# Last week in scylla-cluster-tests.git master (issue #117; 2025-11-21)

**URL:** <https://forum.scylladb.com/t/last-week-in-scylla-cluster-tests-git-master-issue-117-2025-11-21/5186>\
**Category:** ScyllaDB\
**Tags:** git-testing-news\
**Created:** [November 21, 2025, 2:44pm UTC](https://forum.scylladb.com/t/last-week-in-scylla-cluster-tests-git-master-issue-117-2025-11-21/5186 "2025-11-21T14:44:17Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![soyacz](https://sea2.discourse-cdn.com/flex016/user_avatar/forum.scylladb.com/soyacz/32/557_2.png) [@soyacz](https://forum.scylladb.com/u/soyacz)\
**Post date:** [November 21, 2025, 2:44pm UTC](https://forum.scylladb.com/t/last-week-in-scylla-cluster-tests-git-master-issue-117-2025-11-21/5186/1 "2025-11-21T14:44:17Z")

</div>

This short report brings to light some interesting commits to [scylla-cluster-tests.git master](https://github.com/scylladb/scylla-cluster-tests) from the last week. Commits in the 466a75db…f2ba4869 range are covered.

There were 37 non-merge commits from 13 authors in that period. Some notable commits:

[Gradual throughput growth via latte `$throttle`](https://github.com/scylladb/scylla-cluster-tests/commit/6e2d8dd66449d3877eec698088fbfc1c079f0a3d) is now supported in `PerformanceRegressionPredefinedStepsTest` , allowing the same scenario to be driven by latte in addition to cassandra-stress.

Serverless cleanup: [removed the deprecated `cloud_config` option](https://github.com/scylladb/scylla-cluster-tests/commit/0b4a70a1239648891caafbc08ffad6315851ce92) to align with the upcoming driver change.

Upgrade tests gained a switch to [disable Gemini during rolling upgrades](https://github.com/scylladb/scylla-cluster-tests/commit/82832bddf4c641b30000a3bd0dc8202e15ffa6cf); the new `run_gemini_in_rolling_upgrade` defaults to false for safer baseline runs.

CLI: [xcloud provider support was added to `list-resources`](https://github.com/scylladb/scylla-cluster-tests/commit/fcfd9060cebded561d19af9eaf3fd9fe13ab0aa3), including filtering by test-id/user and nicer table output across lab and staging.

AWS provisioner now provides [verbose error logging for spot/fleet failures](https://github.com/scylladb/scylla-cluster-tests/commit/2e25cad5d1ab342cd013e79deafcc905f8f81ad3) with codes and state hints to speed up debugging.

Rolling-upgrade pipelines were updated to [replace debian-12 with debian-13](https://github.com/scylladb/scylla-cluster-tests/commit/6f5f5772b45f349da1dd67446d6452077ced1d2d), keeping test images current.

[Latte bumped to 0.42.0-scylladb](https://github.com/scylladb/scylla-cluster-tests/commit/f90bca41645c44b4d2922bfb79024e351130bbe8), bringing Rust driver 1.4.1, an OCI image, and helpers for large data files (useful for vector search scenarios).

Tester gained [S3 artifact download functionality](https://github.com/scylladb/scylla-cluster-tests/commit/8f043a02f3dd1a4e3b43f313d785bd41187d0011) via the `download_from_s3` config and a helper method to fetch required artifacts.

Vector Search: a latte-based [sanity test and script were added](https://github.com/scylladb/scylla-cluster-tests/commit/050488dd5439473947a5e632c6c0db4c59535349) to create the schema, load data, and validate ANN results.

GCE upgrade jobs were migrated to [z3-highmem-8-highlssd instances](https://github.com/scylladb/scylla-cluster-tests/commit/a73e3d206868814a641de0cf188af12d9e0f4012) for more consistent performance and lower costs.

Housekeeping: [updated CODEOWNERS](https://github.com/scylladb/scylla-cluster-tests/commit/8fd983ed44baed392a828bd056f3a604b14074e4) to reflect current infra ownership.

See you in the next issue of last week in scylla-cluster-tests.git master!