You are viewing a plain text version of this content. The canonical link for it is here.

Posted to dev@gobblin.apache.org by "Kip Kohn (Jira)" <ji...@apache.org> on 2021/08/12 19:50:00 UTC

[jira] [Closed] (GOBBLIN-1509) Ensure flows transition to FAILED and not stuck in COMPILED upon DagManager::addDag error

     [ https://issues.apache.org/jira/browse/GOBBLIN-1509?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Kip Kohn closed GOBBLIN-1509.
-----------------------------
    Resolution: Fixed

my fix was accepted in https://github.com/apache/gobblin/pull/3357

> Ensure flows transition to FAILED and not stuck in COMPILED upon DagManager::addDag error
> -----------------------------------------------------------------------------------------
>
>                 Key: GOBBLIN-1509
>                 URL: https://issues.apache.org/jira/browse/GOBBLIN-1509
>             Project: Apache Gobblin
>          Issue Type: Bug
>            Reporter: Kip Kohn
>            Priority: Major
>          Time Spent: 2h 10m
>  Remaining Estimate: 0h
>
> Presently, addDag failure leaves the flow marooned in the COMPILED state, as the warranted FLOW_FAILED event is never sent.  Particularly insidious is that scheduled flows with their execution stuck in COMPILED miss their next execution, unless `flow.allowConcurrentExecutions` is set.  Thus the scheduled flow is stuck in its entirety, not merely a single execution.
> One observed cause of addDag failure is when the DagStateStore is backed by a replicated DB (e.g. MySqlDagStateStore) that just switched leaders. Cached connections in the pool may suddently point to a read-only follower unable to DagStateStore::writeCheckpoint.



--
This message was sent by Atlassian Jira
(v8.3.4#803005)