spark created RANGER-5140:
-----------------------------
Summary: Local Ranger-Build with ranger-2.5 (ranger-hadoop service
not starting up with ranger_in_docker compose script)
Key: RANGER-5140
URL: https://issues.apache.org/jira/browse/RANGER-5140
Project: Ranger
Issue Type: Bug
Components: Ranger
Affects Versions: 2.5.0
Reporter: spark
Attachments: Screenshot 2025-02-11 at 2.41.19 PM.png
Basically, I am testing a local build and trying to bring up ranger core
services and plugin services. It seems that for ranger hadoop service, the
namenode process fails to start up and exits due to java.net.ConnectException.
The error message I get is:
*Call From ranger-hadoop/10.89.0.10 to localhost:9000 failed on connection
exception*
I checked the /etc/hosts file within the container filesystem and saw the file
was empty (with no DNS resolved hostname/IP pairs). It seems that when primary
namenode is in standby state and shuts down, it brings up another secondary
namenode with exact same host-name/ip pairing, where connection exception issue
arises.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
