On Mon, 10 Jul 2023 21:08:14 +0100 "Adam D. Barratt" <a...@adam-barratt.org.uk> wrote: > Control: tags -1 + bookworm moreinfo > > On Mon, 2023-07-10 at 21:55 +0200, Ervin Hegedüs wrote: > > nmu libnginx-mod-http-modsecurity_1.0.3-1+b1 . ANY . bookworm . -m > > "Closes: 1037226" > > *Please* include information in binNMU requests that allows us to > identify the actual problem without having to dig around. As an > additional note, the text in the "message" argument (-m) is copied > verbatim into changelogs, which will then be visible to users and > anyone else viewing the changelog. It needs to be more descriptive than > a bug number. > > The BTS metadata for #1037226 indicates that it also affects unstable > and testing. Is that correct? If so, it needs to be fixed there first. > If not, please explain why unstable is not affected despite having > exactly the same package version. >
I was in contact with Ervin and suggested him to file the binnmu -- for stable *and* for unstable to get #1037226 fixed. the mentioned pr is related to a new nginx version and orthogonal to the problem in said bug report, beside that the the pr improves the packaging in that aspect. for now a rebuild suffices also for unstable. summary of #1037226: That lead to the issue: - nginx uploaded with OLD PCRE - libnginx-mod-http-modsecurity entered NEW and had been accepted - it uses the OLD PCRE, as it is compiled against libmodsecurity3, which uses PCRE at that time - nginx uploaded with NEW PCRE2 - modsecurity uploaded with PCRE2 Situation: nginx -> PCRE2 modsecurity -> PCRE2 libnginx-mod-http-modsecurity -> OLD PCRE --> a binnmu will rectify that. please do a binnmu for bookworm: nmu libnginx-mod-http-modsecurity_1.0.3-1+b1 . ANY . bookworm . -m "rebuild against pcre2 (Closes: #1037226)" > Regards, > > Adam > > >
