Cezar Moise

> If it's not too late for this, or perhaps it could be done in a follow-up, I've already discussed with @pehala a more interesting and efficient way to test...

With latest version, does the test no longer fill the cluster between the scale-out and scale-in?

## Results https://argus.scylladb.com/tests/scylla-cluster-tests/e0cd5744-57da-4f37-ba40-394550919fad Run date: 2025-03-04 Scylla version: 2025.2.0~dev-20250302.0343235aa269 Initial cluster: 3 racks with one i4i.8xlarge per rack Write to 90%, RF=3, initial tablets=1024 Run mixed workload at about 40%...

Added first result @pehala @swasik After tables balanced, the small instance is at 91%, then when writting data it gets to 100 and out of space error 

Oppened https://github.com/scylladb/scylladb/issues/22028

Replicated again, but it looks like https://github.com/scylladb/scylladb/commit/b74565e83fac845912e7e983bff8b61d2e54ab8b is not in enterprise yet. Will re-run with master instead of enterprise  https://argus.scylladb.com/tests/scylla-cluster-tests/b227928c-e2a2-4044-8301-1e06684bf61f

Using master:latest it looks ok  https://argus.scylladb.com/tests/scylla-cluster-tests/8598cd23-e820-4b22-9103-b51f455a7ad2 Will update when it finishes

Updated results comment. With the fix, no more out of space errors. However, I think the fix also improved how much the initial space was used, so with same stress...

Ran again, with 90% this time, got out of space errors again. https://argus.scylladb.com/tests/scylla-cluster-tests/70132ede-4e2d-4657-a367-f4229df5fe6d 

Run on 2025.2, with 80% initial utilization 90% not possible because instances don't have the same disk space per shard, due to larger instances reserving 2 vCPU for non-shard jobs...