Closed dugalh closed 9 months ago
Thank you for reporting this, sadly, I am not immediately reproducing, but don't doubt there is some issue here.
$ python
Python 3.11.5 | packaged by conda-forge | (main, Aug 27 2023, 03:34:09) [GCC 12.3.0] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> import rasterio as rio
>>> crs = rio.CRS.from_string('EPSG:32634+5773')
>>>
highlights of my env
gdal 3.7.2 py311h815a124_3 conda-forge
rasterio 1.3.8 py311h40fbdff_2 conda-forge
proj 9.3.0 h1d62c97_1 conda-forge
Do you get any dmesg
/ syslog output or any other messages when the segfault occurs?
@dugalh , do you get a segfault with a simple and new environment? I don't :(
$ mamba create -n dev python=3.11 rasterio
$ conda activate dev
$ python
Python 3.11.5 | packaged by conda-forge | (main, Aug 27 2023, 03:34:09) [GCC 12.3.0] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> import rasterio as rio
>>> crs = rio.CRS.from_string('EPSG:32634+5773')
>>>
@akrherz Thanks for the quick reply. It only segfaults when the CRS is used e.g. on print(crs)
. I do see it in a new rasterio
only environment.
This seems to be the relevant part of the dmesg output:
kern :info : [Sep23 13:38] python[48228]: segfault at 0 ip 00007ff6a82a49bd sp 00007ffcc3818cb8 error 4 in libc.so.6[7ff6a81>
kern :info : [ +0.000030] Code: f8 77 c3 66 2e 0f 1f 84 00 00 00 00 00 f3 0f 1e fa 89 f8 48 89 fa c5 f9 ef c0 25 ff 0f 00 0>
kern :info : [ +0.000007] potentially unexpected fatal signal 11.
kern :warn : [ +0.000001] CPU: 4 PID: 48228 Comm: python Not tainted 5.15.90.1-microsoft-standard-WSL2 #1
kern :warn : [ +0.000017] RIP: 0033:0x7ff6a82a49bd
kern :warn : [ +0.000004] Code: f8 77 c3 66 2e 0f 1f 84 00 00 00 00 00 f3 0f 1e fa 89 f8 48 89 fa c5 f9 ef c0 25 ff 0f 00 0>
kern :warn : [ +0.000001] RSP: 002b:00007ffcc3818cb8 EFLAGS: 00010283
kern :warn : [ +0.000002] RAX: 0000000000000000 RBX: 0000000000000000 RCX: 00005602029b2918
kern :warn : [ +0.000001] RDX: 0000000000000000 RSI: 00007ff660861770 RDI: 0000000000000000
kern :warn : [ +0.000047] RBP: 0000000000000064 R08: 00007ff6a7e5c000 R09: f4dca66451297fe6
kern :warn : [ +0.000002] R10: 0000000000000001 R11: 0000000000000001 R12: 0000000000000000
kern :warn : [ +0.000001] R13: 00005602037c6010 R14: 00007ff69ef19b40 R15: 00007ff660861780
kern :warn : [ +0.000001] FS: 00007ff6a80fd740 GS: 0000000000000000
It only segfaults when the CRS is used e.g. on
print(crs)
I'm sorry, I totally missed that and indeed, that reproduces :(
I'm sure #275 is at blame and will backport a patch with rasterio/rasterio#2875 included now.
Thank you...
Thank you...
Thank you for reporting, this was totally my fault. Hopefully will get a green build shortly and get an updated build onto conda-forge later today.
The moment of truth :)
$ mamba install rasterio=1.3.8=py311h40fbdff_3
$ python
Python 3.11.5 | packaged by conda-forge | (main, Aug 27 2023, 03:34:09) [GCC 12.3.0] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> import rasterio as rio
>>> crs = rio.CRS.from_string('EPSG:32634+5773')
>>> print(crs)
COMPD_CS["WGS 84 / UTM zone 34N + EGM96 height",PROJCS["WGS 84 / UTM zone 34N",GEOGCS["WGS 84",DATUM["WGS_1984",SPHEROID["WGS 84",6378137,298.257223563,AUTHORITY["EPSG","7030"]],AUTHORITY["EPSG","6326"]],PRIMEM["Greenwich",0,AUTHORITY["EPSG","8901"]],UNIT["degree",0.0174532925199433,AUTHORITY["EPSG","9122"]],AUTHORITY["EPSG","4326"]],PROJECTION["Transverse_Mercator"],PARAMETER["latitude_of_origin",0],PARAMETER["central_meridian",21],PARAMETER["scale_factor",0.9996],PARAMETER["false_easting",500000],PARAMETER["false_northing",0],UNIT["metre",1,AUTHORITY["EPSG","9001"]],AXIS["Easting",EAST],AXIS["Northing",NORTH],AUTHORITY["EPSG","32634"]],VERT_CS["EGM96 height",VERT_DATUM["EGM96 geoid",2005,AUTHORITY["EPSG","5171"]],UNIT["metre",1,AUTHORITY["EPSG","9001"]],AXIS["Gravity-related height",UP],AUTHORITY["EPSG","5773"]]]
>>>
Great - thanks :)
Solution to issue cannot be found in the documentation.
Issue
Hello. I get a segmentation fault when using a compound CRS with the latest conda-forge
rasterio
.It works OK with pyproj.
I didn't have this problem with previous versions of conda-forge
rasterio
, and don't see it with the latest PyPIrasterio
, so am raising it here...Installed packages
Environment info