Re: CTI heads-up

On Jul  6, 2026, "Carlos O'Donell" <[email protected]> wrote:

> Why do you consider a mirror update issue without data loss a
> "severe mishap?"

A single mistake left mirrors all over severely outdated and in need of
a lengthy recovery process.  It exposed several severe flaws in the
service architecture:

Local snapshots at the mirrors could have significantly sped up the
recovery.

Disabling the rsync server or pointing it at a complete tree during
recovery could have spared downstream mirrors from syncing the mishap
and having to re-download it all as the servers recovered, which further
contributing to the proble.

Publishing severely outdated information at the front page while the
up-to-date information was available throughout in git made people
vulnerable while showing a flaw in the architecture that gets such
critical information published.


Working hard, promptly and transparently to recover from an unfortunate
mistake is only a small part of what I expect from a serious service
provider.  It's a given that mistakes happen, but the responses to
mitigate and avoid further occurrences and their lasting consequences
are a far more important tell.  I'd have expected such measures ahead of
time, but since they visibly weren't there, maybe they will come soon,
if there's seriousness, willingness, and spare capacity to absorb
further projects.

-- 
Alexandre Oliva, happy hacker            https://blog.lx.oliva.nom.br/
Free Software Activist     FSFLA co-founder     GNU Toolchain Engineer
Learn the truth about Richard Stallman at https://stallmansupport.org/