big-data/hadoop
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

YARN-10651. CapacityScheduler crashed with NPE in AbstractYarnScheduler.updateNodeResource(). Contributed by Haibo Chen

Browse Source 
(cherry picked from commit f348ab3f2f468751af329a1ffce4917cb000fcbf)
This commit is contained in:
Jonathan Hung 2021-02-25 15:08:43 -08:00
parent db457b056a
commit be6e99963d
1 changed files with 5 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

5
hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/AbstractYarnScheduler.java
View File

@ -842,6 +842,11 @@ public void updateNodeResource(RMNode nm,
writeLock.lock();
try {
SchedulerNode node = getSchedulerNode(nm.getNodeID());
if (node == null) {
LOG.info("Node: " + nm.getNodeID() + " has already been taken out of " +
"scheduling. Skip updating its resource");
return;
}
Resource newResource = resourceOption.getResource();
final int timeout = resourceOption.getOverCommitTimeout();
Resource oldResource = node.getTotalResource();