Azure Function Unexpected Stop – Log Storage and Size Limit Clarification (Basic B3 Plan)

Question

Azure Function Unexpected Stop – Log Storage and Size Limit Clarification (Basic B3 Plan)

Dcm 20

Current Azure Function Details

Plan: Basic B3
Trigger: Batch execution every 5 minutes
Log Storage Location: /home/site/<custom_folder>
Operating System : Linux

Issue Summary:

We are storing logs in a custom folder under /home/site/. Previously, we encountered a similar issue in App Service, so this time we implemented log rotation to ensure the function does not stop unexpectedly. The log rotation is configured to generate approximately 100 MB per day.

From our observations, the function continues to work fine up to a total log size of around 500 MB (approximately 4–5 days of logs).

The code includes try-catch blocks at every level, and we do not see any exceptions or errors in the logs. Since the Basic B3 plan offers 7 GB memory and 10 GB remote storage, we reviewed the metrics:

Average memory working set: 1–1.5 GB
File system usage: up to 2 GB

Both values appear to be within acceptable limits.

When the log size is around 500 MB, the application functions normally without any issues.

Questions:

What is the maximum allowable storage capacity for logs in the /home/site directory under the Basic B3 plan?
How can we ensure that Azure Function execution does not stop unexpectedly or midway?

Loknathsatyasaivarma Mahali 2,690 Reputation points Moderator

2025-06-03T05:07:30.76+00:00

Hello Dcm,

Thanks for reaching out, Your Azure Function on the Basic B3 plan is likely stopping unexpectedly not because you're exceeding the 10 GB storage limit for /home/site (Azure App Service Plan Storage Limits) (as your 500MB-2GB usage is well within this), but rather due to how your custom logs are being written to this network-attached storage (Understanding /home as Network-Attached Storage). Frequent or inefficient file I/O operations and issues with log rotation can slow things down and even cause functions to stop.

The most effective solutions are to migrate you're logging to Application Insights (Monitor executions in Azure Functions , How to configure monitoring for Azure Functions), which is designed for robust telemetry, or if custom logs are retained, to significantly optimize your logging strategy.
Dcm 20 Reputation points

2025-06-03T05:43:37.8033333+00:00
Hello @Loknathsatyasaivarma Mahali
Thank you for suggesting Application Insights solution.
As this is long term solution, is there any tentative solution or workaround if we want to proceed with custom logs.

Keeping max 400 MB logs and deleting old logs.

Trigger - Batch execution set to X minutes.

Any other recommendation.

Please suggest suitable option.
Anonymous

2025-06-03T15:22:10.52+00:00

@Dcm ,

Thanks for your patience.

Since you're planning to stick with custom logging for now, a good short-term workaround would be to write logs to the local /tmp directory during execution, and then move them to /home/site/<custom_folder> at the end of each run. This helps reduce the number of file write operations directly on the network-attached storage, which can sometimes impact performance.

Also, keeping the total log size under 400 MB and cleaning up older files regularly — as you mentioned — should definitely help maintain stability.
Loknathsatyasaivarma Mahali 2,690 Reputation points Moderator

2025-06-04T05:38:59.3266667+00:00

Hello Dcm,

Adding to the Khadeer Ali answer to help you better understand the file system structure and storage limits in Azure App Service, please refer to the following GitHub resource: Understanding the Azure App Service File System

Since your Function App is hosted on an App Service Plan, one approach you can consider is creating a separate Timer Triggered Function. This function can periodically check the size of your custom folder, and if it exceeds 500 MB, it can automatically purge or clean up the files.

This setup can help manage disk usage proactively and ensure the stability of your app.
Dcm 20 Reputation points

2025-06-04T09:36:09.65+00:00

Hello @Loknathsatyasaivarma Mahali, @Anonymous

Appreciate your response, and thank you for suggesting the solution.

At present, we are planning to retain a maximum of 400 MB to 500 MB in a custom directory. These files are then moved to Azure Storage by another Azure Function, which is scheduled to run once daily.

However, we recently experienced a sudden stop of the function, and we are trying to identify the root cause. Could this interruption be related to heavy logging activity, or might there be another underlying issue?

For reference, resource usage appears to be within normal limits:

Average memory working set: 1–1.5 GB

File system usage: Up to 2 GB

We would appreciate your guidance on what could potentially trigger such unexpected behavior.
Loknathsatyasaivarma Mahali 2,690 Reputation points Moderator

2025-06-05T00:48:48.1766667+00:00

Hello Dcm,

To investigate the issue further can you please check the private message and provide us with the requested details?

Here is the reference link on How to access & Data Privacy policy of private messages in Microsoft Q&A.
Loknathsatyasaivarma Mahali 2,690 Reputation points Moderator

2025-06-06T09:22:28.21+00:00

Hello Dcm,

I am following up to see if you had an opportunity to review my previous message and private message, and to provide the requested details to further troubleshoot the issue.
Loknathsatyasaivarma Mahali 2,690 Reputation points Moderator

2025-06-09T01:00:49.03+00:00

Hello Dcm,

We wanted to follow up on the issue you encountered. Please let us know if it's resolved or if you need further assistance.
Anonymous

2025-06-10T03:56:17.0366667+00:00
Dcm,

Could this interruption be related to heavy logging activity, or might there be another underlying issue?

The issue could be triggered by operational conditions or function runtime factors.

Make sure functionTimeout in host.json is configured properly(if required).

Navigate to Azure Portal => _Function App_ => Diagnose and Solve Problems => check for restarts, crash logs, or auto-heal events.

Check if there are any issues in execution time, failed invocations, or silent exceptions.

Add instrumentation to track batch job duration and detect if there are any unexpected spikes.

Temporarily disable file logging to check if logging I/O is the root cause. If you observe any such, avoid synchronous disk writes. Use asynchronous file logging to reduce I/O latency.

In Function App=>Auto-Heal, set rules to restart the app or collect memory dumps if execution duration exceeds a threshold or if specific errors occur.
Anonymous

2025-06-11T03:45:05.9566667+00:00

Dcm,

We haven’t heard from you on the last response and was just checking back to see if you have a resolution yet. In case if you have any resolution, please do share that same with the community as it can be helpful to others. Otherwise, will respond with more details and we will try to help.
Dcm 20 Reputation points

2025-10-03T02:14:00.2966667+00:00

Hi @Loknathsatyasaivarma Mahali
As you mentioned earlier, Frequent file I/O operations can slow things down and even cause functions to stop.
Currently log file size is approx. 6 to 8 MB max but still it function gets stopped intermittently. Any clue ?
Note we are using Plan: Basic B3

Your answer

Loknathsatyasaivarma Mahali 2,690 Reputation points Moderator

2025-06-03T05:07:30.76+00:00

Hello Dcm,

Thanks for reaching out, Your Azure Function on the Basic B3 plan is likely stopping unexpectedly not because you're exceeding the 10 GB storage limit for /home/site (Azure App Service Plan Storage Limits) (as your 500MB-2GB usage is well within this), but rather due to how your custom logs are being written to this network-attached storage (Understanding /home as Network-Attached Storage). Frequent or inefficient file I/O operations and issues with log rotation can slow things down and even cause functions to stop.

The most effective solutions are to migrate you're logging to Application Insights (Monitor executions in Azure Functions , How to configure monitoring for Azure Functions), which is designed for robust telemetry, or if custom logs are retained, to significantly optimize your logging strategy.
Dcm 20 Reputation points

2025-06-03T05:43:37.8033333+00:00

Hello @Loknathsatyasaivarma Mahali
Thank you for suggesting Application Insights solution.
As this is long term solution, is there any tentative solution or workaround if we want to proceed with custom logs.

Keeping max 400 MB logs and deleting old logs.

Trigger - Batch execution set to X minutes.

Any other recommendation.

Please suggest suitable option.
Anonymous

2025-06-03T15:22:10.52+00:00

@Dcm ,

Thanks for your patience.

Since you're planning to stick with custom logging for now, a good short-term workaround would be to write logs to the local /tmp directory during execution, and then move them to /home/site/<custom_folder> at the end of each run. This helps reduce the number of file write operations directly on the network-attached storage, which can sometimes impact performance.

Also, keeping the total log size under 400 MB and cleaning up older files regularly — as you mentioned — should definitely help maintain stability.
Loknathsatyasaivarma Mahali 2,690 Reputation points Moderator

2025-06-04T05:38:59.3266667+00:00

Hello Dcm,

Adding to the Khadeer Ali answer to help you better understand the file system structure and storage limits in Azure App Service, please refer to the following GitHub resource: Understanding the Azure App Service File System

Since your Function App is hosted on an App Service Plan, one approach you can consider is creating a separate Timer Triggered Function. This function can periodically check the size of your custom folder, and if it exceeds 500 MB, it can automatically purge or clean up the files.

This setup can help manage disk usage proactively and ensure the stability of your app.
Dcm 20 Reputation points

2025-06-04T09:36:09.65+00:00

Hello @Loknathsatyasaivarma Mahali, @Anonymous

Appreciate your response, and thank you for suggesting the solution.

At present, we are planning to retain a maximum of 400 MB to 500 MB in a custom directory. These files are then moved to Azure Storage by another Azure Function, which is scheduled to run once daily.

However, we recently experienced a sudden stop of the function, and we are trying to identify the root cause. Could this interruption be related to heavy logging activity, or might there be another underlying issue?

For reference, resource usage appears to be within normal limits:

Average memory working set: 1–1.5 GB

File system usage: Up to 2 GB

We would appreciate your guidance on what could potentially trigger such unexpected behavior.
Loknathsatyasaivarma Mahali 2,690 Reputation points Moderator

2025-06-05T00:48:48.1766667+00:00

Hello Dcm,

To investigate the issue further can you please check the private message and provide us with the requested details?

Here is the reference link on How to access & Data Privacy policy of private messages in Microsoft Q&A.
Loknathsatyasaivarma Mahali 2,690 Reputation points Moderator

2025-06-06T09:22:28.21+00:00

Hello Dcm,

I am following up to see if you had an opportunity to review my previous message and private message, and to provide the requested details to further troubleshoot the issue.
Loknathsatyasaivarma Mahali 2,690 Reputation points Moderator

2025-06-09T01:00:49.03+00:00

Hello Dcm,

We wanted to follow up on the issue you encountered. Please let us know if it's resolved or if you need further assistance.
Anonymous

2025-06-10T03:56:17.0366667+00:00

Dcm,

Could this interruption be related to heavy logging activity, or might there be another underlying issue?

The issue could be triggered by operational conditions or function runtime factors.

Make sure functionTimeout in host.json is configured properly(if required).

Navigate to Azure Portal => _Function App_ => Diagnose and Solve Problems => check for restarts, crash logs, or auto-heal events.

Check if there are any issues in execution time, failed invocations, or silent exceptions.

Add instrumentation to track batch job duration and detect if there are any unexpected spikes.

Temporarily disable file logging to check if logging I/O is the root cause. If you observe any such, avoid synchronous disk writes. Use asynchronous file logging to reduce I/O latency.

In Function App=>Auto-Heal, set rules to restart the app or collect memory dumps if execution duration exceeds a threshold or if specific errors occur.
Anonymous

2025-06-11T03:45:05.9566667+00:00

Dcm,

We haven’t heard from you on the last response and was just checking back to see if you have a resolution yet. In case if you have any resolution, please do share that same with the community as it can be helpful to others. Otherwise, will respond with more details and we will try to help.
Dcm 20 Reputation points

2025-10-03T02:14:00.2966667+00:00

Hi @Loknathsatyasaivarma Mahali
As you mentioned earlier, Frequent file I/O operations can slow things down and even cause functions to stop.
Currently log file size is approx. 6 to 8 MB max but still it function gets stopped intermittently. Any clue ?
Note we are using Plan: Basic B3