ambari-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Jonathan Hurley" <jhur...@hortonworks.com>
Subject Re: Review Request 27688: Alerts: Convert Bigtop Stack Nagios Alerts
Date Thu, 06 Nov 2014 23:49:56 GMT


> On Nov. 6, 2014, 5:02 p.m., Nate Cole wrote:
> > ambari-server/src/main/java/org/apache/ambari/server/api/services/AmbariMetaInfo.java,
lines 1278-1282
> > <https://reviews.apache.org/r/27688/diff/1/?file=751770#file751770line1278>
> >
> >     Isn't it the case that this could happen?

The continue actually doesn't do anything beneficial aside from skipping some enhanced-for
loops. What was actually happening is that the agent alerts were being skipped b/c of the
continue. I tested this new logic in a cluster with 0 alert definitions and 1 agent alert
definition.


- Jonathan


-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/27688/#review60257
-----------------------------------------------------------


On Nov. 6, 2014, 12:48 p.m., Jonathan Hurley wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/27688/
> -----------------------------------------------------------
> 
> (Updated Nov. 6, 2014, 12:48 p.m.)
> 
> 
> Review request for Ambari, Newton Alex, Nate Cole, and Tom Beerbower.
> 
> 
> Bugs: AMBARI-8084
>     https://issues.apache.org/jira/browse/AMBARI-8084
> 
> 
> Repository: ambari
> 
> 
> Description
> -------
> 
> Update the BT stack to start using the new Ambari alerting system.
> 
> With Ambari 2.0, a new alerts framework has been implemented with the intent to replace
Nagios. If you would like to learn more about this feature, the design documents can be found
in the Jira epic at https://issues.apache.org/jira/browse/AMBARI-6354. 
> 
> The work has been completed to move the HDP 2.x stack over from Nagios to the internal
alerts framework. However, HDP 1.x and Bigtop have not yet been migrated. I would like to
propose that we move all stacks off of Nagios so that we can remove that dependency from Ambari
completely. 
> 
> This also means that upon upgrading from a prior Nagios-enabled release of Ambari, Nagios
will be removed as a dependent service. It will no longer be a part of the managed cluster.
> 
> This will allow us to also remove any Nagios-specific areas of the codebase that would
otherwise need to remain for legacy purposes. It would also relieve the requirement on clients
to have intimate knowledge of which stack/version can be Nagios-enabled.
> 
> 
> Diffs
> -----
> 
>   ambari-server/src/main/java/org/apache/ambari/server/api/services/AmbariMetaInfo.java
3d67fe9 
>   ambari-server/src/main/java/org/apache/ambari/server/controller/internal/AlertDefinitionResourceProvider.java
a8a7f67 
>   ambari-server/src/main/java/org/apache/ambari/server/controller/internal/AlertGroupResourceProvider.java
50820a7 
>   ambari-server/src/main/resources/stacks/BIGTOP/0.8/services/FLUME/alerts.json PRE-CREATION

>   ambari-server/src/main/resources/stacks/BIGTOP/0.8/services/FLUME/package/files/alert_flume_agent_status.py
PRE-CREATION 
>   ambari-server/src/main/resources/stacks/BIGTOP/0.8/services/GANGLIA/alerts.json PRE-CREATION

>   ambari-server/src/main/resources/stacks/BIGTOP/0.8/services/HBASE/alerts.json PRE-CREATION

>   ambari-server/src/main/resources/stacks/BIGTOP/0.8/services/HDFS/alerts.json PRE-CREATION

>   ambari-server/src/main/resources/stacks/BIGTOP/0.8/services/HDFS/package/files/alert_checkpoint_time.py
PRE-CREATION 
>   ambari-server/src/main/resources/stacks/BIGTOP/0.8/services/HDFS/package/files/alert_ha_namenode_health.py
PRE-CREATION 
>   ambari-server/src/main/resources/stacks/BIGTOP/0.8/services/HIVE/alerts.json PRE-CREATION

>   ambari-server/src/main/resources/stacks/BIGTOP/0.8/services/HIVE/package/files/alert_hive_thrift_port.py
PRE-CREATION 
>   ambari-server/src/main/resources/stacks/BIGTOP/0.8/services/OOZIE/alerts.json PRE-CREATION

>   ambari-server/src/main/resources/stacks/BIGTOP/0.8/services/OOZIE/package/files/alert_check_oozie_server.py
PRE-CREATION 
>   ambari-server/src/main/resources/stacks/BIGTOP/0.8/services/WEBHCAT/alerts.json PRE-CREATION

>   ambari-server/src/main/resources/stacks/BIGTOP/0.8/services/WEBHCAT/package/files/alert_webhcat_server.py
PRE-CREATION 
>   ambari-server/src/main/resources/stacks/BIGTOP/0.8/services/YARN/alerts.json PRE-CREATION

>   ambari-server/src/main/resources/stacks/BIGTOP/0.8/services/YARN/package/files/alert_nodemanager_health.py
PRE-CREATION 
>   ambari-server/src/main/resources/stacks/BIGTOP/0.8/services/ZOOKEEPER/alerts.json PRE-CREATION

>   ambari-server/src/test/java/org/apache/ambari/server/controller/internal/AlertDefinitionResourceProviderTest.java
d4f5fb4 
>   ambari-server/src/test/java/org/apache/ambari/server/controller/internal/AlertGroupResourceProviderTest.java
9aad9a1 
> 
> Diff: https://reviews.apache.org/r/27688/diff/
> 
> 
> Testing
> -------
> 
> Added test cases found for issues during BT stack alert implementation. Installed a BT
0.8 cluster with all services, verified alerts are working correctly.
> 
> 
> Thanks,
> 
> Jonathan Hurley
> 
>


Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message