I've a weird glitch with Seqera (see image) where it is saying

Unknown pipeline repository or expired Git credentials: <https://github.com/nf-core/rnaseq>

when I try to modify an existing pipeline, create a new pipeline or

launch a run without a configuration

I get this warning for all of the nf-core repositories. My own repositories are working just fine. Any thoughts why this is suddenly happening?

This usually means that the GitHub PAT that you have set in your profile has expired and you need to set a new one

That's resolved the issue. Strange, I've never had a PAT for nfcore before. Thank you 🙂

It's not specific to nf-core, it's just to access the GitHub API to get anything

I'm just wondering why it suddenly has stopped working, I'm not aware that I had a PAT covering nfcore (I've got loads of other credentials for my own repositories). Clearly I did and then it's expired without me noticing.

nf-core is public, so technically it's not needed, but it's still good to have higher rate limits

Thanks for you help, this was causing me a massive headache today.

It might be that you have multiple credentials in platform, one is for a workspace with nf-core pipelines or something?

That might be wise. I've made a note for myself so hopefully I don't ask the same question again... (can't promise).

Actually I think we have another project in the works which will proactively warn you when a credential expires, so that's an even better solution

Yes, and indeed just a warning that you need to update/add a PAT will probably direct users in the right direction.

Hi. New to seqera. I'm attempting to run a workflow from seqera and after about 13 minutes of no logs, it finally spits out: Unknown error accessing project -- Repository may be corrupted However, when i run this same git repo via nextflow locally, it executes just fine. A couple questions: 1. Why does it take so long to spit out any logs? 2. What is the best way to diagnose the unknown error?

1. You're probably running on cloud, eg. AWS Batch. This uses "batch squared" where the head Nextflow job is itself a Batch job. This from zero to lots very well and is cost efficient, but you have to wait for AWS Batch to schedule and run your head job, which is slow. You can toggle "use fargate for the head job" to make it faster, or use a faster compute environment type - eg. AWS Cloud instead of Batch.

Sorry, I didn't realize I added to someone else's thread. I'll start another and continue it there