[arch-dev-public] [arch-commits] Commit in fail2ban/trunk (PKGBUILD)

Sébastien Luttringer seblu at seblu.net
Wed May 15 16:15:17 EDT 2013

On Wed, May 15, 2013 at 1:02 PM, Bartłomiej Piotrowski <b at bpiotrowski.pl> wrote:
> On 2013-05-14 22:37, Sébastien Luttringer wrote:
>> On Tue, May 14, 2013 at 10:22 PM, Bartłomiej Piotrowski
>> <bpiotrowski at nymeria.archlinux.org> wrote:
>>>     Date: Tuesday, May 14, 2013 @ 22:22:08
>>>   Author: bpiotrowski
>>> Revision: 90846
> SMTP forwarders aren't as crucial as bash is
Without judging what you say, It's not a reason to break them!

> I don't see any reason to revert changes or delay moving binaries to /usr/bin. Just
> message maintainer that his package is broken due to recent changes.

You're not alone to provide a smtp forwarder "virtual" package and the
path provided by all must be the same.
We cannot ask to users and others packagers to update their path if it
is not the same accross smtp forwarders.
Some packages will works with exim and others will not works with
postfix, openstmpd, ssmtp, etc.
The fail2ban package is a good example.
This let no time to others to rebuild against the new path without
breaking things and make us rush because you decide.

This is *strong* reasons to revert your change.

My suggestion is to wait Allan runs the final /usr move, because
fixing the path in all _not_ packaged scripts or webapps  is boring
and not necessary.
And we will need to be synchronized for the final step, so it will be
easy to fixed package at this moment.
Let's be pragmatic.


Sébastien "Seblu" Luttringer
GPG: 0x2072D77A

More information about the arch-dev-public mailing list