Table of Contents
Advertisement
It is also key to manage any node failures in the startup process and continue when possible.
An issue with some part of the cluster that starts might exist that does not affect other parts of
the cluster. When this condition occurs, you must continue with the boot process for all areas
that are successful and retry or diagnose the section with the failure. Continuing with the boot
process allows the rest of the cluster to continue to start, which is more efficient than holding
up the start of the entire cluster.
Optional power-on hardware
During the cluster shutdown process, there is an optional task for disconnecting power. If you
turned off breakers or disconnected power to the management rack or the 775 frames during
the cluster shutdown process, you must continue with this step to connect power.
Perform the following steps only if any of the breakers are shut off or power is disconnected:
1. Turn on breakers or connect power to the management rack.
2. Turn on breakers or connect power to the 775 frames.
Powering on external disks attached to EMS
Power on any external disks that are used for dual-EMS support. This power-on is required
before starting the primary EMS.
Power on EMS and HMCs
After the EMS shared disk drives are up, it is time to power on the primary EMS and the
HMCs.
The backup EMS is started after the cold start is complete and the cluster is operational. The
backup EMS is not needed for the cluster start, and spending time to start it takes away from
the limited time for the entire cluster start process.
Starting the primary EMS and the HMCs is a manual step that requires the administrator to
push the power button on each of these systems to start the boot process. The EMS and
HMCs are started at the same time because they do not have a dependency on each other.
Primary EMS start process
The administrator must execute multiple tasks that work with the primary xCAT EMS. Confirm
that all of the local and external attached disks are started and available to the xCAT EMS.
Important: Do not start the backup EMS or perform any steps on a backup EMS now. The
backup EMS must be started after the Power 775 cluster start process is complete and
working with the primary xCAT EMS.
The administrator must ensure that all of the files systems are mounted properly, including file
systems on external shared disks. The expectation is that some directory names might vary
depending on the site, as shown in Example 5-1.
Example 5-1 Checking the mounted file systems
$ mount ? /etc/xcat
$ mount ? /install
$ mount ?~/.xcat
$ mount ? /databaseloc</pre>
The administrator must ensure that the DB2 environment is enabled on the xCAT EMS. This
verification includes validating that the DB2 monitoring daemon is running, and that the xCAT
DB instance is set up, as shown in Example 5-2 on page 277.
276
IBM Power Systems 775 for AIX and Linux HPC Solution
Advertisement
Table of Contents
