Timothee, Can you post the entire exception stacktrace? -Vinay ...
Can you post the entire exception stacktrace?
Hello HDP community,
i am currently working on building a secure Hadoop platform. My goal is to set up all security measures we know about in the hadoop world.
So after a fresh install of HDP-2.0.6 with ambari, i stopped all services, spawned a Kerberos KDC, followed all steps to secure the HDFS service (for the first step in security). Everything is fine, my HDFS service (1 NN, 1 SecondNN, and 10 DN) is running with KDC auth, and i’m happy with it. To complete the security around HDFS, i wanted to configure HTTPS for the webUI , then 1way SSL and then 2ways SSL.
But unfortunately, i’m stuck with the UI and cannot start the hdfs service with ssl enabled.
I will describe the changes in configuration i made :
hadoop.ssl.enabled = true
dfs.client.https.need-auth = false
dfs.https.enable = true
That’s all. everything else is default conf or working conf with kerberos enabled.
Of course, on all my nodes working the hdfs service, i put the ssl-server.xml file with the appropriate configuration (jks store, location and password). The store file is also on my nodes.
However, on service start, i get this error on every nodes :
2014-03-24 10:50:50,516 WARN mortbay.log (Slf4jLog.java:warn(76)) – java.lang.NullPointerException
2014-03-24 10:50:50,516 WARN mortbay.log (Slf4jLog.java:warn(76)) – failed SslSocketConnector@namenode.cluster.hdp:50470: java.io.IOException: !JsseListener: java.lang.NullPointerException
2014-03-24 10:50:50,516 WARN mortbay.log (Slf4jLog.java:warn(76)) – failed Server@3e5b38d7: java.io.IOException: !JsseListener: java.lang.NullPointerException
2014-03-24 10:50:50,517 INFO http.HttpServer (HttpServer.java:start(690)) – HttpServer.start() threw a non Bind IOException
java.io.IOException: !JsseListener: java.lang.NullPointerException
I have crawled the web to find an answer but i’m stuck here. Do you guys have any ideas ? Thanks a lot !
A HDP Support Subscription connects you experts with deep experience running Apache Hadoop in production, at-scale on the most demanding workloads.
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