-
-
Notifications
You must be signed in to change notification settings - Fork 1.7k
ci: fix yaml (again) #7107
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
ci: fix yaml (again) #7107
Conversation
@@ -21,7 +21,7 @@ | |||
"clean": "yarn rimraf sentry-ember-*.tgz dist tmp build .node_modules.ember-try package.json.ember-try", | |||
"lint": "run-p lint:js lint:hbs lint:ts", | |||
"lint:hbs": "ember-template-lint .", | |||
"lint:js": "eslint . --cache --cache-location '../../eslintcache/'", |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Is this intentional?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Yes, we removed all of these in a previous PR, just forgot about this. (Stuff I find when trying to debug stuff locally)
I also added a workflow we can run to delete all caches on GHA. Maybe that helps... |
@@ -172,7 +174,6 @@ jobs: | |||
key: ${{ steps.compute_lockfile_hash.outputs.hash }} | |||
|
|||
- name: Install dependencies | |||
if: steps.cache_dependencies.outputs.cache-hit == '' || needs.job_get_metadata.outputs.force_skip_cache == 'true' |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Actually I don't think we need this, when we hit the cache this step will be more or less instantenous anyhow. May help here or there with an inconsistency.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
🚀
@@ -1,7 +1,7 @@ | |||
{ | |||
"private": true, | |||
"scripts": { | |||
"build": "node ./scripts/verify-packages-versions.js && lerna run build:types,build:transpile,build:bundle", | |||
"build": "node ./scripts/verify-packages-versions.js && run-s build:types build:transpile build:bundle", |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
cc @JamesHenry I reverted this here, as it seems that we sometimes ended up with an inconsistent state because of this (??) TBH we are not quite sure what is going on, but in multiple tests in weird ways, e.g. https://github.com/getsentry/sentry-javascript/actions/runs/4132316633/jobs/7140902414
I am not really sure that this is the actual problem, but it's my best guess so far that some internal race condition or whatever leads to something being overwritten or not properly build... Will investigate some more!
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Sorry about that @mydea - I'll take a look.
FYI I invited you on Slack Connect, sent to your sentry email, so you can also reach me there
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@mydea At a quick glance it looks like tests depend on builds having run already, at the very least build:types
?
We can naturally extend our existing task pipelines to let the lerna task runner handle coordinating this for us. I'll put up a PR
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
There are also a couple of missing relationships from the package graph right now (because the relevant package.json files are missing them)
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
No worries, again I am not even really sure this was the problem, we seem to have an issue with a bunch of CI caching stuff coming together. Thanks for helping us figure this out!
I've put a request with our IT to get a shared slack channel, it is pending - looking forward to it going through!
I was also thinking about getting rid of our CI build cache and just leverage Nx cache. It would probably streamline stuff quite a bit!
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I've put a request with our IT to get a shared slack channel, it is pending - looking forward to it going through!
Ah ok - I had sent the Slack Connect invite from our side your sentry email - did you receive it? Did you forward it on to your IT, or did you initialize the process separately from your side?
I was also thinking about getting rid of our CI build cache and just leverage Nx cache. It would probably streamline stuff quite a bit!
💯 and I can help set this up for you. As an OSS project you get unlimited remote caching for free
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Cacheable tests PR which should hopefully resolve the issues you've been seeing as well: #7108
yaml is a terrible language, I hope this is now actually correct!