Geoffrey Beausire

Possible deadlock when using mesh with high contention on the queues

1

When taking the sharding example: https://github.com/DataDog/glommio/blob/master/examples/sharding.rs If you increase the number of messages in the stream or reduce the channel size, the program will never finish. Step to reproduce: ```...

Add full support for HostNetwork pod IPs

1

**Description of your changes:** This PR add full support for HostNetworking on scylla clusters. The reason why HostNetworking is needed is when the clients are not inside the same Kubernetes...

This is necessary to implement TLS as the "service" info only report the clear service port. It will be easy to expand the "services_name" to return peers-tls Required for #8

Removenode not working in 5.0.0 with rbno

3

With 5.0.0, we don't seem to be able to perform removenode anymore. # Steps to reproduce * Configure with: `allowed_repair_based_node_ops: "replace,bootstrap,removenode,decommission,rebuild"` * Start new cluster spawning 2 nodes in two...

Changes proposed in this PR: If consulClient.Catalog().NodeServiceList fails for more than 15 minutes (max duration of NewExponentialBackOff()), the loop will get stuck because `minWaitCh` is not rearmed. `case

Alleviate race conditions in roll restart reconciler

5

### Description When a Roll Restart is triggered with at least 2 pools, a race condition can trigger the roll restart of a pods in each pools. This can lead...