Suppose you have to restart a process due to some error condition as a reaction to a fired alert. Sometimes, the process does not come up again immediately (e.g. some needed network resource is not present). In this case we should allow to re-issue the operation n times with x seconds of pause in between. If the process is still down after n iterations, RHQ should give up trying. Actually, this behaviour is not Alert specific and should be available in plain operation scheduling as well.
so this would be an "operation retry" mechanism ?
This bug was previously known as http://jira.rhq-project.org/browse/RHQ-2046
Actually that should go into a separate escalation framework and not into a single plugin, as retrying could be applicable for other alert senders too. See also http://rhq-project.org/display/RHQ/Design+-+Alerts+future+enhancements