Scale out Blazor Server App using Azure cache Redis as backplane

Kevin Seng 0

I am current hosting Blazor server app in Azure App Service. I found an issue when my application scale out. So i change to use Azure cache Redis for SignalR backplane. I also enable session affinity with my App Service configuration. Most of the time I see it work. but sometimes I get the error "No Connection with that ID". My question if I already use Redis as SignalR backplane. Why the connection get lost?

https://domain.com/_blazor?id=LyHnNyvFmFyWwaeNQjx5cQ&_=1726649919956

No Connection with that ID: Status code '404'

Sai Raghunadh M 1,605 Reputation points Microsoft Vendor

2024-09-19T14:33:53.41+00:00

Hi @Kevin Seng,

Thanks for the question and using MS Q&A platform

In addition to @ Bruce (SqlWork.com).

Please refer to this document that might help you: Host and deploy server-side Blazor apps.

For Additional information on Redis Cluster.

I hope this information helps, please do let us know if you have any Queries.
Sai Raghunadh M 1,605 Reputation points Microsoft Vendor

2024-09-20T14:37:32.6633333+00:00

Hi @Kevin Seng,

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.
Sai Raghunadh M 1,605 Reputation points Microsoft Vendor

2024-09-23T07:51:51.0933333+00:00

Hi @Kevin Seng,

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.
Kevin Seng 0 Reputation points

2024-09-24T03:38:17.5566667+00:00

I responsed below.
Sai Raghunadh M 1,605 Reputation points Microsoft Vendor

2024-09-25T15:24:37.5733333+00:00

Hi @Kevin Seng,

Just checking in to see if the below answer provided by Bruce (SqlWork.com) helped.

If this answers your query, do click Accept Answer and Yes for was this answer helpful. And, if you have any further query do let us know.

2 answers

Bruce (SqlWork.com) 68,486 Reputation points

2024-09-18T15:52:12.67+00:00

If the Blazor server app is hosted in Azure, Azure Signal/r is recommended over using Redis cache. Anyway blazor server requires a constant websocket connection. If lost the server has a couple minute timeout (~4 minutes, configurable) on the circuit, after which the circuit context is closed, and the client must reload.

Browsers are getting more aggressive in shutting down websockets (especially mobile, but desktop is following the trend). If the tab is not the active tab, the websocket may be closed. The same may happen if the browser is not the active application.

you can counteract the browser behavior, by setting a higher circuit timeout. This will of course use more server resources, as a tab close or navigation away would leave the circuit open longer on the server.
Please sign in to rate this answer.

1 person found this answer helpful.
Kevin Seng 0 Reputation points

2024-09-24T03:37:41.05+00:00

Thanks for you answer, Bruce.

I plan to use SignalR, but I encountered a challenge. When I switched to SignalR, the HttpContext was null. To minimize code changes, I decided to use Redis as the SignalR backplane

IHttpContextAccessor httpContextAccessor = d.GetService<IHttpContextAccessor>() ?? throw new Exception("Need to add AddHttpContextAccessor"); var httpContext = httpContextAccessor.HttpContext; // null

"Hub": { "ClientTimeoutInterval": 15, "HandshakeTimeout": 15 },

services.AddServerSideBlazor() .AddHubOptions(options => { options.ClientTimeoutInterval = TimeSpan.FromSeconds(int.Parse(configuration["Hub:ClientTimeoutInterval"]!)); options.HandshakeTimeout = TimeSpan.FromSeconds(int.Parse(configuration["Hub:HandshakeTimeout"]!)); }) .AddCircuitOptions(option => { option.DetailedErrors = configuration.GetValue<bool>("DetailedErrors"); }) .AddMicrosoftIdentityConsentHandler();

I noticed that when I only have one instance in my App Service, I don't encounter this error frequently. However, if I increase the number of instances, I'm more likely to experience the issue and need to refresh once or twice to resolve it.

Previously, I observed that my requests were switching instances even with sticky sessions enabled.

Today, I increased the number of instances to two and started testing again. It seems that the App Service Load Balancer's sticky sessions are now working as intended, and my requests are no longer switching instances.

I'll continue to monitor the situation and keep you updated.
Sign in to comment

Use comments to ask for clarification, additional information, or improvements to the question.
kevin 0 Reputation points

2024-09-26T05:51:10.81+00:00

I checked again, and the error disappeared when I added more instances. I noticed that the ARRAffinity value changes sometimes, but before, it wasn't changing automatically.
Please sign in to rate this answer.
Sai Raghunadh M 1,605 Reputation points Microsoft Vendor

2024-09-27T10:03:42.38+00:00

Hi @Kevin Seng,

Just checking in to see if the above answer provided by Bruce (SqlWork.com) helped.

If this answers your query, do click Accept Answer and Yes for was this answer helpful. And, if you have any further query do let us know.
Sign in to comment

Use comments to ask for clarification, additional information, or improvements to the question.

Share via

Scale out Blazor Server App using Azure cache Redis as backplane

2 answers

Your answer