<div dir="ltr"><div><div><div><div><div>This is admittedly a shot in the dark, and with limited repro details, but asking if somebody else has seen this.<br></div><br></div>We run our clustered RabbitMQ 3.0.2 brokers in production without restarts for months at a time. Everything is looking good - No memory leaks, no out-of-disk-space, etc...<br>
<br></div>However, in looking at the rabbit@<node>.log and rabbit@<node>-sasl.log files, I noticed that at some point (e.g. a month earlier), no new entries were being added to the log file. Normally there's a steady stream of new connections and connection drops.<br>
<br>I've listed the open file handles from the rabbitmq beam instance and I still see live file handles for the log files.<br><br></div>Anybody have any idea what might be happening? And better still, how we might kick things without restarting the broker?<br>
<br></div>Thanks,<br><br>Matt<br><div><div><br></div></div></div>