Unanswered: Monitoring health via with snapdbm & snapdb
Hello, fellow DB2 enthusiasts. I am a fairly new DB2 DBA. I have decided to start analyzing the output from SYSIBMADM.SNAPDBM & SYSIBMADM.SNAPDB on a weekly basis. My hope in doing this is establishing a "what's normal" baseline, but I also want it to help guide me in what I need to work on to improve the health and performance of my databases.
My questions for you experienced DB2 users:
1. Does it make sense to approach my 'what is normal and what needs to be addressed' quest in this manner or is there a better method?
2. Are there a handful of key items I should be watching? (i.e., deadlocks, sort overflows, agents stolen, etc.
3. Are there generally accepted guidelines/best practices for the results ?
Also, if there are any good best practices guides or sometime similar you can point me to, I would greatly appreciate it.
Thanks for any help you may offer in helping this excited girl learn more about the wonders of DB2!
Those snapshots are more of a 50,000 foot view of the database / instance. There are a lot more snapshots that give more detail info on what is going on. You need to research what your needs are and go from there.
I would suggest that you write the output from any snapshot to a table so that you have a history of them. This will allow you to determine trends in the health of the databases.