Re: Status of new Debbugs deployment

[Michael Albinus]

Felix Lechner via "General discussion for the tracker at
debbugs.gnu.org" <help-debbugs@gnu.org> writes:

> Hi,

Hi Felix,

> I synchronized our 'gnu-reconstruction' repo with what's deployed on
> debbugs.gnu.org. The changes were small but noticable on the web. [1]
>
> Now both sites produce nearly the same HTML. The differences for bug
> reports (below) can probably be traced to the treatment of hash keys in
> the different Perl versions. [2]
>
> My server runs 5.36.0 instead of 5.20.2, which was released seven years
> earlier. The most significant changes to hash keys were made in Perl
> 5.18.0, so I cannot currently explain the differences but I'm not sure
> it's worth the effort to track it down.
>
> If nobody has any objections, I will ask the FSF admins to repurpose
> debbugs2p.gnu.org for my experimental deployment. Please speak up now!
>
> What does the "2p" stand for anyway, please?
>
> There is some resistance to my use of GNU Guix. I hope the experimental
> server will put those doubts to rest. Guix is superior to any combinaion
> of Ansible and Trisquel, as you will all see shortly.
>
> The server will be maintained in a single flat file, with atomic
> rollbacks if something did not work out as planned..

No objections from my side. However, debbugs2p was installed by both Bob
Proulx and Noam Postavsky. Noam does not work any longer actively on
debbugs, but I would pretty much appreciate if Bob could comment. He
knows the sysadmin basis of debbugs2p better than anybody else.

> Kind regards,
> Felix
>
> [1] 
> https://codeberg.org/lechner/debbugs-gnu/commit/356ec62a8702a77726609723e8ed0246236dd51d
> [2] 
> https://codeberg.org/lechner/debbugs-gnu/src/commit/356ec62a8702a77726609723e8ed0246236dd51d/Debbugs/CGI/Bugreport.pm#L351-L352

Best regards, Michael.