Troy Wu wrote
Howdy, all.
New here. Just built an OSM tile-server manually against Ubuntu 12.04
LTS, and wanted to have multiple rendering pipelines.
Can you describe a bit more about what you are trying to achieve with those
multiple rendering pipelines? There are some special cases I can
Hi,
I am not sure I fully understand what you are trying to do, but the part
of the code that seems most likely what you are looking for is the
function pgsql_out_relation (
https://github.com/openstreetmap/osm2pgsql/blob/master/output-pgsql.c#L693
). First I thought you might be able to use the
On 01/04/2014 10:53 AM, yvecai wrote:
My question is simple:
Is it possible to disable tile throttling for a particular client ?
Sort of. Mod_tile does have an internal white-list where you can add
ip-addresses that are excluded from the tile throttling.
However, currently there isn't a way to
Hello everyone,
I am planning on tagging another release of osm2pgsql and would like to
ask if anyone is aware of any issues with the current osm2pgsql that
would prevent it from being a suitable candidate for a stable release.
The last release was done in mid April.
The main new feature of
Hello everyone,
I would like to mention that I have committed a new feature to mod_tile
/ renderd and would appreciate any feedback or comments.
Mod_tile and renderd are now prepared for the possibility to
parameterize the mapnik style-sheet on the fly on a (meta)tile by
(metatile) basis.
For
, Kai Krueger wrote:
Hello everyone,
I would like to mention that I have committed a new feature to mod_tile
/ renderd and would appreciate any feedback or comments.
Mod_tile and renderd are now prepared for the possibility to
parameterize the mapnik style-sheet on the fly on a (meta)tile
Hello everyone,
It appears that the default method of JOSM to authenticate to the OSM
API is still Basic Auth. Although JOSM does allow for OAuth, the way
the authentication dialog in the settings is structured, I would imagine
from a usability perspecitve that a good many users of JOSM will
Nick Whitelegg-2 wrote
Is there a way to tell osm2pgsql to *only* import the ways into
planet_osm_line and not put anything into planet_osm_point at all?
One option is to use the lua tag processing features. That gives you a lot
of flexibility in determining what to import into the database.
Christoph Hormann wrote
It always amazes me how much the map style influences the mapping
practice, not only by obvious tagging for the renderer but also in the
form of subtle priorities. Having two separate styles - a 'data
verification' style and a 'presentation style' would help
On 07/15/2013 02:03 AM, Andy Allan wrote:
I realise I'm in danger of sounding like a broken record here, but it
would be extremely valuable to work on a unit test framework for
osm2pgsql. The lack of confidence in whether refactored code gives the
same output as battle-hardened existing code
serialisation code.
We'll keep our eyes open for other possible sources of failure.
Please do. Hopefully the added error reporting I included in renderd
might give a hint as to which exception is actually thrown which should
give a clue to why.
Kai
On Sat, Aug 17, 2013 at 1:47 PM, Kai
to my query and I will definitely continue to
look into using the latest mod_tile/renderd code to work - in my
specific case - with British National Grid projection, 27700.
Jason
On 26 March 2013 04:44, Kai Krueger kakrue...@gmail.com
mailto:kakrue...@gmail.com wrote
Does running ./autogen.sh before running ./configure help?
--
View this message in context:
http://gis.19327.n5.nabble.com/problem-with-newest-compile-of-renderd-tp5773826p5773845.html
Sent from the Developer Discussion mailing list archive at Nabble.com.
Peter Körner wrote
Hi
Am 24.07.2013 06:52, schrieb Kai Krueger:
I was curious to see how it compares in speed and quality of calculated
routes to the other engines, so I took the liberty to add it to
http://apmon.dev.openstreetmap.org/routing
It seems the labeling is wrong. When using
Ander Pijoan wrote
[...]
An improvement could be to add a route ( /tiles/:z/:x/:y.json ) to get
when
someone asked for a tile and if it doesn't exist, generate and return it.
Would it be possible to write it as a tirex plug-in? Then all of that
(automatic generation, caching, load management,
Hi,
this looks like an interesting addition to the set of routing engines
already available for OSM data.
I was curious to see how it compares in speed and quality of calculated
routes to the other engines, so I took the liberty to add it to
http://apmon.dev.openstreetmap.org/routing
That page
NopMap wrote
Kai Krueger wrote
I was curious to see how it compares in speed and quality of calculated
routes to the other engines, so I took the liberty to add it to
http://apmon.dev.openstreetmap.org/routing
I took the liberty of comparing the result on that page to the direct
call. How
gurpinder chahal wrote
[...]
But nothing changed in my map. Do I need to import gurmukhi fonts
separately or
I need to do something else?
Please suggest something.
It probably depends on what program you use to drive the mapnik library. If
you are using renderd, then you need to make sure
On 05/22/2013 11:04 AM, Tom Hughes wrote:
On 22/05/13 17:47, Kai Krueger wrote:
So, unless you are using the memcached backend, this commit in theory
should be fine and I have done a bunch of load testing with ab to
confirm this. (Although limited ab testing can't properly simulate real
world
On 05/22/2013 11:27 AM, Tom Hughes wrote:
On 22/05/13 18:20, Kai Krueger wrote:
On 05/22/2013 11:04 AM, Tom Hughes wrote:
So if a new child is started, and multiple requests arrive more or
less simultaneously to different threads in that process, then they
will both try and allocate
On 05/22/2013 01:25 PM, Sven Geggus wrote:
Kai Krueger kakrue...@gmail.com wrote:
I'll try and fix this tonight and hopefully that will then indeed solve
the instability issues Sven and Andy have seen.
With tuning Apache to use many threads per process and set the number of
requests before
ref, so
you can't replace the individual ways with a single relation line string.
Kai
regards
Peter
Am 09.05.2013 23:02, schrieb Kai Krueger:
Hello everyone,
In response to the previous thread, I have cleaned up the initial proof of
concept patch, expanded it somewhat and I have now
much appreciate any feedback on the new lua
tag-transform possibilities.
Kai
Kai Krueger wrote
In response to Richard's suggestion, I hacked up a proof of concept
implementation of a tag filter in lua last week.
Currently it allows you to write a filter function (one for nodes, one
Sven Geggus wrote
Kai Krueger lt;
kakrueger@
gt; wrote:
The question now is what to do with this? Is there enough interest for it
to
be worth cleaning it up and committing it?
I once did a somewhat ugly hardcoded hack for converting width and height
tags to useful floating point
In response to Richard's suggestion, I hacked up a proof of concept
implementation of a tag filter in lua last week.
Currently it allows you to write a filter function (one for nodes, one for
ways and one for relations) in lua that takes in a set of tags and returns a
(potentially) transformed
RainerU-2 wrote
Hello Kai,
Am 14.04.2013 21:47, schrieb Kai Krueger:
Perhaps I don't quite understand what you need, but does osm2pgsql not
already do this?
osm2psql creates a way entry in the planet_osm_line table for each member
of a
route relation. The name tag of the relation
, I will tag the current snapshot (
064e2267afff0706e3c208664fdcb45243ed982e ) as a release and mark it as
version 0.82.0 later today.
Kai
On 04/08/2013 12:26 AM, Sarah Hoffmann wrote:
On Sat, Apr 06, 2013 at 05:40:35PM -0600, Kai Krueger wrote:
Hello everyone,
as just mentioned
RainerU-2 wrote
Hello,
For map rendering I needed the local names of route relations such as
name:de,
name:fr. As osm2pgsql does only copy the name tag from the relation to the
related ways, I modified the program to copy all the name:xx tags to
route_name:xx on the ways.
Perhaps I don't
Hello everyone,
In the past the support for OSM software in the various package
repositories have not been great. Either the software has not been
included at all, or it has been rather out of date, making installing
things like OSM tileservers unnecessarily complex. One key complaint I
have
SimonPoole wrote
Besides all the comments that have already been made, I don't think that
unilaterally declaring the licence to be GPL is a particularly good idea.
There is a COPYING file in the mod_tile directory with the full legal text
of the GPL2 and that has been there since 2007 according
Hello everyone,
as many people seem to prefer using git over svn, we have decided to
convert the master repository of osm2pgsql and mod_tile over to git.
Therefore, from now on the svn is no longer the upstream repository.
Instead these will be
https://github.com/openstreetmap/osm2pgsql
On 04/03/2013 11:08 AM, Andy Allan wrote:
On 3 April 2013 16:10, Kai Krueger kakrue...@gmail.com wrote:
Hello everyone,
as many people seem to prefer using git over svn, we have decided to convert
the master repository of osm2pgsql and mod_tile over to git.
That's fantastic news!
Previously
On 04/03/2013 10:34 AM, SomeoneElse wrote:
Kai Krueger wrote:
Therefore, from now on the svn is no longer the upstream repository.
Instead these will be
https://github.com/openstreetmap/osm2pgsql
https://github.com/openstreetmap/mod_tile/
Is it planned to keep SVN around (if not up
Sven Geggus wrote
Jochen Topf lt;
jochen@
gt; wrote:
Frederik uses Tirex and it works reliable for him
As do I on tile.openstreetmap.de. NOP is also using Tirex on
wanderreitkarte.de
This is only solved by somebody taking up the torch and running with it.
Jepp!
I currently
Bernard Fouché wrote
...
as a simple user of Open Street Map, Nominatim is needed as much as tile
rendering. Currently I (stupidly?) have osm2pgqsl to run twice, once for
Nominatim, once for Mapnik. Maybe a single run is possible?
Currently that is not possible. The tile rendering output
On 03/28/2013 09:50 AM, Simone Cortesi wrote:
On Thu, Mar 28, 2013 at 4:48 PM, Jochen Topf joc...@remote.org wrote:
Tirex is only a replacement for renderd, unfortunately it still needs mod_tile.
s/mod_tile/renderd in my question.
is there a simple way/tutorial to replace renderd with tirex?
On 03/28/2013 10:42 AM, Simone Cortesi wrote:
On Thu, Mar 28, 2013 at 5:39 PM, Jochen Topf joc...@remote.org wrote:
Nothing simple about running a tile server. :-)
Thanks Jochen, this is the only thing I got sound clear :)
Installing and running a standard tile server, at least on ubuntu,
On 03/28/2013 10:45 AM, Simone Cortesi wrote:
On Thu, Mar 28, 2013 at 5:43 PM, Kai Krueger kakrue...@gmail.com wrote:
Is there a particular reason though you do want to replace renderd with
tirex?
having two or more styles based on the same data. this is something I
believe tirex will allow me
On 03/28/2013 11:59 AM, Simone Cortesi wrote:
On Thu, Mar 28, 2013 at 5:52 PM, Kai Krueger kakrue...@gmail.com
mailto:kakrue...@gmail.com wrote:
Renderd can do that just fine as well. Currently there is a hard
coded limit
of 10 in the source code of renderd, but that would be easy enough
On 03/27/2013 03:12 AM, Bernard Fouché wrote:
Le 26/03/2013 17:16, Kai Krueger a écrit :
There have been some major refactorings a couple of days ago in
mod_tile / renderd to support new storage backends. That is when the
error you reported got introduced. So if you take a snapshot from
Andy Allan wrote
On 27 March 2013 14:09, Pieren lt;
pieren3@
gt; wrote:
OSM data tools are mostly developed by volunteers on their free
time.
And I'm sure we all strive to make high-quality tools. If someone
encounters bugs, like crashes and broken documentation, then even just
Hello everyone,
I would like to announce a new OSM mailinglist called tile-serving [1]
to discuss and aid the development of the OSM tile serving infrastructure.
Its purpose is modelled after the rails-dev, geocoding or osrm
mailinglists that already exist.
One of its primary purposes is
On 03/26/2013 04:52 AM, Bernard Fouché wrote:
Hi All,
I'm new to openstreetmap and I've nearly succeeded in having a working
tile server on a Fedora 18 box.
However I experience problems with mod_tile/renderd:
- all wiki pages I find simply state for instance:
Get and install mod_tile
On 03/07/2013 08:36 AM, Jason Lee wrote:
Hi all,
It might seem an odd question but I'll give it a shot - would
mod_tile/renderd support the dynamic generation of map tiles in a map
projection other than web mercator? If not, then would it be
possible/how difficult would it be to modify the
Hello everyone,
I just wanted to let you know that I have committed a new feature to
mod_tile / renderd that might be of interest to some.
So far both mod_tile and renderd had the assumption that (meta)tiles
would be stored on a posix filesystem deeply baked into their
architecture. While
On 03/07/2013 08:36 AM, Jason Lee wrote:
Hi all,
It might seem an odd question but I'll give it a shot - would
mod_tile/renderd support the dynamic generation of map tiles in a map
projection other than web mercator? If not, then would it be
possible/how difficult would it be to modify the
On 02/09/2013 04:13 PM, Lynn W. Deffenbaugh (Mr) wrote:
I'm running Osmosis Version 0.34 to pull and apply minutely updates to
my database for use by mod_tile/renderd. These updates began failing
today with the following information:
sudo apt-get install osmosis
Any hints on the command to
Matt Patterson wrote
Homebrew were packaging osm2pgsql, which was super handy.
Yes, it would be nice to get the osm rendering tool chain into as many
packaging repositories as possible to make it easier for people to install
and use it on various platforms.
Matt Patterson wrote
They removed
On 12/19/2012 06:19 AM, Svavar Kjarrval wrote:
Hi.
I was wondering what methods you use to make osm2pgsql be quiet when run
by crontab. It is very annoying to receive a everything-is-normal
message every time I run it. I run it from within a shellscript since I
need to execute other commands
On 12/14/2012 01:08 PM, Dominik Perpeet wrote:
The idea behind the same location was so that any documentation or
links that still point to that location would automatically be
up-to-date again. That would probably be easier than finding any and
all documentation for windows and update them.
Hello everyone,
in order to try and help ensure and improve the quality of the osm
rendering stack, I have registered osm2pglsql and mod_tile / renderd on
Coverity[1] for their static analysis tools.
Static analysis tools check the source code during compilation time to
try and find various
On 12/12/2012 11:01 AM, Dominik Perpeet wrote:
Would anyone be interested in a current osm2pgsql Windows build (Win32 +
x64, both with 32 bit id space)?
In the near future I will link the zipped binary package on the wiki:
http://wiki.openstreetmap.org/wiki/Osm2pgsql
Great, thank you.
It
On 12/13/2012 11:04 AM, Dominik Perpeet wrote:
It seems like osm2pgsql for windows used to be kept at
http://tile.openstreetmap.org/osm2pgsql.zip and is still from 2010.
I wonder if your version can be put up as a replacement for this at
the same location?
Maybe eventually. For now I have
Matt Amos wrote
i'd sound a note of caution about having separate clean and detailed
styles. we sort-of did that before with mapnik and osmarender respectively
and... well, we don't have osmarender any more.
I doubt that having to maintain two styles was what killed osmarender. It
was a
Dane Springmeyer wrote
I'm sorry about not providing Windows binaries yet for Mapnik 2.x. The
holdup is that I have a dev environment working that is running Visual
Studio 2010, and I need to get a parallel setup running Visual Studio 2008
for support compiling the python bindings. Its a silly
Tom MacWright wrote
The biggest problem with the Mapnik stylesheet right now is that it's in
SVN. Not the technology, but the fact that this gives people without
commit
access to that repository no clear way to contribute. There is no way to
'just do it' until the style is actually maintained
Stefan Elspaß wrote
Hi Kai,
If I read your renderd log correctly, mod_tile is sending rendering
requests
with priority dirty rather than the missing priority. The dirty
priority
is a background priority in which mod_tile does not wait for the result
and
immediately returns the tile (or
Stefan Elspaß wrote
I compiled mod_tile again with svn-r28921 containing today's change from
apmon which writes the actual load into error.log. The result is strange:
[info] [client xx.xx.xx.xx] Load (1202590843) larger max_load_missing
(50). Return HTTP_NOT_FOUND., referer:
If I read your renderd log correctly, mod_tile is sending rendering requests
with priority dirty rather than the missing priority. The dirty priority
is a background priority in which mod_tile does not wait for the result and
immediately returns the tile (or in case of a missing tile a 404).
So
Ákos Maróy wrote
On 07/11/12 22:12, sly (sylvain letuffe) wrote:
osm2pgsql only uses it's multipolygon building ability when relations
have a tag type with value multipolygon or boundary (unless
patched) relation n°25 for exemple hasn't.
it hasn't on purpose. the relations I use are not used
Hello everyone,
I was playing around with vector tile rendering and in particular with
Cover[1], which was a GSoC project this year. It can act as a tirex
backend plugin and thus enable the standard tool chain of mod_tile/tirex
to serve json vector tiles, including the caching and expiry,
On 10/21/2012 04:15 PM, Dane Springmeyer wrote:
On Oct 18, 2012, at 5:05 PM, Kai Krueger wrote:
[...]
I am not sure if this is essential for getting the rendering stack back up
and running on windows, but afaik the most recent version of osm2pgsql on
windows is from early 2010. It would
On 10/17/2012 09:04 PM, sly (sylvain letuffe) wrote:
However, it is not limited anymore to what people think should go into the
next API 0.7 version (I think it's even worse when you ask non-techies people
to find a solution themself) but to something gathering ideas of wishes they
find usefull
Dane Springmeyer wrote
I believe I'm right in saying that he doesn't have a working test setup
at the moment since we moved to mapnik 2 and we need to sort that out but
equally moving to git may be an issue because I believe Windows support
in git is a bit patchy.
Yes, this is my
Tom Hughes-3 wrote
On 17/10/12 22:17, Andy Allan wrote:
On 16 October 2012 00:17, Tom Hughes lt;
tom@
gt; wrote:
The way it works is this - you don't worry about translations as such at
all. You just make sure strings are translatable and we commit that and
then
the translators get to
Mikel Maron wrote
* Create a new map style intended to be the default face of OSM, but
leave the current OSM.org Mapnik style as-is. It works beautifully as an
editor's basemap due to the dense inclusion of
all data. Keep it, but add a new one that's for non-editors to look at.
My suggestion
mod_tile by default uses a directory hash which mixes X and Y coordinates
into a 40 bit hash function and then splits that into a 5 level deep
directory structure.
Why it normally breaks only at Z23 and not at Z21 I am not entirely sure,
but possibly if you look at Australia, it might already
On 10/09/2012 02:31 PM, Tom Hughes wrote:
On 09/10/12 21:24, Tom MacWright wrote:
All those are independent third party sites created by individuals
and are not directly related to core site.
Aren't they using the same database somehow?
No idea.
What we were talking about in
On Sun, Sep 16, 2012 at 2:17 PM, mar...@gmx.eu wrote:
Hello Kai,
I have updated the package for 10.04 as well now. It will take another
couple of hours though to propagate through the system.
Thank you! Did you update just your repository (ppa:kakrueger/openstreetmap)
or is the new
to
ODbL). But if you run osm2pgsql in --create mode, it will
automatically delete the old database for you and change the schema.
So you don't have to worry about that.
Kai
Grüße
Markus
Original-Nachricht
Datum: Thu, 13 Sep 2012 16:58:45 -0600
Von: Kai Krueger kakrue
On Sat, Sep 15, 2012 at 11:15 AM, mar...@gmx.eu wrote:
Hello Kai,
thanks for this quick reply.
What version of Ubuntu do you use? I have updated the packages for
12.04LTS (precise), but I haven't updated them for the other versions
of Ubuntu yet.
Oh, I see. Is there any chance you might
Hello everyone,
I have just changed the default compile of osm2pgsql to use 64bit IDs
instead of 32bit IDs.
The current highest node ID is 1.9 billion. As signed 32 bit ints can
only hold ~2.1 billion IDs, it is likely that the (signed) 32bit ID
space will run out in a couple of months.
Hi,
on what platform are you trying to compile osm2pgsl?
Some systems that don't have lseek64 seem to use a 64bit off_t pointer
anyway. So the normal lseek is fine for large files. I have added a
compile check for this condition a couple of days ago [1].
If your system doesn't support large
figure out how to build osm2pgsql on Windows...
Kai
Thanks for the explanation.
Cheers,
Steven
Op zondag 2 september 2012 18:09:11 schreef Kai Krueger:
Hi,
on what platform are you trying to compile osm2pgsl?
Some systems that don't have lseek64 seem to use a 64bit off_t pointer
On 7/15/2012 6:30, Frederik Ramm wrote:
Hi,
On 15.07.2012 02:23, Kai Krueger wrote:
If one chooses to use the AddTileConfig or AddTileMimeConfig
directives in the Apache site config instead,
Which is the usual procedure for Tirex-based systems since they lack a
renderd.conf (even though
On 07/12/2012 01:10 AM, Jochen Topf wrote:
On Thu, Jul 12, 2012 at 08:45:38AM +0200, Igor Brejc wrote:
Why not use TileJSON? http://mapbox.com/developers/tilejson/
+1
I hadn't known about this before but just looked at it and it seems to be
a well thought-out and documented standard.
OK,
On 07/14/2012 05:12 PM, Frederik Ramm wrote:
Hi,
On 14.07.2012 22:57, Kai Krueger wrote:
The values for description, attribution and tiles is taken from new
parameters in the renderd.conf file.
This means that it is unsuitable for anyone running mod_tile without
renderd, e.g. those using
On 07/10/2012 01:38 PM, Lynn W. Deffenbaugh (Mr) wrote:
I've been wondering if it would be possible to put a fixed URL on the
tile and/or API servers that application programs could fetch to
retrieve the current attribution string for that particular tile
server? Something like 0/0/0.txt or
SimonPoole wrote
RSN a large number of sites using OSM data will be reloading their
databases, due to a certain well known change :-).
Well, this time has come closer now...
SimonPoole wrote
It seems as if it would really make sense to make the 64bit ID version
of osm2pgsql the
There are two main components to the storage system of a tile server,
each of which can have different requirements depending on the circumstances
1) Tile storage cache
For the tile storage usually one needs quite a bit of space, but
performance isn't quite as critical. For a general purpose
Евгений wrote
1) I have several ideas about how to implement this project.
Great. Demonstrating that you have a thorough understanding of the project
idea by being able to expand on it with your own ideas is likely a benefit
in the application process.
Евгений wrote
The first thing that
On 03/23/2012 01:49 AM, Valery N. wrote:
So according all logs - all fine, but I recieve 404 :(
I have committed a few more debug messages for the logs on the serving
path. Could you download the latest svn and recompile?
Kai
2012/3/23 Kai Krueger kakrue...@gmail.com mailto:kakrue
Valery N. wrote
For some unknown reasons request like
http://mapserv.test.maps.local/tiles/mapnik/1/1/-1.png raise* 404 error*
on
web server
Well, that is not a valid request. A negative y coordinate is not allowed
and so it will correctly give a 404 error.
Also the rest of you mod_tile
yvecai wrote
I have trouble to access modtile behind an apache alias.
I have tiles here:
http://192.168.1.3/14/8468/5802.pngOK (==
/var/www/14/8468/5802.png)
But if I do:
ServerName dev-yves.dyndns.org
alias /tiles /var/www
Skye Book wrote
Hi Peter,
I was under the impression that Tirex sat behind mod_tile.. It's meant to
replace it?
There is often a lot of confusion about the naming around mod_tile and what
it is.
mod_tile it self is an apache module that is responsible for serving tiles
and deciding which
Hi,
this sounds like the mod_tile module was not actually installed in
apache, which is where LoadTileConfigFile gets defined. Check if the
LoadModule tile_module /usr/lib/apache2/modules/mod_tile.so is loaded,
e.g. in the /etc/apache2/mods-enabled directory.
How did you install mod_tile? Is
On 01/-10/-28163 12:59 PM, Mikel Maron wrote:
Hi
[...]
2) A packaged solution, like a chef recipe, to install everything needed.
We should try and get more of the rendering tool stack into the standard
repositories of Ubuntu and Feodora. That would probably make it easier
to install a tile
Hi,
On 01/-10/-28163 08:59 PM, Jaak Laineste wrote:
Hello,
I'm working on keeping planet database up-to-date. Not successful
yet, it tends to lag behind due to IO speed/configuration (on raid-5
with sata 10K disks - it should work, right?), 12-hour update takes
11-14 hours for me. Anyway,
On 01/-10/-28163 12:59 PM, Peter Körner wrote:
Am 16.12.2011 12:47, schrieb Hartmut Holzgraefe:
On 16.12.2011 11:21, Peter Körner wrote:
At one of the Hack-Weekends someone played around with distributing the
SQL-Commands issued by osm2pgsql via XMPP.
with the SQL command execution,
On 01/-10/-28163 12:59 PM, sly (sylvain letuffe) wrote:
Hi,
We're working
hard on getting the relevant hardware in place to start trialling this
out, but it's a big project.
Many thanks for the insight
The original topic was about replication for rendering, so a comment
on that
Ákos Maróy wrote
on the OSM wiki, I found a number of update options, and I have to say
I'm a bit confused. which is the best option if I want to get all the
updates for the whole planet file, say, each week?
If you are intending to only update the diffs every week or less (as opposed
to
Thanks for catching this regression. As my test database was always set to
fsync=off, I didn't notice this performance regression during the
development of the parallelisation work.
The problem is that in order to allow multiple threads to work through the
pending ways and thus potentially speed
quot;Petr Morávek [Xificurk]quot; wrote
1) fsync = off, synchronous_commit = off
1990 s = 33 min
2) fsync = on, synchronous_commit = off
3075 s = 51 min
Interestingly, most of the time difference comes from the initial import
of relations.
Yes, I noticed that as well, that the initial
On 01/-10/-28163 12:59 PM, Jukka Rahkonen wrote:
Frederik Ramm wrote:
[...]
After this succeeded, I wanted to try to replicate this database, so I
created a pg_dump using the -Fc switch
This is a bad idea because a significant amount of osm2pgsql import time
is spent building indexes, and
Jukka Rahkonen-2 wrote
[...]
For me it takes many hours with the Finnish dataset and if it fails it
happens in some Going over pending ways phase. I will need to make some
further tests some day so I can give you better information.
If it is at the very beginning of the Going over pending
sylvain letuffe wrote
Nice catch.
revision 26892 ran that phase for the same extract at 6.36k/s instead of
0.14k/s
6.4k/s is much more what I would expect from such a small extract. So yes
something is wrong there, but I haven't seen that behavior before and I
can't currently reproduce
Frederik Ramm wrote
I found out that the culprit is in the multipolygon code, where after
finding out that an one-way outer ring is tagged the same as the
multipolgon relation itself, a delete_way_from_output is issued,
presumably to remove that already-generated ring. This leads to a
Frederik Ramm wrote
I think it would be great to share results of osm2pgsql runs among users
- how long does it take to import X on infrastructure Y?
I've made a start here, please add/modify as you see fit:
http://wiki.openstreetmap.org/wiki/Osm2pgsql/Benchmarks
Great, that could be
If you look at e.g. app/controllers/changeset_controller.rb you will find
lines like before_filter :authorize_web, :only = [:list] and
before_filter :authorize, :only = [:create, :update, :delete, :upload,
:include, :close]. The code to those functions are in
Jaak Laineste wrote:
I do not have diff update for my osm2pgsql database, just to make
renderd happy I created /var/lib/mod_tile/planet-import-complete file
with yesterdays date and started to prerender tiles with ./render_list
-m default -a -z 0 -Z 10
But now I noticed that when I
1 - 100 of 143 matches
Mail list logo
