Expand my Community achievements bar.

Guidelines for the Responsible Use of Generative AI in the Experience Cloud Community.
SOLVED

AEM Splunk

Avatar

Level 7

Hi all,

 

I watched the post Using Splunk for Log Search & Monitoring on AEM as a Cloud Service at https://blog.arborydigital.com/splunk-setup-aem-cloud-service-aemaacs.

 

What are we exactly monitoring here?

I see we are collecting the Application logs.

 

So, we are monitoring aemaacs? How about analytics, target, etc?

 

What about the infra logs?

Or Splunk is limited to software application logs only and infrastructure logs are out of scope?

 

I remember we use different products for monitoring the infra to know uptime of servers, alerting admins in case of server failures, etc. Ex: LogicMonitor.

Are these products of different class?

 

I highly appreciate all your replies.

 

Thanks,

RK.

1 Accepted Solution

Avatar

Correct answer by
Level 4

Hi @nsvsrk,

feeding your AEMaaCS logs into Splunk is just the first step, it just gives you access to the logs via Splunk and the usage of Splunk query language. For monitoring purposes, I suggest setting up dashboards and email alerts in Splunk based on the data in logs. Here are a few things I would usually monitor:

  • Number of HTTP response codes on AEM Publisher and Dispatcher
  • Top X paths with the most requests on AEM
  • Top X slowest response times from AEM
  • Top X errors appearing in the Dispatcher logs
  • Top X errors appearing in the AEM logs

On AEMaaCS, Adobe is responsible for the infrastructure monitoring and you can get read access to the Adobe-managed New Relic. There, you can find different monitors and reports providing you insight into:

  • Different JVMs running on different pods (response time, throughput, error rate, CPU and memory utilization)
  • Most time-consuming web transactions and information like average response times and throughput
    • Detailed transaction traces are available for slow transactions
  • Most time-consuming database operations/queries
  • Response time, throughput, and error rates for calls towards external services
  • And more...

Please note that you cannot create alerts in the Adobe-managed New Relic, you only get read access for APM features.

 

Hope this helps,

Daniel

View solution in original post

2 Replies

Avatar

Community Advisor

Hi @nsvsrk ,

 

Q: "What are we exactly monitoring here?"
A: We are monitoring log files based on the environment's log level (trace, warn, info, errors, debug, etc.).

 

Q: "We are monitoring AEMaaCS? How about Analytics, Target, etc.?"
A: Yes, we are just monitoring AEMaaCS here. Analytics and Target logs can also be set up separately to monitor their own logs. If you want to see Analytics reports on Splunk, I believe that’s possible too. Check these links:
Using Splunk to Monitor, Aggregate, and Analyze Adobe I/O Events
DataVirtuality Adobe Analytics Connector for Splunk

 

Q: "What about the infra logs?"
A: You can set those up as well. As far as I know, all setups are done separately. AEMaaCS only monitors application-level logs (AEM only).

- Sravan

 

Avatar

Correct answer by
Level 4

Hi @nsvsrk,

feeding your AEMaaCS logs into Splunk is just the first step, it just gives you access to the logs via Splunk and the usage of Splunk query language. For monitoring purposes, I suggest setting up dashboards and email alerts in Splunk based on the data in logs. Here are a few things I would usually monitor:

  • Number of HTTP response codes on AEM Publisher and Dispatcher
  • Top X paths with the most requests on AEM
  • Top X slowest response times from AEM
  • Top X errors appearing in the Dispatcher logs
  • Top X errors appearing in the AEM logs

On AEMaaCS, Adobe is responsible for the infrastructure monitoring and you can get read access to the Adobe-managed New Relic. There, you can find different monitors and reports providing you insight into:

  • Different JVMs running on different pods (response time, throughput, error rate, CPU and memory utilization)
  • Most time-consuming web transactions and information like average response times and throughput
    • Detailed transaction traces are available for slow transactions
  • Most time-consuming database operations/queries
  • Response time, throughput, and error rates for calls towards external services
  • And more...

Please note that you cannot create alerts in the Adobe-managed New Relic, you only get read access for APM features.

 

Hope this helps,

Daniel