Segment 5
In this sub-session, we encountered port conflicts when attempting host network mode for the test cluster, which required reverting to the default bridge network. We then stabilized the environment by fixing the configuration to include required settings (EXTERNAL_LOCALWEB_URL) and updating the loadtest to send the 'x-amz-content-sha256: UNSIGNED-PAYLOAD' header, eliminating false corruption reports. The previously developed CQL batcher for high-throughput S3 metadata writes, along with loadtest distinction between timeouts and real corruption, was committed along with test-cluster configuration fixes (RIBS_RETRIEVALBLE_REPAIR_THRESHOLD and semicolon-separated kuri init command). The single-node test cluster is now operational, and a compaction was triggered, leaving only documentation updates as uncommitted changes.
From Host Network Collapse to Clean Working State: A Distributed S3 Test Cluster's Debugging Odyssey