fix(test): The new test case PagebufferReader5 introduced an error. #1936
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
billprovince
requested review from
akon-dey,
joshua-goldstein and
skrdgraph
as code owners
skrdgraph
reviewed
May 4, 2023
| @@ -4,7 +4,7 @@ on: | |||
| branches: | |||
| - main | |||
| - 'release/v*' | |||
| pull_request_target: | |||
| pull_request: | |||
There was a problem hiding this comment.
hey @billprovince I made this fix to your PR.
There are 2 issues with our CI here:
pull_request_targetis what is causing this issue, and this was added per recommendation from coveralls to support coveragecachecleanup was also done on GitHub Actions (as we suspected an issue there).
cc: @adityasadalage and @joshua-goldstein ... lets do a check on these items tomorrow.
There was a problem hiding this comment.
@skrdgraph Thanks for catching this and diagnosing the problem... I suggest we merge the CI fix in #1944 first, and then we can merge Bill's PR (and hopefully CI will run as expected on here)
joshua-goldstein
previously approved these changes
May 5, 2023
joshua-goldstein
added a commit
that referenced
this pull request
May 8, 2023
In #1936 we observed that because we are using `pull_request_target` without a specific reference, we are not in fact checking out the PR branch. This PR splits the workflow into two, one only to run badger CI tests, and the other to run tests & report coverage results (using pr_target with the appropriate commit). This has the benefit that any issues with coverage are isolated to one workflow, which we could disable later if problems arise. This does create duplication of compute (all tests run x2) but we are using github hosted runners so we do not pay for this cost. This also aligns with our dgraph repo workflow setup.
joshua-goldstein
force-pushed
the
FixPagebufferReader5
branch
from
4cbeef1 to
17f28a7
Compare
joshua-goldstein
approved these changes
May 8, 2023
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Problem
Fixes DGRAPHCORE-199
The intended fix for the CI issue for randomly getting EOF failures in PagebufferReader2 test introduced the test case FixPagebufferReader5 -- intended to avoid regressions. However, it did something that is not expected by the PageBuffer struct: it created a new PageBufferReader based on an empty PageBuffer. It seems that this is not allowed.
Solution
Fix the test case for PagebufferReader5 by no longer attempting to create a new PageBufferReader based on an empty PageBuffer.
Note
The CI build will still fail to allow this to pass because it will continue to use the prior version of the tests without the fix. However, locally tested, I get this result:
~/gitspace/github.com/dgraph-io/badger/y$ go test -run PagebufferReader5
PASS
ok github.com/dgraph-io/badger/v4/y 0.335s