Jasbir Dhillon created AMBARI-14049:
---------------------------------------
Summary: Installing cluster with accumulo using blueprints causes
accumulo tserver startup failure
Key: AMBARI-14049
URL: https://issues.apache.org/jira/browse/AMBARI-14049
Project: Ambari
Issue Type: Bug
Components: blueprints
Affects Versions: 2.1.2
Environment: CentOS 6
Reporter: Jasbir Dhillon
Priority: Critical
Installing accumulo on a cluster using blueprints causes accumulo tserver
startup failure.
We have three host_groups.
Host_group_1 consists of the components: PIG, KAFKA_BROKER, HISTORYSERVER,
OOZIE_CLIENT, NAMENODE, HCAT, METRICS_MONITOR, AMBARI_SERVER, ACCUMULO_MASTER,
APP_TIMELINE_SERVER, ACCUMULO_MONITOR, HIVE_CLIENT, HDFS_CLIENT,
METRICS_COLLECTOR, FLUME_HANDLER, DATANODE, WEBHCAT_SERVER, ZOOKEEPER_SERVER,
ZOOKEEPER_CLIENT, HIVE_SERVER, OOZIE_SERVER, TEZ_CLIENT, HIVE_METASTORE,
ACCUMULO_TRACER, YARN_CLIENT, ACCUMULO_GC, MAPREDUCE2_CLIENT, MYSQL_SERVER,
ACCUMULO_CLIENT
Host_group_2 consists of the components: ZOOKEEPER_SERVER, NODEMANAGER,
METRICS_MONITOR, SECONDARY_NAMENODE, DATANODE, ACCUMULO_TSERVER.
Host_group_3 consists of the components: ZOOKEEPER_SERVER, NODEMANAGER,
METRICS_MONITOR, DATANODE, ACCUMULO_TSERVER
The accumulo tservers fail due to the following error. The accumulo tserver
start up task should wait for the namenode to start up which is running on host
group 1. I tested the above configuration without blueprints and it works fine
as host group 2 and host group 3 wait for tasks on host group 1 to complete
prior to starting up the components.
2015-11-24 18:59:05,807 [zookeeper.ZooUtil] ERROR: Problem reading instance id
out of hdfs at hdfs://<host group 1 server
name>:8020/apps/accumulo/data/instance_id
java.net.ConnectException: Call From <host group 1 server name>/<ip address> to
<host group 1 server name>:8020 failed on connection exception:
java.net.ConnectException: Connection refused; For more details see:
http://wiki.apache.org/hadoop/ConnectionRefused
at sun.reflect.NativeConstructorAccessorImpl.newInstance0(Native Method)
at
sun.reflect.NativeConstructorAccessorImpl.newInstance(NativeConstructorAccessorImpl.java:57)
at
sun.reflect.DelegatingConstructorAccessorImpl.newInstance(DelegatingConstructorAccessorImpl.java:45)
at java.lang.reflect.Constructor.newInstance(Constructor.java:526)
at org.apache.hadoop.net.NetUtils.wrapWithMessage(NetUtils.java:792)
at org.apache.hadoop.net.NetUtils.wrapException(NetUtils.java:732)
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
