[Pkg-postgresql-public] Bug#759725: #759725: postgresql-common: non-synchronous service postgresql start/stop/reload

[beuc at debian.org]

Hi,

On Mon, Apr 13, 2015 at 05:50:54PM +0200, Christoph Berg wrote:
> Re: beuc at debian.org 2015-04-13 <20150413103816.GA15842 at mail.beuc.net>
> > Control: reopen -1
> > 
> > I explained in detail that the bug is only partially fixed and how
> > this is causing issues.
> > 
> > I provided 2 fixes:
> > - reverting back to init.d is a fix.
> > - applying the systemd rules from RHEL is another fix.
> > 
> > Please explain your behavior, I can't find a good faith explanation
> > for it.
> 
> Reverting is not an option. And throwing in a lengthy file from a
> different operating system is not a patch.
> 
> > On Mon, Apr 13, 2015 at 12:24:31PM +0200, Christoph Berg wrote:
> > > Re: beuc at debian.org 2015-04-08 <20150408154517.GA12244 at mail.beuc.net>
> > > > Here's my patch:
> > > >   rm -f /lib/systemd/system/postgresql*
> > > > and fallback to init.d/.
> 
> Sarcastic comments like this are not particularly helpful.

I don't allow you to call me sarcastic when I'm not.

Also if you want to discuss this more precisely, I suggest elaborating
more than "Reverting is not an option" (because I don't see why not).

Last but not least, I don't know enough about systemd and postgresql
to provide a patch.  Nonetheless, there is a Debian-specific
regression, hence this bug needs to stay open.


> So let's get back on topic. What do you think should be fixed in
> detail in the package?

Let me know what is unclear in:
https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=759725#37
(namely: "- However 'stop' is not fixed (still async).")
completed with a use case in:
https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=759725#46
(namely: "Can you make the 'stop' procedure synchronous as well ?")
and I'll elaborate.

Cheers!
Sylvain