Azure Database for PostgreSQL Flexible Server (Australia East) has been stuck in "Updating" state for over 4 hours since approximately 08:35 UTC on April 6, 2026

Question

Azure Database for PostgreSQL Flexible Server (Australia East) has been stuck in "Updating" state for over 4 hours since approximately 08:35 UTC on April 6, 2026

Mohamed Abdelaziz 0

My production Azure Database for PostgreSQL Flexible Server has been stuck in "Updating" state for over 4 hours and I cannot perform any operations on it.

Resource details:

Region: Australia East
SKU: Burstable B2s
PostgreSQL version: 17

What happened:

Server had CPU credits fully exhausted (only ~2 credits remaining from midnight)
Azure Insights reported 100% CPU at 08:35 UTC on April 6, 2026
Server became unreachable — all connections refused on port 5432
Azure entered the server into Updating state — likely an internal recovery/maintenance operation
The server has been updating for 18+ hours with no progress

What I tried:

restart → ServerIsNotReady
stop → ServerIsNotReady
parameter set → ServerIsBusy
restore to new server → InternalServerError
delete restored server → succeeded, but the original is still stuck

Azure Service Health previously flagged an active Critical incident:

"Your Azure Database for PostgreSQL - Flexible server is currently unavailable. We are working to resolve the problem."

Questions:

Is there any way to force-cancel or roll back the stuck internal Updating operation?
Can a restore be initiated while the source server is in Updating state?
How long should a WAL recovery operation take on a Burstable B2s with depleted CPU credits?
Is there a way to escalate this without a paid support plan?My production Azure Database for PostgreSQL Flexible Server has been stuck in "Updating" state for over 4 hours and I cannot perform any operations on it. Resource details:
- - Region: Australia East
- SKU: Burstable B2s
  - PostgreSQL version: 17
What happened:
1. Server had CPU credits fully exhausted (only ~2 credits remaining from midnight)
2. Azure Insights reported 100% CPU at 08:35 UTC on April 6, 2026
3. Server became unreachable — all connections refused on port 5432
4. Azure entered the server into Updating state — likely an internal recovery/maintenance operation
5. The server has been Updating for 18+ hours with no progress
What I tried:
- restart → ServerIsNotReady
- stop → ServerIsNotReady
- parameter set → ServerIsBusy
- restore to new server → InternalServerError
- delete restored server → succeeded, but original still stuck
Azure Service Health previously flagged an active Critical incident:

"Your Azure Database for PostgreSQL - Flexible server is currently unavailable. We are working to resolve the problem."

Questions:
1. Is there any way to force-cancel or roll back the stuck internal Updating operation?
2. Can a restore be initiated while the source server is in Updating state?
3. How long should a WAL recovery operation take on a Burstable B2s with depleted CPU credits?

SAI JAGADEESH KUDIPUDI 1,750 Reputation points Microsoft External Staff Moderator

2026-04-07T06:32:47.8833333+00:00

Hi Mohamed Abdelaziz,
Thank you for reaching out to the Microsoft Q&A team and for sharing the details.
Could you please share the requested details in a private message This will allow us to investigate further into this request. Also, could you please confirm whether the server is still stuck in the “Updating” state at this moment?****
Mohamed Abdelaziz 0 Reputation points

2026-04-07T06:35:41.5666667+00:00

Hi SAI JAGADEESH KUDIPUDI,

Thank you for the quick response. I will send the details via private message shortly.

To confirm, yes, the server is still stuck in the "Updating" state as of right now. It has been in this state for over 21 hours since approximately 08:35 UTC on April 6, 2026, with no change.

Port 5432 remains completely unreachable, and our production system is still fully down.

Please find the private message with all the requested details. We urgently need a backend engineer to force-reset the server provisioning state.

Thank you,
SAI JAGADEESH KUDIPUDI 1,750 Reputation points Microsoft External Staff Moderator

2026-04-07T06:41:24.3466667+00:00

Thank you for the update, Mohamed Abdelaziz, and for sharing the details,
We completely understand the impact and your concern. We are working with the backend team to mitigate this issue as soon as possible and are treating this on priority.