#839 Errors during downloading metadata for repository 'baseos' (aarch64)
Closed: Duplicate by smooge. Opened by weatherchnl.

Hi, we have a CI that provisions an aarch64 host with Centos Stream 9 using ansible. The other day one of the provisions failed with this:

STDOUT:

CentOS Stream 9 - BaseOS                        593  B/s | 3.9 kB     00:06    


STDERR:

Errors during downloading metadata for repository 'baseos':
  - Downloading successful, but checksum doesn't match. Calculated: 92f44febb5fbf0f731af498feefb653d5b516e7184a23ab6e63b71c1faa43f18cb10a39bea249ee9467dfc010a51acd9c0314dcd69ea5e1d1e86c5bd48ed94db(sha512)  Expected: 7ce2adbd012d321cabb2bf28fb3fa67e4b31f8f1f21e91643b8b79d8ea18215711e6eadb72f792e1eeff2a39bb22346eb9594078b824ebc45599962a431995e2(sha512) 
Error: Failed to download metadata for repo 'baseos': Cannot download repomd.xml: Cannot download repodata/repomd.xml: All mirrors were tried

This broke the CI run so I took a quick peruse of the mirrors and found some have stale data (and have the checksum found in the error message above). Could these stale mirrors be breaking dnf updates randomly for our provisions? Is this of any concern or is it business as usual and we're missing something in our provisioning that normally negates this issue?

Mirrors with checksum: 92f44

https://lon.mirror.rackspace.com/centos-stream
https://hkg.mirror.rackspace.com/centos-stream
https://mirror.01link.hk/centos-stream
https://iad.mirror.rackspace.com/centos-stream
https://ord.mirror.rackspace.com/centos-stream

Other mirrors with different stale repomd.xml files:

https://humirror.lonyai.com/centos-stream
https://kartolo.sby.datautama.net.id/centos-stream
https://linux.domainesia.com/centos-stream
https://mirror.nevacloud.com/centos-stream
https://epel.mirror.net.in/centos-stream
https://mirror.ihost.md/centos-stream
https://semirror.lonyai.com/centos-stream
https://ftp.upjs.sk/pub/centos-stream
https://fastmirror.pp.ua/centos-stream
https://uscamirror.lonyai.com/centos-stream

Thank you for opening a ticket, and our apologies in the problem. The issue is being tracked in https://pagure.io/fedora-infrastructure/issue/10801 and is due to various factors outlined in the later parts of that ticket.

Thanks @smooge , feel free to close this ticket as duplicate then. Sorry for the noise!

I don't think this was noise. You as a consumer of CentOS Stream should not have to know about N different ticket trackers for your problem. I was just wanting to make sure you could track where this will be worked on longer term.

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

Log in to comment on this ticket.

Metadata