hadoop-hdfs-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Matt Narrell <matt.narr...@gmail.com>
Subject Re: YARN HA Active ResourceManager failover when machine is stopped
Date Fri, 24 Apr 2015 17:39:23 GMT
Ah, yes.  Ok please see below:

Scenario one: Stop the Active ResourceManager process (leaving the VM running)
Active ResourceManager:  https://gist.github.com/mnarrell/157c8e1b82d40541cd88 <https://gist.github.com/mnarrell/157c8e1b82d40541cd88>
Standby ResourceManager: https://gist.github.com/mnarrell/b6ad01d2f4b900b42e6d <https://gist.github.com/mnarrell/b6ad01d2f4b900b42e6d>

Scenario two: Shutdown the VM ($ shutdown -h now)
Active ResourceManager:  https://gist.github.com/mnarrell/95b35cc8be0ed817cf1b <https://gist.github.com/mnarrell/95b35cc8be0ed817cf1b>
Standby ResourceManager:  https://gist.github.com/mnarrell/68a778e0d0d213e1b2cf <https://gist.github.com/mnarrell/68a778e0d0d213e1b2cf>

Here is the yarn-site.xml
https://gist.github.com/mnarrell/115a3eff03bbef947a57 <https://gist.github.com/mnarrell/115a3eff03bbef947a57>

We have some suspicion that this could be related to fencing?  We speculate that when the
machine is shutdown, the NodeManagers do not see the NoRouteToHost exception as a failover
situation?  We have a pretty vanilla configuration of YARN, mostly Ambari defaults, and have
compared our configuration to the YARN ResourceManager HA documentation from Apache and Hortonworks.


> On Apr 24, 2015, at 1:50 AM, Drake민영근 <drake.min@nexr.com> wrote:
> Hi, Matt
> The second log file looks like node manager's log, not the standby resource manager.
> Thanks.
> Drake 민영근 Ph.D
> kt NexR
> On Fri, Apr 24, 2015 at 11:39 AM, Matt Narrell <matt.narrell@gmail.com <mailto:matt.narrell@gmail.com>>
> Active ResourceManager:  http://pastebin.com/hE0ppmnb <http://pastebin.com/hE0ppmnb>
> Standby ResourceManager: http://pastebin.com/DB8VjHqA <http://pastebin.com/DB8VjHqA>
> Oppressively chatty and not much valuable info contained therein.
>> On Apr 23, 2015, at 4:25 PM, Vinod Kumar Vavilapalli <vinodkv@hortonworks.com
<mailto:vinodkv@hortonworks.com>> wrote:
>> I have run into this offline with someone else too but couldn't root-cause it.
>> Will you be able to share your active/standby ResourceManager logs via pastebin or
>> +Vinod
>> On Apr 23, 2015, at 9:41 AM, Matt Narrell <matt.narrell@gmail.com <mailto:matt.narrell@gmail.com>>
>>> I’m using Hadoop 2.6.0 from HDP 2.2.4 installed via Ambari 2.0
>>> I’m testing the YARN HA ResourceManager failover. If I STOP the active ResourceManager
(shut the machine off), the standby ResourceManager is elected to active, but the NodeManagers
do not register themselves with the newly elected active ResourceManager. If I restart the
machine (but DO NOT resume the YARN services) the NodeManagers register with the newly elected
ResourceManager and my jobs resume. I assume I have some bad configuration, as this produces
a SPOF, and is not HA in the sense I’m expecting.
>>> Thanks,
>>> mn

View raw message