I am getting the following error while extracting my private collection (running it in Terrascope jupyter environment) -
RetryError: HTTPSConnectionPool(host=‘ewoc-rdm-api.iiasa.ac.at’, port=443): Max retries exceeded with url: … (Caused by ResponseError(‘too many 500 error responses’))
Is the server down at this time, or is it some other error which I am not addressing
Hi there,
there’s indeed ongoing issues at the CDSE side, which is reported here. Even though it’s mentioned that the issue is resolved, we see it reoccurring frequently in the past few days, which may explain the issue you’re experiencing.
At this point, we can only recommend to keep an eye on the official CDSE channels and retry the extraction workflow until it works again. Apologies for the inconvenience!
It seems to be running fine locally, but when i run it locally on Windows, it throws me module not found error for module “grp” in run_extractions code.
When I run it on Terrascope, only then I am facing this issue.
What do you mean by running locally? Extractions will run on CDSE for access to the satellite and auxiliary data. Which part of the code is giving you the issue?
Here, running locally means in my local machine on VSCode. Following is the snapshot of the error that I get -
As i found out, that grp module is not available on Windows systems, as it is specific to Unix-like operating systems
we’re investigating the use of this module, but I’m a bit confused as this thread originally was started for the server issues. Your other thread is on the grp module. They’re probably unrelated?
Yes, thats correct. Apologies for the cross post. I’ll put the same message in the correct thread.
Hi! We indeed experienced some connectivity issues when fetching data from our Reference Data Module. On our side, things are working smoothly again. Could you try again and notify us if this connection error persists?
FYI: after uploading your private dataset into RDM, it takes a couple of minutes before the data is ready for download, so trying again after a while for a freshly uploaded dataset might also resolve your issue.
Hope this helps!
There is no connection error now. Thanks. Although I am having similar issues when I run it on Terrascope
When i run the code after clicking on “Run on Terrascope”, and I login with CDSE account through which I uploaded the private collection, the private collection is not getting listed.
Hi there. Normally you should be able to see all datasets which you have uploaded. Can you confirm that you do see the dataset under “User Collections” when you login directly to the RDM (https://rdm.esa-worldcereal.org/)?
Yes, when I run it locally ( local run meaning as explained in starting steps in private extractions notebook) , I can see my private collection, but when I run it on Terrascope, it shows me the message that no private collection is present.
I can see the private collection in ‘User Collections’ in RDM module
Hi! We have performed an update to our Terrascope notebooks environment.
It should be working now there as well!
Jeroen
Thanks. I am now able to access the private collections in Terrascope
Glad to hear, thank you for confirming!
While running the map inference job on CDSE through openEO within the Terrascope environment, I am getting the following error in ‘worlcereal_custom_croptype.ipynb’ -
“OpenEoClientException: DataCube.execute_batch() with explicit output format ‘GTiff’, but the process graph already has save_result node(s) which is ambiguous and should not be combined.”
The related github issues are - execute_batch silently adds save_result even if there is already one · Issue #401 · Open-EO/openeo-python-client · GitHub and duplicate `save_result`s after `datacube_from_flat_graph` · Issue #732 · Open-EO/openeo-python-client · GitHub
Thank you once again for reporting and apologies for the bumpy ride you are experiencing.
We have indeed noticed this issue ourselves and are heavily investigating.
We will get back to you once the issue has been resolved.
For your reference, tracking this here and already working on a fix: Fix bug in inference process graph · Issue #350 · WorldCereal/worldcereal-classification · GitHub
We just pushed a new release containing a fix for this issue: Release WorldCereal classification module v2.2.2 · WorldCereal/worldcereal-classification · GitHub
You can either:
Thanks! Its working smoothly now on Terrascope. Local run still has an overflow error, probably a Windows issue.
