We are trying to set up Server and Java monitor for our atlassian tools. We are having issues with the Zabbix JMX monitoring for confluence (and fisheye),
Has any had this issue? Need some help.
We are trying to do the following JMX Query
INFO [main] org.apache.coyote.AbstractProtocol.init Initializing ProtocolHandler ["http-nio-8090"]
jmx["Catalina:type=ThreadPool,name=\"http-bo-8060\"",currentThreadsBusy]
We are getting the following error:
jmx["Catalina:type=ThreadPool,name=\"http-{$CONNECTOR_MODE}-{$HTTP_PORT}\"",currentThreadsBusy]" became not supported: Catalina:type=ThreadPool,name="http-bio-8060"
Yes I enabled that and will look up different things to monitor with Jetty. Thanks
I am looking for max thread counts and active threads, this will allow us to validate that our servers are not being over loaded.
Hi Dan O'Halloran,
Bear in mind FishEye does not use Catalina as a servlet container, it uses embedded Jetty instead. FishEye does not expose any JMX beans for Jetty by default, but there is an undocumented feature you can use:
FISHEYE_OPTS=-Dfisheye.jetty.jmx.enable=true
(see https://confluence.atlassian.com/display/FISHEYE/Environment+variables for information how to use FISHEYE_OPTS environment variable)
Once you enable it you would notice number of JMX beans exposed to control Jetty server.
Hope that helps,
Piotr
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.