enable by default: server_address and server_port attributes

[mariomac]

The above attributes should not increase cardinality too much, and letting them by default would facilitate the landing on Beyla+Asserts.

The user still hs the possibility to disable them by means of the attributes selection YAML section.

[codecov-commenter]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 80.67%. Comparing base (84af0d1) to head (794df40).

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #989      +/-   ##
==========================================
+ Coverage   80.59%   80.67%   +0.08%     
==========================================
  Files         134      134              
  Lines       10688    10688              
==========================================
+ Hits         8614     8623       +9     
+ Misses       1565     1560       -5     
+ Partials      509      505       -4     

Flag	Coverage Δ
integration-test	56.01% <100.00%> (+0.05%)	⬆️
k8s-integration-test	59.35% <100.00%> (+0.09%)	⬆️
oats-test	36.35% <100.00%> (ø)
unittests	50.63% <100.00%> (+0.07%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[grcevski]

LGTM!

[ludov04]

I might be missing something but for me these 2 attribute ended up in infinite cardianlity, as ServerAddr end up being the ip of any client in the world that connect to my server, how do I fix this?

[grcevski]

Hi @ludov04, oh interesting, so I wonder if we misinterpret what's a client and what's a server here then. You can disable the attributes by specifying that you don't want them in Beyla's configuration, like this:

attributes:
  select:
    http_server_request_duration_seconds_count:
      exclude: ["server_address"]

http_server_request_duration_seconds_count is just an example of a series, you can specify the series you are having trouble with and define more than one.

I'd be curious to know which series experienced this problem, is it related to the network metrics or application metrics?