You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@helix.apache.org by "Karthiek (JIRA)" <ji...@apache.org> on 2014/11/12 22:28:34 UTC

[jira] [Created] (HELIX-545) Scheduling a job to Helix job queue when the job's target resource doesn't exist, leaves the queue in inconsistent state

Karthiek created HELIX-545:
------------------------------

             Summary: Scheduling a job to Helix job queue when the job's target resource doesn't exist, leaves the queue in inconsistent state
                 Key: HELIX-545
                 URL: https://issues.apache.org/jira/browse/HELIX-545
             Project: Apache Helix
          Issue Type: Bug
            Reporter: Karthiek


In Helix job queue, if we schedule a job for a target resource that doesn't exist, the queue gets into an inconsistent state. Workflow DAG gets modified to include the new job, but a new Z-node for job is never created. When this job reaches head of queue, Helix is unable to schedule it and gets stuck, preventing other jobs in queue from being scheduled. 

We just need a sanity check in Helix, to ensure that a wrong job config does not cause the queue to stall.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)