Ambari Forum

Having trouble with agent/server version mismatch

  • #33714
    M F

    I’ve had a working ambari deployment on a longstanding Amazon EC2 cluster for about a month now. Recently I went to add new hosts to the cluster, and received the following error message in the logs:

    SEVERE: The exception contained within MappableContainerException could not be mapped to a response, re-throwing to the HTTP container
    org.apache.ambari.server.AmbariException: Cannot register host with non compatible agent version, hostname=ip-10-165-2-215.ec2.internal, agentVersion=, serverVersion=1.3.0

    The unfortunate thing is no matter how hard I look I can’t seem to find a way to upgrade the agent version to 1.3.0 so that I don’t see this problem. Does anybody know where I can either find the source for compile? For instance, looking through my yum list | grep ambari I only see versions. Updating doesn’t reveal a 1.3.0.

    Thanks in advance.

to create new topics or reply. | New User Registration

  • Author
  • #33729
    M F

    Any mod that sees this can close the topic. I figured out what’s going on:

    Looking at the /var/lib/ambari-server/data directory reveals a “version” file. This was reporting 1.3 to me on the slave nodes (ones running a DataNode and JobTracker). Given that the error above says the server version is 1.3, that matched up and made sense to me.

    It turned out that ambari-agent and ambari-server were *actually*, and this was revealed doing a yum list installed | grep ambari

    So once I realized this, I was able to downgrade the new node and it added to the cluster properly. This caused a new problem, but that problem is not relevant to this thread.

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

Support from the Experts

A HDP Support Subscription connects you experts with deep experience running Apache Hadoop in production, at-scale on the most demanding workloads.

Enterprise Support »

Become HDP Certified

Real world training designed by the core architects of Hadoop. Scenario-based training courses are available in-classroom or online from anywhere in the world

Training »

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.