coordinate successful and failing fetches #6699
Conversation
make sure that the servers all receive their requests before confirming that a successful response from one server cancels the shared context in the fetch code.
💚 Build Succeeded
Expand to view the summary
Build stats
Test stats 🧪
🤖 GitHub commentsTo re-run your PR in the CI, just comment with:
|
| resc <- response{consumer, err} | ||
| }() | ||
| // Make sure every server has received a request | ||
| wg.Wait() |
There was a problem hiding this comment.
Can't we remove this wait group? It seems to me that the channels are already making the test wait until we've received all the signals. What is this wait group adding on top of that?
There was a problem hiding this comment.
The waitgroup is making sure that each handler has received a request before we attempt to read from any of the channels. Without that, we could potentially read from errc and successc without the third handler ever being invoked.
There was a problem hiding this comment.
got it, feel free to merge, thanks for the explanation
make sure that the servers all receive their requests before confirming that a successful response from one server cancels the shared context in the fetch code. (cherry picked from commit 039bffa)
make sure that the servers all receive their requests before confirming that a successful response from one server cancels the shared context in the fetch code. (cherry picked from commit 039bffa) Co-authored-by: stuart nelson <[email protected]>
Motivation/summary
make sure that the servers all receive their
requests before confirming that a successful
response from one server cancels the shared
context in the fetch code.
How to test these changes
non-functional change
Related issues
closes #5836