Currently, if you use the voldemort-admin-tool.sh --set-metadata to change/add
a <retention-days> policy to a store, you must restart voldemort on every node
in the cluster for the scheduler to pickup the change and update the cleanup
jobs. Exposing a JMX operation to allow the scheduler rescan the stores.xml
and recreate the cleanup jobs would be very helpful.
Original issue reported on code.google.com by dremlok on 14 Sep 2011 at 5:49
Original issue reported on code.google.com by
dremlok
on 14 Sep 2011 at 5:49