Does federate support request compression?

[bgeesaman]

What did you do?
Attempting to scrape 10 targets of 40MB per /federate?metrics target every 30 seconds

What did you expect to see?
The requests to use compression

What did you see instead? Under which circumstances?
The requests don't appear to use compression based on the nginx logs (each cluster has a pod of nginx in front of promethus-core for logging)

Environment
Centos 7 running a standalone prometheus image prom/prometheus:latest scraping from the same version of prometheus inside multiple kube 1.5.3 clusters

• System information:

Linux 3.10.0 el7 x86_64

• Prometheus version:

Centos 7 running a standalone prometheus image prom/prometheus:latest

I am publishing each cluster's prometheus service behind an nginx container behind an ELB to allow for the federated prometheus system to scrape them. When I use curl without the --compressed flag, I see the /federate?metrics bring down a ~40MB file. With that flag, about 5MB. Searched issues/repo for "gzip" and "zip" but found no hard data to support this. Given that every 30s I'm scraping 400+MB but could be scraping ~50MB, it seems like a nice option.

[bgeesaman]

Nevermind. I don't know why I didn't see the code section with that enabled before. Apologies. Appears my nginx configuration (providing basic auth) wasn't "proxying" the compression.

[lock]

This thread has been automatically locked since there has not been any recent activity after it was closed. Please open a new issue for related bugs.