001st001
3. Wait for the Storwize V7000 Unified system to stop the file module at the
clustered trivial database (CTDB) level. The command does not unmount any
mounted file systems.
A stopped file module does not participate in the cluster and does not host any
records for the clustered trivial database. The IP address of an file module is
taken over by another file module and no services are hosted.
You can log on as root and issue the onnode all service ctdb status
command to view the state of the file module in comparison to another file
module in the system. Or you can issue the lsnode -r command as a non-root
user to view the state of the file module.
The results from running the lsnode -r command are similar to the following
example:
# lsnode -r
Hostname
IP
mgmt001st001
10.254.8.2
mgmt002st001
10.254.8.3
Product Version Connection status GPFS status CTDB status
1.3.0.2-02
OK
1.3.0.2-02
SUSPEND
Description
Role
active management node management,interface,storage
passive management node management,interface,storage
active
active
active
SUSPEND_MAINTENANCE 1/17/12 4:39 PM
Last updated
1/17/12 4:39 PM
Chapter 4. File module
57
Advertisement
Table of Contents
Troubleshooting
