-
Notifications
You must be signed in to change notification settings - Fork 25.6k
Open
Labels
:Search Foundations/SearchCatch all for Search FoundationsCatch all for Search Foundations>bugTeam:Search FoundationsMeta label for the Search Foundations team in ElasticsearchMeta label for the Search Foundations team in Elasticsearchpriority:highA label for assessing bug priority to be used by ES engineersA label for assessing bug priority to be used by ES engineers
Description
Elasticsearch Version
8.16+
Installed Plugins
No response
Java Version
bundled
OS Version
any
Problem Description
Restricted the free_context action to a single thread. This, however, can cause PIT closures to backup tremendously if there is a burst and this might eventually cause a "thundering herd" type of problem as PITs keep being created and never actually closed over hours.
To help debug this type of problem, we should expose the queueing information of this throttle pool in diagnostics.
My thought is that it should just be part of our threadpool output, but it can be done in any old way.
Steps to Reproduce
Its impossible to receive information about the free context action...
Logs (if relevant)
Note, in hot-threads, the output will be in the "generic" thread pool as that is the actual runner
Metadata
Metadata
Assignees
Labels
:Search Foundations/SearchCatch all for Search FoundationsCatch all for Search Foundations>bugTeam:Search FoundationsMeta label for the Search Foundations team in ElasticsearchMeta label for the Search Foundations team in Elasticsearchpriority:highA label for assessing bug priority to be used by ES engineersA label for assessing bug priority to be used by ES engineers