I am receiving HTTP 502 (Bad gateway) when trying to view this URL: https://packages.gentoo.org/packages/sys-kernel/vanilla-sources With some other randomly picked packages the error does not manifest, I see normal pages generated from the packages database. With sys-kernel/vanilla-sources the error persists already for a couple of weeks (or at least from 5.10.11 upstream kernel release), it is 100% reproducible all the time. infra-status does not indicate any problems (except for the packages.gentoo.org Atom feeds being broken which does not seem to be relevant).
(In reply to Alexander Bezrukov from comment #0) > I am receiving HTTP 502 (Bad gateway) when trying to view this URL: > https://packages.gentoo.org/packages/sys-kernel/vanilla-sources > > With some other randomly picked packages the error does not manifest, I see > normal pages generated from the packages database. With > sys-kernel/vanilla-sources the error persists already for a couple of weeks > (or at least from 5.10.11 upstream kernel release), it is 100% reproducible > all the time. > > infra-status does not indicate any problems (except for the > packages.gentoo.org Atom feeds being broken which does not seem to be > relevant). Ack, the ingestor is broken; I'm taking a looksee. -A
(In reply to Alec Warner from comment #1) > (In reply to Alexander Bezrukov from comment #0) > > I am receiving HTTP 502 (Bad gateway) when trying to view this URL: > > https://packages.gentoo.org/packages/sys-kernel/vanilla-sources > > > > With some other randomly picked packages the error does not manifest, I see > > normal pages generated from the packages database. With > > sys-kernel/vanilla-sources the error persists already for a couple of weeks > > (or at least from 5.10.11 upstream kernel release), it is 100% reproducible > > all the time. > > > > infra-status does not indicate any problems (except for the > > packages.gentoo.org Atom feeds being broken which does not seem to be > > relevant). > > Ack, the ingestor is broken; I'm taking a looksee. > > -A Our current theory is that an earlier fix was not deployed properly due to exciting branch semantics. I've merged the branches and I'll check back in 1h to see if the changes made it into prod. -A
This is now fixed.
(In reply to Emily Rowlands from comment #3) > This is now fixed. I confirm.