Wilfred Spiegelenburg created YUNIKORN-1065:
-----------------------------------------------
Summary: panic in nodeUpdate when node disappears
Key: YUNIKORN-1065
URL: https://issues.apache.org/jira/browse/YUNIKORN-1065
Project: Apache YuniKorn
Issue Type: Improvement
Components: core - scheduler
Reporter: Wilfred Spiegelenburg
Assignee: Wilfred Spiegelenburg
Based on a message in the chat channel:
{code:java}
panic: runtime error:
invalid memory address or nil pointer dereference signal SIGSEGV:
segmentation violation code=0x1 addr=0x0 pc=0x9f4da0]
goroutine 16 [running]:
github.com/apache/incubator-yunikorn-core/pkg/scheduler.(*ClusterContext).updateNode(0xc00014d2c0,
0xc020451f10)
/go/pkg/mod/github.com/apache/[email protected]/pkg/scheduler/context.go:620
+0x3a0
github.com/apache/incubator-yunikorn-core/pkg/scheduler.(*ClusterContext).processNodes(0xc00014d2c0,
0xc0213e0410)
/go/pkg/mod/github.com/apache/[email protected]/pkg/scheduler/context.go:230
+0x405
github.com/apache/incubator-yunikorn-core/pkg/scheduler.(*ClusterContext).handleRMUpdateNodeEvent(...)
/go/pkg/mod/github.com/apache/[email protected]/pkg/scheduler/context.go:206{code}
Thanks to Ben Roubicek for finding this
--
This message was sent by Atlassian Jira
(v8.20.1#820001)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
