Navneet Kaur created NIFI-11847:
-----------------------------------
Summary: Processor: ActivityMonitor: Time monitoring is incorrect
Key: NIFI-11847
URL: https://issues.apache.org/jira/browse/NIFI-11847
Project: Apache NiFi
Issue Type: Improvement
Components: Core Framework
Affects Versions: 1.20.0
Reporter: Navneet Kaur
I have been using the ActivityMonitor processor to monitor the flowfiles for a
particular time. An Alert is generated if the time mentioned is exceeded.
I have observed that the time monitored in the background triggers the alert
incorrectly.
Example-1: (AWS)
1. I have an S3 bucket (A), the file is written to bucket A every minute.
Bucket A sends the event notification via SQS, and NiFi Processes this
notification, fetch the file and copies it to S3 bucket B.
In AWS Console, the file is copied to bucket B every minute. Activity monitor
is used for GetSQS, and for 5 minutes of monitoring time, at least one alert is
received for lacking activity!
Example-2: (AWS)
1. We have put monitoring on NiFi for 2160 minutes, but it's observed that at
least one file is sent every 1440 minutes, we still get the alerts via
ActivityMonitor processor for lacking activity.
Processor settings:
Threshold Duration. =. 2160 minContinually Send Message s
false
Inactivity Message
index=rule_engine sourcetype=rulemonitorlog; Host=${hostname(false)}; Lacking
activity at Splunk Processor as of time: ${now():format('yyyy/MM/dd
HH:mm:ss')}; Splunk Processor has not send data for
${inactivityDurationMillis:toNumber():divide(60000)} minutes
Activity Restored Message
index=rule_engine sourcetype=rulemonitorlog; Host=${hostname(false)}; Activity
restored at Splunk Processor at time: ${now():format('yyyy/MM/dd HH:mm:ss')};
Splunk Processor has started sending data after being inactive for
${inactivityDurationMillis:toNumber():divide(60000)} minutes
Copy Attributes
true
Monitoring Scope
cluster
Reporting Node
primary
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
