flink-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "ASF GitHub Bot (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (FLINK-2354) Recover running jobs on JobManager failure
Date Fri, 02 Oct 2015 09:57:27 GMT

    [ https://issues.apache.org/jira/browse/FLINK-2354?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14940974#comment-14940974

ASF GitHub Bot commented on FLINK-2354:

Github user uce commented on the pull request:

    While working on FLINK-2804 I've noticed an issue: the user jars are only uploaded to
the leading job manager at submission time and then are not available to the other job managers
on recovery. A simple solution is to make the user jars available in the file state backend
as well.

> Recover running jobs on JobManager failure
> ------------------------------------------
>                 Key: FLINK-2354
>                 URL: https://issues.apache.org/jira/browse/FLINK-2354
>             Project: Flink
>          Issue Type: Sub-task
>          Components: JobManager
>    Affects Versions: master
>            Reporter: Ufuk Celebi
>            Assignee: Ufuk Celebi
>             Fix For: 0.10
> tl;dr Persist JobGraphs in state backend and coordinate reference to state handle via
> Problem: When running multiple JobManagers in high availability mode, the leading job
manager looses all running jobs when it fails. After a new leading job manager is elected,
it is not possible to recover any previously running jobs.
> Solution: The leading job manager, which receives the job graph writes 1) the job graph
to a state backend, and 2) a reference to the respective state handle to ZooKeeper. In general,
job graphs can become large (multiple MBs, because they include closures etc.). ZooKeeper
is not designed for data of this size. The level of indirection via the reference to the state
backend keeps the data in ZooKeeper small.
> Proposed ZooKeeper layout:
> /flink (default)
>   +- currentJobs
>        +- job id i
>             +- state handle reference of job graph i
> The 'currentJobs' node needs to be persistent to allow recovery of jobs between job managers.
The currentJobs node needs to satisfy the following invariant: There is a reference to a job
graph with id i IFF the respective job graph needs to be recovered by a newly elected job
manager leader.
> With this in place, jobs will be recovered from their initial state (as if resubmitted).
The next step is to backup the runtime state handles of checkpoints in a similar manner.
> ---
> This work will be based on [~trohrmann@apache.org]'s implementation of FLINK-2291. The
leader election service notifies the job manager about granted/revoked leadership. This notification
happens via Akka and thus serially *per* job manager, but results in eventually consistent
state between job managers. For some snapshots of time it is possible to have a new leader
granted leadership, before the old one has been revoked its leadership.
> [~trohrmann@apache.org], can you confirm that leadership does not guarantee mutually
exclusive access to the shared 'currentJobs' state?
> For example, the following can happen:
> - JM 1 is leader, JM 2 is standby
> - JOB i is running (and hence /flink/currentJobs/i exists)
> - ZK notifies leader election service (LES) of JM 1 and JM 2
> - LES 2 immediately notifies JM 2 about granted leadership, but LES 1 notification revoking
leadership takes longer
> - JOB i finishes (TMs don't notice leadership change yet) and JM 1 receives final JobStatusChange
> - JM 2 resubmits the job /flink/currentJobs/i
> - JM 1 removes /flink/currentJobs/i, because it is now finished
> => inconsistent state (wrt the specified invariant above)
> If it is indeed a problem, we can circumvent this with a Curator recipe for [shared locks|http://curator.apache.org/curator-recipes/shared-lock.html]
to coordinate the access to currentJobs. The lock needs to be acquired on leadership.
> ---
> Minimum required tests:
> - Unit tests for job graph serialization and writing to state backend and ZooKeeper with
expected nodes
> - Unit tests for job submission to job manager in leader/non-leader state
> - Unit tests for leadership granting/revoking and job submission/restarting interleavings
> - Process failure integration tests with single and multiple running jobs

This message was sent by Atlassian JIRA

View raw message