Hi, those ids only available for viewing for staff. For us non staff would be better if you show the full logs

Sure, I'm attaching the build logs for the failing redeploy both in plain text and in JSON

1459585126873960532
1459585127167426652

A quick fix would be disabling the metal builder in the service settings, I will tag in @Noah so he can take a closer look at why the metal builder is failing.

Thank you!

Could I get a link to the service having this issue?

Thank you both for looking into this!

https://railway.com/project/f6d25e17-9bb2-457f-8ce2-e55b5ce1dcd8/service/8c4e4e87-7cf3-4ab2-9ec2-b6cc41db7b5b?environmentId=8c3e7efe-326f-4453-a885-2412ed143f36

oops, forgot to ask

All good!

Are you able to trigger a redeploy so I could look at it live? No worries if not I think I have enough

I'll do it right away

Thanks!

Awesome I see what the issue is. Going to go do some poking around on my own service to see if I could repro.
Could I get some info about the service itself?

Sure, the project is fully open source so you may have a deep dive at https://github.com/Stickerifier/Stickerify, I can also share any detail needed

Oh wonderful thats perfect. Appreciate it!

Short term, can you try disabling metal builder to see if this happens again?

Would love to rule out metal builder on this

On it

The redeploy completed successfully just now, weird

On standard builder. Alright this is good to know thank you

I meant this one, that was still on metal

The standard builder one went fine as well

Copy. Alright I'll do some more digging and get back to you on Monday about this. If it gets blocking please ping and let me know!

Should I leave it on standard builder for the time being?
This isn't a critical service so please, have no rush and enjoy your weekend ❤️

We have a somewhat inconsistent bug related to metal builder we're actively working on. We also have a project we're working on which should improve a whole bunch of the flow from "push my code" -> "get it live".
This is definitely odd so I do want to get it sorted but if using standard builder works I'd switch to that temporarily.

!remind me monday at 9 am to check on this if I havent already

Very sorry you're running into this!

Oh, no worries! I just reported it because it felt strange to have it fail consistently altough I didn't change anything build-related

Hello, is there any update on this issue?

There is a successful Dockerfile deploy 5 hours ago?

I've been on the classic builder ever since Noah asked me to, the issue was with the metal environment

Can you switch back to Metal so I can see the latest error?

I just switched to metal and it worked, it was consistently failing last Saturday so we can mark this as solved unless further investigations are needed

I'm pretty sure we made some adjustments