Hi Emma,
that's a weird situation EPSG:900913 is not actually part of the EPSG
database, it's something
that the GeoServer community made up (it's "Google" written using numbers)
and as such, it
can only exist if it's defined in the user_projections/epsg.properties file.
I'm guessing you started with a completely empty data directory, rather
than the demo ones?
In that case GeoServer would start using a epsg.properties file found in
the GeoServer own libraries,
but would then stop doing that once you define your own
explicit epsg.properties, losing the EPSG:900913
definition.
The epsg.properties would normally have these contents, you should be
adding your own extra projection to it:
https://github.com/geoserver/geoserver/blob/main/src/main/src/main/java/org/vfny/geoserver/crs/user_epsg.properties
I can't find a report of this issue in the ticket tracker, could you please
add one?
https://osgeo-org.atlassian.net/projects/GEOS/summary
Cheers
Andrea
On Wed, Dec 20, 2023 at 10:18 PM Emma Flukes
wrote:
> We recently needed to add a custom SRS for GDA2020 / Australian Albers. We
> did this successfully by creating the user_projections folder and
> epsg.properties file and adding the Geoserver one line definition from
> https://epsg.io/9473
> Since then, GWC started failing for layers using Google Global Mercator
> (EPSG:900913). The behaviour was that no new tiles were generated on disc
> when interacting with cached layers at new zoom levels (no errors in logs).
> Seeding the cache also created no new tiles and this was the only error in
> the logs.
> *ERROR [seed.MTSeeder] - Problem communicating with GeoServer*
> *org.geowebcache.GeoWebCacheException: Problem communicating with
> GeoServer*
> We realised 900913 had disappeared from Geoserver's known SRS list since
> adding the custom SRS. Other common SRS such as EPSG:4326 that we also use
> for GWC continue to work fine.
>
> We then added EPSG:900913 as an additional custom definition to the
> epsg.properties file, using the one-line Geoserver definition from
> https://epsg.io/900913. Google Global Mercator was again available from
> the SRS list and could be used GWC seeding as usual. We confirmed that new
> tiles were again being created on disc in the GWC directory. However, this
> custom defined 900913 SRS appears to be 'shifted' by approximately 15km due
> north (at -22 latitude), but only for tiles rendered by GWC. Any layers
> served out via WMS in 900913 without tile caching switched on are rendered
> fine, after the addition of the custom 900913 definition.
>
> GWC is using the inbuilt gridset for 900913 (which has previously worked
> fine).
>
> This is the exact contents of our epsg.properties file:
> 900913=PROJCS["Google_Maps_Global_Mercator",GEOGCS["WGS
> 84",DATUM["WGS_1984",SPHEROID["WGS
> 84",6378137,298.257223563]],PRIMEM["Greenwich",0,AUTHORITY["EPSG","8901"]],UNIT["degree",0.0174532925199433,AUTHORITY["EPSG","9122"]],AUTHORITY["EPSG","4326"]],PROJECTION["Mercator_1SP"],PARAMETER["central_meridian",0],PARAMETER["scale_factor",1],PARAMETER["false_easting",0],PARAMETER["false_northing",0],UNIT["metre",1,AUTHORITY["EPSG","9001"]],AXIS["Easting",EAST],AXIS["Northing",NORTH],EXTENSION["PROJ4","+proj=merc
> +a=6378137 +b=6378137 +lat_ts=0 +lon_0=0 +x_0=0 +y_0=0 +k=1 +units=m
> +nadgrids=@null +wktext +no_defs"],AUTHORITY["EPSG","900913"]]
> 9473=PROJCS["GDA2020 / Australian
> Albers",GEOGCS["GDA2020",DATUM["Geocentric_Datum_of_Australia_2020",SPHEROID["GRS
> 1980",6378137,298.257222101],TOWGS84[0,0,0,0,0,0,0]],PRIMEM["Greenwich",0,AUTHORITY["EPSG","8901"]],UNIT["degree",0.0174532925199433,AUTHORITY["EPSG","9122"]],AUTHORITY["EPSG","7844"]],PROJECTION["Albers_Conic_Equal_Area"],PARAMETER["latitude_of_center",0],PARAMETER["longitude_of_center",132],PARAMETER["standard_parallel_1",-18],PARAMETER["standard_parallel_2",-36],PARAMETER["false_easting",0],PARAMETER["false_northing",0],UNIT["metre",1,AUTHORITY["EPSG","9001"]],AXIS["Easting",EAST],AXIS["Northing",NORTH],AUTHORITY["EPSG","9473"]]
>
> Specific Qs:
> Why has adding the custom SRS for GDA2020 caused Google Mercator to
> disappear from the SRS list? Have we missed any steps or defined this
> incorrectly? Is this a bug?
> Why has re-adding Google Mercator as a custom SRS worked fine for WMS, but
> GWC now has a 'shifted' projection issue? Have we defined this incorrectly?
> Is anyone successfully using both 9437 and 900913 with GWC?
>
> This is the first custom SRS we've defined so I'm not sure whether the
> same disappearing Google Mercator issue would occur with an alternate SRS.
>
> We're running Geoserver version 2.24.0
>
> Thanks in advance for your help, this one has me stumped!
>
>
>
> This email is confidential, and is for the intended recipient only.
> Access, disclosure, copying, distribution, or reliance on any of it by
> anyone outside the intended recipient organisation is prohibited and may be
> a criminal offence. Please delete if obtained in error and email
> confir
