Open mneedham opened 10 months ago
As we checked, many connections were unstable during that period of time. A lot of retries happened. I did not find here the problem.
But I need to check what would happen with the query if that retries happen at AWSHttpResourceClient
. And what timeouts/tries does it use.
related issue https://github.com/ClickHouse/ClickHouse/issues/58160
Here I have to introduce set of metrics which helps to unswer to the questions:
Also:
I'm trying to query some log data in an S3 bucket, but it seems to be trying to lookup security credentials and then times out.
The bucket contains 10 files with 100,000 lines each.
This is the trace output:
So it finds the rows quite quickly, but then waits another 5 minutes before completing. I'm having the same issue when trying to process this bucket using the S3Queue table engine.