tests.r-b.o/debian: archiving buster and bullseye

Mattia Rizzolo mattia at debian.org
Wed Nov 15 18:00:16 GMT 2023


On Wed, Nov 15, 2023 at 05:15:50PM +0000, Holger Levsen wrote:
> eleven days ago I disabled testing Debian buster on tests.r-b.o/debian
> and noone really complained nor noticed.

That's fine for me.  Including archival to reduce disk space usage.

> Tomorrow I intend to also disable testing Debian bullseye which
> development ceased in August 2021.

I disagree with this: I believe we should at least keep testing until
oldstable is handed over to the LTS team and the Release Team will stop
touching it (which I believe will happen around the next Summer).

> The current stable Debian release is bookworm, released in June 2023
> and basically not in any development anymore. bookworm is still
> being tested by us and I think we should keep it that way until the next
> release, trixie, has been published sometime in 2025 probably.

Again, I'd keep testing until stable is touched.  We should archive the
releases with the full data, at least testing all package versions that
have been introduced across their lifetime.

> Keeping buster and bullseye (and previously also stretch) has costs,
> both in diskspace on all machines were using, as well as as higher
> loads for both maintenance jobs as well as the html creation jobs.
> It *also* causes mental load on everyone looking at the results.
> 
> I also doubt that those olds results are useful today more than
> "very rarely".

"very rare" is still reasonable, considering the resources, IMHO.

After all, we have already seen people randomly use our data in random
papers, in way that we were not thinking about.  I don't think we need
to make this type of work, that is not our work, harder than it needs to
be.

> Like for stretch, the results will stay in the postgresql database,
> and I will also backup the generated diffoscope diffs, though on
> a different host.

That's already done, but the old data are still on jenkins.  what were
you thinking about?  I don't think the stretch tarball takes too much
space?

-- 
regards,
                        Mattia Rizzolo

GPG Key: 66AE 2B4A FCCF 3F52 DA18  4D18 4B04 3FCD B944 4540      .''`.
More about me:  https://mapreri.org                             : :'  :
Launchpad user: https://launchpad.net/~mapreri                  `. `'`
Debian QA page: https://qa.debian.org/developer.php?login=mattia  `-
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 833 bytes
Desc: not available
URL: <http://alioth-lists.debian.net/pipermail/reproducible-builds/attachments/20231115/2189c6ad/attachment.sig>


More information about the Reproducible-builds mailing list