hubcio opened a new pull request, #3737: URL: https://github.com/apache/iggy/pull/3737
Benchmarks against a VSR cluster were indistinguishable from single-node runs: they shared the params_identifier trend key and the results directory (overwriting each other), and neither report.json nor the dashboard carried any topology information. iggy-bench now asks the server for cluster metadata after a run. Real clusters (more than one node, or a named roster) land as an optional cluster field on the report and get a _clusterN suffix on both the trend identifier and the results directory, so cluster and single-node series never mix. Single-node output stays byte-identical, the field is omitted when absent, and a failing metadata call is treated as single node, keeping old reports, old dashboards and old servers compatible. The dashboard renders a node-count badge and topology chip, adds an all/single-node/cluster filter, and sweep charts no longer pool runs with differing topologies. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
