The services the agent uses to report operation status are guaranteed
delivery, so they should get reported when the server comes back up.
The server should come back up when the storage node directory move is
complete? Or is that the problem, the storage node will not be
restarted unless it processes the operation results? When the server
is in MM it still executes quartz jobs, I think. Perhaps that could be
leveraged, or perhaps the new storage workflow mechanism will inherently
solve the problem. Certainly I don't think we want to muck with the comm
layer stuff.
On 7/18/2014 1:32 PM, John Sanda wrote:
On Jul 18, 2014, at 1:10 PM, John Mazzitelli <mazz(a)redhat.com>
wrote:
>> As I understand it, if the server throws an exception when the agent sends a
>> measurement report, the agent will immediately try to resend. We could add a
>> delay in there so that the agent wouldn’t keep hammering the server.
> Actually, it won't sent a retry immediately. the comm layer has a built in
delay.
Is the delay fixed? Since we already have a delay and if it sufficiently large, then this
might be the simplest solution.
_______________________________________________
rhq-devel mailing list
rhq-devel(a)lists.fedorahosted.org
https://lists.fedorahosted.org/mailman/listinfo/rhq-devel