> On June 19, 2016, 1:02 a.m., Don Bosco Durai wrote:
> > ambari-logsearch/ambari-logsearch-logfeeder/src/main/java/org/apache/ambari/logfeeder/output/OutputSolr.java,
> >  line 146
> > <https://reviews.apache.org/r/48793/diff/2/?file=1423331#file1423331line146>
> >
> >     I don't not sure whether this is the right place to put this check? 
> > Output is generally only for writing. We should probably move the code to 
> > FetchConfigFromSolr and call it from InputMgr.init(). And also, move     
> > LogfeederScheduler.INSTANCE.start(); from LogFeeder.java before 
> > inputMgr.init() is called.
> >     
> >     More important, it should be only called if Solr Zookeeper/URL and 
> > config collection is provided. In offline, mode, these might not be set.
> 
> Miklos Gergely wrote:
>     This may not be the best place, but the whole filter applying is a total 
> mess, it will be refactored soon. Till then It can be here.

It can't go into FetchConfigFromSolr because it is a deamon thread, and this 
check must hold all the uploading to Solr.


- Miklos


-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/48793/#review138424
-----------------------------------------------------------


On June 19, 2016, 10:39 p.m., Miklos Gergely wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/48793/
> -----------------------------------------------------------
> 
> (Updated June 19, 2016, 10:39 p.m.)
> 
> 
> Review request for Ambari, Don Bosco Durai, Dharmesh Makwana, Oliver Szabo, 
> Robert Nettleton, and Sumit Mohanty.
> 
> 
> Bugs: Ambari-17277
>     https://issues.apache.org/jira/browse/Ambari-17277
> 
> 
> Repository: ambari
> 
> 
> Description
> -------
> 
> After Log Search is started it sends the filters to the Solr which describes 
> what log levels should be persisted. In the meantime Logfeeders start to push 
> their data, which is not filtered. The result is misleading, the user may 
> find it add that some specific level logs are loaded, others are not.
> 
> 
> Diffs
> -----
> 
>   
> ambari-logsearch/ambari-logsearch-logfeeder/src/main/java/org/apache/ambari/logfeeder/output/OutputSolr.java
>  b14c273 
>   
> ambari-logsearch/ambari-logsearch-logfeeder/src/main/java/org/apache/ambari/logfeeder/util/SolrUtil.java
>  29feef7 
>   
> ambari-logsearch/ambari-logsearch-logfeeder/src/main/resources/logfeeder.properties
>  076c09c 
>   
> ambari-logsearch/ambari-logsearch-logfeeder/src/test/java/org/apache/ambari/logfeeder/output/OutputSolrTest.java
>  afbccca 
>   
> ambari-server/src/main/resources/common-services/LOGSEARCH/0.5.0/package/templates/logfeeder.properties.j2
>  6a52708 
> 
> Diff: https://reviews.apache.org/r/48793/diff/
> 
> 
> Testing
> -------
> 
> logfeeder:
> Tests run: 35, Failures: 0, Errors: 0, Skipped: 0
> 
> ambari-server
> Total run:1062
> Total errors:0
> Total failures:0
> 
> 
> Thanks,
> 
> Miklos Gergely
> 
>

Reply via email to