Seeing a lot of 404s to /admin/host/ping after turning on automatic scale out in App Service

Question

Seeing a lot of 404s to /admin/host/ping after turning on automatic scale out in App Service

Ryan McGee 70

I enabled automatic scale out and enforced a scale out limit on some of my app services. Looking at app insights, I see a very steady GET to "/admin/host/ping". I'm assuming that is part of the automatic scale out infrastructure since it started at the same time, but if I turn automatic scale out back off, it continues.

Everything else seems to be working as expected.

Does anyone know how to update the "/admin/host/ping" to hit a different path or any details on what it's doing? I'd like to continue using automatic scale out but this ping has me a little worried.

Manus Hand 0 Reputation points

2023-12-06T19:53:20.2033333+00:00

We have the same question, but clicking "I have the same question" (many of us have done it) initially shows "1" in the count of people who've clicked it, but then on refresh goes back to 0. Just so that Microsoft (I hope) knows, a lot of people want to know what's up with this one. Please and thank you.

EDIT: Now I come back to the page and it's showing 1, not 0. But it's not me, because when I click it, it ups it to 2. Then I refresh the page and it's back to 1.

EDIT 2: After adding that edit, the page is now back to showing 0. How counterproductive.
SnehaAgrawal-MSFT 22,706 Reputation points Moderator

2023-12-07T06:54:07.89+00:00
@Ryan McGee Thanks for reaching here!

The "/admin/host/ping" endpoint is used by Azure App Service to monitor the health of your application instances. When you enable automatic scale out, App Service creates additional instances of your application to handle increased traffic. The "/admin/host/ping" endpoint is used to check the health of these instances and ensure that they are ready to handle requests.

The reason is due to Azure Application Insights' "Availability" monitoring feature, which regularly sends HTTP requests to your application's endpoint to assess its availability and responsiveness.

To prevent these availability pings from reaching your application, you can deactivate the "Availability" monitoring feature in Azure Application Insights.

To do so, access your application insights within your function app, go to the Investigate section, and click on Availability for further instructions.

Enter the URL of your app endpoint that you want to ping for availability and set the test's "Enabled" option to "Disable".

Modify the availability test to use a different endpoint that does not exist or responds with an HTTP 200 status code.

You may see this doc link might be helpful- https://learn.microsoft.com/en-us/azure/app-service/monitor-instances-health-check?tabs=dotnet

Please let us know if you have further query.
SnehaAgrawal-MSFT 22,706 Reputation points Moderator

2023-12-11T05:47:28.45+00:00

@Ryan McGee Just checking if you have chance to see recent response.

Please let us know if further query or issue remains.

Please accept as "Yes" if the answer provided is useful , so that you can help others in the community looking for remediation for similar issues.
Ryan McGee 70 Reputation points

2023-12-27T18:36:57.2766667+00:00
Sorry, things got a little busy with the holiday.

I was able to try the suggestion, but unfortunately, it did not resolve the issue. I have a classic test created to hit the root, and that is working as expected. I even went as far as to run this PowerShell to verify that I hadn't accidentally setup a test on another app insights to hit that URL, but I don't

search-azgraph -Query 'resources | where type =~ "Microsoft.Insights/webtests" | project id, name, config=properties.Configuration' | Select-Object -ExpandProperty Data | %{ [pscustomobject]@{ Url = ([xml]$_.config.WebTest).WebTest.Items.Request.Url Name = $_.name Id = $_.id } } | sort Url | ft name, url, id -AutoSize
leesejoh@roomstogo.com 10 Reputation points

2024-11-13T20:21:55.5366667+00:00

We are still seeing this in our production environment. Has this been fixed? Is Automatic scaling still in preview? Thanks

Accepted answer

0 additional answers

Your answer

Manus Hand 0 Reputation points

2023-12-06T19:53:20.2033333+00:00

We have the same question, but clicking "I have the same question" (many of us have done it) initially shows "1" in the count of people who've clicked it, but then on refresh goes back to 0. Just so that Microsoft (I hope) knows, a lot of people want to know what's up with this one. Please and thank you.

EDIT: Now I come back to the page and it's showing 1, not 0. But it's not me, because when I click it, it ups it to 2. Then I refresh the page and it's back to 1.

EDIT 2: After adding that edit, the page is now back to showing 0. How counterproductive.
SnehaAgrawal-MSFT 22,706 Reputation points Moderator

2023-12-07T06:54:07.89+00:00

@Ryan McGee Thanks for reaching here!

The "/admin/host/ping" endpoint is used by Azure App Service to monitor the health of your application instances. When you enable automatic scale out, App Service creates additional instances of your application to handle increased traffic. The "/admin/host/ping" endpoint is used to check the health of these instances and ensure that they are ready to handle requests.

The reason is due to Azure Application Insights' "Availability" monitoring feature, which regularly sends HTTP requests to your application's endpoint to assess its availability and responsiveness.

To prevent these availability pings from reaching your application, you can deactivate the "Availability" monitoring feature in Azure Application Insights.

To do so, access your application insights within your function app, go to the Investigate section, and click on Availability for further instructions.

Enter the URL of your app endpoint that you want to ping for availability and set the test's "Enabled" option to "Disable".

Modify the availability test to use a different endpoint that does not exist or responds with an HTTP 200 status code.

You may see this doc link might be helpful- https://learn.microsoft.com/en-us/azure/app-service/monitor-instances-health-check?tabs=dotnet

Please let us know if you have further query.
SnehaAgrawal-MSFT 22,706 Reputation points Moderator

2023-12-11T05:47:28.45+00:00

@Ryan McGee Just checking if you have chance to see recent response.

Please let us know if further query or issue remains.

Please accept as "Yes" if the answer provided is useful , so that you can help others in the community looking for remediation for similar issues.
Ryan McGee 70 Reputation points

2023-12-27T18:36:57.2766667+00:00

Sorry, things got a little busy with the holiday.

I was able to try the suggestion, but unfortunately, it did not resolve the issue. I have a classic test created to hit the root, and that is working as expected. I even went as far as to run this PowerShell to verify that I hadn't accidentally setup a test on another app insights to hit that URL, but I don't

search-azgraph -Query 'resources | where type =~ "Microsoft.Insights/webtests" | project id, name, config=properties.Configuration' | Select-Object -ExpandProperty Data | %{ [pscustomobject]@{ Url = ([xml]$_.config.WebTest).WebTest.Items.Request.Url Name = $_.name Id = $_.id } } | sort Url | ft name, url, id -AutoSize
leesejoh@roomstogo.com 10 Reputation points

2024-11-13T20:21:55.5366667+00:00

We are still seeing this in our production environment. Has this been fixed? Is Automatic scaling still in preview? Thanks

Answer 1

SnehaAgrawal-MSFT 22,706 Moderator

@Ryan McGee Thank you for your reply! I did some more research and found out that -

The 404s are caused by enabling Automatic Scaling and pinging the same endpoint for both Function Apps and Web Apps.

The issue of 404s to the /admin/host/ping endpoint is expected behavior as this feature is still in Preview. It will not affect scale-out. Only the status code 429 will impact scale-out. Otherwise, scale-out is determined by latency.

This issue will be fixed in a future release, which will reduce the noise.

In the meantime, the 404s should not have any negative effect on scaling, except for the noise. These pings are also not a security risk for the app, as they come from the same scale unit that hosts the app.

I hope this helps.

Please let us know if further query on this.

Ryan McGee 70 Reputation points

2023-12-28T19:58:27.73+00:00

Thank you for looking into that. It's not the answer I was hoping for but now I know. The noise it makes is unfortunate and throws off some of our monitoring.

Is there a public place, like Github, where bugs like this are tracked or is that a Microsoft internal tool?
SnehaAgrawal-MSFT 22,706 Reputation points Moderator

2024-01-02T06:05:23.2166667+00:00

@Ryan McGee Sure, you can post here: https://github.com/Azure/static-web-apps/issues

- All of the feedback, you share in these forums will be monitored and reviewed by the Microsoft engineering teams responsible for building Azure.
Brian Sweeney 0 Reputation points

2024-02-07T12:42:35.75+00:00

Is the purpose of that request to determine if an instance is usable? And if so, does the 404 response mean the Automatic Scaling feature does not work with WebApps?
Adam Schildmeyer 0 Reputation points

2024-02-15T17:15:14.6133333+00:00

I'd imagine it's pinging something that is supposed to be a success to gauge network latency to gauge if the application instance is having an issue that deems autoscaling should take place. Like it's common for high CPU usage to translate to network waits that translates to network latency.

However, if it's getting a 404 how does it adequately gauge latency to autoscaling? I think it's another half-baked feature, hence preview and it shouldn't be used for the time being. Just got back to the original scale type plan until it gets fully baked. That is what I am going to do at least.
Chris Spanellis 26 Reputation points

2024-02-16T17:39:51.11+00:00

These requests cause my CMS to make database calls to see if the route has a rewrite. I can programmatically intercept these and route them.

@SnehaAgrawal-MSFT My question is... will it disrupt the auto scale these requests get a response of 200, or should it be kept as a 404?
Brandon Parker 0 Reputation points

2024-02-25T02:54:11.1133333+00:00

Is there somewhere this issue is being tracked? We recently scaled to a second instance using the preview auto scaling and are experiencing the same issue, however only one of our three webapps
Chris Spanellis 26 Reputation points

2024-02-25T04:20:10.09+00:00
FYI I rerouted these requests with a Rewriter I was already using at StartUp. This is paraphrasing, but hopefully helps.

app.UseRewriter(new RewriteOptions().AddRedirect("/admin/host/ping", "doesntmatter", 200);

Share via

Seeing a lot of 404s to /admin/host/ping after turning on automatic scale out in App Service

0 additional answers

Your answer