5
1
I am currently running wordpress on an Amazon ec2 instance, which runs fine with no problems most of the time, but occasionally outside connection to the instance shuts down. I can't access it over either http or ssh, and the server seems completely unresponsive (but the status checks pass). However, everything is fine if I simply stop and start the instance (not restart though). Although this problem is manageable, it is obviously extremely inconvenient and a problem for production. Does anyone know why this might happen? Thanks so much in advance!
How do CPU utilization and memory look, in the instance metrics in the console? Is this a micro instance? – Michael - sqlbot – 2014-09-15T03:27:30.050
The metrics all look like I'd expect -- nothing extraordinary when I can connect to the instance, and obviously everything's dead when the connections fail. It's a t2.micro instance. – cderwin – 2014-09-15T04:44:06.647
Well, not quite "obviously" if the status checks are still passing. You're saying all of the Cloudwatch metrics (in the AWS web console) stop updating? – Michael - sqlbot – 2014-09-15T11:08:12.333
No they update, but given that nothing can connect to the server and I had very little running, network in/out, memory write/read, and cpu utilization were all either near 0 or 0. – cderwin – 2014-09-15T17:03:45.083
I experience the same thing and yes it's a t2.micro instance. CPU utilization percent scale is 0 to 1...I assume that's 0% to 100% and if so, I am seeing a sustained +85% CPU before the instance tanks. Then I can no longer SSH to the instance. Have to launch a new one from AMI. – SomethingOn – 2015-03-09T19:38:34.273