Forem: Victorio Berra

Build an Aspire API Using Microsoft OpenAI, Scalar, OpenRouter, Structured Output, and Custom Headers

Victorio Berra — Fri, 03 Oct 2025 18:34:51 +0000

[
  {
    "date": "2025-10-04",
    "temperatureC": 8,
    "summary": "2025-10-04: Crisp 8°C with intermittent interdimensional drizzle — expect marmalade rain and the occasional polite wormhole. 40% chance of phosphorescent fog veins, 20% chance of localized anti-gravity gusts (hold onto your hat; it may orbit you), and a small probability of spontaneous teleporting puddles that prefer not to be stepped in. Bring an umbrella and a treaty offer — umbrellas may demand a cup of tea before cooperating.",
    "temperatureF": 46
  }
]

Bootstraping

Create a new Aspire solution:

Install Microsoft.Extensions.AI.OpenAI pre-release in your ApiService project. Install non-prerelease Microsoft.Extensions.AI as well. Things like structured output extensions are in Microsoft.Extensions.AI.

Install Scalar.Aspire in your AppHost project.

Scalar Aspire

Microsoft is moving away from including Swashbuckle UI stuff in their templates and docs. They now have their own OpenApi (not to be confused with OpenAi!) Swagger document generation. The code will have this already set up with a link to learn more:

// Learn more about configuring OpenAPI at https://aka.ms/aspnet/openapi
builder.Services.AddOpenApi();

// ...

app.MapOpenApi();

So now we want a Swagger UI, there are options, we could use the battle tested Swashbuckle, the popular ReDoc, or the FOSS Scalar https://github.com/scalar/scalar however they do have a cloud offering. If you pay you get the following:

Write your API documentation and publish your API references (free)
Get SSL and a super cool *.apidocumentation.com subdomain (free)
Write free text documentation (paid)
Collaborate with your whole team (paid)
Use any domain (paid)
SDK generation (paid)

To set Scalar up, do the following in your Aspire AppHost:

// Add Scalar API Reference
var scalar = builder.AddScalarApiReference(options =>
{
    options.WithTheme(ScalarTheme.Laserwave);
});

// Register services with the API Reference
scalar
    .WithApiReference(apiService);

One really cool thing to note, under the hood this uses their proxy which gets around CORS issues!

Scalar for Aspire includes a built-in proxy that is enabled by default to provide seamless integration with your services.

Proxy Deep Dive

Aspire code is here https://github.com/scalar/scalar/tree/main/integrations/aspire

They have a custom endpoint they register at /scalar-proxy which uses YARP https://learn.microsoft.com/en-us/aspnet/core/fundamentals/servers/yarp/getting-started?view=aspnetcore-9.0

They also have a hosted service at: https://proxy.scalar.com/. Unrelated but other services they seem to have for development: https://void.scalar.com/, https://galaxy.scalar.com/

Back at The Ranch

There are a variety of other Aspire and configuration options you can check out: https://guides.scalar.com/scalar/scalar-api-references/integrations/net-aspire

At this point, you should be able to run your app and see that the Scalar service is up and running:

Clicking the link gets you to the docs:

I am not the worlds biggest fan of the theme but I set WithTheme to show off the capability.

AI

This tutorial uses OpenRouter. If you are not familiar with it, its essentially a gateway to thousands of other models through a single interface. You load your account with credits and they bill you depending on the provider you want. For example throw $10 in your account and you can call Grok, Gemini, Claude, and you don't need accounts with them or to manage multiple keys. LinqPad recently added built-in support for OpenRouter as well.

Head over to OpenRouter and generate a key https://openrouter.ai/settings/keys

Once you have that, right click your ApiService, click "User Secrets" and paste this:

{
  "OpenAIKey": "sk-or-v1-supa-secret"
}

Add the OpenAI IChatClient. Also, for an overview of the current Microsoft .NET AI landscape this blog is awesome https://roxeem.com/2025/09/08/how-to-correctly-build-ai-features-in-dotnet/

And this is awesome too: https://learn.microsoft.com/en-us/dotnet/ai/quickstarts/build-chat-app?pivots=openai

I highlighted some of the more interesting areas of the docs. If you spent a few hours reading all of that you would be pretty ramped up on AI concepts across the board. Inch-deep, mile-wide as they say. Knowing just enough to be dangerous.

Add the chat service:

var config = new ConfigurationBuilder().AddUserSecrets<Program>().Build();
string key = config["OpenAIKey"];

builder.Services.AddTransient<CustomHeadersHandler>(); //this is a standard 'DelegatingHandler' implementation that adds HTTP headers
builder.Services.AddHttpClient("OpenAiHttpClient")
.AddHttpMessageHandler<CustomHeadersHandler>();

builder.Services.AddSingleton(services =>
{
    var httpClientFactory = services.GetRequiredService<IHttpClientFactory>();
    var httpClient = httpClientFactory.CreateClient("OpenAiHttpClient");

    // Create the IChatClient
    IChatClient chatClient =
        new OpenAIClient(
            new ApiKeyCredential(key),
            new OpenAIClientOptions
            {
                Endpoint = new Uri("https://openrouter.ai/api/v1"),
                Transport = new HttpClientPipelineTransport(httpClient)
            })
        .GetChatClient("openrouter/auto")
        .AsIChatClient();

    return chatClient;
});

CustomHeadersHandler is entirely optional:

// Thanks https://stackoverflow.com/questions/79645038/define-custom-headers-using-net-6-openai-sdk
internal class CustomHeadersHandler : DelegatingHandler
{
    protected override async Task<HttpResponseMessage> SendAsync(HttpRequestMessage request, CancellationToken cancellationToken)
    {
        // Optional. Site URL for rankings on openrouter.ai
        //request.Headers.Add("HTTP-Referer", "MyCustomHeaderValue");

        // Optional. Site Title for rankings on openrouter.ai
        //request.Headers.Add("X-Title", "AnotherHeaderValue");

        // Call the inner handler
        return await base.SendAsync(request, cancellationToken);
    }
}

I wanted to show how we could do what is easily done in NodeJS/TS

import OpenAI from 'openai';
const openai = new OpenAI({
  baseURL: 'https://openrouter.ai/api/v1',
  apiKey: '<OPENROUTER_API_KEY>',
  defaultHeaders: {
    'HTTP-Referer': '<YOUR_SITE_URL>', // Optional. Site URL for rankings on openrouter.ai.
    'X-Title': '<YOUR_SITE_NAME>', // Optional. Site title for rankings on openrouter.ai.
  },
});

Almost to the payoff! Lets use our new Chat Client!

Replace the current WeatherForecast endpoint with the following:

app.MapGet("/weatherforecast", async ([FromServices] IChatClient chatClient) =>
{
    var forecast = new List<WeatherForecast>();

    for (int index = 1; index <= 1; index++)
    {
        var targetDate = DateOnly.FromDateTime(DateTime.Now.AddDays(index));

        // Generate complete weather data using AI with structured output
        var prompt = $"""
            Generate realistic weather data for {targetDate:yyyy-MM-dd}.
            Create temperature in Celsius (between -30 and 50) and weather conditions.
            Use humor to create absurd weather conditions that may only exist on other planets or universes.
            Be creative with the conditions - think interdimensional weather, alien phenomena, or impossible physics.
            """;

        var messages = new List<ChatMessage>
        {
            new ChatMessage(ChatRole.User, prompt)
        };

        var response = await chatClient.GetResponseAsync<WeatherData>(prompt);
        var weatherData = response.Result;

        forecast.Add(new WeatherForecast(
            targetDate,
            weatherData.TemperatureC,
            weatherData.Conditions));
    }

    return forecast.ToArray();
})
.WithName("GetWeatherForecast");

Some cool tidbits about the above code. GetResponseAsync<WeatherData> is special as the response is a C# object! This is thanks to the rather transparent handling of structured output from the LLM.

Finally, open your Scalar UI, scroll down to the /weatherforecast endpoint, click "Test Request", click "Send".

Pocketbase API Rules

Victorio Berra — Mon, 21 Jul 2025 13:46:46 +0000

While learning PocketBase, many times there are patterns and helpful things to know that are either not in the docs, are not obvious.

Namely, API rules are hard. People have built websites to help you build API rules for example.

One thing I recently learned, is that an API rule's are translated to SQL and can create joins in unexpected (but needed ways). For example:

Client call: pb.collection('login_bonus_earned').create({ user: 'abc124', loginBonus: 'hij876' });

API rule:

@request.auth.id != "" &&
@request.body.user = @request.auth.id &&
@request.body.loginBonus.day >= @todayStart &&
@request.body.loginBonus.day <= @todayEnd

@request.body.loginBonus.day at first glance, day is NOT in the body of the request. It automatically triggers a join due to loginBonus being a relation, which matches the foreign key provided by the client. Resulting in this join:

WITH login_bonus_earned__pb_create__OKR3fS AS (
  SELECT
    '__temp_id____pb_create__OKR3fS' AS id,
    'qa5d6btux2iqwwb' AS user,
    '38ivyu8qk4tenot' AS loginBonus,
    '' AS created,
    '' AS updated
)
SELECT DISTINCT (1)
FROM login_bonus_earned__pb_create__OKR3fS
LEFT JOIN login_bonus AS __data_login_bonus_loginBonus
  ON __data_login_bonus_loginBonus.id = '38ivyu8qk4tenot'
WHERE (
  'qa5d6btux2iqwwb' IS NOT '' AND
  'qa5d6btux2iqwwb' = 'qa5d6btux2iqwwb' AND
  __data_login_bonus_loginBonus.day >= '2025-07-20 00:00:00.000Z' AND
  __data_login_bonus_loginBonus.day <= '2025-07-20 23:59:59.999Z'
)
LIMIT 1

Smart Home MCP Server with VSCode Copilot

Victorio Berra — Wed, 16 Apr 2025 18:19:09 +0000

What is MCP?

My short definition is: A protocol that lets LLMs talk to your code.

For example, if you have an API that returns a list of warehouses, how could you get Copilot to call your API and return a list of those warehouses? Or use the list internally to answer your questions about your warehouses?

The answer is MCP. These clients can use it https://modelcontextprotocol.io/clients

VSCode is in that list. They have added support for MCP https://code.visualstudio.com/docs/copilot/chat/mcp-servers

What can I write my MCP server in?

Python SDK
TypeScript SDK
Java SDK
Kotlin SDK
C# SDK

Using CSharp

I followed this tutorial, I will simplify here and adapt for my Smart Home

https://devblogs.microsoft.com/dotnet/build-a-model-context-protocol-mcp-server-in-csharp/

Create a .NET app:

dotnet new console -n SmartHomeMCPServer
dotnet add package ModelContextProtocol
dotnet add package Microsoft.Extensions.Logging.Console
dotnet add package Microsoft.Extensions.Hosting
dotnet add package RestEase (http client helper https://github.com/canton7/RestEase)

I am using a Hubitat with an API plugin https://hubitat.com/.

This is what the API plugin shows me:

I used Copilot edit mode to generate my app after feeding in the API spec above. I also fed it the RestEase docs which is just the entire README.md. This is the HTTP client it generated with RestEase:

public interface IHubitatApi
{
    [Get("apps/api/60/devices/all")]
    Task<List<Device>> GetAllDevicesAsync([Query("access_token")] string accessToken);

    [Get("apps/api/60/devices/{deviceId}/{command}/{secondaryValue}")]
    Task SendDeviceCommandAsync(
        [Path("deviceId")] string deviceId,
        [Path("command")] string command,
        [Path("secondaryValue")] string secondaryValue,
        [Query("access_token")] string accessToken
    );

    [Get("apps/api/60/devices/{deviceId}/{command}")]
    Task SendDeviceCommandAsync(
        [Path("deviceId")] string deviceId,
        [Path("command")] string command,
        [Query("access_token")] string accessToken
    );
}

public class Device
{
    public string Id { get; set; }
    public string Name { get; set; }
    public string Label { get; set; }
    public string Type { get; set; }
    public string Room { get; set; }
    public string Date { get; set; }
    public string Model { get; set; }
    public string Manufacturer { get; set; }
    public List<string> Capabilities { get; set; }
    public Attributes Attributes { get; set; }
    public List<Command> Commands { get; set; }
}

// ...

After that, I wrapped the client around an MCP server:

[McpServerToolType]
public static class HubitatTool
{
    [McpServerTool, Description("Fetches all devices from the Hubitat API.")]
    public static async Task<List<Device>> GetAllDevices()
    {
        const string baseUrl = "http://192.168.1.120";
        const string accessToken = "YOUR-TOKEN";

        var api = RestClient.For<IHubitatApi>(baseUrl);
        return await api.GetAllDevicesAsync(accessToken);
    }

    [McpServerTool, Description("Send a command to a specific device.")]
    public static async Task SendDeviceCommand(string deviceId, string command, string secondaryValue = null)
    {
        const string baseUrl = "http://192.168.1.120";
        const string accessToken = "YOUR-TOKEN";

        var api = RestClient.For<IHubitatApi>(baseUrl);

        // Ensure secondaryValue is handled properly
        if (string.IsNullOrWhiteSpace(secondaryValue))
        {
            await api.SendDeviceCommandAsync(deviceId, command, accessToken);
        }
        else
        {
            await api.SendDeviceCommandAsync(deviceId, command, secondaryValue.Trim(), accessToken);
        }
    }
}

I "vibe coded" the above, in a real world scenario id have a .env file for tokens and URLs.

Finally, you can ask the LLM to list your devices!

Easy Car Wash Tips

Victorio Berra — Fri, 14 Mar 2025 18:11:29 +0000

Tools of the trade:

None of the below are affiliate links! And any of these can probably be purchased cheaper at your local hardware store. Amazon links just for reference and images.

Initial rinse

Use the 40 degree nozzle about 2 feet away from the paint. Important, if it hurts your hand, it hurts the paint! Do not pressure wash damaged areas of a car, and aim to avoid seals and gaps where pressurized water would enter in places rain normally would not. Also I would not use anything less than 40 degrees as anything tighter could damage the paint or increase damages in problem areas (rock chips, scratches, rust).

Wash top-down. I usually start at the windows and work my way down. A lot of times for bird-poop and other stuck on times an initial once-over softens it, and then come back a after rinsing other parts and it blasts right off. Don't try to laser dry/hardened right off the bat.

Its safe to hit your rims, and carefully wash out your wheel wells as well. Knocks down a lot of salt and dirt.

Soap

Make sure you test the foam cannon on the ground first. I noticed unless you adjust the cannon I linked, it will be in laser mode that can blast a line into concrete.

Some cannons demand a specific soap and ratio. Mr Pink seems to work great in anything and you may want to play with ratios to nail down what works best.

Do the cars baseboards and near the tires last! Use an entirely different glove to wash your tires and VS your car paint and store them separate. The reason being brake dust is basically sandpaper and it will destroy your paint. Plus all the horrible road dust, dirt, gravel, asphalt. Not stuff you want to go over in a glove and then buff into your paint.

Wax

Any wax is going to work well. I have used Turtle Wax and Meguiar's. I am now trying the wax I linked above by Chemical-Guys which is liquid. I think I prefer solid but it works well.

You may not have a choice but its best to wax in the shade. You want the wax to find its way into all those swirls, scratches and imperfections of the paint. The sun will insta-dry it out.

Common questions: yes its totally OK to wax your headlights and taillights! This will protect them against UV damage and bug splatters. Yes its totally OK to wax your rims! Obviously avoid the caliper/rotors and make sure you wash the rims WELL. As waxing on top of brake pad dust is literally sandpaper.

The best part about waxing is you can do a few "maintenance washes" with just water later and knock down so much dirt and bugs that slide right off the wax.

Plastic Restorer

This stuff is awesome. It can make 10 year old plastic brand new. It protects from UV damage and deepens and darkens the black on your plastic. Cant recommend enough. Check out pics on Amazon.

Rubber Seal Restorer

I haven't used these too much but as your seals age and rubber ages, a plasticizer is a helpful treatment to restore your cars seals. Door seals, window seals, trunk seals, etc. Personally id only use this for older seals until I have seen data that products like these do not actually increase the rate at which seals dry out without frequent re-applications of products.

Wiper Blade Tips

Buy some silicone grease, I like Super Lube, and wipe down your wiper blades with it. RainX sells blades they call Silicone AdvantEdge™️ this is literally just silicone grease on the blade. And they charge a LOT more for these blades.

Cleanup

Make sure you add pump protector before you store your pressure washer for the season. If you have a garage, I have seen dudes mount theirs to the wall and have a little shelf for different nozzles and their soap blaster and stuff.

Wash out your cannons with fresh water. Soap can dry and harden and ruin nozzles and cannon siphons and filters.

Send all your mitts through the washing machine. Hand-wash wax soaked rags in warm soapy water. You do not want your washing machine pump soaked in wax water.

Dapper mappings, which is best?

Victorio Berra — Wed, 15 May 2024 16:03:50 +0000

Like many, our Database columns look like this customer_id. In CSharp, guidelines suggest using CustomerId for properties.

The first thing people will tell you is to just alias the column in SQL. IE: SELECT customer_id AS CustomerId .... This works great... unless you are calling a SPROC. In many environments, making changes to SPROCS that could be called by other SPROCS or countless downstream apps is completely out of the question.

There are 3 ways I have learned to map.

Use dynamic.
Use Dapper.FluentMap
Use custom [Column] attribute

1. Use dynamic

var queryResult = await connection
    .QueryAsync(
        new CommandDefinition(
            commandText: sql,
            parameters: p,
            commandType: CommandType.StoredProcedure,
            cancellationToken: cancellationToken));

return queryResult?.Select(x => new SponsorGuestRequest
{
    CustomerId = x.customer_id,
}).ToList();

Above, x is dynamic, which is what .QueryAsync() returns if you do not give it a type.

Pros:

No external libs, no additional configuration.

Cons:

Not using the Generic QueryAsync<> means you're manually doing all the mapping. You could forget to map a column. This might as well be a step above ADO.NET if you aren't using Dapper to turn SQL into POCOs.

2. Use Dapper.FluentMap

public class CustomerMap : EntityMap<Product>
{
    public CustomerMap()
    {
        Map(p => p.CustomerId)
            .ToColumn("customer_id");
    }
}

FluentMapper.Initialize(config =>
{
    config.AddMap(new CustomerMap());
});

Pros:

Very familiar syntax to anyone who has written mappers, or used FluentValidation, or used Entity Framework.

Cons:

You have to remember to call .Initialize() and provide all entity maps, you could maybe use Scrutor to scan and auto-register them? Mapping logic is now less discoverable and spread out from the classes themselves. Registration is static and global, not sure how to explain it but AutoMapper has me inject IMapper. Doing any sort of global static setup of dependencies feels dirty in 2024.

3. Use custom `[Column]` attribute

// Via https://stackoverflow.com/questions/8902674/manually-map-column-names-with-class-properties
public class ColumnAttributeTypeMapper<T> : FallbackTypeMapper
{
    public ColumnAttributeTypeMapper()
        : base(new SqlMapper.ITypeMap[]
            {
                new CustomPropertyTypeMap(
                    typeof(T),
                    (type, columnName) =>
                        type.GetProperties().FirstOrDefault(prop =>
                            prop.GetCustomAttributes(false)
                                .OfType<ColumnAttribute>()
                                .Any(attr => attr.Name == columnName)
                            )
                    ),
                new DefaultTypeMap(typeof(T))
            })
    {
    }
}

public class Customer
{
    [Column("customer_id")]
    public int CustomerId { get; set; }
}

Pros:

Highly familiar [Column()] syntax. Highly discoverable, mapping is right on the model.

Cons:

Custom code must now be maintained, and probably tested. Why is this not just in Dapper?

Personally, I like the [Column()] feature. What are you using?

AWS Lightsail Container Services with Reverse Proxy

Victorio Berra — Sun, 01 Jan 2023 22:17:00 +0000

Including: SSL, custom domains and Route 53

Today I will show you how to leverage AWS Lightsail Container Services to create an environment where you have 2 public services behind a reverse proxy, and a database.

I host my domain in AWS Route 53. I use Lightsail Custom Domains to both assign custom domains to my container service, and to create a certificate.

Caddy? Logto?

The reverse proxy is Caddy

Caddy will proxy 2 apps:

An app called LogTo think self-hosted Auth0.
An app called whoami. This is a dead simple go app that spits back header and IP information.

In the real world, you would host an app instead of whoami that is protected with logto.

Lastly a Postgres container for this example. You should probably use the Lightsail Postgres database offering, at time of writing it is $10/mo.

Prerequisites

Grab the AWS CLI, configure it, and the lightsail cli plugin.
Install docker, I did this all on Windows 10 with WSL2.

I use a mix of the GUI and CLI. You could do it all in the GUI, or CLI.

I specify my AWS region as Ohio, because when I created my container service, it created it in Ohio (us-east-2). You might have to change your commands to reflect your region.

Buy a domain on Route53

Do not delete NS, or SOA records.

Create a container service

Head over to lightsail and create a container service, or use the CLI. Beware windows users, for the CLI I had to follow a guide here, and the CLI still wouldn't work for me. Ended up moving the lightsail exe to another bin folder for something like git to get it to work. I might have clobbered my environment vars with their setx command recommendation too...

A micro container service worked just fine for me. $10/mo at time of writing.

Create a cert, and some custom domains

For the cert and custom domains: Create example.com, www.example.com, logto.example.com and reserved.example.com. I created reserved in case I needed a new domain.

Create the DNS records, go to your lightsail instance and grab the Public domain (<service-name>.<random-guid>.<aws-region-name>.cs.amazonLightsail.com) you will need the lightsail hosted zone id, list here: https://lightsail.aws.amazon.com/ls/docs/en_us/articles/amazon-lightsail-route-53-alias-record-for-container-service for example, for US East (Ohio) (us-east-2) it is Z10362273VJ548563IY84.

Create a file change-resource-record-sets.json this will be used by the AWS CLI to create a bunch of DNS records.

{
  "Comment": "Comment",
  "Changes": [
    {
      "Action": "CREATE",
      "ResourceRecordSet": {
        "Name": "example.com",
        "Type": "A",
        "AliasTarget": {
          "HostedZoneId": "MYHOSTEDZONEID",
          "DNSName": "<service-name>.<random-guid>.<aws-region-name>.cs.amazonLightsail.com",
          "EvaluateTargetHealth": true
        }
      }
    },
    {
      "Action": "CREATE",
      "ResourceRecordSet": {
        "Name": "www.example.com",
        "Type": "A",
        "AliasTarget": {
          "HostedZoneId": "MYHOSTEDZONEID",
          "DNSName": "<service-name>.<random-guid>.<aws-region-name>.cs.amazonLightsail.com",
          "EvaluateTargetHealth": true
        }
      }
    },
    {
      "Action": "CREATE",
      "ResourceRecordSet": {
        "Name": "logto.example.com",
        "Type": "A",
        "AliasTarget": {
          "HostedZoneId": "MYHOSTEDZONEID",
          "DNSName": "<service-name>.<random-guid>.<aws-region-name>.cs.amazonLightsail.com",
          "EvaluateTargetHealth": true
        }
      }
    },
    {
      "Action": "CREATE",
      "ResourceRecordSet": {
        "Name": "reserved.example.com",
        "Type": "A",
        "AliasTarget": {
          "HostedZoneId": "MYHOSTEDZONEID",
          "DNSName": "<service-name>.<random-guid>.<aws-region-name>.cs.amazonLightsail.com",
          "EvaluateTargetHealth": true
        }
      }
    }
  ]
}

deploy with aws route53 change-resource-record-sets --hosted-zone-id MYHOSTEDZONEID --change-batch file://change-resource-record-sets.json

Don't forget to attach the cert to your instance!

Build and publish images to Lightsail Container Services Image Store

Create a Caddy, Logto, and Postgres docker file each in their own folders.

Caddy

Create a folder for /Caddy, create a file called Caddyfile, put the following in:

example.com:80 {
    reverse_proxy localhost:3001 {
        trusted_proxies private_ranges
    }
}

logto.example.com:80 {
    reverse_proxy localhost:3002 {
        trusted_proxies private_ranges
    }
}

Create the docker file /Caddy/Dockerfile

FROM caddy:latest

COPY Caddyfile /etc/caddy/Caddyfile

Logto

Create a docker file /Logto/Dockerfile

FROM ghcr.io/logto-io/logto:prerelease

ENTRYPOINT ["sh", "-c", "npm run cli db seed -- --swe && npm start"]

Notes: You need to call db seed --swe with LogTo so it will connect to Postgres, and create/seed the database. --swe means "skip when exists".

Postgres

Create a docker file /Postgres/Dockerfile

FROM postgres:14-alpine

Note: Yeah... I know. I had to do this. I could NOT just use registry.hub.docker.com/postgres:14-alpine, I had to create and push my own containers. Otherwise, Lightsail Container Services will refuse to start, the error log looks like this:

[1/Jan/2022:01:03:21] [deployment:8] Creating your deployment
[1/Jan/2022:01:04:31] [deployment:8] Started 1 new node
[1/Jan/2022:01:05:25] [deployment:8] Started 1 new node
[1/Jan/2022:01:06:34] [deployment:8] Started 1 new node
[1/Jan/2022:01:07:12] [deployment:8] Canceled

All the googling I did was unhelpful. I saw someone mention I should publish container images to my Lightsail container services image repo... and that worked.

Build

cd Caddy; docker build -t caddy-container .
cd Postgres; docker build -t postgres-container .
cd Logto; docker build -t logto-container .

Publish the images to Lightsail Container Registry

aws lightsail push-container-image --region us-east-2 --service-name myservice --label caddy-container --image caddy-container
aws lightsail push-container-image --region us-east-2 --service-name myservice --label postgres-container --image postgres-container
aws lightsail push-container-image --region us-east-2 --service-name myservice --label logto-container --image logto-container

Finally, create the container service

Create 2 JSON files. containers.json and public-endpoint.json

For containers.json:

NOTE: Environment variables are super important! Change them to whatever you need. Also image names!

{
  "whoami-3001": {
    "image": "registry.hub.docker.com/containous/whoami",
    "command": ["--port", "3001"]
  },
  "postgres": {
    "image": ":myservice.postgres-container.NUMBERHERE",
    "environment" : {
        "POSTGRES_USER": "postgresuser",
        "POSTGRES_PASSWORD": "yoursecretpassword1234"
    }
  },
  "logto": {
    "image": ":myservice.logto-container.NUMBERHERE",
    "environment" : {
        "TRUST_PROXY_HEADER": "1",
        "DB_URL": "postgres://postgresuser:yoursecretpassword1234@localhost:5432/logto",
        "ENDPOINT": "https://logto.example.com",
        "PORT": "3002",
        "NODE_ENV": "production"
    }
  },
  "caddy": {
    "image": ":myservice.caddy-container.NUMBERHERE",
    "ports": {
      "80": "HTTP"
    }
  }
}

for public-endpoint.json

{
  "containerName": "caddy",
  "containerPort": 80
}

Deploy!

aws lightsail create-container-service-deployment --region us-east-2 --service-name myservice --containers file://containers.json --public-endpoint file://public-endpoint.json

Questions/Pain-points

No way in UI to delete a certificate? What if I need to add a new subdomain to my cert? I saw the CLI has delete-certificate, I tried and got Distribution-related APIs are only available in the us-east-1 Region. Please set your Region configuration to us-east-1 to create, view, or edit distribution resources....
Installing the Amazon Lightsail container services plugin for Windows is difficult. Compare this to installing an Azure CLI extension. az extension add --name <extension-name>...
Postgres would not launch inside Lightsail when using public image (registry.hub.docker.com/postgres:14-alpine). I had to create my own blank docker file as shown above, and deploy it to the Lightsail container registry and it worked perfect. Still super confused by this one...
The public endpoint is doing SSL termination, and setting X-Forwarded-* headers in addition to allowing traffic from my custom domain. This means there is effectively a load balancer in front of my containers. But there does not seem to be docs explaining this. For example, Caddy is listening on port 80, and I expose HTTP port 80 as my public pot, but ALL traffic to my app comes in through SSL 443, and X-Forwarded-Proto is even https.
You can only have 4 custom domains pointed to your container service... if I hosted two more public apps in my container service, id be SOL.
What is a public port on a container exactly? My containers seem to be able to talk without me needing to use public ports. For example, logto can talk to Postgres without issue. Same with the "whoami" container. If I can set multiple public ports, then why can I only set a single public endpoint to one of these ports? What does public ports even do?

DNS notes

Route 53 is an easy powerful way to maintain your domain. Docs will advise you use Lightsail Container Services to host the domain, but unless you register your domain using Lightsail in the first place, you will need to wait 48 hours for the nameservers to update for Lightsail to have control.

I chose to keep my DNS on Route 53. But I did have issues with my Apex domain. I tried following the docs here https://Lightsail.aws.amazon.com/ls/docs/en_us/articles/amazon-Lightsail-route-53-alias-record-for-container-service

Somewhat odd they say:

"You can route traffic for a registered domain, such as example.com, to the applications running on a Lightsail container service. You do this by adding an alias record to the hosted zone of your domain that points to the default domain of your Lightsail container service. You can do this only by using the AWS Command Line Interface (AWS CLI). It cannot be done using the Route 53 console.".

They also say:

"To use the root of your domain with your Lightsail container service, you must specify an @ symbol in the subdomain space of your domain (for example, @.example.com)."

I tried this, and it did not work for me at all. In fact, simply creating an A record pointing to my Lightsail alias <service-name>.<random-guid>.<aws-region-name>.cs.amazonLightsail.com worked for me.

I created an issue on the AWS form (re:Post) and have not yet received a reply. https://repost.aws/questions/QU-y7P6lmzQZKj-QasRd8MyA/lightsail-container-services-with-an-apex-domain-in-route-53

Auth0 Custom Google Social Login Connection

Victorio Berra — Wed, 16 Feb 2022 02:13:57 +0000

The Auth0 free tier only lets you have a couple social connections. If you use their Google one, then you have used up one of your connections.

We can create a custom social connection, and connect it to Google.

Head to GCP

Click API & Services
Click OAuth Consent Screen
Fill out the requested info, put whatever you want for the URLs, just make sure you add the domain auth0.com
Add the top 3 basic scopes: profile, email, openid
Add test users
save it all, and click "Credentials" on the left navigation menu
Create an OAuth Client Id, for redirect URI use: https://YOUR_TENANT.us.auth0.com/login/callback

Now head over to Auth0, click "Authentication" -> "Social" -> Create

Scroll to bottom of all the options, click "Create Custom"
name it "GCP-whatever-you-want", might be a good idea to put your GCP project name or something
Use the URLs they gave you on the Credentials screen in GCP
Authorization URL: https://accounts.google.com/o/oauth2/auth
Token: https://oauth2.googleapis.com/token

Finally, use this user info script:

function(accessToken, ctx, cb) {
  var i = jwt.decode(ctx.id_token);
  var p = {
    user_id: i.sub,
    email: i.email,
  };
  cb(null,p);
  }

The ID_Token from Google has the two most important things we need: sub, and email.

the access_token wont however.

Happy developing!

Use the .NET Core AWS SDK on Blazor WASM

Victorio Berra — Mon, 25 Jan 2021 15:38:41 +0000

The .NET Core AWS SDK has released an update that allows it to run fully in the browser thanks to WASM.

In this post, id like to show an example that enumerates S3 objects fully client side! You should also be able to utilize services like Cognito and DynamoDB.

Before we get started, id like to point out some caveats that I ran into while testing this functionality.

⚠ .NET 5 not yet supported! This is because Crypto libraries no longer exist in .NET 5 WASM, so AWS can't sign requests. More info here.
⚠ Today, you must use the UseAlternateUserAgentHeader wherever possible. More on this below.

Lets get started!

Prerequisites

Grab the latest Core 3.1 SDK: https://dotnet.microsoft.com/download/dotnet-core/3.1

AWS (bucket & user)

Create an S3 bucket (or use an existing)
Add a few objects

❗ Add a CORS policy. Remember, you're in the browser now! The AWS SDK requests to the AWS API will show up right in the networking tab of your browser's dev tools. Very cool!

[
    {
        "AllowedHeaders": [
            "*"
        ],
        "AllowedMethods": [
            "GET",
            "HEAD"
        ],
        "AllowedOrigins": [
            "*"
        ],
        "ExposeHeaders": [],
        "MaxAgeSeconds": 3000
    }
]

Create an IAM user and select "Programmatic access", apply an existing policy called AmazonS3ReadOnlyAccess to the user, download the resulting credentials csv or copy the access key ID and secret and store somewhere.

🎉 You are ready to code! 🎊

Code

Create a folder, and inside place a global.json with the following contents:

{
  "sdk": {
    "version": "3.1.403"
  }
}

This tells the dotnet CLI that any dotnet commands you run inside that folder should target 3.1.

Next, scaffold out a new blazor project by running the following commands:

dotnet new blazorwasm --name s3blazorwasm
cd s3blazorwasm
dotnet add package AWSSDK.S3

AWSSDK.S3 has a dependency on AWSSDK.Core.

In _imports.razor import the AWS stuff:

@using Amazon
@using Amazon.S3
@using Amazon.S3.Model

For speed and brevity, open up Counter.razor clear out everything and place the following:

@page "/counter"

<h1>S3 List Objects</h1>

@if (bucketObjectNames == null)
{
    <p><em>Loading...</em></p>
}
else
{
    <ul>
        @foreach (var bucketObjectName in bucketObjectNames)
        {
            <li>@bucketObjectName</li>
        }
    </ul>
}

@code {
    private List<string> bucketObjectNames;

    protected override async Task OnInitializedAsync()
    {
        var config = new AmazonS3Config
        {
            RegionEndpoint = RegionEndpoint.USEast1,
            UseAlternateUserAgentHeader = true
        };

        var s3Client = new AmazonS3Client("YOURKEY", "YOURSECRET", config);

        var bucketObjectResponse = await s3Client.ListObjectsV2Async(new ListObjectsV2Request
        {
            BucketName = "YOURBUCKET"
        });

        bucketObjectNames = bucketObjectResponse.S3Objects.Select(x => x.Key).ToList();
    }
}

ℹ Special note on UseAlternateUserAgentHeader. Internally, the AWS SDK generates a user agent on its own and includes it as part of the signature. When your browser makes a request it ends up using the User-Agent that Chrome generates. This fails the signature check. You need UseAlternateUserAgentHeader to tell AWS to use the generated header (looks like: x-amz-user-agent: aws-sdk-dotnet-netstandard/3.5.7.8 aws-sdk-dotnet-core/3.5.2.0 .NET_Core/(3.2-wasm/1a6e64a9381) OS/web ClientAsync) which gets signed, and sent in the request and validated by AWS properly.

Fire up your app, you should see a list of your objects when you navigate to /counter! Of course feel free to update the page name, route and navigation.

Hopefully this helps you get started, I noticed as this is so new, there are not many docs or information available on any of this yet.

Also, don't forget to delete the user you created and/or remove policies from it, as well as remove the CORS policy from your bucket.

Open GitExtensions from Visual Studio

Victorio Berra — Tue, 24 Mar 2020 16:13:06 +0000

If you haven't used GitExtensions as your Git GUI, I highly recommend it. Unlike other Git GUIs, it does not attempt to hold your hand. It is feature rich, under very active development.

I try and automate or script everything I can. Something I find myself doing a lot is opening a solution using a shortcut, and then having to open a CLI or explorer and having to navigate to my solution dir and open GitExtensions.

I recently learned about "External Tools" in Visual Studio.

It is as simple as it sounds:

You can call external tools from inside Visual Studio by using the Tools menu. A few default tools are available from the Tools menu, and you can customize the menu by adding other executables of your own.

For GitExtensions, open "Tools" -> "External Tools..." and then enter the following info in the dialog:

And when youre done, click the Tools menu, and you should see this:

GitHub Actions Mini Tutorial - Only Deploy Master

Victorio Berra — Fri, 28 Feb 2020 22:14:48 +0000

I thought id share a little info on how to have a conditional statement for your actions that ONLY executes if you are on master.

Firstly, only run on pushes to master, and PRs to master.

on:
  # Trigger the workflow on push or pull request for master only.
  push:
    branches:
      - master
  pull_request:
    branches:
      - master

Then finally, in any action:

    # Upload to S3
    - name: sync s3
      # Do not deploy anything other than master
      if: github.ref == 'refs/heads/master'    
      uses: jakejarvis/s3-sync-action@master
      with:
        args: --acl public-read --follow-symlinks --delete
      env:
        AWS_S3_BUCKET: ${{ secrets.AWS_S3_BUCKET }}
        AWS_ACCESS_KEY_ID: ${{ secrets.AWS_ACCESS_KEY_ID }}
        AWS_SECRET_ACCESS_KEY: ${{ secrets.AWS_SECRET_ACCESS_KEY }}
        AWS_REGION: 'us-east-1'
        SOURCE_DIR: 'dist'

The key above being: if: github.ref == 'refs/heads/master'

Personally I wish they would make the branch name an env variable to make this much cleaner but Actions are still new.

Happy deploying!

Gotchas when deploying a WebJob to Azure.

Victorio Berra — Wed, 05 Feb 2020 18:41:28 +0000

Azure App Services - Deploying a .NET Core 3.1 worker as a schedule triggered WebJob via DevOps

This article will cover some issues I ran into creating a new .NET Core 3.1 worker as a scheduled triggered WebJob on Azure App Services (AAS)

You might need to be a little familiar with: .NET Core, Azure DevOps, AAS, Kudu.

Goals

Lean into conventions as much as possible. Avoid complicated customization to the pipelines YAML, or deployment pipeline or unnecessary extra scripts and files in the repo.

Getting Started

Environment

Create the AAS as Windows only because WebJobs are not yet available on Linux.
Enable Always-On (so the scheduler works). You may see Core 3.1 Runtime is not on the Windows AAS according to the Azure Portal GUI but it actually is.

Pipeline

Creating the app from scratch is easy: dotnet new worker --name MyFirst.Worker

Create a very basic .NET Core azure-pipeline.yaml, all it needs to do is run the core publish task and then publish the artifact. The core template for DevOps does this out of the box.

Use this guide: https://docs.microsoft.com/en-us/azure/devops/pipelines/ecosystems/dotnet-core?view=azure-devops

steps:

# ...

- task: DotNetCoreCLI@2
  inputs:
    command: publish
    publishWebProjects: True
    arguments: '--configuration $(BuildConfiguration) --output $(Build.ArtifactStagingDirectory)'
    zipAfterPublish: True

# this code takes all the files in $(Build.ArtifactStagingDirectory) and uploads them as an artifact of your build.
- task: PublishBuildArtifacts@1
  inputs:
    pathtoPublish: '$(Build.ArtifactStagingDirectory)' 
    artifactName: 'MyFirstWorker'

Run the build and verify your worker was published.

First gotcha:

Always generate an Exe. Kudu looks for specific files to run. If you use the DevOps Ubuntu agent, you wont get an Exe. So how can we ALWAYS get one?

You need to edit the publish arguments in your pipeline YAML and add -r win-x86 --self-contained false. this will force Core to spit out an Exe for windows, even when publishing on Ubuntu. WITHOUT creating a self contained deployment. Although, there is nothing wrong with creating an SCD if you want. might even be a good idea, but for this tutorial I want to go from dotnet new to a working deployment without lighting up every feature in Core and without adding more scripts and files to my repo.

NOTE: To see how Kudu picks something to run, check the Wiki for WebJobsKudu wiki.

Second gotcha:

The path you deploy to in AAS for web jobs is highly specific. It MUST be in one of the following:

For a triggered (or scheduled) job, the folder is wwwroot\app_data\jobs\triggered\{job name}, and for a continuous job, it's wwwroot\app_data\jobs\continuous\{job name}. This is easily achieved by setting the --output flag on your dotnet publish to publish into a directory like that.

Mark Heath has a great blog post on a very similar topic:

Why the app_data folder? Well that's a special ASP.NET folder that is intended for storing your application data. The web server will not serve up the contents of this folder, so everything in there is safe. It's also considered a special case for deployments - since it might contain application generated data files, its contents won't get deleted or reset when you deploy a new version of your app.

Via Mark Heath - https://markheath.net/post/managing-webjobs-with-kudu-api

To expand on Marks last line, in Azure DevOps Pipelines -> Deployments -> Deploy AAS Task there is actually a checkbox to blow away the app_data folder if you want. I thought that was interesting.

The Kudu WebJob wiki also says:

As an alternative, d:\home\site\jobs\... can be used instead of d:\home\site\wwwroot\app_data\jobs\.... This is useful when using Run-From-Zip, which makes d:\home\site\wwwroot become read-only.

But I haven't tested this.

Third gotcha:

Make sure to copy always the settings.job file. AND it uses https://github.com/atifaziz/NCrontab which requires a seconds field. I first tried 5 0 * * * and then had to go back and do 0 5 0 * * * this means run 5 minutes after midnight every day.

This is trivial to verify, go to your AAS in the portal, and click the WebJob blade, you should see your schedule.

Full YAML

trigger:
- master

pool:
  vmImage: 'ubuntu-latest'

variables:
  buildConfiguration: 'Release'

steps:

- task: DotNetCoreCLI@2
  inputs:
    command: publish
    publishWebProjects: False
    arguments: '--configuration $(BuildConfiguration) --output $(Build.ArtifactStagingDirectory)'
    zipAfterPublish: True

- task: PublishBuildArtifacts@1
  inputs:
    pathtoPublish: '$(Build.ArtifactStagingDirectory)' 
    artifactName: 'myWebsiteName'

Things id like to know more about:

How to use Run-From-Zip WebJobs
How to add a WebApp to my repo, and deploy that with my WebJob at the same time.

SEO Tags: Azure,App,Services,Deploying,.NET,Core,3.1,worker,schedule,triggered,WebJob,DevOps

GUI for services like https://patchbay.pub/

Victorio Berra — Tue, 07 Jan 2020 21:47:57 +0000

https://patchme.link

Read patchbay.pub for the best example of how this works. The real benefit you get from this GUI is the ability for local notifications. That is to say, you can get a little ping on your desktop when a subscriber resolves. To see this in action, click "Add" and then execute the CURL command and accept notifications when prompted.

Forem: Victorio Berra

Build an Aspire API Using Microsoft OpenAI, Scalar, OpenRouter, Structured Output, and Custom Headers

Bootstraping

Scalar Aspire

Proxy Deep Dive

Back at The Ranch

AI

Pocketbase API Rules

Smart Home MCP Server with VSCode Copilot

What is MCP?

Using CSharp

Easy Car Wash Tips

Tools of the trade:

Washing

Waxing

Nice to haves

Initial rinse

Soap

Wax

Plastic Restorer

Rubber Seal Restorer

Wiper Blade Tips

Cleanup

Dapper mappings, which is best?

1. Use dynamic

Pros:

Cons:

2. Use Dapper.FluentMap

Pros:

Cons:

3. Use custom [Column] attribute

Pros:

Cons:

AWS Lightsail Container Services with Reverse Proxy

Including: SSL, custom domains and Route 53

Caddy? Logto?

Prerequisites

Buy a domain on Route53

Create a container service

Create a cert, and some custom domains

Build and publish images to Lightsail Container Services Image Store

Caddy

Logto

Postgres

Build

Publish the images to Lightsail Container Registry

Finally, create the container service

Deploy!

Questions/Pain-points

DNS notes

Auth0 Custom Google Social Login Connection

Use the .NET Core AWS SDK on Blazor WASM

Prerequisites

AWS (bucket & user)

Code

Open GitExtensions from Visual Studio

GitHub Actions Mini Tutorial - Only Deploy Master

Gotchas when deploying a WebJob to Azure.

Azure App Services - Deploying a .NET Core 3.1 worker as a schedule triggered WebJob via DevOps

Goals

Getting Started

Environment

Pipeline

First gotcha:

Second gotcha:

Third gotcha:

Full YAML

GUI for services like https://patchbay.pub/

3. Use custom `[Column]` attribute