Hello Gunnar, others,
On Wed 19 Aug 2020 at 12:31PM -05, Gunnar Wolf wrote:
> Maybe we could improve on the problem putting it upside down: What if
> systemd stated "Provides:" for their main interfaces? While not every
> provided binary would qualify as a "main interface", I think a line
> such
Hello world,
Dmitry Smirnov Tue, 14 Jul 2020 18:04:45 +1000:
> Not yet, unfortunately. Sorry for inconvenience. I'm going to seek
> CTTE advise on #959174...
I am joining the conversation as an individual (so I'm not wearing any
tech-ctte hat yet), prompted by this. Do note that this has _not
Hello,
My 2ยข on this: speech-dispatcher build-depends on systemd in order to
get systemd.pc and read systemdsystemunitdir from it, but that fails on
experimental buildds because they currently have systemctl installed
instead of systemd:
On Tuesday, 14 July 2020 3:10:39 PM AEST Ross Vandegrift wrote:
> Hi Dmitry - systemctl is still breaking builds in experimental, any
> updates?
Not yet, unfortunately. Sorry for inconvenience. I'm going to seek CTTE
advise on #959174...
--
All the best,
Dmitry Smirnov.
---
Someone once
On Wed, May 20, 2020 at 12:48:18PM -0700, Ross Vandegrift wrote:
> On Thu, May 07, 2020 at 07:52:31PM +0200, Julien Cristau wrote:
> > On Thu, May 07, 2020 at 09:48:34PM +1000, Dmitry Smirnov wrote:
> > > On Thursday, 7 May 2020 7:04:17 PM AEST Julien Cristau wrote:
> > > > This use of Provides is
On Thu, May 07, 2020 at 07:52:31PM +0200, Julien Cristau wrote:
> On Thu, May 07, 2020 at 09:48:34PM +1000, Dmitry Smirnov wrote:
> > On Thursday, 7 May 2020 7:04:17 PM AEST Julien Cristau wrote:
> > > This use of Provides is not acceptable. The systemctl package does not
> > > in any way provide
On Friday, 8 May 2020 3:52:31 AM AEST Julien Cristau wrote:
> That's not what "Provides" means though. What you're saying is
> "systemctl provides a subset of the systemd package's functionality".
> That's not good enough. Provides is much stronger than "there are cases
> where this might be
On Friday, 8 May 2020 1:17:59 AM AEST Shengjing Zhu wrote:
> I can see both php-fpm and systemctl maintainers have good reasons to
> do what they did.
No, php-fpm maintainer had a bad reason to do what he did. His insistence on
using "systemd-tmpfiles" in SysV init script is a _demand_ to
On Thu, May 07, 2020 at 09:48:34PM +1000, Dmitry Smirnov wrote:
> On Thursday, 7 May 2020 7:04:17 PM AEST Julien Cristau wrote:
> > This use of Provides is not acceptable. The systemctl package does not
> > in any way provide the same functionality / interfaces as the systemd
> > package, and as
I can see both php-fpm and systemctl maintainers have good reasons to
do what they did.
So taking one step back, does it make sense to dpkg-divert /usr/bin/systemctl?
On Thursday, 7 May 2020 7:04:17 PM AEST Julien Cristau wrote:
> This use of Provides is not acceptable. The systemctl package does not
> in any way provide the same functionality / interfaces as the systemd
> package, and as such it does not get to pretend that it does and cause
> problems to
On Thursday, 7 May 2020 7:04:17 PM AEST Julien Cristau wrote:
> This use of Provides is not acceptable. The systemctl package does not
> in any way provide the same functionality / interfaces as the systemd
> package, and as such it does not get to pretend that it does and cause
> problems to
On Thursday, 7 May 2020 6:58:49 PM AEST Ansgar wrote:
> So all packages that want systemd-tmpfiles need "Depends: systemd" and
> "Conflicts: systemctl" now? (Same for anything else provided by systemd,
> but not systemctl.)
You are over-reacting. "systemctl" is a niche package that is used only
Control: severity -1 serious
Justification: in the release manager's opinion, makes the package unsuitable
for release
On Thu, May 07, 2020 at 08:19:48AM +1000, Dmitry Smirnov wrote:
> We do not have a problem with "Provides: systemd" in "systemctl" package on
> systems where "systemd" is
Dmitry Smirnov writes:
> On Wednesday, 6 May 2020 9:58:03 PM AEST Ansgar wrote:
>> > Clearly there is a problem in "php7.4-fpm" which should not depend on
>> > "systemd" in first place because it is perfectly functional without
>> > "systemd- tmpfiles" and without "systemd" as far as I can tell.
On Wednesday, 6 May 2020 9:58:03 PM AEST Ansgar wrote:
> > Clearly there is a problem in "php7.4-fpm" which should not depend on
> > "systemd" in first place because it is perfectly functional without
> > "systemd- tmpfiles" and without "systemd" as far as I can tell.
>
> It might or might not
Dmitry Smirnov writes:
>> `php7.4-fpm` has "Depends: systemd" because it wants `systemd-tmpfiles` to
>> be available, not for the `systemctl` program. `systemctl` doesn't
>> provide `systemd-tmpfiles`, so that is an example why the Provides that
>> was added in the last upload is not correct.
>
>
On Wednesday, 6 May 2020 7:48:16 PM AEST Ansgar wrote:
> Control: severity -1 serious
Why that again? Didn't I explain that "systemctl" provides enough
functionality to suffice as a lightweight init and service management system
for application containers implementing sufficient subset of
Control: severity -1 serious
Dmitry Smirnov writes:
> On Wednesday, 6 May 2020 8:28:09 AM AEST Ansgar wrote:
>> The last upload adds `Provides: systemd`, but `systemctl` obviously
>> doesn't provide the same (or similar) functionality as the `systemd`
>> package.
>
> It does provide similar
On Wednesday, 6 May 2020 8:28:09 AM AEST Ansgar wrote:
> The last upload adds `Provides: systemd`, but `systemctl` obviously
> doesn't provide the same (or similar) functionality as the `systemd`
> package.
It does provide similar functionality under certain circumstances such as
inside
Package: systemctl
Version: 1.4.4147-2
Severity: serious
The last upload adds `Provides: systemd`, but `systemctl` obviously
doesn't provide the same (or similar) functionality as the `systemd`
package.
As a random example, systemd-tmpfiles and other utilities are not
provided by `systemctl`.
21 matches
Mail list logo
