FWIW, I got my toolserver account two weeks ago. Colin
2009/7/21, Ævar Arnfjörð Bjarmason <avarab(a)gmail.com>om>:
On Tue, Jul 21, 2009 at 2:01 PM,
Aude<aude.wiki(a)gmail.com> wrote:
On 7/19/09, Ævar Arnfjörð Bjarmason
<avarab(a)gmail.com> wrote:
On Sat, Jul 18, 2009 at 5:45 PM, Colin
Marquardt<cmarqu42(a)googlemail.com>
wrote:
2009/7/18 Ævar Arnfjörð Bjarmason
<avarab(a)gmail.com>om>:
> Cassini is back up now thanks to the Toolserver admins, since it's
> inactive at the moment I'm going to set it up with a rendering stack
> that'll run until Wikimania, of course if the database server pops
> back up at the beginning of august we may actually have something
> production ready by that time.
>
> So what I'm going to do with it (pending objections) is:
>
> * Set up a Planet.osm -> Cassini mirror
> * Set up a multilingular PostGIS database (see
>
https://bugzilla.wikimedia.org/show_bug.cgi?id=19668). This should
> take around 50GB of space from the 345G /sql partition
For my hiking map (
http://opentiles.com/cmarqu/), I would need some
more keys imported with osm2pgsql. My default.style file is
http://code.google.com/p/mapnik-utils/source/browse/sandbox/cascadenik/hike…
Not sure if you want to support that already, but I thought I'd mention
it.
What I'm setting up now on Cassini is just a temporary hacky system
that rightly should be setup on the database server / tileserver,
it'll be taken down again once the rest of this stuff is up.
I'm taking notes on what I'm doing here as I go along:
https://wiki.toolserver.org/view/Cassini/Setup_notes
and here are some scripts in MW svn I'm writing:
http://svn.wikimedia.org/viewvc/mediawiki/trunk/tools/osm-tools/
It would be easy to add stuff the hiking map needs, but this is all a
temporary setup anyway so we shouldn't move any production stuff to it
I think.
_______________________________________________
Maps-l mailing list
Maps-l(a)lists.wikimedia.org
https://lists.wikimedia.org/mailman/listinfo/maps-l
Is there anything you need help with, such as setting up the database.
These are the notes I'm taking when setting it up:
http://www.mediawiki.org/wiki/Extension:SlippyMap/Cassini/Setup_notes
and here are the scripts I'm writing to make things easier:
http://svn.wikimedia.org/viewvc/mediawiki/trunk/tools/osm-tools/osm2pgsql-s…
What needs to be done is:
* Hack up an export script:
https://bugzilla.wikimedia.org/show_bug.cgi?id=19860
https://bugzilla.wikimedia.org/show_bug.cgi?id=19861
This is an easy task someone else could do.
* Set up apache/mod_tile
I'll try to do this when the osm2pgsql --slim import finishes.
* Set up planet.osm/osm2pgsql mirroring so that it's automatically updated.
I haven't looked into how to do this. It would be very helpful if
someone would read up on the optimal way to configure postgres / how
to do this updating so that I can stick something in cron that'll keep
it up-to-date.
* Set up squid in front of the cgi-bin/export script & tile server
Someone just needs to give me a squid config I can use, or
alternatively it would probably be better to adapt the tile cache Jens
set up so that it can cache cassini too.
For the setup on the toolserver, I am fine with
trying things and see
how they work. But, I hope that we won't be locked into using one
set of tools for rendering and working with the OSM data. It would be
good to try alternative ways of styling the data that fit with what
Wikimedia needs, beyond just supporting multiple languages.
Oh yes, what I'm doing now is just a hacky way of getting something to
run. We'll want some alternative styles later (if we can get someone
to maintain it!) but in the meantime I'm just poking the box because
it's better than having it idle.
Of course for some osm2pgsql styles (like the hikingmap) we'll need to
add extra fields to the postgis database.
Whether we're going to need one "production" rendering DB that does
the main rendering Wikipedia wants to use and another one with a lot
of fields or just shove everything in the DB & replicate it or
something else is something we're going to have to figure out once we
actually have a database server and some users to run tools.
Speaking of which why don't we have any users yet? Is everything
stalled on getting the DB machine up or are there some tools that can
just use Cassini as-is? How is Unix-account approval being managed
(who's taking care of that?).
_______________________________________________
Maps-l mailing list
Maps-l(a)lists.wikimedia.org
https://lists.wikimedia.org/mailman/listinfo/maps-l