Home Forums YARN yarn-resource-manager having nodes in unhealthy

Tagged: 

This topic contains 3 replies, has 4 voices, and was last updated by  Atul Chaudhari 4 months, 1 week ago.

  • Creator
    Topic
  • #28189

    Software: HDP 2.0
    OS: CentOS 6.4
    Cluster size : 1-NN,RM, 1-SN, 5 data node – test cluster

    I am seeing all the nodes reported as unhealthy in the Application Manager Cluster. Due to this ,the jobs are not getting executed. I am able to upload the files successfully from hadoop fs command line.

    >>> Following the snippet of the yarn-yarn-resourcemanager-xxxx.log
    qatstvm16.sensage.com:45454 Node Transitioned from NEW to RUNNING
    2013-06-24 20:04:48,999 INFO capacity.CapacityScheduler (CapacityScheduler.java:addNode(710)) – Added node qatstvm16.sensage.com:45454 clusterResource:
    2013-06-24 20:04:50,022 INFO rmnode.RMNodeImpl (RMNodeImpl.java:handle(324)) – qatstvm16.sensage.com:45454 Node Transitioned from RUNNING to UNHEALTHY
    2013-06-24 20:04:50,023 INFO capacity.CapacityScheduler (CapacityScheduler.java:removeNode(744)) – Removed node qatstvm16.sensage.com:45454 clusterResource:
    <<<<

    Any pointers to debug/fix this is much appreciated.

Viewing 3 replies - 1 through 3 (of 3 total)

You must be logged in to reply to this topic.

  • Author
    Replies
  • #45530

    was this resolved? The ticket seems to be open for long time.
    What was the resolution?

    Collapse
    #36172

    runeetv
    Member

    Do you see anything on the resourcemanager web-UI, specifically the nodes’ page? The node’s page should show you the reason why it became unhealthy. Do you have any node health script configured?

    Collapse
    #28672

    tedr
    Moderator

    Hi Satish,

    I am researching this, will get back to you when I have an answer.

    Thanks,
    Ted.

    Collapse
Viewing 3 replies - 1 through 3 (of 3 total)