Within a pipeline a stage is considered  as "Failed" if it failed the first time, even though a successful retry( ) has taken place afterwards. It ends-up with pipeline builds marked as Failed although all stages have finally succeeded.

This behavior entirely defeats the purpose of retry() blocks, which is precisely to cope with transient errors in complex pipelines stages, allowing them to complete successfully.

Thanks,

Phil.

PS: This works correctly on older versions of the plugin, retaining the status of the last retried step rather than the worst one...