@hello4usharath,   Since the stderr.log shows that the supervisor already tried starting Zookeeper and the agent/supervisor is written in python, the exception must be coming from Zookeeper itself.  I'd recommend checking the zookeeper logs on the host where it failed to start.  Also you might check "stdout.log" for the process as it sometimes contains output including role exceptions.   -Ben ... View more

@tadeuuuuu and @shiv003:   In @shiv003's problem we see that the agent cannot heartbeat to Cloudera Manager due to " [Errno 111] Connection refused".  This means that the agent could not reach Cloudera Manager's host on port 7182, likely since Cloudera Manager is not running on that host.   @tadeuuuuu, while you are seeing the same error about failed heartbeating, the cause could be different.  Please share the stack trace that accompanies the heartbeat failure so we can review. ... View more

@kchaitanya, Indeed, if agent_tls is enabled even without web_tls and there is a problem with the certificate or trust store, then you would need to turn off agent_tls.   A big thanks for sharing that solution!   ... View more

@kchaitanya,   If that did not help, then it is likely there is another problem. Please start a new thread and outline what you are trying to do and what is happening. Also, make sure to review your /var/log/cloudera-scm-server/cloudera-scm-server.log for clues about what problem is occurring.   -Ben ... View more

@Vitali1,     If you have "Dump Heap When Out of Memory" enabled, a heap dump will be generated if the process runs out of heap. The files generated can be deleted as they are only for debugging purposes.  Navigator does not require them.   That said, you can if you are running out of  heap, increasing the heap size is likely a good first try to resolve that issue.   -Ben ... View more

Hello @ManuelAllh,   Sorry to hear you hit some problems installing and running. The stderr.log and stdout.log do not show anything unsual, so we should take a look at the agent log on the host where you are trying to start the services.  The log is /var/log/cloudera-scm-agent/cloudera-scm-agent.log   You might tail -f that file while trying to start the services that are not starting. If you don't see anything obvious that leads you to a solution, share the logging with us.   Also, it is possible there is a failure that is captured in the role log.  In your case, it appears the failure is during the start of the Host Monitor.  I recommend checking the Host Monitor log to see if there is anything useful there:   /var/log/cloudera-scm-firehose/mgmt-cmf-mgmt-HOSTMONITOR-vengland.log.out   The information in the agent and role log should help point us toward the cause.   -Ben   ... View more

Hi @hadoopuser2000,   Based on your screenshots, it appears you are attempting to install on an OS that is not yet supported (Xenial)   Currently, only Trusty and Precise Ubuntu are supported as noted here: https://www.cloudera.com/documentation/enterprise/release-notes/topics/rn_consolidated_pcm.html#os510   Trusty 14.04 - Long-Term Support (LTS) Precise 12.04 - Long-Term Support (LTS)   When there is an attempt to download the xenial rpms from the Cloudera archive, that fails since there is no such directory or rpms.   Ubuntu 16.04 support is currently targeted for CM/CDH 5.11   -Ben   Ben ... View more

Hi @maragon,   You mentioned that you tried to use psql to connect; what command did you use and what was the result?   It would be unusual that you can connect but Cloudera Manager cannot.  Nonetheless, you may want to try restarting the embedded database:   service cloudera-scm-server-db restart   Then try:   service cloudera-scm-server start     ... View more

Hello @mjarnold,   Thank you for pointing that out.  I have added this post to internal Cloudera Jira DOCS-1862 where we are discussing documentation.   Regards,   Ben ... View more