[Slackbuilds-users] postgresql

[Donald Allen]

On Fri, Apr 2, 2010 at 3:15 PM, Robby Workman <rworkman at slackbuilds.org> wrote:
> On Fri, 2 Apr 2010 14:29:47 -0400
> Donald Allen <donaldcallen at gmail.com> wrote:
>
>> On Fri, Apr 2, 2010 at 1:53 PM, /dev/rob0 <rob0 at slackbuilds.org>
>> wrote:
>> > On Fri, Apr 02, 2010 at 01:44:58PM -0400, Donald Allen wrote:
>> >> I do note that the README.SBo says "make sure rc.postgresql is
>> >> executable", which I overlooked on first reading. I still think
>> >> this is a bug, or at least an inconvenience. Why require the user
>> >> to add execute permission? The installation should just take care
>> >> of it.
>> >
>> > No, this is how Slackware works. Most server packages do not come
>> > enabled out of the box. You have to take explicit steps to enable
>> > them. The setup has a section which does this for included server
>> > packages.
>> >
>> > A Slackbuild builds a Slackware package, so it would be wrong for
>> > builds to assume that the service should be enabled by default.
>>
>> Understood, but I knew that at the outset and what I'm saying doesn't
>> violate that. I am neither assuming nor asking for the postgres server
>> to be started by default. I'm saying that the postgres slackbuild goes
>> to the trouble of creating an /etc/rc.d/rc.postgresql, which the user
>> either chooses to execute from /etc/rc.d/rc.local or he doesn't. But
>> if the user *does* choose to do this, my point is that the execute bit
>> ought to be set for him, rather than requiring him to do the chmod
>> manually, which is the case now. If the user doesn't start the server
>> at boot time, the execute permission does no harm and may do some good
>> -- allows him to do
>>
>> /etc/rc.d/rc.postgresql start
>>
>> from a shell. All benefit, no cost, one less manual thing the user
>> doesn't have to do.
>
>
> For what it's worth, this is the way that we *try* to handle
> things (or at least *I* have been doing them this way).  I
> think I was the last person to touch the postgresql build,
> so I guess I missed it on that one.
>
> Basically, the init script should be install with mode 0755 but
> as a rc.postgresql.new; the config() in doinst.sh then checks
> to see if there's already a rc.postgresql on the system, and if
> so, it sets the perms of the .new script to the same as what's
> currently on the system -- I think that will give us the best
> results regardless.

Makes sense.

>
> I'll try to get that fixed up soonish.

Great. This is a minor detail, but, as we know, that's where the devil is.

/Don


>
>
>> > Welcome back to Slackware, BTW.
>>
>> Thanks, that's very nice of you. It's come a long way since 1994 or
>> thereabouts! I'm hoping I've finally found a Linux distribution that
>> will rival the stability and attention to QA and documentation of
>> OpenBSD, but give me the performance and friendly user forums that
>> OpenBSD lacks.
>
>
> Here's hoping :-)
>
> -RW
>
> _______________________________________________
> SlackBuilds-users mailing list
> SlackBuilds-users at slackbuilds.org
> http://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users
> Archives - http://lists.slackbuilds.org/pipermail/slackbuilds-users/
> FAQ - http://slackbuilds.org/faq/
>
>
>