Zookeeper failed to start after reboot

This topic contains 3 replies, has 3 voices, and was last updated by  Larry Liu 2 years ago.

  • Creator
    Topic
  • #14735

    Hi :

    I have a HDP1.2 cluster with 10 nodes, and have installed HDP1.2 successfully with my cluster 10 days ago. My cluster worked fine for theses days until yesterday.
    Unfortunately, after reboot my cluster yesterday, I found I can only started HDFS and MapRed, but zookeeper service. Totally,I have 3 zookeepers with name as host001,host002 and host003.
    zookeepers in host002, and host003 seemed to be successfully started, but Zookeeper in host001 failed to be restarted. I am confused. What’s wrong with me ? What’s the correct reboot procedure to HDP1.2 ?

    PS1. I restarted amberi-serever in host001 and restarted ambari-client in host002-host010.
    PS2.Because I can not find zookeeper log file in /var/log/hadoop/zookeeper, so I do not know what to do now.

    Thanks for any help.

    Regards,

    Jeff

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

You must be to reply to this topic. | Create Account

  • Author
    Replies
  • #14889

    Larry Liu
    Moderator

    Hi, Jeff

    One thing you can try is to try to stop zookeeper from ambari web UI and start them if you installed your cluster with Ambari.

    Larry

    Collapse
    #14815

    Hi tedr:

    Thanks for your help.

    Actually I have manually restarted zookeeper on hosty001 by “sudo -u zookeeper /usr/lib/zookeeper/bin/zkServer.sh start /etc/zookeeper/conf/zoo.cfg”

    I will try your suggestion next time while I restart cluster again.

    Thanks

    Regards,

    Jeff

    Collapse
    #14738

    tedr
    Member

    Hi Jeff,

    It looks like all that you may have missed is restarting the ambari-agent on host001. Ambari-agent needs to be started on all hosts that have any hadoop services running on them.

    Thanks,
    Ted.

    Collapse
Viewing 3 replies - 1 through 3 (of 3 total)
Hortonworks Data Platform
The Hortonworks Data Platform is a 100% open source distribution of Apache Hadoop that is truly enterprise grade having been built, tested and hardened with enterprise rigor.
Get started with Sandbox
Hortonworks Sandbox is a self-contained virtual machine with Apache Hadoop pre-configured alongside a set of hands-on, step-by-step Hadoop tutorials.
Modern Data Architecture
Tackle the challenges of big data. Hadoop integrates with existing EDW, RDBMS and MPP systems to deliver lower cost, higher capacity infrastructure.