You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Relevant part of the error (the full traceback is 200+ lines):
ConnectionError: HTTPConnectionPool(host='ftp.ebi.ac.uk', port=80): Max retries exceeded with url: /pub/databases/chebi/ontology/chebi.obo (Caused by NewConnectionError('<urllib3.connection.HTTPConnection object at 0x7f5dcb7bcb50>: Failed to establish a new connection: [Errno 111] Connection refused'))
One possible lead on this issue I think is the port used. Because we use port 80 according to the traceback, the full URL should be http://ftp.ebi.ac.uk/pub/databases/chebi/ontology/chebi.obo, note the http here because port 80 is used, while going to that URL in the browser, you're redirected to https://ftp.ebi.ac.uk/pub/databases/chebi/ontology/chebi.obo, again note https which implies port 443 should be used. My guess here is that somewhere, the http URL is requested, but that it's only the https URL actually works, hence the ConnectionError. I haven't been able to find where the http request would come from though.
My setup:
bioversions version: 0.5.560 (I run a git cloned setup)
Python version: 3.11.7
OS (output from uname -a): Linux 5.15.0-126-generic #136~20.04.1-Ubuntu SMP Thu Nov 14 16:38:05 UTC 2024 x86_64 x86_64 x86_64 GNU/Linux
The text was updated successfully, but these errors were encountered:
I think this URL is an artifact of what's inside the OBO PURL service. I was trying to track it down yesterday and added a few improvements that will make logging easier, but I'm not done yet
Error can be reproduced by:
Relevant part of the error (the full traceback is 200+ lines):
One possible lead on this issue I think is the port used. Because we use port 80 according to the traceback, the full URL should be
http://ftp.ebi.ac.uk/pub/databases/chebi/ontology/chebi.obo
, note thehttp
here because port 80 is used, while going to that URL in the browser, you're redirected tohttps://ftp.ebi.ac.uk/pub/databases/chebi/ontology/chebi.obo
, again notehttps
which implies port 443 should be used. My guess here is that somewhere, thehttp
URL is requested, but that it's only thehttps
URL actually works, hence theConnectionError
. I haven't been able to find where the http request would come from though.My setup:
bioversions version:
0.5.560
(I run a git cloned setup)Python version:
3.11.7
OS (output from
uname -a
):Linux 5.15.0-126-generic #136~20.04.1-Ubuntu SMP Thu Nov 14 16:38:05 UTC 2024 x86_64 x86_64 x86_64 GNU/Linux
The text was updated successfully, but these errors were encountered: