Tamas Domok created YARN-11422:
----------------------------------

             Summary: fs2cs failes to generate valid capacity-scheduler.xml
                 Key: YARN-11422
                 URL: https://issues.apache.org/jira/browse/YARN-11422
             Project: Hadoop YARN
          Issue Type: Bug
          Components: capacityscheduler
    Affects Versions: 3.4.0
            Reporter: Tamas Domok
            Assignee: Tamas Domok
         Attachments: fair-scheduler-issue.xml, yarn-site.xml

Attachments:
 - [^fair-scheduler-issue.xml]
 - [^yarn-site.xml]

Save them to /tmp/fair-scheduler.xml and /tmp/yarn-site.xml.

Run the following command:
{code}
rm -rf /tmp/output/*; yarn fs2cs --cluster-resource 
memory-mb=36201037824,vcores=4914 --no-terminal-rule-check -y 
/tmp/yarn-site.xml -f /tmp/fair-scheduler.xml -o  /tmp/output
{code}

The conversion fails with the following error:
{code}
2023-01-20 13:37:42,250 INFO service.AbstractService: Service 
org.apache.hadoop.yarn.server.resourcemanager.scheduler.capacity.CapacityScheduler
 failed in state STOPPED
java.lang.NullPointerException
        at 
org.apache.hadoop.yarn.server.resourcemanager.scheduler.capacity.CapacityScheduler.serviceStop(CapacityScheduler.java:428)
        at 
org.apache.hadoop.service.AbstractService.stop(AbstractService.java:221)
        at 
org.apache.hadoop.service.AbstractService.close(AbstractService.java:248)
        at 
org.apache.hadoop.yarn.server.resourcemanager.scheduler.fair.converter.ConvertedConfigValidator.validateConvertedConfig(ConvertedConfigValidator.java:74)
        at 
org.apache.hadoop.yarn.server.resourcemanager.scheduler.fair.converter.FSConfigToCSConfigArgumentHandler.parseAndConvert(FSConfigToCSConfigArgumentHandler.java:179)
        at 
org.apache.hadoop.yarn.server.resourcemanager.scheduler.fair.converter.FSConfigToCSConfigConverterMain.main(FSConfigToCSConfigConverterMain.java:47)
2023-01-20 13:37:42,251 ERROR converter.ConvertedConfigValidator: Could not 
start Capacity Scheduler
org.apache.hadoop.yarn.exceptions.YarnException: Failed to initialize queues
        at 
org.apache.hadoop.yarn.server.resourcemanager.scheduler.capacity.CapacityScheduler.initializeQueues(CapacityScheduler.java:807)
        at 
org.apache.hadoop.yarn.server.resourcemanager.scheduler.capacity.CapacityScheduler.initScheduler(CapacityScheduler.java:318)
        at 
org.apache.hadoop.yarn.server.resourcemanager.scheduler.capacity.CapacityScheduler.serviceInit(CapacityScheduler.java:412)
        at 
org.apache.hadoop.yarn.server.resourcemanager.scheduler.fair.converter.ConvertedConfigValidator.validateConvertedConfig(ConvertedConfigValidator.java:70)
        at 
org.apache.hadoop.yarn.server.resourcemanager.scheduler.fair.converter.FSConfigToCSConfigArgumentHandler.parseAndConvert(FSConfigToCSConfigArgumentHandler.java:179)
        at 
org.apache.hadoop.yarn.server.resourcemanager.scheduler.fair.converter.FSConfigToCSConfigConverterMain.main(FSConfigToCSConfigConverterMain.java:47)
        Suppressed: java.lang.NullPointerException
                at 
org.apache.hadoop.yarn.server.resourcemanager.scheduler.capacity.CapacityScheduler.serviceStop(CapacityScheduler.java:428)
                at 
org.apache.hadoop.service.AbstractService.stop(AbstractService.java:221)
                at 
org.apache.hadoop.service.AbstractService.close(AbstractService.java:248)
                at 
org.apache.hadoop.yarn.server.resourcemanager.scheduler.fair.converter.ConvertedConfigValidator.validateConvertedConfig(ConvertedConfigValidator.java:74)
                ... 2 more
Caused by: org.apache.hadoop.yarn.exceptions.YarnRuntimeException: Unable to 
construct queue ordering policy=fair queue=root.default
        at 
org.apache.hadoop.yarn.server.resourcemanager.scheduler.capacity.CapacitySchedulerConfiguration.getQueueOrderingPolicy(CapacitySchedulerConfiguration.java:2053)
        at 
org.apache.hadoop.yarn.server.resourcemanager.scheduler.capacity.ParentQueue.setupQueueConfigs(ParentQueue.java:172)
        at 
org.apache.hadoop.yarn.server.resourcemanager.scheduler.capacity.ParentQueue.<init>(ParentQueue.java:140)
        at 
org.apache.hadoop.yarn.server.resourcemanager.scheduler.capacity.ParentQueue.<init>(ParentQueue.java:115)
        at 
org.apache.hadoop.yarn.server.resourcemanager.scheduler.capacity.CapacitySchedulerQueueManager.parseQueue(CapacitySchedulerQueueManager.java:276)
        at 
org.apache.hadoop.yarn.server.resourcemanager.scheduler.capacity.CapacitySchedulerQueueManager.parseQueue(CapacitySchedulerQueueManager.java:283)
        at 
org.apache.hadoop.yarn.server.resourcemanager.scheduler.capacity.CapacitySchedulerQueueManager.initializeQueues(CapacitySchedulerQueueManager.java:176)
        at 
org.apache.hadoop.yarn.server.resourcemanager.scheduler.capacity.CapacityScheduler.initializeQueues(CapacityScheduler.java:798)
        ... 5 more
Caused by: java.lang.ClassNotFoundException: fair
        at java.net.URLClassLoader.findClass(URLClassLoader.java:382)
        at java.lang.ClassLoader.loadClass(ClassLoader.java:418)
        at sun.misc.Launcher$AppClassLoader.loadClass(Launcher.java:352)
        at java.lang.ClassLoader.loadClass(ClassLoader.java:351)
        at java.lang.Class.forName0(Native Method)
        at java.lang.Class.forName(Class.java:264)
        at 
org.apache.hadoop.yarn.server.resourcemanager.scheduler.capacity.CapacitySchedulerConfiguration.getQueueOrderingPolicy(CapacitySchedulerConfiguration.java:2049)
        ... 12 more
2023-01-20 13:37:42,251 ERROR converter.FSConfigToCSConfigArgumentHandler: 
Verification failed: Failed to initialize queues
Verification failed: Failed to initialize queues
2023-01-20 13:37:42,251 ERROR converter.FSConfigToCSConfigConverterMain: Error 
while starting FS configuration conversion, see previous error messages for 
details!
{code}

With the following manual changes on the generated capacity-scheduler.xml:
 - add: yarn.scheduler.capacity.root.users.auto-queue-creation-v2.enabled=true
 - remove: 
yarn.scheduler.capacity.root.default.auto-queue-creation-v2.enabled=true
 - remove: yarn.scheduler.capacity.root.users.ordering-policy=fair

RM can start up.




--
This message was sent by Atlassian Jira
(v8.20.10#820010)

---------------------------------------------------------------------
To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org

Reply via email to