All Apps and Add-ons

Splunk AWS Add-On (Cloudwatch) takes excessive amount of time to poll AWS for new EC2 Instances.

specora
Explorer

Description of issue:
When a new EC2 instance is created in AWS, Splunk takes an excessive amount of time (upwards of an hour) to recognize the newly created instance and begin collecting data. In addition once Splunk finally recognizes the new EC2 instances, data collection is not consistent and does adhere to the specified interval in the cloud watch settings. One thing to note, a restart of the Splunk service allows data collection to commence.

Components: Splunk Cloud, Splunk App for AWS, Splunk Add On for AWS

Versions:
- Splunk 6.4.1.2
- AWS App 4.2.1

Enabled Data Sources:
- Cloud Watch

Cloud Watch Input Settings:

- EC2 enabled (all other services disabled)
- Granularity = 5 minutes
- Interval = Granularity x 1 = 5 minutes

Questions:
1. Can the frequency in which Splunk queries for newly created/terminated EC2 instances be increased?
2. Are there any known bugs related to Cloud Watch data collection, specifically pertaining to CPU metric

hnelsonit
Explorer

Just in case anyone was wondering, splunk finally added this feature in 4.6 of the aws addon.

0 Karma

hnelsonit
Explorer

I have also noticed this issue and for the most part I am always missing cw metrics from new EC2 metrics. I have opened up several cases with splunk about this and every time they seem to indicate that they don't have an ongoing discovery method after the inputs are initially setup. As you said, restart splunk "fixes" the issue. But this is a terrible method for cloud environment monitoring. I was recommended to refresh the inputs with a rest call that I have set to run on a cron schedule every 6 hours. I space it that far because every time I do this, it takes about half an hour for all CW metrics to fully collect again which results in a gap of the metric data.

Here is the REST call: curl -k -u userID:pwd https://localhost:8089/servicesNS/nobody/Splunk_TA_aws/data/inputs/aws_cloudwatch//enable?output_mod... -X POST

Really looking for splunk to step up and catch up to their competitors in this regard. Auto discovery of new metric dimensions should have been a day 1 feature.

0 Karma
Get Updates on the Splunk Community!

.conf24 | Day 0

Hello Splunk Community! My name is Chris, and I'm based in Canberra, Australia's capital, and I travelled for ...

Enhance Security Visibility with Splunk Enterprise Security 7.1 through Threat ...

 (view in My Videos)Struggling with alert fatigue, lack of context, and prioritization around security ...

Troubleshooting the OpenTelemetry Collector

  In this tech talk, you’ll learn how to troubleshoot the OpenTelemetry collector - from checking the ...