Internal server error 500 when fetching large changesets #5247
What steps will reproduce the problem?
What is the expected result?
Should successfully download past changeset data without warning messages.
What happens instead?
A error message appears in roughly half of the attempts and every time at different points.
It is always a 500 Internal Server Error.
'''If any option is chosen an error message appears stating it's not possible to complete the request.
The text was updated successfully, but these errors were encountered:
This is (unfortunately) normal currently - they are just too big and complicated and in order to protect server resources we have a timeout in place which is what you are seeing.
Hopefully one day the relation/full call will be moved into cgimap and we'll be able to handle larger relations.
relation/full is in cgimap, TomH presumably means the changeset download API call.
It's an invitation to upload "unwanted" stuff and give people who have to fix it a hard time. What would be the solution when this happens?
Maybe restrict the number of objects in one changeset to what can still be downloaded with