koji

#3074 Chromium running out of disk space on builds
Closed: Invalid by tkopecek. Opened by spot.

Closed: Invalid
Reopen Issue

Chromium keeps getting larger. Specifically, the latest builds have gotten so big that on some targets, it actually fails because it runs out of disk space. See:

https://koji.fedoraproject.org/koji/getfile?taskID=77205278&volume=DEFAULT&name=build.log&offset=-4000

For some targets, I can work around this by turning the debugging level down (from -g2 to -g0), but even that was not enough for the EPEL-8 build to succeed.

I'm not sure what the default disk allocation is, but is there any way to increase it for chromium builds (heavybuilder)?

There is some urgency here, because there are actively exploited vulnerabilities in Chromium that need an update to fix.

This should very likely be a downstream fedora-infrastructure ticket. :)

That said, it's not simple.

The aarch64 'heavybuilders' have a single 230GB disk in them. Then, builders keep buildroots/failed builds around for some time. Builders also sometimes have multiple builds on them running at the same time.

That said, looking into it, I see a ton of space taken up by old nvram files on aarch64.
I've deleted those, so can you try again?

Metadata Update from @kevin:
- Custom field Size adjusted to None

We don't maintain fedora's instance as Kevin mentioned. From dev side we're working on rewriting scheduler, so build will be able to reserve diskspace/mem/cpus. It is planned +- to first quarter of 2022.

Metadata Update from @tkopecek:
- Issue close_status updated to: Invalid
- Issue status updated to: Closed (was: Open)

This issue has been migrated to Fedora Forge:
https://forge.fedoraproject.org/koji/koji/issues/3074

Please continue any further discussion there.
Metadata
None
None
None
None
Low
None
Static archive. pagure.io was sunset in 2026; this is a read-only snapshot.
Red Hat Logo

Fedora is sponsored by Red Hat. Learn more about the relationship between Red Hat and Fedora »

© Red Hat, Inc. and others.