Prepare 0.6.2 release

Change-Id: I4f1dfa72403aaa110e029ef7d0ab2950abd47724

deployment/README.md

@@ -5,7 +5,7 @@ This directory contains a [Terraform Module](https://www.terraform.io/docs/modul
 **Note:** The `main` branch may be updated at any time with the latest changes which could be breaking. You should always configure your module to use a release. This can be configured in the modules Terraform Configuration block.
 
 ```
-source = "github.com/GoogleCloudPlatform/knfsd-cache-utils//deployment/terraform-module-knfsd?ref=v0.6.1"
+source = "github.com/GoogleCloudPlatform/knfsd-cache-utils//deployment/terraform-module-knfsd?ref=v0.6.2"
 ```
 
 ## Prerequisites
@@ -92,7 +92,7 @@ Basic usage of this module is as follows:
 
 ```terraform
 module "nfs_proxy" {
-    source = "github.com/GoogleCloudPlatform/knfsd-cache-utils//deployment/terraform-module-knfsd?ref=v0.6.1"
+    source = "github.com/GoogleCloudPlatform/knfsd-cache-utils//deployment/terraform-module-knfsd?ref=v0.6.2"
 
     # Google Cloud Project Configuration
     PROJECT                        = "my-gcp-project"
@@ -137,7 +137,7 @@ provider "google" {
 }
 
 module "nfs_proxy" {
-    source = "github.com/GoogleCloudPlatform/knfsd-cache-utils//deployment/terraform-module-knfsd?ref=v0.6.1"
+    source = "github.com/GoogleCloudPlatform/knfsd-cache-utils//deployment/terraform-module-knfsd?ref=v0.6.2"
 
     # Network Configuration
     NETWORK                        = "my-vpc"

docs/changes/0.6.2.md

@@ -0,0 +1,37 @@
+# v0.6.2
+
+* Increase health check interval to 60 seconds
+* Add parameters to configure health checks
+* Support deploying metrics as a Terraform module
+* Remove per-mount stats (aggregate by source server)
+* Exclude support when resolving knfsd proxy instance name
+
+## Increase health check interval to 60 seconds
+
+This allows 2 minutes (with the default health check values) to reboot a knfsd proxy instance without the managed instance group replacing the instance.
+
+## Add parameters to configure health checks
+
+This allows overriding various parameters used by the health checks. For example, if you do not encounter the culling issue you might want to reduce the `HEALTHCHECK_INTERVAL_SECONDS` so that failed instances are detected more quickly.
+
+If you have a lot of volumes, or high latency between the source and the proxy causing a startup time slower than 10 minutes (600 seconds), you might want to increase the `HEALTHCHECK_INITIAL_DELAY_SECONDS`. Conversely, if you know your proxy starts up in less than 5 minutes, you can reduce the initial delay so that instances that fail to start up correctly are detected and replaced more quickly.
+
+## Support deploying metrics as a Terraform module
+
+Support deploying the metrics as a Terraform module so that the metrics can be deployed without needing to clone the Terraform configuration from git.
+
+## Remove per-mount stats (aggregate by source server)
+
+Reporting on the stats per-mount generates a lot of logging data when the source has 50 or more volumes.
+
+Secondly, the stats cannot be reliably aggregated later because multiple mounts can share the same NFS client. All mounts sharing the same NFS clients will have identical stats that are an aggregate of all the mounts sharing the NFS client. If these per-mount stats are then summed together on a dashboard it leads to the stats being multiplied by the number of mounts that share the same NFS client.
+
+However, because some mounts might have a separate NFS client, and thus separate stats, it becomes impossible to view an accurate total on a dashboard when the stats are reported per-mount.
+
+## Exclude support when resolving knfsd proxy instance name
+
+Support a list of excluded servers and/or local paths when `query_proxy_instances` is enabled.
+
+When collecting metrics from clients with `query_proxy_instances` enabled by default the collector will probe every NFS server that is mounted by the client. This can cause issues if the client is mounting a mixture of knfsd proxies and other NFS servers.
+
+The metrics collector now has an `exclude` configuration section for `query_proxy_instances`.

docs/changes/next.md

@@ -1,37 +1 @@
 # Next
-
-* Increase health check interval to 60 seconds
-* Add parameters to configure health checks
-* Support deploying metrics as a Terraform module
-* Remove per-mount stats (aggregate by source server)
-* Exclude support when resolving knfsd proxy instance name
-
-## Increase health check interval to 60 seconds
-
-This allows 2 minutes (with the default health check values) to reboot a knfsd proxy instance without the managed instance group replacing the instance.
-
-## Add parameters to configure health checks
-
-This allows overriding various parameters used by the health checks. For example, if you do not encounter the culling issue you might want to reduce the `HEALTHCHECK_INTERVAL_SECONDS` so that failed instances are detected more quickly.
-
-If you have a lot of volumes, or high latency between the source and the proxy causing a startup time slower than 10 minutes (600 seconds), you might want to increase the `HEALTHCHECK_INITIAL_DELAY_SECONDS`. Conversely, if you know your proxy starts up in less than 5 minutes, you can reduce the initial delay so that instances that fail to start up correctly are detected and replaced more quickly.
-
-## Support deploying metrics as a Terraform module
-
-Support deploying the metrics as a Terraform module so that the metrics can be deployed without needing to clone the Terraform configuration from git.
-
-## Remove per-mount stats (aggregate by source server)
-
-Reporting on the stats per-mount generates a lot of logging data when the source has 50 or more volumes.
-
-Secondly, the stats cannot be reliably aggregated later because multiple mounts can share the same NFS client. All mounts sharing the same NFS clients will have identical stats that are an aggregate of all the mounts sharing the NFS client. If these per-mount stats are then summed together on a dashboard it leads to the stats being multiplied by the number of mounts that share the same NFS client.
-
-However, because some mounts might have a separate NFS client, and thus separate stats, it becomes impossible to view an accurate total on a dashboard when the stats are reported per-mount.
-
-## Exclude support when resolving knfsd proxy instance name
-
-Support a list of excluded servers and/or local paths when `query_proxy_instances` is enabled.
-
-When collecting metrics from clients with `query_proxy_instances` enabled by default the collector will probe every NFS server that is mounted by the client. This can cause issues if the client is mounting a mixture of knfsd proxies and other NFS servers.
-
-The metrics collector now has an `exclude` configuration section for `query_proxy_instances`.

image/smoke-tests/terraform/main.tf

@@ -27,7 +27,7 @@ resource "google_filestore_instance" "source" {
 }
 
 module "proxy" {
-  source = "github.com/GoogleCloudPlatform/knfsd-cache-utils//deployment/terraform-module-knfsd?ref=v0.6.1"
+  source = "github.com/GoogleCloudPlatform/knfsd-cache-utils//deployment/terraform-module-knfsd?ref=v0.6.2"
 
   PROJECT = var.project
   REGION  = var.region