Upgrade Raft to v1.3.9 for saturation metrics

.changelog/12865.txt

@@ -0,0 +1,3 @@
+```release-note:improvement
+telemetry: Added `consul.raft.thread.main.saturation` and `consul.raft.thread.fsm.saturation` metrics to measure approximate saturation of the Raft goroutines
+```

go.mod

@@ -53,7 +53,7 @@ require (
 	github.com/hashicorp/hcl v1.0.0
 	github.com/hashicorp/hil v0.0.0-20200423225030-a18a1cd20038
 	github.com/hashicorp/memberlist v0.3.1
-	github.com/hashicorp/raft v1.3.8
+	github.com/hashicorp/raft v1.3.9
 	github.com/hashicorp/raft-autopilot v0.1.6
 	github.com/hashicorp/raft-boltdb v0.0.0-20211202195631-7d34b9fb3f42 // indirect
 	github.com/hashicorp/raft-boltdb/v2 v2.2.2

go.sum

@@ -368,8 +368,8 @@ github.com/hashicorp/memberlist v0.3.1/go.mod h1:MS2lj3INKhZjWNqd3N0m3J+Jxf3DAOn
 github.com/hashicorp/raft v1.1.0/go.mod h1:4Ak7FSPnuvmb0GV6vgIAJ4vYT4bek9bb6Q+7HVbyzqM=
 github.com/hashicorp/raft v1.1.1/go.mod h1:vPAJM8Asw6u8LxC3eJCUZmRP/E4QmUGE1R7g7k8sG/8=
 github.com/hashicorp/raft v1.2.0/go.mod h1:vPAJM8Asw6u8LxC3eJCUZmRP/E4QmUGE1R7g7k8sG/8=
-github.com/hashicorp/raft v1.3.8 h1:lrhx4wesQLOSv3ERX/pK4cwfzQ0J2RgzsvAkBxHe1bA=
-github.com/hashicorp/raft v1.3.8/go.mod h1:4Ak7FSPnuvmb0GV6vgIAJ4vYT4bek9bb6Q+7HVbyzqM=
+github.com/hashicorp/raft v1.3.9 h1:9yuo1aR0bFTr1cw7pj3S2Bk6MhJCsnr2NAxvIBrP2x4=
+github.com/hashicorp/raft v1.3.9/go.mod h1:4Ak7FSPnuvmb0GV6vgIAJ4vYT4bek9bb6Q+7HVbyzqM=
 github.com/hashicorp/raft-autopilot v0.1.6 h1:C1q3RNF2FfXNZfHWbvVAu0QixaQK8K5pX4O5lh+9z4I=
 github.com/hashicorp/raft-autopilot v0.1.6/go.mod h1:Af4jZBwaNOI+tXfIqIdbcAnh/UyyqIMj/pOISIfhArw=
 github.com/hashicorp/raft-boltdb v0.0.0-20171010151810-6e5ba93211ea/go.mod h1:pNv7Wc3ycL6F5oOWn+tPGo2gWD4a5X+yp/ntwdKLjRk=

website/content/docs/agent/telemetry.mdx

@@ -149,6 +149,28 @@ you will need to apply a function such as InfluxDB's [`non_negative_difference()
 Sudden large changes to the `consul.client.rpc` metrics (greater than 50% deviation from baseline).
 `consul.client.rpc.exceeded` or `consul.client.rpc.failed` count > 0, as it implies that an agent is being rate-limited or fails to make an RPC request to a Consul server
 
+### Raft Thread Saturation
+
+| Metric Name                          | Description                                                                                                              | Unit       | Type   |
+| :----------------------------------- | :----------------------------------------------------------------------------------------------------------------------- | :--------- | :----- |
+| `consul.raft.thread.main.saturation` | An approximate measurement of the proportion of time the main Raft goroutine is busy and unavailable to accept new work. | percentage | sample |
+| `consul.raft.thread.fsm.saturation`  | An approximate measurement of the proportion of time the Raft FSM goroutine is busy and unavailable to accept new work.  | percentage | sample |
+
+**Why they're important:** These measurements are a useful proxy for how much
+capacity a Consul server has to accept additional write load. High saturation
+of the Raft goroutines can lead to elevated latency in the rest of the system
+and cause cluster instability.
+
+**What to look for:** Generally, a server's steady-state saturation should be
+less than 50%.
+
+**NOTE:** These metrics are approximate and under extremely heavy load won't
+give a perfect fine-grained view of how much headroom a server has available.
+Instead, treat them as an early warning sign.
+
+** Requirements: **
+* Consul 1.13.0+
+
 ### Raft Replication Capacity Issues
 
 | Metric Name                  | Description                                                                                                                                                                       | Unit     | Type    |