peer.service adjustment for sql propagation with DBM #3127
Conversation
Codecov Report
@@ Coverage Diff @@
## master #3127 +/- ##
=======================================
Coverage 98.16% 98.16%
=======================================
Files 1323 1323
Lines 75292 75318 +26
Branches 3476 3476
=======================================
+ Hits 73908 73934 +26
Misses 1384 1384
📣 We’re building smart automated test selection to slash your CI/CD build times. Learn more |
spec/datadog/tracing/contrib/sql_comment_propagation_examples.rb
Outdated
Show resolved
Hide resolved
| @@ -102,6 +103,63 @@ | |||
| end | |||
| end | |||
|
|
|||
| context 'when tracing is enabled with peer.service' do | |||
There was a problem hiding this comment.
The description does not match the scenario, it only need a span_op with a specific tag
There was a problem hiding this comment.
I disagree, the main thing here is to mock a situation when tracing is enabled with peer.service, ensuring that the peer.service value is getting set as the dddbs tag.
it only need a span_op with a specific tag
That's what the peer.service feature does. It adds a tag
There was a problem hiding this comment.
Sorry, I meant that the test could be simplified, for example,
We don't need
Datadog.configure do |c|
c.env = 'production'
c.service = "Traders' Joe"
c.version = '1.0.0'
end
Because, the method is stateless, only depends on the input.
There was a problem hiding this comment.
The description should be
context 'when given a span operation tagged with`peer.service`' do
This reverts commit 1c5860d.
This PR addresses the issue of DBM relying on
service_namefor their own service names prior topeer.service. Now since all services are flattened, the DBM map is inaccurate and a better tag to create a more accurate map is to usepeer.service