Merge pull request #202709 from sr-msft/fspg-ha-updates-0623

FSPG HA same-zone update

Author: PRMerger3

articles/postgresql/flexible-server/concepts-high-availability.md

@@ -45,6 +45,9 @@ This model of high availability deployment enables Flexible server to be highly
 Automatic backups are performed periodically from the primary database server, while the transaction logs are continuously archived to the backup storage from the standby replica. If the region supports availability zones, then backup data is stored on zone-redundant storage (ZRS). In regions that doesn't support availability zones, backup data is stored on local redundant storage (LRS).   
 :::image type="content" source="./media/business-continuity/concepts-same-zone-high-availability-architecture.png" alt-text="Same-zone high availability"::: 
 
+>[!NOTE]
+> See the [HA limitation section](#high-availability---limitations) for a current restriction with same-zone HA deployment.
+
 ## Components and workflow
 
 ### Transaction completion
@@ -217,6 +220,9 @@ Flexible servers that are configured with high availability, log data is replica
 
 ## High availability - limitations
 
+>[!NOTE]
+> New server creates with **Same-zone HA** are currently restricted when you choose the primary server's AZ.  Workarounds are to (a) create your same-zone HA server without choosing the primary AZ, or (b) create as a single instance (non-HA) server and then enable same-zone HA after the server is created. 
+
 * High availability is not supported with burstable compute tier.
 * High availability is supported only in regions where multiple zones are available.
 * Due to synchronous replication to the standby server, especially with zone-redundant HA, applications can experience elevated write and commit latency.

articles/postgresql/flexible-server/how-to-manage-high-availability-portal.md

@@ -6,7 +6,7 @@ author: sr-msft
 ms.service: postgresql
 ms.subservice: flexible-server
 ms.topic: how-to
-ms.date: 06/16/2022
+ms.date: 06/23/2022
 ---
 
 # Manage high availability in Flexible Server
@@ -43,21 +43,24 @@ This section provides details specifically for HA-related fields. You can follow
 4.  If you chose the Availability zone in step 2 and if you chose zone-redundant HA, then you can choose the standby zone.
     :::image type="content" source="./media/how-to-manage-high-availability-portal/choose-standby-availability-zone.png" alt-text="Screenshot of Standby AZ selection.":::
 
-5.  If you want to change the default compute and storage, click  **Configure server**.
+>[!NOTE]
+> See the [HA limitation section](concepts-high-availability.md#high-availability---limitations) for a current restriction with same-zone HA deployment.  
+
+1.  If you want to change the default compute and storage, click  **Configure server**.
 
     :::image type="content" source="./media/how-to-manage-high-availability-portal/configure-server.png" alt-text="Screenshot of configure compute and storage screen.":::  
 
-6.  If high availability option is checked, the burstable tier will not be available to choose. You can choose either
+2.  If high availability option is checked, the burstable tier will not be available to choose. You can choose either
     **General purpose** or **Memory Optimized** compute tiers. Then you can select **compute size** for your choice from the dropdown.
 
     :::image type="content" source="./media/how-to-manage-high-availability-portal/select-compute.png" alt-text="Compute tier selection screen.":::  
 
 
-7.  Select **storage size** in GiB using the sliding bar and select the **backup retention period** between 7 days and 35 days.
+3.  Select **storage size** in GiB using the sliding bar and select the **backup retention period** between 7 days and 35 days.
 
     :::image type="content" source="./media/how-to-manage-high-availability-portal/storage-backup.png" alt-text="Screenshot of Storage Backup."::: 
 
-8. Click **Save**. 
+4. Click **Save**. 
 
 ## Enable high availability post server creation
 
@@ -140,6 +143,18 @@ Follow these steps to perform a planned failover from your primary to the standb
 >
 > * The overall end-to-end operation time may be longer than the actual downtime experienced by the application. Please measure the downtime from the application perspective.
 
+## Enabling Zone redundant HA after the region supports AZ
+
+There are Azure regions that do not support availability zones. If you have already deployed non-HA servers, you cannot directly enable zone redundant HA on the server, but you can perform restore and enable HA in that server.  Following steps shows how to enable Zone redundant HA for that server.
+
+1. From the overview page of the server, click **Restore** to [perform a PITR](how-to-restore-server-portal.md#restoring-to-the-latest-restore-point). Choose **Latest restore point**. 
+2. Choose a server name, availability zone.
+3. Click **Review+Create**".
+4. A new Flexible server will be created from the backup. 
+5. Once the new server is created, from the overview page of the server, follow the [guide](#enable-high-availability-post-server-creation) to enable HA.
+6. After data verification, you can optionally [delete](how-to-manage-server-portal.md#delete-a-server) the old server. 
+7. Make sure your clients connection strings are modified to point to your new HA-enabled server.
+   
 
 ## Next steps
 

articles/postgresql/flexible-server/overview.md

@@ -6,7 +6,7 @@ author: sunilagarwal
 ms.service: postgresql
 ms.subservice: flexible-server
 ms.topic: overview
-ms.date: 06/15/2022
+ms.date: 06/23/2022
 ms.custom: "mvc, references_regions"
 ---
 
@@ -134,7 +134,7 @@ $ New Zone-redundant high availability deployments are temporarily blocked in th
 
 $$ New server deployments are temporarily blocked in these regions. Already provisioned servers are fully supported.
 
-** Zone-redundant high availability can now be deployed when you provision new servers in these regions. Pre-existing servers deployed in AZ with *no preference* (which you can check on the Azure portal), the standby will be provisioned in the same AZ. To configure zone-redundant high availability, perform a point-in-time restore of the server and enable HA on the restored server.
+** Zone-redundant high availability can now be deployed when you provision new servers in these regions. Any existing servers deployed in AZ with *no preference* (which you can check on the Azure portal) prior to the region started to support AZ, even when you enable zone-redundant HA, the standby will be provisioned in the same AZ (same-zone HA) as the primary server. To enable zone-redundant high availability, [follow the steps.](how-to-manage-high-availability-portal.md#enabling-zone-redundant-ha-after-the-region-supports-az).
 
 <!-- We continue to add more regions for flexible server. -->
 > [!NOTE]