Hi,
This is resulting in accumulation of thousands of node monitor threads in BLOCKED state. I suspect these threads don't go away once the computer goes away.
I was curious if there is a way to either disable these monitors entirely Jenkins wide or disable them for computer type perhaps a method 'needsMonitoring()' on Computer.
eg:
"Monitoring thread for Response Time started on Mon Nov 14 20:55:26 UTC 2016" #43777 daemon prio=5 os_prio=0 tid=0x00002aab5f376800 nid=0x1f4b waiting for monitor entry [0x00002aabd4e3b000]
java.lang.Thread.State: BLOCKED (on object monitor)
at hudson.remoting.Channel.send(Channel.java:577)
- waiting to lock <0x0000000548779748> (a hudson.remoting.Channel)
at hudson.remoting.Request.callAsync(Request.java:205)
at hudson.remoting.Channel.callAsync(Channel.java:823)
at hudson.node_monitors.AbstractAsyncNodeMonitorDescriptor.monitor(AbstractAsyncNodeMonitorDescriptor.java:75)
at hudson.node_monitors.ResponseTimeMonitor$1.monitor(ResponseTimeMonitor.java:57)
at hudson.node_monitors.AbstractNodeMonitorDescriptor$Record.run(AbstractNodeMonitorDescriptor.java:305)