Israel Fruchter

Can you reproduce this issue in a unit test ? The reason we don want to run on slaves is to avoid multiple reports bring sent out, we care only...

We need artifact tests as part of gating, yet again. This should have been worked on before merging such a change

look like the version we are using of java-snappy is quite old and doesn't support arm yet very well

I gave up trying to pass the `reactor_options` into the signal handlers I think of possible two direction 1) pass way is via global variables 2) duplicate the handlers -...

> > align the handlers with the same logic that is used in the reactor stalls backtrace prints > > > > Ref: [scylladb/scylladb#16922](https://github.com/scylladb/scylladb/issues/16922) > > I am missing some...

at least now the compilers are happy, but still didn't found a way to get the template working, to avoid the duplication.

@arjanschaaf looks good, especially the part of switching to newer aws client, and setting the credentials not only from a configuration file.

happened again this week: ## Installation details Kernel Version: 5.15.0-1035-aws Scylla version (or git commit hash): `5.3.0~dev-20230425.124153d43915` with build-id `c084d51a610cd0c9956abf5bed57fbd54301731c` Cluster size: 3 nodes (i3.4xlarge) Scylla Nodes used in this...

> When a new node is added to the cluster, most (all?) requests still go to the old nodes. Only restarting the benchmark will cause new coordinators to receive CQL...

> I suggest you embrace the [new driver](https://github.com/scylladb/scylla-go-driver) in the bechmark, and start maintaining it. I'm not sure that on @avelanarius team roadmap. I think they have other plans for...