-
Notifications
You must be signed in to change notification settings - Fork 4.9k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Remove the HTTP2 and HTTP3 implementations on browser-wasm. #52420
Conversation
This allows for HPackEncoder and QPackEncoder to be trimmed in browser wasm. Contributes to dotnet#44534
Tagging subscribers to this area: @dotnet/ncl Issue DetailsThis allows for HPackEncoder and QPackEncoder to be trimmed in browser wasm. Contributes to #44534
|
Some of this code is shared with aspnetcore. Will the changes require a reaction in aspnetcore? |
I think the only change would be to include the new files in whatever .csproj uses them. For example: |
@dotnet/dnceng - can anyone tell what the problem was with the Linux x64 Debug leg? From the log:
|
Sure, I can talk to this. You've now taken the changes from dotnet/arcade#7310 , which means that when test results can't be reported to Azure Devops, the work item fails. The HTTP 503 would, prior to this change, have resulted in incomplete or non-reporting to AzDO , while leaving the work item's exit code as 0 (Since we now coerce both 1 (failed some tests) and 0 (passed all tests) to 0). Previously, we were suppressing this problem; when it happens, it enables a change that breaks a test, if it gets lucky while Azure DevOps is having a bad day, to "pass". This came because before, we'd exit with the same exit code XUnit did when tests failed, but once it was decided by folks that we would turn all the 1s to 0s, this became necessary to prevent ignoring failure. My preference would be to not coerce the exit codes, but there was some reason this particular behavior was needed for the developer workflow. This is, hopefully, a point-in-time issue; @ChadNedzlek is working on stuff that will put the reporting into the Helix client and remove the need to do it inside the work item, allowing greater control over retry logic and handling of the problem. |
Thanks for the explanation @MattGal. So in summary - we can treat these failures as "flaky infrastructure"? |
Sure, as long as we're counting this particular flakiness as ADO's :) |
This allows for HPackEncoder and QPackEncoder to be trimmed in browser wasm.
Contributes to #44534
.br size of
dotnet new blazorwasm
app: