Red Hat's Global Partner Enablement team, of which i am a member, uses 110 public Openshift accounts of medium sized gears using the jbosseap-6 and postgres-8.4 cartridges.
the past two days the jbosseap-6 process of each of our 110 accounts has gone down and has not recovered (ie: has not automatically come back up)
I can manually bring them back up however i receive the following error message if I attempt to start the app by remoting into the gear and executing : app_ctl.sh start :
"Application is explicitly stopped! Use 'rhc app start -a pfpCore' to start back up"
so i can certainly us the rhc command to start my accounts back up.
I'm wondering what the scenario(s) is that explicitly stops an app in the first place ??
we've had accounts go down off and on over the past few months but the typical pattern has been :
1) the apps are never "explicitly stopped"
2) the apps tend to start back up automatically
3) only groups of accounts tend to go down at any given time ..... not all 110. Specifically, the accounts that go down at the same time tend to all reside on the same AWS node.
Global Partner Enablement
Red Hat (www.redhat.com)