Update README for known issue

Signed-off-by: Fei Su <[email protected]>

README.md

@@ -7,33 +7,34 @@ This repository is the [Terraform Provider of XenServer](https://registry.terraf
 - `xenserver/` The provider, resources, data sources and tests.
 - Miscellaneous meta files.
 
-## Requirements
+## Developing the Provider
+### Requirements
 
 - [Terraform](https://developer.hashicorp.com/terraform/downloads) >= 1.8
 - [Go](https://golang.org/doc/install) >= 1.22.2
 
-## Developing the Provider
 
-### Prepare
+### Prepare Development Environment
 
-If you wish to work on the provider, you'll first need [Go](http://www.golang.org) installed on your machine (see [Requirements](#requirements) above).
+The following instructions are based on Linux system.
 
-To build the provider, you'll need to prepare the local XenServer module for Go. 
-- [Download](https://www.xenserver.com/downloads) the XenServer SDK zip package and unzip
-- Create `goSDK/` directory under `terraform-provider-xenserver/`
-- Copy all source files under `XenServer-SDK/XenServerGo/src/` to `terraform-provider-xenserver/goSDK/` folder
+- Prepare the local XenServer SDK module
+    - [Download](https://www.xenserver.com/downloads) the XenServer SDK zip package and unzip
+    - Create `goSDK/` directory under `terraform-provider-xenserver/`
+    - Copy all source files under `XenServer-SDK/XenServerGo/src/` to `terraform-provider-xenserver/goSDK/` folder
 
 ### Build
 
 Run the commands as follows:
-
 ```shell
 go get -u all
 go mod tidy
 ```
 
 To compile the provider, run `"go install"`. This will build the provider and put the provider binary in the `$GOPATH/bin` directory.
 
+In short, you can just run `make provider` instead.
+
 ### Document
 
 To generate or update documentation, run `make doc`.
@@ -44,10 +45,12 @@ Set up log with `github.com/hashicorp/terraform-plugin-log/tflog`. To enable log
 
 ```shell
 export TF_LOG_PROVIDER="DEBUG"
+export TF_LOG_PATH=/tmp/terraform.log
 ```
-See https://developer.hashicorp.com/terraform/plugin/log/managing.
 
-### Test
+Find more details in [Managing Log Output](https://developer.hashicorp.com/terraform/plugin/log/managing).
+
+### Acceptance Test
 In order to run the full suite of acceptance tests, prepare a local `.env` file like:
 
 ```shell
@@ -61,7 +64,7 @@ export SMB_SERVER_USERNAME=<smb-server-username>
 export SMB_SERVER_PASSWORD=<smb-server-password>
 ```
 
-Run `"make testacc"`. *Note:* Acceptance tests create real resources, and often cost money to run.
+Run `"make testacc"`. *Note:* Acceptance tests generate actual resources and frequently incur costs when run.
 
 ```shell
 make testacc
@@ -74,7 +77,7 @@ Terraform allows to use local provider builds by setting a `dev_overrides` block
 1. Set `GOBIN` to the path where Go installs binaries or use the default path:
 
 ```shell
-export GOBIN=/Users/<Username>/go/bin
+export GOBIN=$HOME/go/bin
 go env GOBIN
 ```
 
@@ -121,6 +124,9 @@ sudo docker run -it -v $(pwd):/app -w /app golangci/golangci-lint bash
 golangci-lint run --config=/app/.golangci.yml
 ```
 
+## Known Issue
+1. If you are using Terraform provider v0.1.1, you might encounter compatibility issues after applying the XenServer 8 updates released to Early Access on 25 September 2024 and Normal on 2 October 2024. Terraform v0.1.2 resolves these compatibility issues.
+
 ## Contributing
 
 See [DEVELOP.md](DEVELOP.md)

docs/resources/snapshot.md

@@ -88,7 +88,7 @@ resource "xenserver_snapshot" "snapshot" {
 
 - `revert` (Boolean) Set to `true` if you want to revert this snapshot to VM, default to be `false`.
 
--> **Note:** When `revert` is true, the snapshot resource will be updated with new configuration first and then revert to VM.
+-> **Note:** `revert` only works after the snapshot resource created. When `revert` is true, the snapshot resource attributes will be updated first, for example `name_label`. And then revert to VM.
 
 ~> **Warning:** After revert, the VM `hard_drive` will be updated. If snapshot revert to the VM resource defined in 'main.tf', it'll cause issue when continue execute terraform commands. There's a suggest solution to resolve this issue, follow the steps: <br>1. run `terraform state show xenserver_snapshot.<snapshot_resource_name>`, get the revert VM's UUID 'vm_uuid' and revert VDIs' UUID 'vdi_uuid'.<br>2. run `terraform state rm xenserver_vm.<vm_resource_name>` to remove the VM resource state.<br>3. run `terraform import xenserver_vm.<vm_resource_name> <vm_uuid>` to import the VM resource new state.<br>4. run `terraform state rm xenserver_vdi.<vdi_resource_name>` to remove the VDI resource state. Be careful, you only need to remove the VDI resource used in above VM resource. If there're multiple VDI resources, remove them all.<br>5. run `terraform import xenserver_vdi.<vdi_resource_name> <vdi_uuid>` to import the VDI resource new state. If there're multiple VDI resources, import them all.<br>
 - `with_memory` (Boolean) True if snapshot with the VM's memory, default to be `false`.

examples/pool-main/main.tf

@@ -16,6 +16,25 @@ provider "xenserver" {
   password = local.env_vars["XENSERVER_PASSWORD"]
 }
 
+# get the existing supporter hosts
+data "xenserver_host" "supporter" {
+  is_coordinator = false
+}
+
+# join a new supporter to the pool
+resource "xenserver_pool" "pool" {
+  name_label   = "pool"
+  join_supporters = [
+    {
+      host = local.env_vars["SUPPORTER_HOST"]
+      username = local.env_vars["SUPPORTER_USERNAME"]
+      password = local.env_vars["SUPPORTER_PASSWORD"]
+    }
+  ]
+  eject_supporters = [ data.xenserver_host.supporter.data_items[0].uuid ]
+}
+
+
 # resource "xenserver_sr_nfs" "nfs" {
 #   name_label       = "NFS shared storage"
 #   name_description = "A test NFS storage repository"
@@ -43,25 +62,8 @@ provider "xenserver" {
 #   }
 # }
 
-# resource "xenserver_pool" "pool" {
+# resource "xenserver_pool" "pool_management" {
 #   name_label   = "pool"
-#   # default_sr = xenserver_sr_nfs.nfs.uuid
-#   # management_network = data.xenserver_pif.pif.data_items[0].network
-#   join_supporters = [
-#     {
-#       host = local.env_vars["SUPPORTER_HOST"]
-#       username = local.env_vars["SUPPORTER_USERNAME"]
-#       password = local.env_vars["SUPPORTER_PASSWORD"]
-#     }
-#   ]
-# }
-
-# comment out the following block for the second run
-data "xenserver_host" "supporter" {
-  is_coordinator = false
-}
-
-resource "xenserver_pool" "pool" {
-  name_label   = "pool"
-  eject_supporters = [ data.xenserver_host.supporter.data_items[0].uuid ]
-}
+#   default_sr = data.xenserver_sr_nfs.nfs.data_items[0].uuid
+#   management_network = data.xenserver_pif.pif.data_items[0].uuid
+# }

xenserver/snapshot_resource.go

@@ -59,7 +59,7 @@ func (r *snapshotResource) Schema(_ context.Context, _ resource.SchemaRequest, r
 			},
 			"revert": schema.BoolAttribute{
 				MarkdownDescription: "Set to `true` if you want to revert this snapshot to VM, default to be `false`." +
-					"\n\n-> **Note:** When `revert` is true, the snapshot resource will be updated with new configuration first and then revert to VM." +
+					"\n\n-> **Note:** `revert` only works after the snapshot resource created. When `revert` is true, the snapshot resource attributes will be updated first, for example `name_label`. And then revert to VM." +
 					"\n\n~> **Warning:** After revert, the VM `hard_drive` will be updated. If snapshot revert to the VM resource defined in 'main.tf', it'll cause issue when continue execute terraform commands. There's a suggest solution to resolve this issue, follow the steps: <br>" +
 					"1. run `terraform state show xenserver_snapshot.<snapshot_resource_name>`, get the revert VM's UUID 'vm_uuid' and revert VDIs' UUID 'vdi_uuid'.<br>" +
 					"2. run `terraform state rm xenserver_vm.<vm_resource_name>` to remove the VM resource state.<br>" +