Benjamin Wang

Improve release process to ensure we have stable & consistent format (mediaType)

4

### What would you like to be added? I checked several etcd released images, different images may have different mediaType. I think the reason could be different release persons have...

### What would you like to be added? We need to try to cleanup (or find alternatives) for the archived and unmaintained dependencies: - https://github.com/etcd-io/etcd/issues/20991 - https://github.com/etcd-io/etcd/issues/14533 - https://github.com/etcd-io/etcd/issues/15145 -...

[WIP] Bootstrap etcd from consistent_index

7

Link to https://github.com/etcd-io/etcd/issues/20187 Still need to do some performance comparison (bootstrap time) Note that this PR can also workaround https://github.com/etcd-io/etcd/issues/20967. Once users run into the issue, they just need to...

Evaluate & potentially replace `https://github.com/soheilhy/cmux`

9

### What would you like to be added? etcdserver depends on https://github.com/soheilhy/cmux to receive both gRPC and REST connections on the same TCP listener/port. But the library isn't well maintained;...

Unify the approach to manage etcd images in Kubernetes

34

### What would you like to be added? The etcd container images are currently built from [cluster/images/etcd](https://github.com/kubernetes/kubernetes/tree/master/cluster/images/etcd), pushed to [registry.k8s.io/etcd](http://registry.k8s.io/etcd), and used by kubeadm, KOps and Kubernetes workflow tests. Concerns...

Update `bump_etcd_version_k8s.md` to mark the step `Build etcd image` deprecated

1

### What would you like to be added? We use etcd officially released image, and we don't build etcd image using the legacy [cluster/images/etcd](https://github.com/kubernetes/kubernetes/tree/master/cluster/images/etcd) anymore. So we need to update...

Update release guide to include the step of bump etcd for K8s

3

cc @ivanvc @joshjms @hakman @ArkaSaha30 Please read https://github.com/etcd-io/etcd/blob/main/CONTRIBUTING.md#contribution-flow.

Add a new metrics etcd_server_request_duration_seconds to track e2e duration for each request

5

Fix https://github.com/etcd-io/etcd/issues/21036 Eventually we may need to deprecate `etcd_server_range_duration_seconds`. cc @lavacat @fuweid @ivanvc @serathius Please read https://github.com/etcd-io/etcd/blob/main/CONTRIBUTING.md#contribution-flow.

Add metrics to track the end-to-end processing duration of each request

3

### What would you like to be added? Currently, etcd has metrics to track the disk I/O latency, - etcd_disk_wal_fsync_duration_seconds - etcd_disk_backend_commit_duration_seconds Also peer-2-peer network round trip, - etcd_network_peer_round_trip_time_seconds But...

[test] support cluster-api style rolling replacement upgrade

4

### What would you like to be added? Currently our e2e framework supports `in-place` upgrade. We just replace the binary, and reuse the existing data. https://github.com/etcd-io/etcd/blob/30452ab110473eda5f60bdb301c046194a154e91/tests/framework/e2e/downgrade.go#L131 However, the cluster-api/Kubeadm upgrades...