Friday, August 10, 2012

Zombie Blackouts From The Past

Recently, I've been doing a cluster-wide maintenance and put all nodes in Full Blackout for that time (Enterprise Manager 11g).
When the blackout was stopped, for some reason older blackouts from as far back as 13 months ago started to activate themselves again on the agent side.
In the console, the affected targets just showed "Status Pending" - but it could be verified that the Blackouts were the root cause via issuing on the target host:

emctl status blackout

Though I haven't found the source of the problem yet, I used a one-liner to stop all those that I want to share for anyone having a similar problem. Doing them one by one just gets tiresome at a certain amount. And it's not the first time I encountered those.

emctl status blackout | grep Blackoutname | awk '{sub(/Blackoutname = /,"",$0); print "emctl stop blackout \""$0"\""}' | bash

Also might come in handy if you'd just like to get rid of several blackouts on a host at once.

Will update once/if I find the root cause and a solution.

No comments:

Post a Comment