IPC sv SE
ASF includes its own unique IPC interface that can be used for further interaction with the process. IPC stands for inter-process communication and in the most simple definition this is "ASF web interface" based on Kestrel HTTP server that can be used for further integration with the process, both as a frontend for end-user (ASF-ui), and backend for third-party integrations (ASF API).
IPC can be used for a lot of different things, depending on your needs and skills. For example, you can use it for fetching status of ASF and all bots, sending ASF commands, fetching and editing global/bot configs, adding new bots, deleting existing bots, submitting keys for BGR or accessing ASF's log file. All of those actions are exposed by our API, which means that you can code your own tools and scripts that will be able to communicate with ASF and influence it during runtime. In addition to that, selected actions (such as sending commands) are also implemented by our ASF-ui which allows you to easily access them through a friendly web interface.
You can enable our IPC interface by enabling IPC global configuration property. ASF will state IPC launch in its log, which you can use for verifying if IPC interface has started properly:
INFO|ASF|Start() Starting IPC server...
INFO|ASF|Start() IPC server ready!
ASF's http server is now listening on selected endpoints. If you didn't provide a custom configuration file for IPC, those will be IPv4-based 127.0.0.1 and IPv6-based [::1] on default 1242 port. You can access our IPC interface by above links, from the same machine as the one running ASF process.
ASF's IPC interface exposes three different ways to access it, depending on your planned usage.
On the lowest level there is ASF API that is the core of our IPC interface and allows everything else to operate. This is what you want to use in your own tools, utilities and projects in order to communicate with ASF directly.
On the medium ground there is our Swagger documentation which acts as a frontend to ASF API. It features a complete documentation of ASF API and also allows you to access it more easily. This is what you want to check if you're planning on writing a tool, utility or other projects that are supposed to communicate with ASF through its API.
On the highest level there is ASF-ui which is based on our ASF API and provides user-friendly way to execute various ASF actions. This is our default IPC interface designed for end-users, and a perfect example of what you can build with ASF API. If you'd like, you can use your own custom web UI to use with ASF, by specifying --path command-line argument and using custom www directory located there.
ASF-ui is a community project that aims to create user-friendly graphical web interface for end-users. In order to achieve that, it acts as a frontend to our ASF API, allowing you to do various actions with ease. This is the default UI that ASF comes with.
As stated above, ASF-ui is a community project that isn't maintained by core ASF developers. It follows its own flow in ASF-ui repo which should be used for all related questions, issues, bug reports and suggestions.
Our ASF API is typical RESTful web API that is based on JSON as its primary data format. We're doing our best to precisely describe response, using both HTTP status codes (where appropriate), as well as a response you can parse yourself in order to know whether the request succeeded, and if not, then why.
Our ASF API can be accessed by sending appropriate requests to appropriate /Api endpoints. You can use those API endpoints to make your own helper scripts, tools, GUIs and alike. This is exactly what our ASF-ui achieves under the hood, and every other tool can achieve the same. ASF API is officially supported and maintained by core ASF team.
For complete documentation of available endpoints, descriptions, requests, responses, http status codes and everything else considering ASF API, please refer to our swagger documentation.
ASF IPC interface by default does not require any sort of authentication, as IPCPassword is set to null. However, if IPCPassword is enabled by being set to any non-empty value, every call to ASF's API requires the password that matches set IPCPassword. If you omit authentication or input wrong password, you'll get 401 - Unauthorized error. If you continue sending requests without authentication, eventually you'll get temporarily blocked with 403 - Forbidden error.
Authentication can be done through two separate ways.
In general you should use HTTP request headers, by setting Authentication field with your password as a value. The way of doing that depends on the actual tool you're using for accessing ASF's IPC interface, for example if you're using curl then you should add -H 'Authentication: MyPassword' as a parameter. This way authentication is passed in the headers of the request, where it in fact should take place.
Alternatively you can append password parameter to the end of the URL you're about to call, for example by calling /Api/ASF?password=MyPassword instead of /Api/ASF alone. This approach is good enough, but obviously it exposes password in the open, which is not necessarily always appropriate. In addition to that it's extra argument in the query string, which complicates the look of the URL and makes it feel like it's URL-specific, while password applies to entire ASF API communication.
Both ways are supported and it's totally up to you which one you want to choose. We recommend to use HTTP headers everywhere where you can, as usage-wise it's more appropriate than query string. However, we support query string as well, mainly because of various limitations related to request headers. A good example includes lack of custom headers while initiating a websocket connection in javascript (even though it's completely valid according to the RFC). In this situation query string is the only way to authenticate.
Our IPC interface, in additon to ASF API and ASF-ui also includes swagger documentation, which is available under /swagger URL. Swagger documentation serves as a middle-man between our API implementation and other tools using it (e.g. ASF-ui). It provides a complete documentation and availability of all API endpoints in OpenAPI specification that can be easily consumed by other projects, allowing you to write and test ASF API with ease.
Apart from using our swagger documentation as a complete specification of ASF API, you can also use it as user-friendly way to execute various API endpoints, mainly those that are not implemented by ASF-ui. Since our swagger documentation is generated automatically from ASF code, you have a guarantee that the documentation will always be up-to-date with the API endpoints that your version of ASF includes.
ASF by default listens only on localhost addresses, which means that accessing ASF IPC from any other machine but your own is impossible. Unless you modify default endpoints, attacker would need a direct access to your own machine in order to access ASF's IPC, therefore it's as secure as it can be and there is no possibility of anybody else accessing it, even from your own LAN.
However, if you decide to change default localhost bind addresses to something else, then you're supposed to set proper firewall rules yourself in order to allow only authorized IPs to access ASF's IPC interface. In addition to doing that, we strongly recommend to set up IPCPassword, that will add another layer of extra security. You may also want to run ASF's IPC interface behind a reverse proxy in this case, which is further explained below.
Yes, this is what ASF API was designed for and you can use anything capable of sending a HTTP request to access it. Local userscripts follow CORS logic, and we allow access from all origins for them (*), as long as IPCPassword is set, as an extra security measure. This allows you to execute various authenticated ASF API requests, without allowing potentially malicious scripts to do that automatically (as they'd need to know your IPCPassword to do that).
Yes, we recommend to use a reverse proxy for that (explained below). This way you can access your web server in typical way, which will then access ASF's IPC on the same machine. Alternatively, if you don't want to run with a reverse proxy, you can use custom configuration with appropriate URL for that. For example, if your machine is in a private VPN with 10.8.0.1 address, then you can set http://10.8.0.1:1242 listening URL in IPC config, which would enable IPC access from within your private VPN, but not from anywhere else.
Yes, our IPC is fully compatible with such setup, so you're free to host it also in front of your own tools for extra security and compatibility, if you'd like to. In general ASF's Kestrel http server is very secure and possesses no risk when being connected directly to the internet, but putting it behind a reverse-proxy such as Apache or Nginx could provide extra functionality that wouldn't be possible to achieve otherwise, such as securing ASF's interface with a basic auth.
Example Nginx configuration can be found below. We included full server block, although you're interested mainly in location ones. Please refer to nginx documentation if you need further explanation.
server {
listen *:443 ssl;
server_name asf.mydomain.com;
ssl_certificate /path/to/your/certificate.crt;
ssl_certificate_key /path/to/your/certificate.key;
location ~* /Api/NLog {
proxy_pass http://127.0.0.1:1242;
# Only if you need to override default host
# proxy_set_header Host 127.0.0.1;
# X-headers should be specified in the situation where nginx is on the same machine as ASF
# They're crucial for proper usage of reverse-proxy, allowing ASF to e.g. ban the actual offenders instead of your nginx server
# Specifying them allows ASF to properly resolve IP addresses of users making requests - making nginx work as a reverse proxy
# Not specifying them will cause ASF to treat your nginx as the client - nginx will act as a traditional proxy in this case
# If you're unable to host nginx service on the same machine as ASF (e.g. different docker container), you have to comment out these
proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
proxy_set_header X-Forwarded-Host $host:$server_port;
proxy_set_header X-Forwarded-Proto $scheme;
proxy_set_header X-Forwarded-Server $host;
proxy_set_header X-Real-IP $remote_addr;
# We add those 3 extra options for websockets proxying, see https://nginx.org/en/docs/http/websocket.html
proxy_http_version 1.1;
proxy_set_header Connection "Upgrade";
proxy_set_header Upgrade $http_upgrade;
}
location / {
proxy_pass http://127.0.0.1:1242;
# Only if you need to override default host
# proxy_set_header Host 127.0.0.1;
# X-headers should be specified in the situation where nginx is on the same machine as ASF
# They're crucial for proper usage of reverse-proxy, allowing ASF to e.g. ban the actual offenders instead of your nginx server
# Specifying them allows ASF to properly resolve IP addresses of users making requests - making nginx work as a reverse proxy
# Not specifying them will cause ASF to treat your nginx as the client - nginx will act as a traditional proxy in this case
# If you're unable to host nginx service on the same machine as ASF (e.g. different docker container), you have to comment out these
proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
proxy_set_header X-Forwarded-Host $host:$server_port;
proxy_set_header X-Forwarded-Proto $scheme;
proxy_set_header X-Forwarded-Server $host;
proxy_set_header X-Real-IP $remote_addr;
}
}Example Apache configuration can be found below. Please refer to apache documentation if you need further explanation.
<IfModule mod_ssl.c>
<VirtualHost *:443>
ServerName asf.mydomain.com
SSLEngine On
SSLCertificateFile /path/to/your/fullchain.pem
SSLCertificateKeyFile /path/to/your/privkey.pem
# TODO: Apache can't do case-insensitive matching properly, so we hardcode two most commonly used cases
ProxyPass "/api/nlog" "ws://127.0.0.1:1242/api/nlog"
ProxyPass "/Api/NLog" "ws://127.0.0.1:1242/Api/NLog"
ProxyPass "/" "http://127.0.0.1:1242/"
</VirtualHost>
</IfModule>Yes, you can achieve it through two different ways. A recommended way would be to use a reverse proxy for that (described above) where you can access your web server through https like usual, and connect through it with ASF's IPC interface on the same machine. This way your traffic is fully encrypted and you don't need to modify IPC in any way to support such setup.
Second way includes specifying a custom config for ASF's IPC interface where you can enable https endpoint and provide appropriate certificate directly to our Kestrel http server. This way is recommended if you're not running any other web server and don't want to run one exclusively for ASF. Otherwise, it's much easier to achieve a satisfying setup by using a reverse proxy mechanism.
Our IPC interface supports extra config file, IPC.config that should be put in standard ASF's config directory.
When available, this file specifies advanced configuration of ASF's Kestrel http server, together with other IPC-related tuning. Unless you have a particular need, there is no reason for you to use this file, as ASF is already using sensible defaults in this case.
The configuration file is based on following JSON structure:
{
"Kestrel": {
"Endpoints": {
"example-http4": {
"Url": "http://127.0.0.1:1242"
},
"example-http6": {
"Url": "http://[::1]:1242"
},
"example-https4": {
"Url": "https://127.0.0.1:1242",
"Certificate": {
"Path": "/path/to/certificate.pfx",
"Password": "passwordToPfxFileAbove"
}
},
"example-https6": {
"Url": "https://[::1]:1242",
"Certificate": {
"Path": "/path/to/certificate.pfx",
"Password": "passwordToPfxFileAbove"
}
}
},
"KnownNetworks": [
"10.0.0.0/8",
"172.16.0.0/12",
"192.168.0.0/16"
],
"PathBase": "/"
}
}There are 2 properties worth explanation/editing, those are Endpoints and PathBase.
Endpoints - This is a collection of endpoints, each endpoint having its own unique name (like example-http4) and Url property that specifies Protocol://Host:Port listening address. By default, ASF listens on IPv4 and IPv6 http addresses, but we've added https examples for you to use, if needed. You should declare only those endpoints that you need, we've included 4 example ones above so you can edit them easier.
Host accepts a variety of values, including * value that binds ASF's http server to all available interfaces. Be extremely careful when you use Host values that allow remote access. Doing so will enable access to ASF's IPC interface from other machines, which may pose a security risk. We strongly recommend to use IPCPassword (and preferably your own firewall too) at a minimum in this case.
KnownNetworks - This variable specifies network addresses which we consider trustworthy. This property is crucial especially in combination with hosting a reverse-proxy to ASF on different machine than ASF itself - in this case, you should declare the machine's IP here, in order for ASF to respect its proxying headers and accept the requests. Specifying this variable is not required if you're not planning to use any sort of reverse-proxy with ASF, or if the reverse-proxy is located on the same machine as ASF (and therefore connecting to ASF's IPC using loopback address of 127.0.0.1). Be extremely careful with the networks you specify here, as it allows a potential IP spoofing attack in case the trusted machine is compromised or wrongly configured.
PathBase - This is base path that will be used by IPC interface. This property is optional, defaults to / and shouldn't be required to modify for majority of use cases. By changing this property you'll host entire IPC interface on a custom prefix, for example http://localhost:1242/MyPrefix instead of http://localhost:1242 alone. Using custom PathBase may be wanted in combination with specific setup of a reverse proxy where you'd like to proxy a specific URL only, for example mydomain.com/ASF instead of entire mydomain.com domain. Normally that would require from you to write a rewrite rule for your web server that would map mydomain.com/ASF/Api/X -> localhost:1242/Api/X, but instead you can define a custom PathBase of /ASF and achieve easier setup of mydomain.com/ASF/Api/X -> localhost:1242/ASF/Api/X.
Unless you truly need to specify a custom base path, it's best to leave it at default.
The following config will allow remote access from all sources, therefore you should ensure that you read and understood our security notice about that, available above.
{
"Kestrel": {
"Endpoints": {
"HTTP": {
"Url": "http://*:1242"
}
}
}
}If you do not require access from all sources, but for example your LAN only, then it's much better idea to use something like 192.168.0.* instead of *. Adapt the network address appropriately if you use a different one.
 |
 |
 |
 |
|---|---|---|---|
 |
 |
 |
 |
|---|---|---|---|
- 🏡 Home
- 🔧 Configuration
- 💬 FAQ
- ⚙️ Setting up (start here)
- 👥 Background games redeemer
- 📢 Commands
- 🛠️ Compatibility
- 🧩 ItemsMatcherPlugin
- 📋 Management
- ⏱️ Performance
- 📡 Remote communication
- 👪 Steam Family Sharing
- 🔄 Trading