Hi all,
I have recently changed all of our checks to use a different set of handlers, removing an older “hipchat” handler, and adding a new one. After applying the configuration and restarting the server, any new alerts seem to continue using the old set of handlers. The json listed here is the log entry on the master showing what it’s going to do with this check. The second block is the configuration block output on server startup, limited to this check.
{
“timestamp”: “2015-08-04T20:09:18.961191+0000”,
“level”: “info”,
“message”: “processing event”,
“event”: {
“id”: “ba5ba045-abf2-405b-acd2-7c5b84edb091”,
“client”: {
“name”: “host.erchinger.org”,
“address”: “10.208.22.113”,
“subscriptions”: [
“atd”
],
“version”: “0.19.2”,
“timestamp”: 1438718944
},
“check”: {
“command”: “/etc/sensu/plugins/processes/check-procs.rb -p atd -C 1”,
“dependencies”: [
“sensu-server.erchinger.org/alerts-on_check”
],
“handlers”: [
“mailer”,
“hipchat”
],
“interval”: 60,
“occurrences”: 2,
“refresh”: 86400,
“subscribers”: [
“atd”
],
“name”: “atd_check”,
“issued”: 1438718958,
“executed”: 1438718958,
“duration”: 0.181,
“output”: “CheckProcs OK: Found 1 matching processes; cmd /atd/\n”,
“status”: 0,
“history”: [
“2”,
“2”,
“2”,
“0”
],
“total_state_change”: 11
},
“occurrences”: 3,
“action”: “resolve”,
“timestamp”: 1438718958
}
}
{ “atd_check”: [
null,
{
“command”: “/etc/sensu/plugins/processes/check-procs.rb -p atd -C 1”,
“dependencies”: [
“sensu-server.erchinger.org/alerts-on_check”
],
“handlers”: [
“pagerduty”
],
“interval”: 60,
“occurrences”: 2,
“refresh”: 86400,
“subscribers”: [
“atd”
]
}
]}
Why wouldn’t the alert log entry shown first not adhere to the configuration read on startup? Do I have to clear something out of rabbitmq, or am I missing something else? For what it’s worth, I don’t even have anything with the word hipchat in the /etc/sensu directory anymore.
Thoughts?
Thanks,
Ethan