Monitoring in a Db2 pureScale environment
For example, you might have a hardware problem that causes a member or cluster caching facility (also known as a CF) to fail over repeatedly to another host. However, because recovery is in most cases automatic, you might never be aware of the problem. If the problem goes undetected and uncorrected, you will not be realizing the full performance potential of your Db2 pureScale environment .
- Are all components of my Db2 pureScale instance running?
- If a member or CF has failed, was it able to restart successfully?
- Is my CF running on its preferred primary host? Or has it failed over to another host?
- Is another CF in a state where it is ready to take over if the primary CF fails?
A good place to start when you want to see the overall status of your Db2 pureScale environment is to examine the operational status of the hosts, members and CFs in your instance. By examining state and alert information reported for each of these entities, you can form an overall picture of how well your Db2 pureScale instance is functioning.
Beyond looking at the overall state of your Db2 pureScale instance, you can also use the database monitoring infrastructure to examine monitor elements that provide information about specific aspects of the Db2 pureScale instance. This information can help you to better understand where configuration and application design issues might detract from overall system performance.