Closed phuong-leeo closed 1 year ago
@phuong-leeo , if you are still experiencing this issue, could you take a jstack of the master process while the CPU is high.
Even better, you could try to create a flamegraph
A jstack is probably the easiest method right now.
tks @ZacBlanco , I should give it a try
This issue has been automatically marked as stale because it has not had recent activity. It will be closed in two weeks if no further activity occurs. Thank you for your contributions.
Will close it for now, feel free to reopen it and contact us if this is still valid.
Alluxio Version: V2.2.0
AWS EMR version: 5.25.0
Presto version: 0.220
Cluster specs 1: Master c5.4xlarge 1: Core r5a.xlarge 10: Task r5a.xlarge
Describe the bug After Alluxio had been cached approx 40GB data. CPU started getting high load on Master node we thought it was caused by storing metastore on ROCKS, we then turnover to HEAP, but it did not help us solve this issue. We did not face kinda this situation on Alluxio v1.8
Error log
alluxio-site.properties
To Reproduce Filled the cache at least 40GB
Expected behavior reduce CPU load on Master node, firgure out why
Journal flush failed
happenedUrgency high