Activity monitor process comes down after start

[#1]

Activity monitor process comes down immedidately after start. Please let me know how to resolve the issue.

Exception in thread "main" java.lang.ExceptionInInitializerError
at com.mchange.v2.c3p0.DataSources.<clinit>(DataSources.java:59)
at org.hibernate.service.jdbc.connections.internal.C3P0ConnectionProvider.configure(C3P0ConnectionProvider.java:197)
at org.hibernate.service.internal.StandardServiceRegistryImpl.configureService(StandardServiceRegistryImpl.java:75)
at org.hibernate.service.internal.AbstractServiceRegistryImpl.initializeService(AbstractServiceRegistryImpl.java:159)
at org.hibernate.service.internal.AbstractServiceRegistryImpl.getService(AbstractServiceRegistryImpl.java:131)
at org.hibernate.engine.jdbc.internal.JdbcServicesImpl.buildJdbcConnectionAccess(JdbcServicesImpl.java:223)
at org.hibernate.engine.jdbc.internal.JdbcServicesImpl.configure(JdbcServicesImpl.java:89)
at org.hibernate.service.internal.StandardServiceRegistryImpl.configureService(StandardServiceRegistryImpl.java:75)
at org.hibernate.service.internal.AbstractServiceRegistryImpl.initializeService(AbstractServiceRegistryImpl.java:159)
at org.hibernate.service.internal.AbstractServiceRegistryImpl.getService(AbstractServiceRegistryImpl.java:131)
at org.hibernate.cfg.Configuration.buildTypeRegistrations(Configuration.java:1797)
at org.hibernate.cfg.Configuration.buildSessionFactory(Configuration.java:1755)
at org.hibernate.ejb.EntityManagerFactoryImpl.<init>(EntityManagerFactoryImpl.java:96)
at org.hibernate.ejb.Ejb3Configuration.buildEntityManagerFactory(Ejb3Configuration.java:914)
at org.hibernate.ejb.Ejb3Configuration.buildEntityManagerFactory(Ejb3Configuration.java:899)
at org.hibernate.ejb.HibernatePersistence.createEntityManagerFactory(HibernatePersistence.java:59)
at javax.persistence.Persistence.createEntityManagerFactory(Persistence.java:63)
at com.cloudera.cmon.FhDatabaseManager.initialize(FhDatabaseManager.java:57)
at com.cloudera.cmon.firehose.Main.main(Main.java:484)
Caused by: java.util.ConcurrentModificationException
at java.util.Hashtable$Enumerator.next(Hashtable.java:1378)
at com.mchange.v2.cfg.BasicMultiPropertiesConfig.extractPrefixMapFromRsrcPathMap(BasicMultiPropertiesConfig.java:236)
at com.mchange.v2.cfg.BasicMultiPropertiesConfig.<init>(BasicMultiPropertiesConfig.java:129)
at com.mchange.v2.cfg.BasicMultiPropertiesConfig.<init>(BasicMultiPropertiesConfig.java:39)
at com.mchange.v2.cfg.MultiPropertiesConfig.readVmConfig(MultiPropertiesConfig.java:102)
at com.mchange.v2.cfg.MultiPropertiesConfig.readVmConfig(MultiPropertiesConfig.java:74)
at com.mchange.v2.log.MLog.<clinit>(MLog.java:48)
... 19 more

[#2 · Accepted solution]

@Krish216,

As mentioned, there is no reason to have Activity Monitor running in your cluster if you are not using MapReduce1.

Rather than troubleshoot the issue, unless you have legacy MR1 (deprecated in C5 and removed in C6) I would vote for just removing Activity Monitor from your deployment.

To remove Activity Monitor:

1. in Cloudera Manager, click on Clusters --> Cloudera Management Service --> Instances
2. Check the box to the left of Activity Monitor.
3. Click the Actions for Selected button and choose Delete.

If you are still using Mapreduce1 in your cluster and need Activity Monitor, please show us the errors or log information you are seeing relative to the hue issue.

[#3]

@Krish216,

Activity Monitor is only useful for MapReduce1.  It is not used if you are using YARN for your jobs.

So, rather than troubleshoot this, unless you are using MR1, you can safely delete the Activity Monitor instance.  It is not used for anything if you are using YARN/MR2 only.

If you are using MR1 after all, is that the complete stack or is there possible more to the logs.

Ben

[#4]

@bgooley

Thank you for the response. Activity monitor logs shows there are a token issue with hue user, which is not the correct thing as Generate credentials was workign fine. This issue is coming after spark being added to the existing , even though situation is completed unreal .. 

if we remove activity monitor and install would help ?

[#5 · Accepted solution]

@Krish216,

As mentioned, there is no reason to have Activity Monitor running in your cluster if you are not using MapReduce1.

Rather than troubleshoot the issue, unless you have legacy MR1 (deprecated in C5 and removed in C6) I would vote for just removing Activity Monitor from your deployment.

To remove Activity Monitor:

1. in Cloudera Manager, click on Clusters --> Cloudera Management Service --> Instances
2. Check the box to the left of Activity Monitor.
3. Click the Actions for Selected button and choose Delete.

If you are still using Mapreduce1 in your cluster and need Activity Monitor, please show us the errors or log information you are seeing relative to the hue issue.

[#6 · Accepted solution]

@bgooley

Yes, uninstalling and re-installing worked. I do agree with the point of usage. 

Thanks for the help.