#I started to see `No space left on
shell knot
looking into this, Mark. Please let us know if you have any more info.
What is your GHA config?
abstract geyser
There's a similar report here: https://discord.com/channels/707636530424053791/1212477059235057664
smoky tusk
Here: https://github.com/openmeterio/openmeter/blob/main/.github/workflows/ci.ya...
Looking at this now. FYI: https://github.com/dagger/dagger/issues/6839
smoky tusk
I am trying to figure out when this started happening, and why.
I've noticed that Free disk space step was introduced as of https://github.com/openmeterio/openmeter/actions/runs/8155484168/job/22291093451 via https://github.com/openmeterio/openmeter/pull/662
I am not able to access the disk-space related failures before that, e.g. https://github.com/openmeterio/openmeter/actions/runs/8155474438/job/22291061466
What would be the easiest way of reproducing this @vestal pawn ?
vestal pawn
Here is one: https://github.com/openmeterio/openmeter/actions/runs/8123963176/job/22236107513#step:2:8790
It just occurred to me that these builds use Dagger Cloud cache
And those probably grow like forever
Chances are the go module cache is so large it just fills the disk
With all those dependabot PRs this is a very real possibilty
sadly the UI doesn't display cache volume sizes, so I can't be sure
smoky tusk
That is a solid assumption! WDYT @fallen smelt?
fallen smelt
That is a solid assumption! WDYT <@336241811179962368>?
yes, this will grow forever if nothing is pruning whatever is in there
vestal pawn
That free disk space step just ran for 11 minutes....
I bumped the cache volume name to see if it solves anything.
Would be nice if I could purge cache from the UI.
shell knot
@covert rapids 👆
covert rapids
Thanks! We have this planned. I'll add your feedback to the issue to help with prioritization.
vestal pawn
Thanks!