[ 
https://issues.apache.org/jira/browse/NIFI-16088?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=18094551#comment-18094551
 ] 

ASF subversion and git services commented on NIFI-16088:
--------------------------------------------------------

Commit 337069a4eb2f57fc194855df42b16d0e8586f474 in nifi's branch 
refs/heads/main from Kevin Doran
[ https://gitbox.apache.org/repos/asf?p=nifi.git;h=337069a4eb2 ]

NIFI-16088 Add scheduling diagnostics for cron primary-node-only processors 
(#11407)

Cron triggers are scheduled as one-shot tasks that reschedule only after a
successful run. If the triggered task throws, or the next trigger fails to be
scheduled, the scheduling chain silently terminates: the component stays
RUNNING but never fires again on its schedule until it is stopped and started,
with no record in the logs. Make that observable:

- CronSchedulingAgent: log a WARN (before propagating) when a cron-scheduled
  trigger throws or when rescheduling the next trigger fails, noting the
  component will not fire again until stopped and started.
- ConnectableTask: log at INFO (was DEBUG) when a cron-driven, primary-node-only
  component is skipped because this is not the primary node, so a skipped
  scheduled fire is visible; timer-driven components stay at DEBUG to avoid 
noise.

Diagnostics only; no scheduling or lifecycle behavior is changed.

> Add logging related to primary node only cron scheduling
> --------------------------------------------------------
>
>                 Key: NIFI-16088
>                 URL: https://issues.apache.org/jira/browse/NIFI-16088
>             Project: Apache NiFi
>          Issue Type: Improvement
>            Reporter: Kevin Doran
>            Assignee: Kevin Doran
>            Priority: Minor
>          Time Spent: 20m
>  Remaining Estimate: 0h
>
> I have seen but cannot reproduce or diagnose that after primary node 
> election, cron scheduled processors that are set to run only on primary node 
> don't fire at the configured time. 
> this ticket is to add logging so that if it reoccurs I can diagnose it better



--
This message was sent by Atlassian Jira
(v8.20.10#820010)

Reply via email to