Per the documentation in the link from my first email, you'll see that message when a destination is removed because it has been inactive (empty) for the inactiveTimeoutBeforeGC period. So if you're getting that message, it means that your destination has been empty for 60 seconds.
This is entirely expected behavior for your configuration, and nothing you've said so far indicates that there's anything to troubleshoot. Unless you've got something that indicates that messages are going outstanding that you didn't include in your original email? To be crystal-clear, the log lines are not in and of themselves an indication that "messages are going outstanding", though if there was some bottleneck somewhere else in your system and you weren't seeing messages come out the far end, the lack of messages flowing through the destination could cause it to be considered inactive and to be GC'ed. But in that scenario, those log lines are a secondary symptom of, not a cause of, the underlying problem. And so far you haven't told us anything that indicates that this is your scenario. So if you don't have some other indication that messages are somehow failing to be delivered within your system, and all you want is to not have these lines in your logs, your options are: 1. Increase the rate at which you send messages to that destination, to ensure that you never have a 60-second interval with no messages enqueued. 2. Increase the timeout to ensure that you don't delete destinations until they've been empty for a longer period of time. 3. Turn off the feature, and leave inactive destinations on the broker. 4. Turn down logging so you don't have your log spammed when inactive destinations are removed. 5. Leave things as they are and ignore these lines in your logs. Obviously there are trade-offs about whether you want to do this GCing of inactive destinations and how often, so the right option from the five I listed will depend on your use case. On Thu, Oct 30, 2014 at 11:00 AM, Lodgen, Brad <brad.lod...@centurylink.com> wrote: > Tim, > > Thanks for that, but I'm trying to get rid of the messages by fixing the > issue of the messages going outstanding, not changing the logging levels or > turning off the feature. :) > > Anyone have a troubleshooting path or tuning recommendation, by chance? > > > Tim Bain <tb...@alumni.duke.edu> wrote: > > Well, you could get rid of those messages by choosing not to delete > inactive destinations (documentation: > http://activemq.apache.org/delete-inactive-destinations.html). > > Or you could tune Log4J to only show WARN level messages for that > Logger.... > > On Thu, Oct 30, 2014 at 10:01 AM, Lodgen, Brad < > brad.lod...@centurylink.com> > wrote: > > > I'm still learning tunables in activemq. Right now, I'm trying to tune > and > > troubleshoot my environment to get rid of the numerous activemq.log > entries > > showing "Inactive for longer than X ms - removing ...". I currently have > > the timeout set to 60000ms, but the messages I'm sending normally respond > > in less than 10 seconds. Can anyone suggest a troubleshooting and/or > tuning > > process to follow to diagnose and fix these messages? > > >