Azure Database for PostgreSQL
An Azure managed PostgreSQL database service for app development and deployment.
Azure Database for PostgreSQL - Flexible server is currently unavailable. We are working to resolve the problem (9 days running)
![](data:image/svg+xml, %3Csvg xmlns='http://www.w3.org/2000/svg' height='64' class='font-weight-bold' style='font: 600 30.11764705882353px "SegoeUI", Arial' width='64'%3E%3Ccircle fill='hsl(108.80000000000001, 37%, 20%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3ERS%3C/text%3E%3C/svg%3E)
Richard Sleegers
1
Reputation point
Initial indication was an expired SSL certificate dated Apr 27, however server cannot be stopped or restarted, and databases are unavailable in the portal (Failed to load PostgreSQL flexible server databases.)
Location East US 2, Availability zone 1
Have upgraded to Developer support, but there are no options to continue the help/support request except to come to this forum.
Azure Database for PostgreSQL
-
![](data:image/svg+xml, %3Csvg xmlns='http://www.w3.org/2000/svg' height='64' class='font-weight-bold' style='font: 600 30.11764705882353px "SegoeUI", Arial' width='64'%3E%3Ccircle fill='hsl(313.6, 43%, 40%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3EIK%3C/text%3E%3C/svg%3E)
Igor Kolibaba 6 Reputation points2026-05-05T20:31:38.9166667+00:00 That message (“Flexible server is currently unavailable. We are working to resolve the problem”) + can’t stop/restart + portal can’t even list databases is almost always an Azure platform/control-plane or stuck maintenance/failover state, not something you can fix from inside PostgreSQL.
Here’s the most effective escalation + recovery playbook.
1) Confirm what Azure thinks is happening (takes 5–10 minutes)
In the server’s blade in Azure Portal:
- Resource Health Look for states like maintenance/fault/unavailable. Microsoft support and internal teams will ask for this first. (This is also recommended in similar “stuck/unavailable” cases.) ([Microsoft Learn](https://learn.microsoft.com/en-us/answers/a/12740593?utm_source=chatgpt.com"Azure Database for PostgreSQL Flexible Server stuck in ..."))
Activity Log (for the resource / resource group) Look for entries around when it went down (patching, failover, restart attempts). ([Microsoft Learn](https://learn.microsoft.com/en-us/answers/a/12740593?utm_source=chatgpt.com"Azure Database for PostgreSQL Flexible Server stuck in ..."))
Service Health (subscription) Check if there’s an ongoing incident in East US 2 affecting PostgreSQL Flexible Server. ([Microsoft Learn](https://learn.microsoft.com/en-us/answers/a/12740593?utm_source=chatgpt.com"Azure Database for PostgreSQL Flexible Server stuck in ..."))
In the server menu: Support + Troubleshooting / Diagnostics Run any built-in diagnostics it offers; Microsoft explicitly directs people to the troubleshooting tool for this exact “unavailable” banner. ([Microsoft Learn](https://learn.microsoft.com/en-us/answers/questions/5840303/your-azure-database-for-postgresql-flexible-server?utm_source=chatgpt.com"Your Azure Database for PostgreSQL - Flexible server is ..."))
2) Don’t get stuck on “restart” (it only appears when status is Ready)
Microsoft’s own doc says the Restart button is shown only when the server status is Ready. If your server is “Unavailable,” restart may be blocked by design. ([Microsoft Learn](https://learn.microsoft.com/en-us/azure/postgresql/configure-maintain/how-to-restart-server?utm_source=chatgpt.com"Restart PostgreSQL engine - Azure"))
3) If High Availability (HA) is enabled: try a failover (often unblocks “stuck” states)
If the server is zone-redundant HA, you can try planned failover via CLI:
az postgres flexible-server restart \ --resource-group <rg> \ --name <server> \ --failover PlannedThis is the documented command pattern for HA failover on Flexible Server. ([Azure Documentation for China](https://docs.azure.cn/en-us/postgresql/high-availability/how-to-configure-high-availability?utm_source=chatgpt.com"Configure High Availability - Azure Database for PostgreSQL"))
If HA is not enabled, failover commands won’t work.
4) Fastest “get me running” mitigation: restore to a new server (Point-in-Time Restore)
When the platform won’t bring the server back, the quickest way to recover service is usually:
Restore to latest restore point (creates a new flexible server from backups) ([Azure Documentation for China](https://docs.azure.cn/en-us/postgresql/backup-restore/how-to-restore-latest-restore-point?utm_source=chatgpt.com"Restore to latest restore point"))
Choose a different Availability Zone (or even a different region) if you suspect a zonal/regional issue. Microsoft docs explicitly mention using PITR to recover from zone-level failures by deploying a new server in another zone. ([GitHub](https://github.com/MicrosoftDocs/azure-databases-docs/blob/main/articles/postgresql/high-availability/concepts-high-availability.md?utm_source=chatgpt.com"azure-databases-docs/articles/postgresql/high-availability/ ..."))
Portal path is typically: Server → Overview → Restore. ([Azure Documentation for China](https://docs.azure.cn/en-us/postgresql/backup-restore/how-to-restore-latest-restore-point?utm_source=chatgpt.com"Restore to latest restore point"))
This is often the best move when it’s been days and you need uptime now.
5) About the “expired SSL certificate” clue
If the Azure Portal itself can’t load the DB list and stop/restart is blocked, it’s unlikely to be only a client TLS trust issue. That said:
Still verify your app isn’t failing due to client CA / SSL settings, but the portal symptoms point to platform-side unavailability. (Microsoft’s troubleshooting guidance distinguishes transient vs persistent issues and recommends opening support when it’s persistent.) ([Microsoft Learn](https://learn.microsoft.com/en-us/azure/postgresql/troubleshoot/how-to-troubleshoot-common-connection-issues?utm_source=chatgpt.com"Troubleshoot connection issues to Azure Database ..."))
6) If the portal won’t let you continue a support request: open a ticket from the right place
Use Microsoft’s official “create a support request” flow:
From the resource: Support + Troubleshooting → Create/New support request ([Microsoft Learn](https://learn.microsoft.com/en-us/azure/azure-portal/supportability/how-to-create-azure-support-request?utm_source=chatgpt.com"How to create an Azure support request - Azure portal"))
If that option is missing/buggy for your plan/tenant, this is a known scenario people hit on Developer support. ([Microsoft Learn](https://learn.microsoft.com/en-us/answers/questions/2167987/unable-to-create-a-support-request-in-azure-portal?utm_source=chatgpt.com"Unable to Create a Support Request in Azure Portal")) Workarounds that usually work:
Start from Help + support at the subscription level (not inside the DB blade)
Use the “New support request” entry point described in the doc above ([Microsoft Learn](https://learn.microsoft.com/en-us/azure/azure-portal/supportability/how-to-create-azure-support-request?utm_source=chatgpt.com"How to create an Azure support request - Azure portal"))
If you have Unified/Services Hub access, you can open requests there too ([Microsoft Learn](https://learn.microsoft.com/en-us/services-hub/unified/support/open-support-requests?utm_source=chatgpt.com"Open Support requests"))
What to put in the ticket (copy/paste)
“PostgreSQL Flexible Server Unavailable for 9 days; cannot stop/restart; portal cannot load databases”
Region East US 2, AZ 1
Include screenshots of Resource Health, relevant Activity Log events, and any Service Health incident IDs.
If you tell me whether HA is enabled and what Resource Health says (exact wording), I’ll recommend the best next step (HA failover vs PITR restore) and give you the exact Azure CLI commands + the safest restore choices (zone/region) to minimize further downtime.That message (“Flexible server is currently unavailable. We are working to resolve the problem”) + can’t stop/restart + portal can’t even list databases is almost always an Azure platform/control-plane or stuck maintenance/failover state, not something you can fix from inside PostgreSQL.
Here’s the most effective escalation + recovery playbook.
1) Confirm what Azure thinks is happening (takes 5–10 minutes)
In the server’s blade in Azure Portal:
Resource Health
Look for states like maintenance/fault/unavailable. Microsoft support and internal teams will ask for this first. (This is also recommended in similar “stuck/unavailable” cases.) ([Microsoft Learn](https://learn.microsoft.com/en-us/answers/a/12740593?utm_source=chatgpt.com"Azure Database for PostgreSQL Flexible Server stuck in ..."))Activity Log (for the resource / resource group)
Look for entries around when it went down (patching, failover, restart attempts). ([Microsoft Learn](https://learn.microsoft.com/en-us/answers/a/12740593?utm_source=chatgpt.com"Azure Database for PostgreSQL Flexible Server stuck in ..."))Service Health (subscription)
Check if there’s an ongoing incident in East US 2 affecting PostgreSQL Flexible Server. ([Microsoft Learn](https://learn.microsoft.com/en-us/answers/a/12740593?utm_source=chatgpt.com"Azure Database for PostgreSQL Flexible Server stuck in ..."))In the server menu: Support + Troubleshooting / Diagnostics
Run any built-in diagnostics it offers; Microsoft explicitly directs people to the troubleshooting tool for this exact “unavailable” banner. ([Microsoft Learn](https://learn.microsoft.com/en-us/answers/questions/5840303/your-azure-database-for-postgresql-flexible-server?utm_source=chatgpt.com"Your Azure Database for PostgreSQL - Flexible server is ..."))2) Don’t get stuck on “restart” (it only appears when status is Ready)
Microsoft’s own doc says the Restart button is shown only when the server status is Ready. If your server is “Unavailable,” restart may be blocked by design. ([Microsoft Learn](https://learn.microsoft.com/en-us/azure/postgresql/configure-maintain/how-to-restart-server?utm_source=chatgpt.com"Restart PostgreSQL engine - Azure"))
3) If High Availability (HA) is enabled: try a failover (often unblocks “stuck” states)
If the server is zone-redundant HA, you can try planned failover via CLI:
az postgres flexible-server restart \ --resource-group <rg> \ --name <server> \ --failover PlannedThis is the documented command pattern for HA failover on Flexible Server. ([Azure Documentation for China](https://docs.azure.cn/en-us/postgresql/high-availability/how-to-configure-high-availability?utm_source=chatgpt.com"Configure High Availability - Azure Database for PostgreSQL"))
If HA is not enabled, failover commands won’t work.
4) Fastest “get me running” mitigation: restore to a new server (Point-in-Time Restore)
When the platform won’t bring the server back, the quickest way to recover service is usually:
Restore to latest restore point (creates a new flexible server from backups) ([Azure Documentation for China](https://docs.azure.cn/en-us/postgresql/backup-restore/how-to-restore-latest-restore-point?utm_source=chatgpt.com"Restore to latest restore point"))
Choose a different Availability Zone (or even a different region) if you suspect a zonal/regional issue. Microsoft docs explicitly mention using PITR to recover from zone-level failures by deploying a new server in another zone. ([GitHub](https://github.com/MicrosoftDocs/azure-databases-docs/blob/main/articles/postgresql/high-availability/concepts-high-availability.md?utm_source=chatgpt.com"azure-databases-docs/articles/postgresql/high-availability/ ..."))
Portal path is typically: Server → Overview → Restore. ([Azure Documentation for China](https://docs.azure.cn/en-us/postgresql/backup-restore/how-to-restore-latest-restore-point?utm_source=chatgpt.com"Restore to latest restore point"))
This is often the best move when it’s been days and you need uptime now.
5) About the “expired SSL certificate” clue
If the Azure Portal itself can’t load the DB list and stop/restart is blocked, it’s unlikely to be only a client TLS trust issue. That said:
Still verify your app isn’t failing due to client CA / SSL settings, but the portal symptoms point to platform-side unavailability. (Microsoft’s troubleshooting guidance distinguishes transient vs persistent issues and recommends opening support when it’s persistent.) ([Microsoft Learn](https://learn.microsoft.com/en-us/azure/postgresql/troubleshoot/how-to-troubleshoot-common-connection-issues?utm_source=chatgpt.com"Troubleshoot connection issues to Azure Database ..."))
6) If the portal won’t let you continue a support request: open a ticket from the right place
Use Microsoft’s official “create a support request” flow:
From the resource: Support + Troubleshooting → Create/New support request ([Microsoft Learn](https://learn.microsoft.com/en-us/azure/azure-portal/supportability/how-to-create-azure-support-request?utm_source=chatgpt.com"How to create an Azure support request - Azure portal"))
If that option is missing/buggy for your plan/tenant, this is a known scenario people hit on Developer support. ([Microsoft Learn](https://learn.microsoft.com/en-us/answers/questions/2167987/unable-to-create-a-support-request-in-azure-portal?utm_source=chatgpt.com"Unable to Create a Support Request in Azure Portal"))
Workarounds that usually work:Start from Help + support at the subscription level (not inside the DB blade)
Use the “New support request” entry point described in the doc above ([Microsoft Learn](https://learn.microsoft.com/en-us/azure/azure-portal/supportability/how-to-create-azure-support-request?utm_source=chatgpt.com"How to create an Azure support request - Azure portal"))
If you have Unified/Services Hub access, you can open requests there too ([Microsoft Learn](https://learn.microsoft.com/en-us/services-hub/unified/support/open-support-requests?utm_source=chatgpt.com"Open Support requests"))
What to put in the ticket (copy/paste)
“PostgreSQL Flexible Server Unavailable for 9 days; cannot stop/restart; portal cannot load databases”
Region East US 2, AZ 1
Include screenshots of Resource Health, relevant Activity Log events, and any Service Health incident IDs.
-
Richard Sleegers 1 Reputation point
2026-05-05T20:44:06.84+00:00 I had tried #3, and received this error:
az postgres flexible-server restart --name ### --resource-group ###(InternalServerError) An unexpected error occured while processing the request. Tracking ID: 'c6f14b44-b665-41c7-9a13-b869ae29b6c9'
Code: InternalServerError
Message: An unexpected error occured while processing the request. Tracking ID: 'c6f14b44-b665-41c7-9a13-b869ae29b6c9'
-
![](data:image/svg+xml, %3Csvg xmlns='http://www.w3.org/2000/svg' height='64' class='font-weight-bold' style='font: 600 30.11764705882353px "SegoeUI", Arial' width='64'%3E%3Ccircle fill='hsl(208, 49%, 30%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3ESJ%3C/text%3E%3C/svg%3E)
SAI JAGADEESH KUDIPUDI 2,950 Reputation points Microsoft External Staff Moderator2026-05-05T20:52:35.2833333+00:00 Hi **Richard Sleegers,
**Could you please share requested details in private message? -
Igor Kolibaba 6 Reputation points
2026-05-05T20:55:40.78+00:00 That InternalServerError + Tracking ID from
az postgres flexible-server restartalmost always means the Azure control plane can’t complete the operation (it’s not a PostgreSQL-level error you can fix with SQL). In other words: the “restart” request reached Azure, but the service hit an internal fault and returned a correlation/Tracking ID for Microsoft to investigate. (This is very similar to other Flexible Server cases where Microsoft had to perform a backend restart.) ([Microsoft Learn](https://learn.microsoft.com/en-us/answers/questions/2243112/azure-postgresql-flexible-server-restart-failed-in"Azure PostgreSQL Flexible Server restart failed – InternalServerError - Microsoft Q&A"))Here’s what to do next, in the order that tends to get people unstuck fastest:
1) Check if HA is enabled — and if yes, force a failover
First, see whether High Availability is on:
az postgres flexible-server show -g <rg> -n <server> --query "highAvailability" -o jsonIf HA is enabled, try failover via restart (this is the documented way; “failover” is integrated into the restart command): ([Microsoft Learn](https://learn.microsoft.com/en-us/answers/questions/2084882/azure-postgresql-flexible-server-az-command-line-t"Azure PostgreSQL Flexible server - AZ command line to failover - Microsoft Q&A"))
# Planned failover (cleaner) az postgres flexible-server restart -g <rg> -n <server> --failover Planned # Forced failover (use when planned won’t run) az postgres flexible-server restart -g <rg> -n <server> --failover ForcedMicrosoft Q&A confirms those Planned/Forced flags are the correct CLI approach for Flexible Server failover. ([Microsoft Learn](https://learn.microsoft.com/en-us/answers/questions/2084882/azure-postgresql-flexible-server-az-command-line-t"Azure PostgreSQL Flexible server - AZ command line to failover - Microsoft Q&A")) (And the reliability docs describe planned vs forced failover behavior.) ([Azure Documentation for China](https://docs.azure.cn/en-us/reliability/reliability-postgresql-flexible-server?utm_source=chatgpt.com"Reliability and High Availability in PostgreSQL - Flexible ..."))
If both planned and forced failover also throw InternalServerError, that’s a strong sign the service is stuck in a backend state that only Microsoft can clear or you need to restore to a new server.
2) Fastest mitigation: restore to a new server (Point-in-Time Restore)
If the server has been “unavailable” for days, the most reliable way to get running again is usually restore to a new server from backups (PITR). Azure CLI supports restoring to “now” or a specific timestamp. ([Microsoft Learn](https://learn.microsoft.com/en-us/cli/azure/postgres/flexible-server?view=azure-cli-latest"az postgres flexible-server | Microsoft Learn"))
# Restore to latest point-in-time (creates a NEW server) az postgres flexible-server restore \ -g <target-rg> -n <new-server-name> \ --source-server <server>Or to a specific time:
az postgres flexible-server restore \ -g <target-rg> -n <new-server-name> \ --source-server <server> \ --restore-time "2026-05-05T20:10:00Z"Tip (important for your “zone 1 / East US 2” scenario): restore into a different Availability Zone if you suspect a zonal issue. The CLI restore supports
--zone(see the parameter list for restore). ([Microsoft Learn](https://learn.microsoft.com/en-us/cli/azure/postgres/flexible-server?view=azure-cli-latest"az postgres flexible-server | Microsoft Learn"))If you need cross-region DR and you have geo-redundant backups, there’s also
az postgres flexible-server geo-restore. ([Microsoft Learn](https://learn.microsoft.com/en-us/cli/azure/postgres/flexible-server?view=azure-cli-latest"az postgres flexible-server | Microsoft Learn"))3) Escalation: you need Microsoft to act on that Tracking ID
That Tracking ID (
c6f14b44-b665-41c7-9a13-b869ae29b6c9) is exactly what support uses to find backend logs.If you’re blocked from creating a support request even after upgrading to Developer Support, two common causes are:
- The support plan isn’t linked to the subscription (needs Access ID / Contract ID), or
You don’t have the needed RBAC role (Owner/Contributor/Support Request Contributor). ([Microsoft Learn](https://learn.microsoft.com/en-nz/answers/questions/5867045/i-have-a-developer-support-plan-unable-to-submit-s"I have a Developer Support Plan, Unable to Submit Support Request - Microsoft Q&A"))
Also, there are known portal flows where people on Developer support report they can’t see the final submit/create step until the above is fixed. ([Microsoft Learn](https://learn.microsoft.com/en-nz/answers/questions/5867045/i-have-a-developer-support-plan-unable-to-submit-s"I have a Developer Support Plan, Unable to Submit Support Request - Microsoft Q&A"))
What I would do in your exact situation
Try Forced failover (only if HA is enabled). ([Microsoft Learn](https://learn.microsoft.com/en-us/answers/questions/2084882/azure-postgresql-flexible-server-az-command-line-t"Azure PostgreSQL Flexible server - AZ command line to failover - Microsoft Q&A"))
If it still errors: do a PITR restore to a new server, ideally into another AZ (and even another region if you have geo-backup). ([Microsoft Learn](https://learn.microsoft.com/en-us/cli/azure/postgres/flexible-server?view=azure-cli-latest"az postgres flexible-server | Microsoft Learn"))
In parallel, open a support request and include the Tracking ID; Microsoft often has to “restart from the backend” in these stuck states. ([Microsoft Learn](https://learn.microsoft.com/en-us/answers/questions/2243112/azure-postgresql-flexible-server-restart-failed-in"Azure PostgreSQL Flexible Server restart failed – InternalServerError - Microsoft Q&A"))
If you paste the output of:
az postgres flexible-server show -g <rg> -n <server> --query "{state:state,ha:highAvailability,zone:availabilityZone,region:location}" -o json…I’ll tell you immediately whether failover is even possible and the safest restore target (zone/region) for quickest recovery.That InternalServerError + Tracking ID from
az postgres flexible-server restartalmost always means the Azure control plane can’t complete the operation (it’s not a PostgreSQL-level error you can fix with SQL). In other words: the “restart” request reached Azure, but the service hit an internal fault and returned a correlation/Tracking ID for Microsoft to investigate. (This is very similar to other Flexible Server cases where Microsoft had to perform a backend restart.) ([Microsoft Learn](https://learn.microsoft.com/en-us/answers/questions/2243112/azure-postgresql-flexible-server-restart-failed-in"Azure PostgreSQL Flexible Server restart failed – InternalServerError - Microsoft Q&A"))Here’s what to do next, in the order that tends to get people unstuck fastest:
1) Check if HA is enabled — and if yes, force a failover
First, see whether High Availability is on:
az postgres flexible-server show -g <rg> -n <server> --query "highAvailability" -o jsonIf HA is enabled, try failover via restart (this is the documented way; “failover” is integrated into the restart command): ([Microsoft Learn](https://learn.microsoft.com/en-us/answers/questions/2084882/azure-postgresql-flexible-server-az-command-line-t"Azure PostgreSQL Flexible server - AZ command line to failover - Microsoft Q&A"))
# Planned failover (cleaner) az postgres flexible-server restart -g <rg> -n <server> --failover Planned # Forced failover (use when planned won’t run) az postgres flexible-server restart -g <rg> -n <server> --failover ForcedMicrosoft Q&A confirms those Planned/Forced flags are the correct CLI approach for Flexible Server failover. ([Microsoft Learn](https://learn.microsoft.com/en-us/answers/questions/2084882/azure-postgresql-flexible-server-az-command-line-t"Azure PostgreSQL Flexible server - AZ command line to failover - Microsoft Q&A"))
(And the reliability docs describe planned vs forced failover behavior.) ([Azure Documentation for China](https://docs.azure.cn/en-us/reliability/reliability-postgresql-flexible-server?utm_source=chatgpt.com"Reliability and High Availability in PostgreSQL - Flexible ..."))If both planned and forced failover also throw InternalServerError, that’s a strong sign the service is stuck in a backend state that only Microsoft can clear or you need to restore to a new server.
2) Fastest mitigation: restore to a new server (Point-in-Time Restore)
If the server has been “unavailable” for days, the most reliable way to get running again is usually restore to a new server from backups (PITR). Azure CLI supports restoring to “now” or a specific timestamp. ([Microsoft Learn](https://learn.microsoft.com/en-us/cli/azure/postgres/flexible-server?view=azure-cli-latest"az postgres flexible-server | Microsoft Learn"))
# Restore to latest point-in-time (creates a NEW server) az postgres flexible-server restore \ -g <target-rg> -n <new-server-name> \ --source-server <server>Or to a specific time:
az postgres flexible-server restore \ -g <target-rg> -n <new-server-name> \ --source-server <server> \ --restore-time "2026-05-05T20:10:00Z"Tip (important for your “zone 1 / East US 2” scenario): restore into a different Availability Zone if you suspect a zonal issue. The CLI restore supports
--zone(see the parameter list for restore). ([Microsoft Learn](https://learn.microsoft.com/en-us/cli/azure/postgres/flexible-server?view=azure-cli-latest"az postgres flexible-server | Microsoft Learn"))If you need cross-region DR and you have geo-redundant backups, there’s also
az postgres flexible-server geo-restore. ([Microsoft Learn](https://learn.microsoft.com/en-us/cli/azure/postgres/flexible-server?view=azure-cli-latest"az postgres flexible-server | Microsoft Learn"))3) Escalation: you need Microsoft to act on that Tracking ID
That Tracking ID (
c6f14b44-b665-41c7-9a13-b869ae29b6c9) is exactly what support uses to find backend logs.If you’re blocked from creating a support request even after upgrading to Developer Support, two common causes are:
The support plan isn’t linked to the subscription (needs Access ID / Contract ID), or
You don’t have the needed RBAC role (Owner/Contributor/Support Request Contributor). ([Microsoft Learn](https://learn.microsoft.com/en-nz/answers/questions/5867045/i-have-a-developer-support-plan-unable-to-submit-s"I have a Developer Support Plan, Unable to Submit Support Request - Microsoft Q&A"))
Also, there are known portal flows where people on Developer support report they can’t see the final submit/create step until the above is fixed. ([Microsoft Learn](https://learn.microsoft.com/en-nz/answers/questions/5867045/i-have-a-developer-support-plan-unable-to-submit-s"I have a Developer Support Plan, Unable to Submit Support Request - Microsoft Q&A"))
What I would do in your exact situation
Try Forced failover (only if HA is enabled). ([Microsoft Learn](https://learn.microsoft.com/en-us/answers/questions/2084882/azure-postgresql-flexible-server-az-command-line-t"Azure PostgreSQL Flexible server - AZ command line to failover - Microsoft Q&A"))
If it still errors: do a PITR restore to a new server, ideally into another AZ (and even another region if you have geo-backup). ([Microsoft Learn](https://learn.microsoft.com/en-us/cli/azure/postgres/flexible-server?view=azure-cli-latest"az postgres flexible-server | Microsoft Learn"))
In parallel, open a support request and include the Tracking ID; Microsoft often has to “restart from the backend” in these stuck states. ([Microsoft Learn](https://learn.microsoft.com/en-us/answers/questions/2243112/azure-postgresql-flexible-server-restart-failed-in"Azure PostgreSQL Flexible Server restart failed – InternalServerError - Microsoft Q&A"))
Please paste the output of:
az postgres flexible-server show -g <rg> -n <server> --query "{state:state,ha:highAvailability,zone:availabilityZone,region:location}" -o jsonand I’ll tell you whether failover is even possible and the safest restore target (zone/region) for quickest recovery.
-
Richard Sleegers 1 Reputation point
2026-05-05T21:17:12.36+00:00 Thanks for the update, I meant to reply that HA was not enabled. It has been more than 7 days, so I do not see a backup available to restore. The server finally stopped and I have initiated a Start which is showing some signs of operation, but not fully in Ready state yet.
Running the az command returns:
(ResourceNotFound) The Resource 'Microsoft.DBforPostgreSQL/flexibleServers/psql-#.postgres.database.azure.com' under resource group 'rg-#' was not found. For more details please go to https://aka.ms/ARMResourceNotFoundFixCode: ResourceNotFound
Message: The Resource 'Microsoft.DBforPostgreSQL/flexibleServers/psql-#.postgres.database.azure.com' under resource group 'rg-#' was not found. For more details please go to https://aka.ms/ARMResourceNotFoundFix
-
Igor Kolibaba 6 Reputation points
2026-05-05T21:51:16.0066667+00:00 Got it — with HA not enabled, this narrows down a lot.
Why you’re getting
ResourceNotFoundIn your CLI error, you’re passing the FQDN as the server name:
psql-#.postgres.database.azure.comBut
az postgres flexible-server ... --nameexpects the resource name only, e.g.:psql-#If you include
.postgres.database.azure.com, Azure Resource Manager won’t find a resource with that name, so you getResourceNotFound. (Your earlier notes also show the same pattern.)Fix it: run these in order
- Make sure you’re in the right subscription
az account show --query "{name:name, id:id}" -o json az account set --subscription "<subscription-id-or-name>"List the flexible servers in that resource group
az postgres flexible-server list -g rg-# -o tableThis will show the actual server resource name(s). Use exactly what appears under Name.
Then “show” using the short name
az postgres flexible-server show -g rg-# -n psql-#If that works, you can also check current lifecycle/provisioning state:
az postgres flexible-server show -g rg-# -n psql-# --query "{state:state, provisioningState:provisioningState}" -o jsonAbout “no backup available after >7 days”
That’s consistent with Azure’s default backup retention:
Default retention is 7 days, configurable up to 35 days. ([Microsoft Learn](https://learn.microsoft.com/en-us/azure/postgresql/backup-restore/concepts-backup-restore?utm_source=chatgpt.com"Backup and restore in Azure Database for PostgreSQL"))
So if your retention was left at default 7 days, and you’re now past that window, PITR restore points older than 7 days won’t be available.
You can verify what your server was configured for (once
showworks):az postgres flexible-server show -g rg-# -n psql-# --query "backup" -o json“Start is running but not Ready yet”
That usually means the resource is in a transitional provisioningState (Starting / Updating / Repairing) and the data plane may still be coming up. The best thing you can do right now is keep checking state and activity:
Activity Log in portal (look for Start/Stop, maintenance, failures)
CLI:
az monitor activity-log list --resource-group rg-# --max-events 20 -o tableIf you still get
ResourceNotFoundeven with the short nameThen it’s almost always one of these:
wrong subscription (most common)
wrong resource group
the server was deleted/recreated and you’re using an old name
you’re logged into a different tenant
The
list -g rg-#step above will reveal that immediately.If you paste the output of:
az postgres flexible-server list -g rg-# -o table(Ignore/substitute sensitive bits), I’ll tell you exactly what name to use and whether Azure currently thinks it’s “Starting”, “Ready”, or stuck again.Got it — with HA not enabled, this narrows down a lot.
Why you’re getting
ResourceNotFoundIn your CLI error, you’re passing the FQDN as the server name:
psql-#.postgres.database.azure.comBut
az postgres flexible-server ... --nameexpects the resource name only, e.g.:psql-#If you include
.postgres.database.azure.com, Azure Resource Manager won’t find a resource with that name, so you getResourceNotFound. (Your earlier notes also show the same pattern.)Fix it: run these in order
Make sure you’re in the right subscription
az account show --query "{name:name, id:id}" -o json az account set --subscription "<subscription-id-or-name>"List the flexible servers in that resource group
az postgres flexible-server list -g rg-# -o tableThis will show the actual server resource name(s). Use exactly what appears under Name.
Then “show” using the short name
az postgres flexible-server show -g rg-# -n psql-#If that works, you can also check current lifecycle/provisioning state:
az postgres flexible-server show -g rg-# -n psql-# --query "{state:state, provisioningState:provisioningState}" -o jsonAbout “no backup available after >7 days”
That’s consistent with Azure’s default backup retention:
Default retention is 7 days, configurable up to 35 days. ([Microsoft Learn](https://learn.microsoft.com/en-us/azure/postgresql/backup-restore/concepts-backup-restore?utm_source=chatgpt.com"Backup and restore in Azure Database for PostgreSQL"))
So if your retention was left at default 7 days, and you’re now past that window, PITR restore points older than 7 days won’t be available.
You can verify what your server was configured for (once
showworks):az postgres flexible-server show -g rg-# -n psql-# --query "backup" -o json“Start is running but not Ready yet”
That usually means the resource is in a transitional provisioningState (Starting / Updating / Repairing) and the data plane may still be coming up. The best thing you can do right now is keep checking state and activity:
Activity Log in portal (look for Start/Stop, maintenance, failures)
CLI:
az monitor activity-log list --resource-group rg-# --max-events 20 -o tableIf you still get
ResourceNotFoundeven with the short nameThen it’s almost always one of these:
wrong subscription (most common)
wrong resource group
the server was deleted/recreated and you’re using an old name
you’re logged into a different tenant
The
list -g rg-#step above will reveal that immediately.If you paste the output of:
az postgres flexible-server list -g rg-# -o table(Ignore/substitute sensitive bits), I’ll tell you exactly what name to use and whether Azure currently thinks it’s “Starting”, “Ready”, or stuck again.
-
Richard Sleegers 1 Reputation point
2026-05-06T15:01:55.53+00:00 Yes, my mistake, some commands require the database name, and some the resource name. Correcting for that, the output is:
az postgres flexible-server show -g rg-# -n psql-# --query "{state:state,ha:highAvailability,zone:availabilityZone,region:location}" -o json{ "ha": { "mode": "Disabled", "standbyAvailabilityZone": null, "state": "NotEnabled" }, "region": "East US 2", "state": "Ready", "zone": "1" }After a time, the resource is now showing "Ready" and the first backup has populated, after I did a complete Stop and Start (waiting 30-60 minutes between each).
Some operations are now functioning, initially the SSL certificate for the domain ebea6f997ae5.database.azure.com was still showing as expired, so I need to confirm certificates are being updated.(Igor, I appreciate your posts, they contain great information from your prompts; be aware your posts are being pasted twice, and the links bleed into the title of the link).
Sign in to comment
Sign in to answer