Unable to increase postgres flexible server plan

Question

Unable to increase postgres flexible server plan

Atramentum 0

Hi,

I’ve tried several times via the Azure Portal and to scale server erp-api-dev-postgres3 from B2ms (2 vCores, 4 GiB) to Standard_D4ds_v5 (4 vCores, 16 GiB, 6400 max IOPS). Each attempt sits in Updating for a few minutes and then flips to Failed with no useful error details (just an Activity ID). I’m using the server’s admin account, so this shouldn’t be a permissions issue.

This is a tiny dev instance that I must enlarge to validate a new feature, and it’s the second time I’ve hit the same problem with Flexible Server—making me nervous about production.

How can I pull real diagnostics for a failed scale-up?
Are there hidden quotas or limits that block a B-series D-series move even when the SKU shows as available?
What workaround would you recommend if the in-place upgrade keeps failing?

operation ID: /subscriptions/4be7eda8-3ab7-4818-ae84-05669f833c9f/resourceGroups/erp-api-dev-rgp/providers/Microsoft.Resources/deployments/Microsoft.PostgreSQLFlexibleServer.UpdateComputeAndStoragePostgr/operations/A7BB0C76D3302D6C

Thanks for any help.

Mahesh Kurva 5,025 Reputation points Microsoft External Staff Moderator

2025-06-12T15:51:20.08+00:00

Hi Atramentum,

I kindly request you to please share the details requested in the private message, which will help us to address your query.
Atramentum 0 Reputation points

2025-06-12T16:50:30.37+00:00

Shared @Mahesh Kurva
Atramentum 0 Reputation points

2025-06-13T09:43:54.8233333+00:00

Any update @Mahesh Kurva ?

Thanks
Mahesh Kurva 5,025 Reputation points Microsoft External Staff Moderator

2025-06-13T10:06:37.3933333+00:00

Hi Atramentum,

Thanks for sharing the requested details.

I kindly request some time to investigate this further, as I am currently checking internally regarding the behavior of this resource.

I will get back to you soon with an update.

Thanks.
Atramentum 0 Reputation points

2025-06-13T15:58:53.94+00:00
The error that the system throws after running for hours:

{ "code": "OperationTimedOut", "message": "The operation timed out and automatically rolled back. Please retry the operation." }

Thanks
Atramentum 0 Reputation points

2025-06-16T09:53:12.5066667+00:00

Any updates on this? I’m trying to scale a database in the cloud, which is supposed to be one of the main advantages over my on-premises infrastructure. Do I really have to wait more than four days for a response? Seriously??? @Mahesh Kurva
Atramentum 0 Reputation points

2025-06-16T10:45:46.35+00:00
I wrote this as an answer on Jun 13, 2025, 5:58 PM. Copy and paste again as comment.

The error that the system throws after running for hours:

YAMLCopy

{ "code": "OperationTimedOut", "message": "The operation timed out and automatically rolled back. Please retry the operation." }

Thanks
Mahesh Kurva 5,025 Reputation points Microsoft External Staff Moderator

2025-06-17T08:05:59.0366667+00:00

Hi Atramentum,

We received an internal update confirming that the issue is due to the subscription not being allowlisted. We've already informed the relevant team, and they will proceed with allowlisting it as soon as possible.

Thank you.
Atramentum 0 Reputation points

2025-06-17T13:27:14.95+00:00

What does it mean "The subscription not being allowlisted"? In a cloud environment, do I need Microsoft’s permission to scale a resource? Could you please clarify this for me?

Thanks.
Mahesh Kurva 5,025 Reputation points Microsoft External Staff Moderator

2025-06-17T16:23:02.09+00:00

Hi Atramentum,

What does it mean "The subscription not being allowlisted"? In a cloud environment, do I need Microsoft’s permission to scale a resource? Could you please clarify this for me?

No, but in some regions, restrictions may apply. In such cases, subscriptions are categorized, and allowlisting is required to scale resources beyond certain limits.

Thank you.
Atramentum 0 Reputation points

2025-06-17T17:07:44.7966667+00:00

Are you telling me that I’m going to run into these problems when running Postgres on Azure in West Europe?? Please show me where this is referenced in the official documentation.

The upgrade operation already shows as ‘Accepted’, yet I still can’t scale my database; this is ridiculous in a cloud environment. In the same zone I’ve been able to scale another Postgres instance without any issue.

I’d like more information, because for me this is a reason to change cloud provider. And the error is conflict https://learn.microsoft.com/en-us/azure/azure-resource-manager/troubleshooting/common-deployment-errors:

"You're requesting an operation that isn't allowed in the resource's current state. For example, disk resizing is allowed only when creating a VM or when the VM is deallocated."
Mahesh Kurva 5,025 Reputation points Microsoft External Staff Moderator

2025-06-18T05:47:42.6566667+00:00

Hi Atramentum,

Could you please try scaling from B2ms (2 vCores, 4 GiB) to Standard_D4ds_v5 (4 vCores, 16 GiB, 6400 max IOPS) now? Kindly let us know if the scaling is successful or if you're still experiencing any issues.
Thank you.

Atramentum 0

The upgrade still fails. Same error:

{
  "code": "OperationTimedOut",
  "message": "The operation timed out and automatically rolled back. Please retry the operation."
}

Mahesh Kurva 5,025 Reputation points Microsoft External Staff Moderator

2025-06-19T05:17:04.7633333+00:00

Hi Atramentum,

Could you please try scaling now and let us know if it is successful or if you encounter any issues?

Thank you.
Atramentum 0 Reputation points

2025-06-19T10:38:23.5633333+00:00

I feel like you’re taking me for a ride. This isn’t a trial-and-error situation. Two hours later and the process is still running—it has to be the same error. I’ve been waiting eight days already, even while paying for the basic support tier—does that actually help at all??? I’m not sure if you realize it, but problems like this drive people to clouds like AWS.
Mahesh Kurva 5,025 Reputation points Microsoft External Staff Moderator

2025-06-19T12:16:05.6066667+00:00
Hi Atramentum,

We apologize for the inconvenience caused.

We got an update from internal that your server is currently deployed in Logical Zone 1, which maps to Availability Zone AZ02. Please note that at this time, scale-up operations are not supported on AZ02 and AZ03.

Given this, we would like to offer you two options to proceed:

Enable High Availability (HA): You can enable HA and create a standby server in Logical Zone 3 (AZ01). During a failover, the standby in AZ01 will become active, and you will then be able to proceed with scaling or upgrade operations in that zone.

Point-in-Time Restore (PITR): Alternatively, you can perform a PITR and restore the server to Logical Zone 3 (AZ01). This would allow you to run your operations in a supported zone for scaling.

I hope this information helps. Please do let us know if you have any further queries.
Atramentum 0 Reputation points

2025-06-19T19:34:36.9133333+00:00

Have you even looked at the case? It’s a Flexible Server, not a regular Azure Database for PostgreSQL—and for this tier, HA isn’t permitted (zone-redundant high availability isn’t supported with the chosen compute tier).

And how am I supposed to know that? If I need to scale my production database and can’t, is it going to be the same story? It's crazy.

Am I really getting official support here?
Atramentum 0 Reputation points

2025-06-24T09:26:38.8+00:00

@Mahesh Kurva 12 days without a solution and rising.
Oury Ba-MSFT 20,911 Reputation points Microsoft Employee Moderator

2025-06-24T18:33:33.99+00:00

Atramentum Thank you for reaching out.

Could you please let us know your availability for call so I can take a look at your environment to further assist you.

Regards,

Oury
Atramentum 0 Reputation points

2025-06-24T21:47:40.3433333+00:00

If I use the cloud, it’s because I don’t want my teams managing infrastructure at the stage the company is in. We also don’t have a fixed resource consumption, so the cloud is the ideal solution. It’s for that latter reason that, as CTO, I’ve decided the higher cost is worth it.

But when, for a simple resource upgrade, I receive no support (the basic paid plan) for 13 days, I decide that’s enough. The migration to AWS is already settled. I hadn’t encountered these problems before in other cloud, and this is the second time it happens with a Postgres database.

We were about to start a new project here that could already be generating around $32-40k annually on Azure (right now, about 15k). If they treat their customers like this, I can’t imagine the rest.

I use the cloud so I’m not stuck monitoring a chat for two weeks. I don’t want small talk. I’ve provided all the data. The steps are clear. I’ve sent the error as well. What do they want to discuss? I’ve been using the cloud for years and I know this isn’t normal usage.

As I said, it’s too late to enter this conversation. As soon as I’m able, I’ll begin the migration to AWS.
Atramentum 0 Reputation points

2025-06-25T00:21:46.3666667+00:00

If I use the cloud, it’s because I don’t want my teams managing infrastructure at the stage the company is in. We also don’t have a fixed resource consumption, so the cloud is the ideal solution. It’s for that latter reason that, as CTO, I’ve decided the higher cost is worth it.

But when, for a simple resource upgrade, I receive no support (the basic paid plan) for 13 days, I decide that’s enough. The migration to AWS is already settled. I hadn’t encountered these problems before in other cloud, and this is the second time it happens with a Postgres database.

We were about to start a new project here that could already be generating around $32-40k annually on Azure (right now, about 15k). If they treat their customers like this, I can’t imagine the rest.

I use the cloud so I’m not stuck monitoring a chat for two weeks. I don’t want small talk. I’ve provided all the data. The steps are clear. I’ve sent the error as well. What do they want to discuss? I’ve been using the cloud for years and I know this isn’t normal usage.

As I said, it’s too late to enter this conversation. As soon as I’m able, I’ll begin the migration to AWS.
Atramentum 0 Reputation points

2025-06-27T13:40:48.89+00:00

Hi @Oury Ba-MSFT ,

I received an email with a reply from you that I believe has since been deleted. It said:

Atramentum Thank you for your feedback. I am sorry to hear about this. I wanted to clarify that I have tried the same by creating an Azure PostgreSQL Flexible server burstable tier with Zone 1. Later, I upgraded to General purpose, and it worked for me…

I honestly don’t quite understand a response like that. Does it help in any way to know that it worked for you, when in my case the upgrade failed more than 10 times over a period of 12 days, following support’s instructions, and providing detailed logs and error traces each time?

If the official stance after all that is essentially “well, it worked for me,” that really says a lot about the level of support Azure is offering.

Best regards,
Hugo

1 answer

Your answer

Mahesh Kurva 5,025 Reputation points Microsoft External Staff Moderator

2025-06-12T15:51:20.08+00:00

Hi Atramentum,

I kindly request you to please share the details requested in the private message, which will help us to address your query.
Atramentum 0 Reputation points

2025-06-12T16:50:30.37+00:00

Shared @Mahesh Kurva
Atramentum 0 Reputation points

2025-06-13T09:43:54.8233333+00:00

Any update @Mahesh Kurva ?

Thanks
Mahesh Kurva 5,025 Reputation points Microsoft External Staff Moderator

2025-06-13T10:06:37.3933333+00:00

Hi Atramentum,

Thanks for sharing the requested details.

I kindly request some time to investigate this further, as I am currently checking internally regarding the behavior of this resource.

I will get back to you soon with an update.

Thanks.
Atramentum 0 Reputation points

2025-06-13T15:58:53.94+00:00

The error that the system throws after running for hours:

{ "code": "OperationTimedOut", "message": "The operation timed out and automatically rolled back. Please retry the operation." }

Thanks
Atramentum 0 Reputation points

2025-06-16T09:53:12.5066667+00:00

Any updates on this? I’m trying to scale a database in the cloud, which is supposed to be one of the main advantages over my on-premises infrastructure. Do I really have to wait more than four days for a response? Seriously??? @Mahesh Kurva
Atramentum 0 Reputation points

2025-06-16T10:45:46.35+00:00

I wrote this as an answer on Jun 13, 2025, 5:58 PM. Copy and paste again as comment.

The error that the system throws after running for hours:

YAMLCopy

{ "code": "OperationTimedOut", "message": "The operation timed out and automatically rolled back. Please retry the operation." }

Thanks
Mahesh Kurva 5,025 Reputation points Microsoft External Staff Moderator

2025-06-17T08:05:59.0366667+00:00

Hi Atramentum,

We received an internal update confirming that the issue is due to the subscription not being allowlisted. We've already informed the relevant team, and they will proceed with allowlisting it as soon as possible.

Thank you.
Atramentum 0 Reputation points

2025-06-17T13:27:14.95+00:00

What does it mean "The subscription not being allowlisted"? In a cloud environment, do I need Microsoft’s permission to scale a resource? Could you please clarify this for me?

Thanks.
Mahesh Kurva 5,025 Reputation points Microsoft External Staff Moderator

2025-06-17T16:23:02.09+00:00

Hi Atramentum,

What does it mean "The subscription not being allowlisted"? In a cloud environment, do I need Microsoft’s permission to scale a resource? Could you please clarify this for me?

No, but in some regions, restrictions may apply. In such cases, subscriptions are categorized, and allowlisting is required to scale resources beyond certain limits.

Thank you.
Atramentum 0 Reputation points

2025-06-17T17:07:44.7966667+00:00

Are you telling me that I’m going to run into these problems when running Postgres on Azure in West Europe?? Please show me where this is referenced in the official documentation.

The upgrade operation already shows as ‘Accepted’, yet I still can’t scale my database; this is ridiculous in a cloud environment. In the same zone I’ve been able to scale another Postgres instance without any issue.

I’d like more information, because for me this is a reason to change cloud provider. And the error is conflict https://learn.microsoft.com/en-us/azure/azure-resource-manager/troubleshooting/common-deployment-errors:

"You're requesting an operation that isn't allowed in the resource's current state. For example, disk resizing is allowed only when creating a VM or when the VM is deallocated."
Mahesh Kurva 5,025 Reputation points Microsoft External Staff Moderator

2025-06-18T05:47:42.6566667+00:00

Hi Atramentum,

Could you please try scaling from B2ms (2 vCores, 4 GiB) to Standard_D4ds_v5 (4 vCores, 16 GiB, 6400 max IOPS) now? Kindly let us know if the scaling is successful or if you're still experiencing any issues.
Thank you.
Atramentum 0 Reputation points

2025-06-18T09:38:14.0033333+00:00

The upgrade still fails. Same error:

{ "code": "OperationTimedOut", "message": "The operation timed out and automatically rolled back. Please retry the operation." }
Mahesh Kurva 5,025 Reputation points Microsoft External Staff Moderator

2025-06-19T05:17:04.7633333+00:00

Hi Atramentum,

Could you please try scaling now and let us know if it is successful or if you encounter any issues?

Thank you.
Atramentum 0 Reputation points

2025-06-19T10:38:23.5633333+00:00

I feel like you’re taking me for a ride. This isn’t a trial-and-error situation. Two hours later and the process is still running—it has to be the same error. I’ve been waiting eight days already, even while paying for the basic support tier—does that actually help at all??? I’m not sure if you realize it, but problems like this drive people to clouds like AWS.
Mahesh Kurva 5,025 Reputation points Microsoft External Staff Moderator

2025-06-19T12:16:05.6066667+00:00

Hi Atramentum,

We apologize for the inconvenience caused.

We got an update from internal that your server is currently deployed in Logical Zone 1, which maps to Availability Zone AZ02. Please note that at this time, scale-up operations are not supported on AZ02 and AZ03.

Given this, we would like to offer you two options to proceed:

Enable High Availability (HA): You can enable HA and create a standby server in Logical Zone 3 (AZ01). During a failover, the standby in AZ01 will become active, and you will then be able to proceed with scaling or upgrade operations in that zone.

Point-in-Time Restore (PITR): Alternatively, you can perform a PITR and restore the server to Logical Zone 3 (AZ01). This would allow you to run your operations in a supported zone for scaling.

I hope this information helps. Please do let us know if you have any further queries.
Atramentum 0 Reputation points

2025-06-19T19:34:36.9133333+00:00

Have you even looked at the case? It’s a Flexible Server, not a regular Azure Database for PostgreSQL—and for this tier, HA isn’t permitted (zone-redundant high availability isn’t supported with the chosen compute tier).

And how am I supposed to know that? If I need to scale my production database and can’t, is it going to be the same story? It's crazy.

Am I really getting official support here?
Atramentum 0 Reputation points

2025-06-24T09:26:38.8+00:00

@Mahesh Kurva 12 days without a solution and rising.
Oury Ba-MSFT 20,911 Reputation points Microsoft Employee Moderator

2025-06-24T18:33:33.99+00:00

Atramentum Thank you for reaching out.

Could you please let us know your availability for call so I can take a look at your environment to further assist you.

Regards,

Oury
Atramentum 0 Reputation points

2025-06-24T21:47:40.3433333+00:00

If I use the cloud, it’s because I don’t want my teams managing infrastructure at the stage the company is in. We also don’t have a fixed resource consumption, so the cloud is the ideal solution. It’s for that latter reason that, as CTO, I’ve decided the higher cost is worth it.

But when, for a simple resource upgrade, I receive no support (the basic paid plan) for 13 days, I decide that’s enough. The migration to AWS is already settled. I hadn’t encountered these problems before in other cloud, and this is the second time it happens with a Postgres database.

We were about to start a new project here that could already be generating around $32-40k annually on Azure (right now, about 15k). If they treat their customers like this, I can’t imagine the rest.

I use the cloud so I’m not stuck monitoring a chat for two weeks. I don’t want small talk. I’ve provided all the data. The steps are clear. I’ve sent the error as well. What do they want to discuss? I’ve been using the cloud for years and I know this isn’t normal usage.

As I said, it’s too late to enter this conversation. As soon as I’m able, I’ll begin the migration to AWS.
Atramentum 0 Reputation points

2025-06-25T00:21:46.3666667+00:00

If I use the cloud, it’s because I don’t want my teams managing infrastructure at the stage the company is in. We also don’t have a fixed resource consumption, so the cloud is the ideal solution. It’s for that latter reason that, as CTO, I’ve decided the higher cost is worth it.

But when, for a simple resource upgrade, I receive no support (the basic paid plan) for 13 days, I decide that’s enough. The migration to AWS is already settled. I hadn’t encountered these problems before in other cloud, and this is the second time it happens with a Postgres database.

We were about to start a new project here that could already be generating around $32-40k annually on Azure (right now, about 15k). If they treat their customers like this, I can’t imagine the rest.

I use the cloud so I’m not stuck monitoring a chat for two weeks. I don’t want small talk. I’ve provided all the data. The steps are clear. I’ve sent the error as well. What do they want to discuss? I’ve been using the cloud for years and I know this isn’t normal usage.

As I said, it’s too late to enter this conversation. As soon as I’m able, I’ll begin the migration to AWS.
Atramentum 0 Reputation points

2025-06-27T13:40:48.89+00:00

Hi @Oury Ba-MSFT ,

I received an email with a reply from you that I believe has since been deleted. It said:

Atramentum Thank you for your feedback. I am sorry to hear about this. I wanted to clarify that I have tried the same by creating an Azure PostgreSQL Flexible server burstable tier with Zone 1. Later, I upgraded to General purpose, and it worked for me…

I honestly don’t quite understand a response like that. Does it help in any way to know that it worked for you, when in my case the upgrade failed more than 10 times over a period of 12 days, following support’s instructions, and providing detailed logs and error traces each time?

If the official stance after all that is essentially “well, it worked for me,” that really says a lot about the level of support Azure is offering.

Best regards,
Hugo

Answer 1

You can close de issue. Leaving Azure.

If I use the cloud, it’s because I don’t want my teams managing infrastructure at the stage the company is in. We also don’t have a fixed resource consumption, so the cloud is the ideal solution. It’s for that latter reason that, as CTO, I’ve decided the higher cost is worth it.

But when, for a simple resource upgrade, I receive no support (the basic paid plan) for 13 days, I decide that’s enough. The migration to AWS is already settled. I hadn’t encountered these problems before in other cloud, and this is the second time it happens with a Postgres database.

We were about to start a new project here that could already be generating around $32-40k annually on Azure (right now, about 15k). If they treat their customers like this, I can’t imagine the rest.

I use the cloud so I’m not stuck monitoring a chat for two weeks. I don’t want small talk. I’ve provided all the data. The steps are clear. I’ve sent the error as well. What do they want to discuss? I’ve been using the cloud for years and I know this isn’t normal usage.

As I said, it’s too late to enter this conversation. As soon as I’m able, I’ll begin the migration to AWS.

Oury Ba-MSFT 20,911 Reputation points Microsoft Employee Moderator

2025-06-25T18:15:17.2466667+00:00

Atramentum Thank you for your feedback.

I am sorry to hear about this.

I wanted to clarify that I have tried the same by creating an Azure PostgreSQL Flexible server burstable tier with Zone 1. Later, I upgraded to General purpose, and it worked for me without any issue. That was the reason why I wanted to get on a call with you to see if I can check your environment and help you fix it. It seems there might be an environmental factor on your end that's preventing the upgrade.

It is sad to hear that you are going for another provider. I believe we can resolve this issue successfully by working together. Your collaboration is key, and I'm here to support you every step of the way.

Please let me know.

Regards,

Oury

Share via

Unable to increase postgres flexible server plan

1 answer

Your answer