04-20-2017 12:45 AM
I spend more than 2 days trying to install a new cluster with installation scripts. Since it wasn't working (I always get the error "Unable to generate config file creds.localjceks"), I try to reinstall all manually.
But even when I do it manually I am still getting the same error :
Completed only 0/1 steps. First failure: Could not create process: com.cloudera.cmf.service.config.ConfigGenException: Unable to generate config file creds.localjceks
It drive me completly crazy... I didn't find a lot of help on the net, only 2 relevent articles :
=> from the Cloudera Knowledge database : https://cloudera-portal.force.com/articles/KB_Article/Not-able-to-start-hive-metastore-configGenExce...
So I have checked my java version (By the way, I am aware of the java bug if I use the java oracle JDK 1.8 121, So I use the Java oracle 1.8 111 version).
here is the java -version result :
java version "1.8.0_111"
Java(TM) SE Runtime Environment (build 1.8.0_111-b14)
Java HotSpot(TM) 64-Bit Server VM (build 25.111-b14, mixed mode)
here is the rpm installed :
And I am getting the same error when I try to launch Hive, Oozie or Hue.
=> And from apache website : https://hadoop.apache.org/docs/current/hadoop-project-dist/hadoop-common/CredentialProviderAPI.html
it seams to be something about SSL.
So what can I do please ? I don't find any other thing about this file creds.localjceks.
Have you a clue please ?
Solved! Go to Solution.
04-20-2017 11:33 AM
04-21-2017 12:49 AM
Yes script was running. CM is installed, I was able to add a new cluster, to add new services, the agent are installed and configured and they are able to talk with CM. But once my services are installed, they are unable to start (not all of them). But Hue, Hive Metastore, Oozie and Sentry are unable to start due to the same error :
Role failed to start due to error com.cloudera.cmf.service.config.ConfigGenException: Unable to generate config file creds.localjceks.
And I get this error when I install it trought the scripts or manually. And yes CM is running as root. When I say I do it manually, I was saying I was doing it without script, so with CM wizard, so it is not completly manually. And CM run as root, so this is not a right concern.
Do it help to target the problem ?
04-25-2017 01:43 AM
I answer to myself : it was coming from the schema. On postgreSQL the table wasn't existing. So after recreating the table, it was working.