STARTED
Jul 17, 12:55 PM (2¼ months ago)
CLOSED
Jul 17, 01:27 PM (2¼ months ago)
DESCRIPTION
Customers on one of our LNSs, z.witless, dropped and reconnected at around 1PM. This was due to the LNS crashing and restarting. Cause is being investigated.
Resolution: The CPU on z.witless hung and the watchdog rebooted the device causing customers connected to it to drop and reconnect - causing an outage for a couple of minutes. We suspect the cause of the CPU problem is something rare (we've not seen this in production before) and one that the FireBrick developers are already working on to understand.