[
https://issues.apache.org/jira/browse/SOLR-12240?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Andrzej Bialecki updated SOLR-12240:
-------------------------------------
Description:
Triggers determine whether to generate events by measuring if the time since
the last event is greater than {{waitFor}} or not. That is not the intended
purpose of {{waitFor}}. They should actually keep checking whether the
violation remains for the duration of the {{waitFor}} time, and only then
generate events.
This is further complicated by the fact that all triggers are paused whenever
an event is being processed (SOLR-11747). It seems that instead they should not
be paused but continue running to properly track the cluster state - they just
should not generate any events, or their events should be IGNORED.
was:
Triggers determine whether to generate events by measuring if the time since
the last event is greater than {{waitFor}} or not. That is not the intended
purpose of {{waitFor}}. They should actually check whether the rate breaches
the threshold and remains so for the {{waitFor}} time.
This is further complicated by the fact that all triggers are paused whenever
an event is being processed (SOLR-11747). It seems that instead they should not
be paused but continue running to properly track the cluster state - they just
should not generate any events, or their events should be IGNORED.
> Triggers don't properly track "waitFor"
> ---------------------------------------
>
> Key: SOLR-12240
> URL: https://issues.apache.org/jira/browse/SOLR-12240
> Project: Solr
> Issue Type: Sub-task
> Security Level: Public(Default Security Level. Issues are Public)
> Components: AutoScaling
> Affects Versions: 7.3, 7.4, master (8.0)
> Reporter: Andrzej Bialecki
> Priority: Major
>
> Triggers determine whether to generate events by measuring if the time since
> the last event is greater than {{waitFor}} or not. That is not the intended
> purpose of {{waitFor}}. They should actually keep checking whether the
> violation remains for the duration of the {{waitFor}} time, and only then
> generate events.
> This is further complicated by the fact that all triggers are paused whenever
> an event is being processed (SOLR-11747). It seems that instead they should
> not be paused but continue running to properly track the cluster state - they
> just should not generate any events, or their events should be IGNORED.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
