-
Notifications
You must be signed in to change notification settings - Fork 85
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
Smoke Tests Bug Fixes #2094
Merged
Merged
Smoke Tests Bug Fixes #2094
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
At the moment we only wait 2 seconds, and we keep the introspection command running for the whole test. We don't need to do the latter, and from empirically running the smokes it seems as though we need slightly more time to ensure writing to the file is complete.
At the moment we're scanning for a result that, while it exists, is irrelevant because of the GitHub Actions does status conditions. This should fix that problem so that this flow only runs on the failure of the previous two steps, rather than being skipped as happens at present.
This way the UI looks much prettier
jonathanrainer
force-pushed
the
jr/smoke-test-bug-fixes
branch
from
September 2, 2024 10:56
87e92fb
to
1125fa9
Compare
nmoutschen
approved these changes
Sep 2, 2024
loshz
approved these changes
Sep 2, 2024
Merged
jonathanrainer
added a commit
that referenced
this pull request
Sep 4, 2024
# [0.26.1] - 2024-09-04 ## 🚀 Features - **Respect the use of `--output` flag in the supergraph binary - @aaronArinder PR #2045** In testing to attempt to reduce the runtime of `supergraph compose` we noticed that a very large proportion of the time spent (in the case of large supergraphs) was spent printing the result to `stdout`. With this change we add an `--output` flag to the `supergraph` binary which means this time can be reduced significantly, leading to much faster compositions. - **Add `--license` flag to `rover dev` - @loshz PR #2078** Adds the ability to pass along an offline enterprise licence to the router when running `rover dev` - **Remove Rayon and reduce usage of Crossbeam - @jonathanrainer PR #2081** Now that `rover` has transitioned to using an asynchronous runtime we don't need to use Rayon any more. This also resolves a bug whereby `rover dev` could lock up if passed a `supergraph.yaml` file with lots of subgraphs in. - **Introduce new print macros - @loshz PR #2090** Adds three new macros to the codebase so that we can still visually distinguish between INFO, WARNING and ERROR log lines without the use of emoji - **Use new print macros in place of emoji - @loshz PR #2096** Updates the locations that previously used emoji to utilise the new macros defined in the previous PR ## 🐛 Fixes - **Stop Windows Installer failing if whitespace is accidentally passed to the `rover install` command - @jonathanrainer PR #1975** In some situations it was possible for whitespace to be passed to the `rover install` command which then caused the installer to fail. A guard has now been added to strip whitespace out before it is passed to the install command. ## 🛠 Maintenance - **Move CI to using newly create Ubuntu images - @jonathanrainer PR #2080** CircleCI is removing support for older Ubuntu machine images, this brings us up to date but does **not** change any of our `glibc` support etc. - **Add check for aarch-64-unknown-linux-musl to installers - @loshz PR #2079** - **Update node.js packages - @jonathanrainer PR #2070** Includes `eslint` to v9.9.1 and `node` to 20.17.0 - **Update `node` CircleCI orb to v5.3.0 - @jonathanrainer PR #2071** - **Update `apollographql/federation-rs` to v2.9.0 - @jonathanrainer PR #1983** - **Update `apollographql/router` to v1.52.1 - @jonathanrainer PR #2077** - **Update `node` Docker Image to v20.17.0 - @jonathanrainer PR #2072** - **Update `apollographql/router` to v1.53.0 - @jonathanrainer PR #2084** - **Update `npm` to v10.8.3 - @jonathanrainer PR #2091** - **Update `slackapi/slack-github-action` to v1.27.0 - @jonathanrainer PR #2092** - **Update `node` CircleCI orb to v6.1.0 - @jonathanrainer PR #2093** - **Fix some bugs in the smoke tests - @jonathanrainer PR #2094** ## 📚 Documentation - **Add `cloud config` docs - @loshz PR #2066**
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.
We've seen some spurious smoke test failures recently because of being unable to decode the schema that comes back when its being watched. My hunch is that this is because we don't quite wait long enough in the test so this increases the timeout. To try and add some extra safety we also shutdown the
watch
command post-waiting so that the state of the file should be fixed before we attempt to read it. This should make this saferThis also includes a fix for the smoke tests so that Slack should actually get messaged when the workflow fails. Beforehand because it was relying on a state of the job looked up explicitly, if a job failed elsewhere it would cancel all other jobs so the messaging job would never run. However in this version and using the special
failure()
syntax we can run the messaging job whenever any task fails without it being skipped.Green Smoke Test Run: https://github.com/apollographql/rover/actions/runs/10665653050