Add Slack/Teams notification plugin docs#3821
Add Slack/Teams notification plugin docs#3821FriederikeHanssen wants to merge 1 commit intonf-core:mainfrom
Conversation
- Core team checklist: add step to invite @Full test notifications bot to new pipeline Slack channels (required for nf-slack CI notifications) - User-facing intro: replace deprecated hook_url tip with pointers to nf-slack and nf-teams plugins Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
✅ Deploy Preview for nf-core-docs ready!
To edit notification comments on pull requests, go to your Netlify project configuration. |
✅ Deploy Preview for nf-core-main-site ready!
To edit notification comments on pull requests, go to your Netlify project configuration. |
mashehu
left a comment
There was a problem hiding this comment.
I don't think we should add these to pipeline channels. These channels are for communication not monitoring. Better to have just add it to the AWS megatests channel
|
people are ignoreing the aws-megatest channel and we have tons and tons of full_size tests broken. At the moment, we cannot completely prevent people from merging to master without the full size test passing and it is treated as an after thought. I think we should give these more attention by making their failures visible. Pipelines that have a dedicated dev channel can just replace the name with |
|
I agree with @FriederikeHanssen here. There are too many channels for an average pipeline dev to monitor, the best way to make people more informed is to post somewhere we can guarantee they check the status, but also to help nudge people to actually fix the issue is to do it publicly (yes it might be a little uncomfortable, but it's for a good cause - the realistic output is a very big plus for nf-core and we should make sure we capitalise on it). |
|
Like one alternative could be to have the lead dev of each pipeline being pinged by the notification on the aws-megatest channel so they go look, but that would require tracking lead Devs which is not static and code owner files were controversial when we proposed it previously |
|
the problem with aws megatest is not that people ignore it, it is that we don't have a good way to see when something fails. if we pump all the failures to one channel, it is easier to ping people (manually, don't need to automate this). |
|
but if they go to the pipeline channel, we don't need to ping them? |
|
but we then need to monitor all the pipeline channels to make sure we don't have a lot of failing tests. Usually debugging these tests anyway needs a bit of a back and forth with a core member, no? |
No, in my experiences, in most cases the failures have been stuff like interrupted downloads, badly configured resources etc . Not stuff that is needed by core members, and we are already having problems with core members not having enough time with other tasks, having to monitor yet another channel and chase people will not work effectively. Also one could argue that a successful pipeline execution on AWS after a release might be a nice way to celebrate the end of a successful release cycle - an automatic 🎉🎉🎉 is always a nice and satisfying thing to see, so there is a human factor to this as well. |
|
okay, let's try it out and if it is too noisy, we can always change this. |
|
Let me test if nf-slack can send to two channels. In any case, this PR is dependent on the tools PR that will open another Pandoras box conceptually: should we keep slack/teams configured by default or just point people to plugins? |
|
Doesn't work yet, but Adam is working on the feature already: seqeralabs/nf-slack#37 |
|
On Slack, you can also notify people or channels using the unique Slack IDs: |
This is requires the template updates to be merged: nf-core/tools#4051
Summary
@Full test notificationsbot to new pipeline Slack channels (required for nf-slack CI notifications)--hook_urlwebhook tip with pointers to nf-slack and nf-teams pluginsTest plan
🤖 Generated with Claude Code
@netlify /docs/checklists/community_governance/core_team