[
https://issues.apache.org/jira/browse/GOBBLIN-1728?focusedWorklogId=818953&page=com.atlassian.jira.plugin.system.issuetabpanels:worklog-tabpanel#worklog-818953
]
ASF GitHub Bot logged work on GOBBLIN-1728:
-------------------------------------------
Author: ASF GitHub Bot
Created on: 20/Oct/22 21:46
Start Date: 20/Oct/22 21:46
Worklog Time Spent: 10m
Work Description: homatthew commented on code in PR #3586:
URL: https://github.com/apache/gobblin/pull/3586#discussion_r1001148782
##########
gobblin-yarn/src/main/java/org/apache/gobblin/yarn/YarnService.java:
##########
@@ -845,8 +877,6 @@ public void onContainersAllocated(List<Container>
containers) {
instanceName = null;
}
}
- allocatedContainerCountMap.put(containerHelixTag,
Review Comment:
Turns out it is thread safe to iterate over concurrent hash set.
https://stackoverflow.com/questions/56099744/java-concurrenthashset-iterating-over-it-in-a-multi-threaded-environment
And the reason for the lock / synchronized block is because if multiple
threads are iterating over the same concurrent hashset and one of the threads
calls remove, then since the iterator is essentially a snapshot they could each
re-use the same helix instance.
The sync block makes the removal process completely thread safe
Issue Time Tracking
-------------------
Worklog Id: (was: 818953)
Time Spent: 4h 10m (was: 4h)
> Yarn Service requests too many containers due to improper calculation
> ---------------------------------------------------------------------
>
> Key: GOBBLIN-1728
> URL: https://issues.apache.org/jira/browse/GOBBLIN-1728
> Project: Apache Gobblin
> Issue Type: New Feature
> Reporter: Matthew Ho
> Priority: Major
> Time Spent: 4h 10m
> Remaining Estimate: 0h
>
> Yarn Service is responsible for calculating the number of instances based on
> the helix tasks. Yarn service tracks the number of instances by asking Yarn
> for the number of resource requests and the number of allocated containers.
>
> It uses this count to determine if it should ask for more containers or
> shrink the number of containers. This calculation is currently done
> improperly and we continue to request containers when we have enough
> requested.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
