Some of our production jobs are being killed ,when i collected yarn logs i found below info
application_14880028******_*** does not exist. Log aggregation has not completed or is not enabled.
The jobs are killed before launching the containers or before nodes are assigned.
HDP version =2.3.4
@Vijay, It looks like containers are not getting launched. If AM is failing to launch, log aggregation is expected to not work. In such failures, Resource manager and Node manager logs can help. Please look at RM and NM logs for above application. The stack trace from these logs can help further debug this issue.