Recently I noticed that I'm not getting substitutes for Mutt. Looking at the ci.guix.gnu.org web interface, it seems that we stopped building Mutt about a month ago: https://ci.guix.gnu.org/search?query=spec%3Amaster+mutt-2.0.7+system%3Ax86_64-linux And `guix weather` confirms that we seem to have stopped building it: ------ $ guix describe Generation 174 Jun 08 2021 14:59:25 (current) guix 74c7055 repository URL: https://git.savannah.gnu.org/git/guix.git commit: 74c70553394e32f7b6d5fa56da3bf3b18623c70f $ guix weather mutt computing 1 package derivations for x86_64-linux... looking for 1 store items on https://ci.guix.gnu.org... https://ci.guix.gnu.org 0.0% substitutes available (0 out of 1) unknown substitute sizes 0.0 MiB on disk (uncompressed) 0.0% (0 out of 1) of the missing items are queued at least 1,000 queued builds x86_64-linux: 906 (90.6%) aarch64-linux: 9 (.9%) i686-linux: 85 (8.5%) build rate: .00 builds per hour x86_64-linux: 0.00 builds per hour i686-linux: 0.00 builds per hour aarch64-linux: 0.00 builds per hour ------ I wonder what is going on?
Hello Leo, > I wonder what is going on? The current mutt package output is: --8<---------------cut here---------------start------------->8--- mathieu@meije ~$ guix build mutt --no-grafts /gnu/store/c0wp5iq5k11mvqrm4bs011qlm7qy48ia-mutt-2.0.7 --8<---------------cut here---------------end--------------->8--- The last mutt build on the master specification is https://ci.guix.gnu.org/build/321231/details, that reports the same output. This build has not been attempted again because its output didn't change. The substitute is however missing as you noticed. That's because the Cuirass GC root expired, and the nar file also probably expired its 90 days TTL. We should maybe increase this TTL, or make sure that every "latest" master derivation is covered by a GC root. WDYT? Thanks, Mathieu
Hi,
Mathieu Othacehe <othacehe@gnu.org> skribis:
> The substitute is however missing as you noticed. That's because the
> Cuirass GC root expired, and the nar file also probably expired its 90
> days TTL.
>
> We should maybe increase this TTL, or make sure that every "latest"
> master derivation is covered by a GC root.
>
> WDYT?
Mathieu, do you have Zabbix graphs showing how we’re doing wrt. storage
on berlin?
If we’re doing well, we should definitely increase the ‘guix publish’ TTL.
Thanks,
Ludo’.
Hey, > Mathieu, do you have Zabbix graphs showing how we’re doing wrt. storage > on berlin? Here's one attached, we have been stable around ~10TiB of free space for the last couple of months. > If we’re doing well, we should definitely increase the ‘guix publish’ TTL. What about increasing it to say 180 days? Thanks, Mathieu
Hi, Mathieu Othacehe <othacehe@gnu.org> skribis: >> Mathieu, do you have Zabbix graphs showing how we’re doing wrt. storage >> on berlin? > > Here's one attached, we have been stable around ~10TiB of free space for > the last couple of months. Interesting. The graph suggests that GC is running everyday to collect probably a few GiBs. Perhaps we should ‘guix gc -F15T’ once a week to reduce GC work on subsequent days? >> If we’re doing well, we should definitely increase the ‘guix publish’ TTL. > > What about increasing it to say 180 days? Sounds good to me. Can you make the change? Thanks, Ludo’.
Hey,
>> What about increasing it to say 180 days?
>
> Sounds good to me. Can you make the change?
That was done with ca6f40272fdf5644597f6ad2e38fa8ab2907d126.
Closing,
Thanks,
Mathieu