Re: [mapserver-users] Mapserver fastcgi process errors / crash
Sorry, 6.4.1 Do you need the spawn-fastcgi script too? On 22 June 2015 at 17:51, Lime, Steve D (MNIT) steve.l...@state.mn.us wrote: What version? *From:* mapserver-users-boun...@lists.osgeo.org [mailto: mapserver-users-boun...@lists.osgeo.org] *On Behalf Of *Wim Vanbelle *Sent:* Monday, June 22, 2015 10:46 AM *To:* mapserver-users@lists.osgeo.org *Subject:* [mapserver-users] Mapserver fastcgi process errors / crash Hi all, Rather urgent production question. If we have some errors with our shapefiles, our fastcgi processes get killed very rapidly. (we believe this is some rapid error protection?) Is there a way to prevent this? We're running nginx + fastcgi mapserver over unix sockets. I'm aware that we need to look into our shapefiles/dbf's but that's not a simple thing because they need to be updated often. Migration to postgres is also not an option. Please let me know if we can tune this somehow. Anyone experience with this? Thanks in advance! best regards, Wim ___ mapserver-users mailing list mapserver-users@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/mapserver-users
Re: [mapserver-users] Mapserver fastcgi process errors / crash
This is the full configuration: MapServer version 6.4.1 OUTPUT=PNG OUTPUT=JPEG OUTPUT=KML SUPPORTS=PROJ SUPPORTS=AGG SUPPORTS=FREETYPE SUPPORTS=ICONV SUPPORTS=WMS_SERVER SUPPORTS=FASTCGI INPUT=JPEG INPUT=POSTGIS INPUT=SHAPEFILE On 22 June 2015 at 17:55, Wim Vanbelle wimvanbe...@gmail.com wrote: Sorry, 6.4.1 Do you need the spawn-fastcgi script too? On 22 June 2015 at 17:51, Lime, Steve D (MNIT) steve.l...@state.mn.us wrote: What version? *From:* mapserver-users-boun...@lists.osgeo.org [mailto: mapserver-users-boun...@lists.osgeo.org] *On Behalf Of *Wim Vanbelle *Sent:* Monday, June 22, 2015 10:46 AM *To:* mapserver-users@lists.osgeo.org *Subject:* [mapserver-users] Mapserver fastcgi process errors / crash Hi all, Rather urgent production question. If we have some errors with our shapefiles, our fastcgi processes get killed very rapidly. (we believe this is some rapid error protection?) Is there a way to prevent this? We're running nginx + fastcgi mapserver over unix sockets. I'm aware that we need to look into our shapefiles/dbf's but that's not a simple thing because they need to be updated often. Migration to postgres is also not an option. Please let me know if we can tune this somehow. Anyone experience with this? Thanks in advance! best regards, Wim ___ mapserver-users mailing list mapserver-users@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/mapserver-users
[mapserver-users] WMS from netCDF returns white/transparent image
Hello, I am trying to generate WMS services from layers in a NetCDF file. WCS services already work well...While a WMS call does return an image (ex. PNG), it is all white or transparent, whereas a WCS call to the same layer returns an image with a range of pixel values. I have run gdalinfo on the layer to get the data range, and have tried using the data range to set the PROCESSING DIRECTIVE to scale the output (e.g. PROCESSING SCALE=311.76470947266 ,423.52941894531 ). I have also used the SCALE=AUTO directive, but both just return a white image. When I try using WMS on this layer in QGIS, it is correctly projected, but invisible (zoom to extents on the imported WMS layer is correct). Below is the output from gdalinfo to get the original data range, and then the mapfile entry for the layer to see how I have set this up gdalinfo NETCDF:flat_sample.nc:I_lw Band 1 Block=170x1 Type=Float32, ColorInterp=Undefined Min=311.765 Max=423.529 Minimum=311.765, Maximum=423.529, Mean=344.247, StdDev=16.599 NoData Value=9.96920996838686905e+36 Metadata: ipw_desc=incoming thermal (long-wave) radiation NETCDF_DIM_time=14 NETCDF_VARNAME=I_lw standard_name=downwelling_longwave_flux_in_air STATISTICS_MAXIMUM=423.52941894531 STATISTICS_MEAN=344.24709697293 STATISTICS_MINIMUM=311.76470947266 STATISTICS_STDDEV=16.598777681661 units=W m-2 LAYER DATA 'NETCDF:/geodata/watershed-data/8c/8c18e10f-44b0-4462-bf4f-1c27c984aece/flat_sample.nc:I_lw' EXTENT -116.142905556 43.729406 -116.13758 43.732697 METADATA wcs_label downwelling_longwave_flux_in_air layer_title downwelling_longwave_flux_in_air raster_selected yes wms_metadataurl_type FGDC-STD-001-1998 ows_keywordlist wcs_formats GEOTIFF_16 background no wcs_bands_description ISNoBal Timesteps 1-16 ows_abstract incoming thermal (long-wave) radiation wcs_rangeset_name Timesteps wms_metadataurl_href http://129.24.63.85//apps/vwp/datasets/05e73f4f-51d3-417c-9757-e376b5969a9c/metadata/FGDC-STD-001-1998.xml; wcs_bands_values 1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16 legend_display yes queryable no wcs_rangeset_label netCDF flat_sample.nc time_sensitive no ows_srs epsg:26911 wcs_bands_name Timestep wcs_bands_rangeitem _bands ows_title incoming thermal (long-wave) radiationstatic no wms_metadataurl_format text/xml wcs_bands_label Timestep Number wcs_rangeset_axes bands wms_encoding UTF-8 wcs_enable_request * layer_data_units W m-2 annotation_name downwelling_longwave_flux_in_air: 2015-06-19 20:30:29.202558 END # METADATANAME downwelling_longwave_flux_in_air PROCESSING CLOSE_CONNECTION=DEFER PROCESSING SCALE=311.76470947266 ,423.52941894531 #PROCESSING “SCALE=AUTO” PROJECTION init=epsg:26911 END # PROJECTION STATUS ON TILEITEM location TYPE RASTER UNITS METERS END # LAYER It seems I am missing something herewould appreciate any insight into how to return a valid image in WMS Thanks, Bill ___ mapserver-users mailing list mapserver-users@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/mapserver-users
Re: [mapserver-users] Mapserver fastcgi process errors / crash
What version? From: mapserver-users-boun...@lists.osgeo.org [mailto:mapserver-users-boun...@lists.osgeo.org] On Behalf Of Wim Vanbelle Sent: Monday, June 22, 2015 10:46 AM To: mapserver-users@lists.osgeo.org Subject: [mapserver-users] Mapserver fastcgi process errors / crash Hi all, Rather urgent production question. If we have some errors with our shapefiles, our fastcgi processes get killed very rapidly. (we believe this is some rapid error protection?) Is there a way to prevent this? We're running nginx + fastcgi mapserver over unix sockets. I'm aware that we need to look into our shapefiles/dbf's but that's not a simple thing because they need to be updated often. Migration to postgres is also not an option. Please let me know if we can tune this somehow. Anyone experience with this? Thanks in advance! best regards, Wim ___ mapserver-users mailing list mapserver-users@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/mapserver-users
[mapserver-users] Mapserver fastcgi process errors / crash
Hi all, Rather urgent production question. If we have some errors with our shapefiles, our fastcgi processes get killed very rapidly. (we believe this is some rapid error protection?) Is there a way to prevent this? We're running nginx + fastcgi mapserver over unix sockets. I'm aware that we need to look into our shapefiles/dbf's but that's not a simple thing because they need to be updated often. Migration to postgres is also not an option. Please let me know if we can tune this somehow. Anyone experience with this? Thanks in advance! best regards, Wim ___ mapserver-users mailing list mapserver-users@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/mapserver-users
Re: [mapserver-users] Mapserver fastcgi process errors / crash
I’m afraid I don’t have any suggestions – I’m not a big FastCGI user. I was concerned this might be related to an issue specific to 7.0 that will be resolved shortly. Are you seeing MapServer specific errors in the logs? Steve From: Wim Vanbelle [mailto:wimvanbe...@gmail.com] Sent: Monday, June 22, 2015 10:56 AM To: Lime, Steve D (MNIT) Cc: mapserver-users@lists.osgeo.org Subject: Re: [mapserver-users] Mapserver fastcgi process errors / crash This is the full configuration: MapServer version 6.4.1 OUTPUT=PNG OUTPUT=JPEG OUTPUT=KML SUPPORTS=PROJ SUPPORTS=AGG SUPPORTS=FREETYPE SUPPORTS=ICONV SUPPORTS=WMS_SERVER SUPPORTS=FASTCGI INPUT=JPEG INPUT=POSTGIS INPUT=SHAPEFILE On 22 June 2015 at 17:55, Wim Vanbelle wimvanbe...@gmail.commailto:wimvanbe...@gmail.com wrote: Sorry, 6.4.1 Do you need the spawn-fastcgi script too? On 22 June 2015 at 17:51, Lime, Steve D (MNIT) steve.l...@state.mn.usmailto:steve.l...@state.mn.us wrote: What version? From: mapserver-users-boun...@lists.osgeo.orgmailto:mapserver-users-boun...@lists.osgeo.org [mailto:mapserver-users-boun...@lists.osgeo.orgmailto:mapserver-users-boun...@lists.osgeo.org] On Behalf Of Wim Vanbelle Sent: Monday, June 22, 2015 10:46 AM To: mapserver-users@lists.osgeo.orgmailto:mapserver-users@lists.osgeo.org Subject: [mapserver-users] Mapserver fastcgi process errors / crash Hi all, Rather urgent production question. If we have some errors with our shapefiles, our fastcgi processes get killed very rapidly. (we believe this is some rapid error protection?) Is there a way to prevent this? We're running nginx + fastcgi mapserver over unix sockets. I'm aware that we need to look into our shapefiles/dbf's but that's not a simple thing because they need to be updated often. Migration to postgres is also not an option. Please let me know if we can tune this somehow. Anyone experience with this? Thanks in advance! best regards, Wim ___ mapserver-users mailing list mapserver-users@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/mapserver-users
Re: [mapserver-users] WMS from netCDF returns white/transparent image
TC Thank you for your help...I have tried building classes into the layer definition with the same results...a white image. Also used the CLASSITEM [pixel] parameter. I can import the layer into QGIS, and it actually shows the colors I define in the legend, and is projected correctly, but the image is invisibleI've not seen something like this before! Bill - Original Message - From: bhudspeth bhudsp...@edac.unm.edu To: mapserver-users@lists.osgeo.org Sent: Monday, June 22, 2015 9:51:16 AM Subject: [mapserver-users] WMS from netCDF returns white/transparent image Hello, I am trying to generate WMS services from layers in a NetCDF file. WCS services already work well...While a WMS call does return an image (ex. PNG), it is all white or transparent, whereas a WCS call to the same layer returns an image with a range of pixel values. I have run gdalinfo on the layer to get the data range, and have tried using the data range to set the PROCESSING DIRECTIVE to scale the output (e.g. PROCESSING SCALE=311.76470947266 ,423.52941894531 ). I have also used the SCALE=AUTO directive, but both just return a white image. When I try using WMS on this layer in QGIS, it is correctly projected, but invisible (zoom to extents on the imported WMS layer is correct). Below is the output from gdalinfo to get the original data range, and then the mapfile entry for the layer to see how I have set this up gdalinfo NETCDF:flat_sample.nc:I_lw Band 1 Block=170x1 Type=Float32, ColorInterp=Undefined Min=311.765 Max=423.529 Minimum=311.765, Maximum=423.529, Mean=344.247, StdDev=16.599 NoData Value=9.96920996838686905e+36 Metadata: ipw_desc=incoming thermal (long-wave) radiation NETCDF_DIM_time=14 NETCDF_VARNAME=I_lw standard_name=downwelling_longwave_flux_in_air STATISTICS_MAXIMUM=423.52941894531 STATISTICS_MEAN=344.24709697293 STATISTICS_MINIMUM=311.76470947266 STATISTICS_STDDEV=16.598777681661 units=W m-2 LAYER DATA 'NETCDF:/geodata/watershed-data/8c/8c18e10f-44b0-4462-bf4f-1c27c984aece/flat_sample.nc:I_lw' EXTENT -116.142905556 43.729406 -116.13758 43.732697 METADATA wcs_label downwelling_longwave_flux_in_air layer_title downwelling_longwave_flux_in_air raster_selected yes wms_metadataurl_type FGDC-STD-001-1998 ows_keywordlist wcs_formats GEOTIFF_16 background no wcs_bands_description ISNoBal Timesteps 1-16 ows_abstract incoming thermal (long-wave) radiation wcs_rangeset_name Timesteps wms_metadataurl_href http://129.24.63.85//apps/vwp/datasets/05e73f4f-51d3-417c-9757-e376b5969a9c/metadata/FGDC-STD-001-1998.xml; wcs_bands_values 1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16 legend_display yes queryable no wcs_rangeset_label netCDF flat_sample.nc time_sensitive no ows_srs epsg:26911 wcs_bands_name Timestep wcs_bands_rangeitem _bands ows_title incoming thermal (long-wave) radiationstatic no wms_metadataurl_format text/xml wcs_bands_label Timestep Number wcs_rangeset_axes bands wms_encoding UTF-8 wcs_enable_request * layer_data_units W m-2 annotation_name downwelling_longwave_flux_in_air: 2015-06-19 20:30:29.202558 END # METADATANAME downwelling_longwave_flux_in_air PROCESSING CLOSE_CONNECTION=DEFER PROCESSING SCALE=311.76470947266 ,423.52941894531 #PROCESSING “SCALE=AUTO” PROJECTION init=epsg:26911 END # PROJECTION STATUS ON TILEITEM location TYPE RASTER UNITS METERS END # LAYER It seems I am missing something herewould appreciate any insight into how to return a valid image in WMS Thanks, Bill ___ mapserver-users mailing list mapserver-users@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/mapserver-users ___ mapserver-users mailing list mapserver-users@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/mapserver-users
Re: [mapserver-users] WMS from netCDF returns white/transparent image
I think there is an issue with the fact that a request for a single layer from a netCDF file may contain multiple bands In QGIS, the WMS services capability brings in a single layer (with multiple bands) as Singleband Color Data. The image is transparent. When I request the layer with a URL, and I use the Bands=1 (or 2,3,4), I get a white image Bill - Original Message - From: bhudspeth bhudsp...@edac.unm.edu To: mapserver-users@lists.osgeo.org Sent: Monday, June 22, 2015 3:35:27 PM Subject: Re: [mapserver-users] WMS from netCDF returns white/transparent image TC Thank you for your help...I have tried building classes into the layer definition with the same results...a white image. Also used the CLASSITEM [pixel] parameter. I can import the layer into QGIS, and it actually shows the colors I define in the legend, and is projected correctly, but the image is invisibleI've not seen something like this before! Bill - Original Message - From: bhudspeth bhudsp...@edac.unm.edu To: mapserver-users@lists.osgeo.org Sent: Monday, June 22, 2015 9:51:16 AM Subject: [mapserver-users] WMS from netCDF returns white/transparent image Hello, I am trying to generate WMS services from layers in a NetCDF file. WCS services already work well...While a WMS call does return an image (ex. PNG), it is all white or transparent, whereas a WCS call to the same layer returns an image with a range of pixel values. I have run gdalinfo on the layer to get the data range, and have tried using the data range to set the PROCESSING DIRECTIVE to scale the output (e.g. PROCESSING SCALE=311.76470947266 ,423.52941894531 ). I have also used the SCALE=AUTO directive, but both just return a white image. When I try using WMS on this layer in QGIS, it is correctly projected, but invisible (zoom to extents on the imported WMS layer is correct). Below is the output from gdalinfo to get the original data range, and then the mapfile entry for the layer to see how I have set this up gdalinfo NETCDF:flat_sample.nc:I_lw Band 1 Block=170x1 Type=Float32, ColorInterp=Undefined Min=311.765 Max=423.529 Minimum=311.765, Maximum=423.529, Mean=344.247, StdDev=16.599 NoData Value=9.96920996838686905e+36 Metadata: ipw_desc=incoming thermal (long-wave) radiation NETCDF_DIM_time=14 NETCDF_VARNAME=I_lw standard_name=downwelling_longwave_flux_in_air STATISTICS_MAXIMUM=423.52941894531 STATISTICS_MEAN=344.24709697293 STATISTICS_MINIMUM=311.76470947266 STATISTICS_STDDEV=16.598777681661 units=W m-2 LAYER DATA 'NETCDF:/geodata/watershed-data/8c/8c18e10f-44b0-4462-bf4f-1c27c984aece/flat_sample.nc:I_lw' EXTENT -116.142905556 43.729406 -116.13758 43.732697 METADATA wcs_label downwelling_longwave_flux_in_air layer_title downwelling_longwave_flux_in_air raster_selected yes wms_metadataurl_type FGDC-STD-001-1998 ows_keywordlist wcs_formats GEOTIFF_16 background no wcs_bands_description ISNoBal Timesteps 1-16 ows_abstract incoming thermal (long-wave) radiation wcs_rangeset_name Timesteps wms_metadataurl_href http://129.24.63.85//apps/vwp/datasets/05e73f4f-51d3-417c-9757-e376b5969a9c/metadata/FGDC-STD-001-1998.xml; wcs_bands_values 1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16 legend_display yes queryable no wcs_rangeset_label netCDF flat_sample.nc time_sensitive no ows_srs epsg:26911 wcs_bands_name Timestep wcs_bands_rangeitem _bands ows_title incoming thermal (long-wave) radiationstatic no wms_metadataurl_format text/xml wcs_bands_label Timestep Number wcs_rangeset_axes bands wms_encoding UTF-8 wcs_enable_request * layer_data_units W m-2 annotation_name downwelling_longwave_flux_in_air: 2015-06-19 20:30:29.202558 END # METADATANAME downwelling_longwave_flux_in_air PROCESSING CLOSE_CONNECTION=DEFER PROCESSING SCALE=311.76470947266 ,423.52941894531 #PROCESSING “SCALE=AUTO” PROJECTION init=epsg:26911 END # PROJECTION STATUS ON TILEITEM location TYPE RASTER UNITS METERS END # LAYER It seems I am missing something herewould appreciate any insight into how to return a valid image in WMS Thanks, Bill ___ mapserver-users mailing list mapserver-users@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/mapserver-users ___ mapserver-users mailing list mapserver-users@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/mapserver-users ___ mapserver-users mailing list mapserver-users@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/mapserver-users
