We find ourselves struggling with:
org.jenkinsci.plugins.workflow.cps.CpsCompilationErrorsException: startup failed:
General error during class generation: Method code too large!
I understand that this is due to underlying jvm limitations…
My question is “does embedded node scripting such as sh blocks” contribute directly to the method size in any meaningful way or are each of these blocks basically just data off to the side that the script references? In other words is moving this script out of groovy and into standalone script files likely to help? In this particular case it is a scripted pipeline, but we run into this problem with declarative also.
On a related note, is there anyway to learn the current ‘method size’ ?