topic Fullstack Monitoring on Hybrid Environment in Open Q&A

Fullstack Monitoring on Hybrid Environment

agylpradipta — Mon, 20 Mar 2023 07:56:06 GMT

Hello all,

I'm currently having environment with topology like this:

Front-end App server (EC2) -> AWS Network Load Balancer (NLB) -> Backend Server (VMWare On-Premise)

So I used AWS NLB provided rule to forward my app traffic to on-premise server.

After I deployed OneAgent in Front-end server and Back-end server I couldn't see connection between them.

Is there any rule or header that I could apply on my AWS Load Balancer so the service flow can be detected?
Or actually by enabling fullstack monitoring on both server are enough?

Thank you

Re: Fullstack Monitoring on Hybrid Environment

AntonioSousa — Fri, 17 Mar 2023 20:47:55 GMT

@agylpradipta,

Are you using Infrastructure mode? If that is the case, it is normal that you don't see requests between them.

Re: Fullstack Monitoring on Hybrid Environment

agylpradipta — Sat, 18 Mar 2023 03:11:50 GMT

Yap after I re-check, server on my on-premise environment still on Infra Only mode. Will try to enable the full-stack monitoring mode.

So actually there is no problem regarding the AWS Load Balancer?

Re: Fullstack Monitoring on Hybrid Environment

Romanenkov_Al3x — Sat, 18 Mar 2023 21:19:34 GMT

There are no problems with any kind of proxies\balancers\web server - balancers if they are not block headers.

If you have some balancers \ proxy which block or delete \ modify header x-dynatrace - there will no correct serviceflow. From Documentation you need only header x-dynatrace, other headers related to RUM\UEM.

I agree with Antonio - problem is related that agents are not in full stack mode. Front web-servers not instrumented. Change monitoring mode, restart frontend process and you will see correct service flow.

Regards,

Romanenkov Alex