Craig Condit created YUNIKORN-2180:
--------------------------------------
Summary: Clean up scheduler state initialization
Key: YUNIKORN-2180
URL: https://issues.apache.org/jira/browse/YUNIKORN-2180
Project: Apache YuniKorn
Issue Type: Sub-task
Reporter: Craig Condit
Assignee: Craig Condit
Scheduler state initialization (otherwise known as recovery) is currently
fragile and somewhat unpredictable since multiple asynchronous processes
coordinate to perform the various init tasks.
Startup initialization should be simplified to the following steps:
# Read all priority classes from the informer and register them with the
scheduler cache
# Read all nodes from the informer and register them (in a drained state) with
the scheduler core
# Read all pods from the informer and register applications and allocations as
necessary, associating existing allocations with nodes from step #2
# Enable the nodes which were originally registered in step #2
# Register and start Kubernetes event handlers
# Re-read priority classes from the informer and remove any that have gone
away since step #1, ensuring we don't miss priority class deletions during init
# Re-read nodes from the informer and remove any that have gone away since
step #2, ensuring we don't miss node deletions during init
# Re-read pods from the informer and remove any that have gone away since step
#3, ensuring we don't miss pod deletions during init
Additionally, this process should be handled entirely by the scheduler context
to avoid mulitple competing concerns.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
