Hi Andy,

Thanks again for the help!  This did the trick.

Regards,
Paul

On Fri, Jun 1, 2018 at 9:54 PM, Andy LoPresto <[email protected]> wrote:

> Hi Paul,
>
> Thanks for the further information. You should be able to use either of
> the two options:
>
> <logger name=“org.apache.nifi.controller.cluster” level=“WARN”/>
>
> <logger name=“org.apache.nifi.controller.cluster.ClusterProtocolHeartbeater”
> level=“WARN”/>
>
> The first is the package that contains the class that is actually logging
> the heartbeat sent message, and the only other log output in that package
> is a ZookeeperClientConfig WARN when an attempt to access time period fails
> because the config value is not in a valid format. The three log events in
> the actual heartbeat class (2 INFO, 1 DEBUG) are the ones you want to
> suppress, so telling logback that the entire class should be at WARN should
> take care of this.
>
> Please note the name value is the package or class you want to affect, so
> in this case, the different package structures for the coordinator and the
> “clients” of the framework are why your earlier attempts did not yield the
> desired outcome. Please let me know if this doesn’t solve your problem.
>
> Andy LoPresto
> [email protected]
> *[email protected] <[email protected]>*
> PGP Fingerprint: 70EC B3E5 98A6 5A3F D3C4  BACE 3C6E F65B 2F7D EF69
>
> On Jun 1, 2018, at 6:09 PM, Paul Riddle <[email protected]> wrote:
>
> Hi Andy,
>
> Thanks for the fast response.  I am trying to supress the standard cluster
> heartbeat messages:
>
> 2018-06-02 01:03:18,280 INFO [Clustering Tasks Thread-1] 
> o.a.n.c.c.ClusterProtocolHeartbeater
> Heartbeat created at 2018-06-02 01:03:18,147 and sent to <node>:<pot> at
> 2018-06-02 01:03:18,280; send took 133 millis.
>
> I tried adding  <logger name=“org.apache.nifi.cluster.coordination.heartbeat"
> level=“WARN”/> but the messages persist.  I also tried <logger
> name="org.apache.nifi.cluster" level="WARN"/> and that suppresses the
> cluster manager INFO messages but not the heartbeat messages.
> Setting <logger name="org.apache.nifi" level="WARN"/> does the trick but I
> wanted to go a little less all inclusive than that.
>
> Regards,
> Paul
>
> On Fri, Jun 1, 2018 at 6:53 PM, Andy LoPresto <[email protected]>
> wrote:
>
>> Hi Paul,
>>
>> Can you provide a specific message you would like to be suppressed? Given
>> a quick examination, all the logger messages in
>> ClusterProtocolHeartbeatMonitor.java are at DEBUG level, so
>> setting <logger name=“org.apache.nifi.cluster.coordination.heartbeat"
>> level=“INFO”/> should be sufficient. If this doesn’t solve your problem,
>> please let us know which messages specifically (you can redact sensitive
>> information like addresses, but provide enough of the message so we can
>> locate it in the code) you would like to be suppressed. Thanks.
>>
>> Andy LoPresto
>> [email protected]
>> *[email protected] <[email protected]>*
>> PGP Fingerprint: 70EC B3E5 98A6 5A3F D3C4  BACE 3C6E F65B 2F7D EF69
>>
>> On Jun 1, 2018, at 12:54 PM, Paul Riddle <[email protected]>
>> wrote:
>>
>> Hello NiFi Team,
>>
>> Is there a setting for controlling the log level of heartbeat messages in
>> logback.xml on NiFi 1.6.0?  While adjusting  <logger
>> name="org.apache.nifi.cluster" level="WARN"/> , only the cluster
>> coordinator messages are affected.  I'd like to be able to silence the
>> cluster node heartbeat messages in my app logs without setting the overall
>> logging to WARN.
>>
>> Any help you can provide is greatly appreciated.
>>
>> Regards,
>> Paul
>>
>>
>>
>
>

Reply via email to