We had an... interesting situation yesterday. One of our admins was changing a log setting on a router and it locked up, blocking off our Kaseya server (and a few other servers) from the rest of the world for about 10 minutes while we contacted the data-center and had the router power cycled.

The Kaseya server seemed to get a really bad case of indigestion from this. Clients were showing off-line when they were in fact on-line and the Agent icon was showing connected. Some machines that were off-line were showing on-line. Remote desktop wasn't working to many of the on-line machines. Most, if not all, of the Agent Logs showed some strange errors that have never come up before, including:
Received 4 OOB requests to checkin in the last 5 min 16 sec.
ERROR: kascSend failed with error -12 in checkFileWithServer while processing...
ERROR: keep-alive in doClientTasks() call to kascSend failed with error -12

The fix was obvious enough, we rebooted the server and all is well (there were some updates to install). But, it provides some insight into a possible glitch with the server thinking the clients are connected while the client thinks the server is down. When these errors start to appear a simple flush of the current connections should clear things up. (We made it a point to keep the server off-line for 5 minutes to ensure the Agents knew it was off.)

After the reboot, the server seemed to remember all of our admin sessions, but it looked like it was pulling cached data. Potential MitM exploit buried in this behavior (we don't use a proxy).

Food for thought.

Legacy Forum Name: Router outage - Kserver went out of sync,
Legacy Posted By Username: dwujcik