[jira] [Created] (AMBARI-15471) Race condition happens when orchestrating hadoop cluster using blueprint

["Chen He (JIRA)" <ji...@apache.org>]

Chen He created AMBARI-15471:
--------------------------------

             Summary: Race condition happens when orchestrating hadoop cluster using blueprint
                 Key: AMBARI-15471
                 URL: https://issues.apache.org/jira/browse/AMBARI-15471
             Project: Ambari
          Issue Type: Bug
    Affects Versions: 2.2.2
            Reporter: Chen He


Assume there are 3 nodes, NN and RM are HA and running on node1 and node2. However, historyserver is on node3. Historyserver is depending on namenode. Once command "cluster create" is issued. We saw following errors:
"  File "/usr/lib/python2.6/site-packages/resource_management/libraries/functions/namenode_ha_utils.py", line 124, in get_property_for_active_namenode
    raise Fail("There is no active namenodes.")
resource_management.core.exceptions.Fail: There is no active namenodes."

Actually, two namenode are starting and succeed. 

It means the historyserver starts before namenodes are ready.  



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)