[
https://issues.apache.org/jira/browse/IGNITE-22979?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Aleksandr Polovtsev updated IGNITE-22979:
-----------------------------------------
Labels: ignite-3 (was: )
> Add message throttling to CriticalWorkerWatchdog
> ------------------------------------------------
>
> Key: IGNITE-22979
> URL: https://issues.apache.org/jira/browse/IGNITE-22979
> Project: Ignite
> Issue Type: Improvement
> Reporter: Aleksandr Polovtsev
> Assignee: Aleksandr Polovtsev
> Priority: Major
> Labels: ignite-3
>
> {{CriticalWorkerWatchdog}} is responsible for monitoring some threads that
> are not intended to be blocked for long periods of time. However, if the
> system is in a bad state for some reason (e.g. memory is being consumed by
> other processes), this watchdog starts spamming with a lot of messages, which
> pollutes the log and makes later investigation more difficult.
> I propose to add an exponential throttling policy: if a thread has been
> blocked more than a configured amount of time and the message has been
> printed, the next check must be performed in an increased time interval
> comparing to the previous check. If no threads are blocked, then the check
> interval can be reset to its default value.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
