Lately, I've been having issues where my scheduled rundeck jobs never terminate. After killing a stalled job, I'm presenting with these logs:
Dec 10 17:21:56 infra-2 build.sofi.com:5000/rundeck[28030]: 2016-12-10 17:21:56,453 - [ameissner] [Monitors] [/Fleet_Monitor] [#605071] [fe49691a-fead-4e94-9449-befa22b03714] - Result: 2147483647
Dec 10 17:21:56 infra-2 build.sofi.com:5000/rundeck[28030]: 2016-12-10 17:21:56,453 project: Monitors execution: 605071 (log, ) /Fleet_Monitor [fe49691a-fead-4e94-9449-befa22b03714] - Result: 2147483647
Dec 10 17:21:56 infra-2 build.sofi.com:5000/rundeck[28030]: 2016-12-10 17:21:56,859 - [ameissner] [Monitors] [/Fleet_Monitor] [#605071] [fe49691a-fead-4e94-9449-befa22b03714] - Failed: NonZeroResultCode: Result code was 2147483647
Dec 10 17:21:56 infra-2 build.sofi.com:5000/rundeck[28030]: 2016-12-10 17:21:56,863 - [] [] [/] [#] [] - Execution failed: 605071 in project Monitors: [Workflow result: , step failures: {1=Dispatch failed on 1 nodes: [internal.sofi.com: NonZeroResultCode: Result code was 2147483647]}, Node failures: {internal.sofi.com=[NonZeroResultCode: Result code was 2147483647]}, status: failed]
Dec 10 17:21:56 infra-2 build.sofi.com:5000/rundeck[28030]: 2016-12-10 17:21:56,863 - [] [] [/] [#] [] - Failed addEvent for StreamingLogWriter(log4jStreamingLogWriterPlugin), disabling. Ambiguous method overloading for method java.util.HashMap#plus.
Dec 10 17:21:56 infra-2 build.sofi.com:5000/rundeck[28030]: Cannot resolve which method to invoke for [null] due to overlapping prototypes between:
Dec 10 17:21:56 infra-2 build.sofi.com:5000/rundeck[28030]: [interface java.util.Collection]
Dec 10 17:21:56 infra-2 build.sofi.com:5000/rundeck[28030]: [interface java.util.Map]
Dec 10 17:21:57 infra-2 build.sofi.com:5000/rundeck[28030]: 2016-12-10 17:21:56,859 project: Monitors execution: 605071 (log, ) /Fleet_Monitor [fe49691a-fead-4e94-9449-befa22b03714] - Failed: NonZeroResultCode: Result code was 2147483647
It looks like the ambiguous reference is maybe preventing rundeck from putting the job in a finished state, successful or otherwise. I've currently got jobs that have been running all night, even though the processes are complete.
Lately, I've been having issues where my scheduled rundeck jobs never terminate. After killing a stalled job, I'm presenting with these logs:
It looks like the ambiguous reference is maybe preventing rundeck from putting the job in a finished state, successful or otherwise. I've currently got jobs that have been running all night, even though the processes are complete.