Issue
I have upgraded my HA/HS instances to release 2.479.1.4
(or higher) and when starting, they fail throwing the following stack trace in the logs:
SEVERE h.i.i.InstallUncaughtExceptionHandler$DefaultUncaughtExceptionHandler#uncaughtException: A thread (Jenkins initialization thread/40) died unexpectedly due to an uncaught exception. This may leave your server corrupted and usually indicates a software bug. java.lang.NoSuchMethodError: 'boolean hudson.model.Computer.isMasterComputer()' at PluginClassLoader for cloudbees-replication//com.cloudbees.jenkins.plugins.replication.ui.ReplicatedComputers.add(ReplicatedComputers.java:87) at PluginClassLoader for cloudbees-replication//com.cloudbees.jenkins.plugins.replication.ui.ReplicatedComputers$ComputerListenerImpl.onOnline(ReplicatedComputers.java:566) at jenkins.model.Jenkins.<init>(Jenkins.java:1056) at hudson.model.Hudson.<init>(Hudson.java:102) at hudson.model.Hudson.<init>(Hudson.java:87) at hudson.WebAppMain$3.run(WebAppMain.java:249) [...]
Resolution
This issue is caused by a non-upgraded ci-plugin:cloudbees-replication when upgrading the HA/HS instance to 2.479.1.4
(or higher) due to the checkbox called Allow automatic upgrades of plugins on restart
being disabled under .
If you are facing this issue, is because you have upgraded your instance to release 2.479.1.4
(or higher) without enabling the aforementioned checkbox Allow automatic upgrades of plugins on restart
.
As your controller is down, in order to fix this issue you have to follow the indications from Allow automatic upgrades of plugins on restart and enable automatic plugin upgrades upon restart by running (tested with version 2.414.2.2):
touch $JENKINS_HOME/cb-envelope/upgrade-plugins.enabled