[ 
https://issues.apache.org/jira/browse/AMBARI-7070?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14116698#comment-14116698
 ] 

Clint Shumack commented on AMBARI-7070:
---------------------------------------

I tested it again with 48 cores (it works) and then with 49 cores (it fails) 
and I found these differences:

ambari-server.log on 48 cores
{quote}
....
INFO [main] AbstractConnector:338 - Started [email protected]:8080
....
{quote}

ambari-server.log on 49 cores
{quote}
....
WARN [main] AbstractConnector:335 - insufficient threads configured for 
[email protected]:8080
INFO [main] AbstractConnector:338 - Started [email protected]:8080
....
{quote}

ambari-agent.log on 48 cores
{quote}
....
Controller.py:271 - Response from server = OK
INFO PythonExecutor.py:78 - Running command ['/usr/bin/python2.6',
 
u'/var/lib/ambari-agent/cache/stacks/HDP/2.1/services/FALCON/package/scripts/falcon_server.py',
 'STATUS',
 '/var/lib/ambari-agent/data/status_command.json',
 u'/var/lib/ambari-agent/cache/stacks/HDP/2.1/services/FALCON/package',
 '/var/lib/ambari-agent/data/structured-out-status.json',
INFO PythonExecutor.py:112 - Result: {'structuredOut': {}, 'stdout': '', 
'stderr': '', 'exitcode': 0}
 ....
_(log file continues like this for a while)_
....
INFO PythonExecutor.py:112 - Result: {'structuredOut': {}, 'stdout': '', 
'stderr': '', 'exitcode': 0}
_(end of log)_
{quote}

ambari-agent.log on 49 cores
{quote}
....
INFO Controller.py:271 - Response from server = OK
INFO Heartbeat.py:76 - Sending heartbeat with response id: 0 and timestamp: 
1409360953048. Command(s) in progress: False. Components mapped: False
_(end of log)_
{quote}

I hope this is helpful.

> Cannot view Ambari web interface with more than 48 CPU cores.
> -------------------------------------------------------------
>
>                 Key: AMBARI-7070
>                 URL: https://issues.apache.org/jira/browse/AMBARI-7070
>             Project: Ambari
>          Issue Type: Bug
>    Affects Versions: 1.6.1
>         Environment: CentOS 6.5 Kernel 2.6.32-431.23.3.el6.x86_64, qemu-kvm 
> virtual machine on Dell PowerEdge C6145 Opteron 6376
>            Reporter: Clint Shumack
>              Labels: newbie
>
> I am running Ambari and the full hadoop stack in a virtual machine on my 
> 64-core server for evaluation before I deploy it to the base OS. When I set 
> it up with a 4-core virtual machine it worked perfectly, however after 
> increasing the core count to 64 I could no longer log in to Ambari. The 
> browser seems to be attempting to load the page but receiving no response. I 
> can't see anything unusual in the logs. I found that it works with up to 48 
> cores, but no more. 
> I have a second identical virtual machine running just the hadoop client 
> software with 64 cores without problems, so this looks like a bug in the 
> Ambari web interface. Has anyone else experienced this?



--
This message was sent by Atlassian JIRA
(v6.2#6252)

Reply via email to