Command line arguments ka GE

Command-line arguments

ASF includes support for several command-line arguments that can affect the program runtime. Those can be used by advanced users in order to specify how program should run. In comparison with default way of ASF.json configuration file, command-line arguments are used for core initialization (e.g. --path), platform-specific settings (e.g. --system-required) or sensitive data (e.g. --cryptkey).

Usage

Usage depends on your OS and ASF flavour.

Generic:

dotnet ArchiSteamFarm.dll --argument --otherOne

Windows:

.\ArchiSteamFarm.exe --argument --otherOne

Linux/OS X

./ArchiSteamFarm --argument --otherOne

Command-line arguments are also supported in generic helper scripts such as ArchiSteamFarm.cmd or ArchiSteamFarm.sh. In addition to that, when using helper scripts you can also use ASF_ARGS environment property, like stated in our docker section.

If your argument includes spaces, don't forget to quote it. Those two are wrong:

./ArchiSteamFarm --path /home/archi/My Downloads/ASF # Bad!
./ArchiSteamFarm --path=/home/archi/My Downloads/ASF # Bad!

However, those two are completely fine:

./ArchiSteamFarm --path "/home/archi/My Downloads/ASF" # OK
./ArchiSteamFarm "--path=/home/archi/My Downloads/ASF" # OK

Arguments

--cryptkey <key> or --cryptkey=<key> - will start ASF with custom cryptographic key of <key> value. This option affects security and will cause ASF to use your custom provided <key> key instead of default one hardcoded into the executable. Since this property affects default encryption key (for encrypting purposes) as well as salt (for hashing purposes), keep in mind that everything encrypted/hashed with this key will require it to be passed on each ASF run.

Due to the nature of this property, it's also possible to set cryptkey by declaring ASF_CRYPTKEY environment variable, which may be more appropriate for people that would want to avoid sensitive details in the process arguments.

--ignore-unsupported-environment - will cause ASF to ignore detection of unsupported environment, which normally is signalized with an error and forced exit. As of now, unsupported environment is classifed as running .NET Framework build on platform that could be running .NET Core build instead. Since we support generic-netf builds only in very limited scenarios (with Mono), using it for other cases (e.g. for launching on win-x64 platform) is not supported. Visit compatibility for more info.

--network-group <group> or --network-group=<group> - will cause ASF to init its limiters with a custom network group of <group> value. This option affects running ASF in multiple instances by signalizing that given instance is dependent only on instances sharing the same network group, and independent of the rest. Typically you want to use this property only if you're routing ASF requests through custom mechanism (e.g. different IP addresses) and you want to set networking groups yourself, without relying on ASF to do it automatically (which currently includes taking into account WebProxy only). Keep in mind that when using a custom network group, this is unique identifier within the local machine, and ASF will not take into account any other details, such as WebProxy value, allowing you to e.g. start two instances with different WebProxy values which are still dependent on each other.

Due to the nature of this property, it's also possible to set the value by declaring ASF_NETWORK_GROUP environment variable, which may be more appropriate for people that would want to avoid sensitive details in the process arguments.

--no-config-migrate - by default ASF will automatically migrate your config files to latest syntax. Migration includes conversion of deprecated properties into latest ones, removing properties with default values (as they have no effect), as well as cleaning up the file in general (correcting indentation and likewise). This is almost always a good idea, but you might have a particular situation where you'd prefer ASF to never overwrite the config files automatically. For example, you might want to chmod 400 your config files (read permission for the owner only) or put chattr +i over them, in result denying write access for everyone, e.g. as a security measure. Usually we recommend to keep the config migration enabled, but if you have a particular reason for disabling it and would instead prefer ASF to not do that, you can use this switch for achieving that purpose.

--no-config-watch - by default ASF sets up a FileSystemWatcher over your config directory in order to listen for events related to file changes, so it can interactively adapt to them. For example, this includes stopping bots on config deletion, restarting bot on config being changed, or loading keys into BGR once you drop them into the config directory. This switch allows you to disable such behaviour, which will cause ASF to completely ignore all the changes in config directory, requiring from you to do such actions manually, if deemed appropriate. Usually we recommend to keep the config events enabled, but if you have a particular reason for disabling them and would instead prefer ASF to not do that, you can use this switch for achieving that purpose.

--no-restart - this switch is mainly used by our docker containers and forces AutoRestart of false. Unless you have a particular need, you should instead configure AutoRestart property directly in your config. This switch is here so our docker script won't need to touch your global config in order to adapt it to its own environment. Of course, if you're running ASF inside a script, you may also make use of this switch (otherwise you're better with global config property).

--path <path> or --path=<path> - ASF always navigates to its own directory on startup. By specifying this argument, ASF will navigate to given directory after initialization, which allows you to use custom path for various application parts (including config, plugins and www directories, as well as NLog.config file), without a need of duplicating binary in the same place. It may come especially useful if you'd like to separate binary from actual config, as it's done in Linux-like packaging - this way you can use one (up-to-date) binary with several different setups. The path can be either relative according to current place of ASF binary, or absolute. Keep in mind that this command points to new "ASF home" - the directory that has the same structure as original ASF, with config directory inside, see below example for explanation.

Due to the nature of this property, it's also possible to set expected path by declaring ASF_PATH environment variable, which may be more appropriate for people that would want to avoid sensitive details in the process arguments.

If you're considering using this command-line argument for running multiple instances of ASF, we recommend reading our compatibility page on this manner.

Examples:

dotnet /opt/ASF/ArchiSteamFarm.dll --path /opt/TargetDirectory # Absolute path
dotnet /opt/ASF/ArchiSteamFarm.dll --path ../TargetDirectory # Relative path works as well
ASF_PATH=/opt/TargetDirectory dotnet /opt/ASF/ArchiSteamFarm.dll # Same as env variable

├── /opt
│     ├── ASF
│     │     ├── ArchiSteamFarm.dll
│     │     └── ...
│     └── TargetDirectory
│           ├── config
│           ├── logs (generated)
│           ├── plugins (optional)
│           ├── www (optional)
│           ├── log.txt (generated)
│           └── NLog.config (optional)
└── ...

--process-required - declaring this switch will disable default ASF behaviour of shutting down when no bots are running. No auto-shutdown behaviour is especially useful in combination with IPC where majority of users would expect their web service to be running regardless of the amount of bots that are enabled. If you're using IPC option or otherwise need ASF process to be running all the time until you close it yourself, this is the right option.

If you do not intend to run IPC, this option will be rather useless for you, as you can just start the process again when needed (as opposed to ASF's web server where you require it listening all the time in order to send commands).

--service - this switch is mainly used by our systemd service and forces Headless of true. Unless you have a particular need, you should instead configure Headless property directly in your config. This switch is here so our systemd service won't need to touch your global config in order to adapt it to its own environment. Of course, if you have a similar need then you may also make use of this switch (otherwise you're better with global config property).

--system-required - declaring this switch will cause ASF to try signalizing the OS that the process requires system to be up and running for its entire lifetime. Currently this switch has effect only on Windows machines where it'll forbid your system from going into sleep mode as long as the process is running. This can be proven especially useful when farming on your PC or laptop during night, as ASF will be able to keep your system awake while it's farming, then, once ASF is done, it'll shutdown itself like usual, making your system allowed to enter into sleep mode again, therefore saving power immediately once farming is finished.

Keep in mind that for proper auto-shutdown of ASF you need other settings - especially avoiding --process-required and ensuring that all your bots are following ShutdownOnFarmingFinished. Of course, auto-shutdown is only a possibility for this feature, not a requirement, since you can also use this flag together with e.g. --process-required, effectively making your system awake infinitely after starting ASF.