Postgresql Flexible Server Restore get stuck

Question

Postgresql Flexible Server Restore get stuck

Rafa Lachica Garrido 40

Good afternoon,

I have created a restore backup from a PostgreSQL Flexible server through Azure portal, using point in time to restore, however, after more than 3 hours, the backup is not yet ready.

I have check the credits from the original DB, and everything seems fine.

Thanks

Pilladi Padma Sai Manisha 9,055 Reputation points Microsoft External Staff Moderator

2026-05-29T06:48:48.5066667+00:00

Hi Rafa Lachica Garrido
Thankyou for reaching microsoft Q&A!
A Point-in-Time Restore (PITR) for Azure PostgreSQL Flexible Server typically completes within a reasonable timeframe, although the duration can vary depending on factors such as database size, storage performance, and regional service load.

To help investigate further, could you please provide the following details?

Source PostgreSQL Flexible Server name and Azure region.

Approximate database size and provisioned storage.

The exact time when the restore operation was initiated (UTC if possible).

Current status shown in the Azure portal for the restore operation.

Any error messages, Activity Log entries, or deployment operation details associated with the restore.

Whether the target restored server was created successfully and is stuck in a provisioning state, or if the restore request itself remains in progress.

A screenshot of the restore operation status from the Azure portal.

You can also check the Activity Log for the restore operation and share any correlation ID or operation ID if available.

With these details, we can better determine whether the restore is still progressing normally or if there is a backend issue requiring further investigation.
Pilladi Padma Sai Manisha 9,055 Reputation points Microsoft External Staff Moderator

2026-06-10T04:09:35.3866667+00:00

Hi
I hope you had a chance to review the information shared earlier, and I hope this information has been helpful! If you still have questions, please let us know what is needed in the comments so the question can be answered.

Answer accepted by question author

Andrew Taylor - COREZENN 980 Volunteer Moderator

Hello @Rafa Lachica Garrido

Thank you for reaching out to Microsoft Q&A. I understand your issue is with PostgreSQL Flexible server backup.

Restore duration for PostgreSQL Flexible Server PITR

This can be normal for a point-in-time restore on Azure Database for PostgreSQL Flexible Server.

Microsoft documents that PITR recovery time usually ranges from a few minutes to a few hours, and it depends mainly on:

how much transaction log data must be replayed after the last snapshot backup
the number of databases being recovered
the restore target and region

For custom/latest restore points, Azure first restores the snapshot backup and then replays WAL/logs to bring the server to the requested time. That log-recovery phase is what usually takes the time.

If your goal is the fastest possible restore, use Restore to full backup (fast restore) or choose a backup timestamp instead of a custom point in time. Fast restore skips log replay and is therefore quicker.

Also, keep in mind that a restore creates a new server; it does not overwrite the existing one. You can monitor the restore operation in the Activity log or under Jobs in the portal.

If the operation is still running after 3 hours, it is worth checking the Activity log to confirm that the restore is progressing. If there is no progress or the restore eventually fails, open a support request with the restore operation details.

I hope this was helpful. Please mark this as an "accepted" answer if you found this helpful and follow me here on Microsoft Q&A. Have a wonderful day!

Best regards,
Andrew Taylor

Reference: Microsoft Learn — Backup and restore in Azure Database for PostgreSQL

0 comments

1 additional answer

Your answer

Pilladi Padma Sai Manisha 9,055 Reputation points Microsoft External Staff Moderator

2026-05-29T06:48:48.5066667+00:00

Hi Rafa Lachica Garrido
Thankyou for reaching microsoft Q&A!
A Point-in-Time Restore (PITR) for Azure PostgreSQL Flexible Server typically completes within a reasonable timeframe, although the duration can vary depending on factors such as database size, storage performance, and regional service load.

To help investigate further, could you please provide the following details?

Source PostgreSQL Flexible Server name and Azure region.

Approximate database size and provisioned storage.

The exact time when the restore operation was initiated (UTC if possible).

Current status shown in the Azure portal for the restore operation.

Any error messages, Activity Log entries, or deployment operation details associated with the restore.

Whether the target restored server was created successfully and is stuck in a provisioning state, or if the restore request itself remains in progress.

A screenshot of the restore operation status from the Azure portal.

You can also check the Activity Log for the restore operation and share any correlation ID or operation ID if available.

With these details, we can better determine whether the restore is still progressing normally or if there is a backend issue requiring further investigation.
Pilladi Padma Sai Manisha 9,055 Reputation points Microsoft External Staff Moderator

2026-06-10T04:09:35.3866667+00:00

Hi
I hope you had a chance to review the information shared earlier, and I hope this information has been helpful! If you still have questions, please let us know what is needed in the comments so the question can be answered.

Answer 1

Rafa Lachica Garrido 40

It almost took 5h based on the image below:

User's image

0 comments

Share via

Postgresql Flexible Server Restore get stuck

Restore duration for PostgreSQL Flexible Server PITR

1 additional answer

Your answer