Hi Simon, zimoun <zimon.touto...@gmail.com> writes:
> Hi Maxim, > > On Wed, 23 Mar 2022 at 11:50, zimoun <zimon.touto...@gmail.com> wrote: >> On Wed, 18 Aug 2021 at 22:25, Maxim Cournoyer <maxim.courno...@gmail.com> >> wrote: > >>> Extra note: the problems reported earlier (hang or backtrace instead of >>> graceful fallback to other substitute servers) also affect the scenario >>> where substitutes are fetched from uDNS discovered substitute servers (I >>> just tried). >> >> I guess this old bug [1] about the daemon hanging is still there, right? >> >> How such issue could be tackled? >> >> 1: <http://issues.guix.gnu.org/issue/30290> > > What could be actionable for tackling this old bug? I can't seem to reproduce the original issue, which was that an offline server would slow down fetching of substitutes by repeatedly trying to reach it, at least when setting e.g.: --8<---------------cut here---------------start------------->8--- 127.0.0.2 ci.guix.gnu.org --8<---------------cut here---------------end--------------->8--- In my /etc/hosts file to simulate it's unreachable, with at least another substitute server available in my --substitute-urls list. It goes right to that second server without loosing any time. Closing for now, we can always reopen the issue if it manifests itself in the future. -- Thanks, Maxim