R2 query through API returns every objects from bucket 24 times

I’ve successfully configured 5 workers to work with logpush connected to my R2 bucket dh-log. Currently there are only 3 object in it, containing totally 8 log entries.

If I query the objects through API with curl I get every objects exactly 24 time in the result, so 192 log entries instead of 8. I think this is a bug or amI wrong?

The command executed:

$ curl -s -g -X GET ‘https://api.cloudflare.com/client/v4/accounts/b39aecfdb64beb24a44bddea48c2adc0/logs/retrieve?start=2023-07-28T00:00:00Z&end=2023-07-28T23:59:59Z&bucket=dh-log
-H “X-Auth-Email: [email protected]
-H “X-Auth-Key: xxx”
-H “R2-Access-Key-Id: xxx”
-H “R2-Secret-Access-Key: xxx” | jq . > /tmp/logs.json

$ grep RayID /tmp/logs.json | sort | uniq -c
24 “RayID”: “7edd8f90dbb538c8”,
24 “RayID”: “7edddc007fc92f6c”,
24 “RayID”: “7edddc24e9b32f6c”,
24 “RayID”: “7eddf41e3dd478ac”,
24 “RayID”: “7eddf431bfd878ac”,
24 “RayID”: “7eddf4437e5e78ac”,
24 “RayID”: “7eddf45d6c5e78ac”,
24 “RayID”: “7eddf46ca91678ac”,

I’ve given up at the end. The method to get R2 object list as described here is totally buggy.
I started now with the S3 API to acces my buckets in R2.

The problem is not waht is described above, but queriing a list of object from a bucket with more 1000 entries returns a worker error after about 40 seconds: resource limit exceeded.