A Termux environment packaged into Docker image. Environment doesn't have Android runtime components, so certain things will not be available (DalvikVM, OpenSLES, etc...).
You should have a properly configured and running Docker or Podman container systems. Further instructions will provide examples only for Docker.
This will start interactive login shell. Everything will look like in a normal Termux installation.
docker run -it ghcr.io/termux-user-repository/termux-docker:latest
When using the tag latest
, container will be 32 bit (i686 architecture).
Other architecture can be installed using a different tags. Available tags:
aarch64
arm
i686
(latest
)x86_64
If architecture is not compatible with host, the additional setup will be needed. Read this document further to learn how you can run containers of incompatible CPU architecture.
Important note: do not pass --user
option to Docker command line.
The initial user of container must be root. Otherwise DNS will be broken
because of dnsmasq
server failure.
In order to run AArch64 container on x86(64) host, you need to setup QEMU emulator through binfmt_misc. This can be easily done by one command:
docker run --rm --privileged aptman/qus -s -- -p aarch64 arm
Note that AArch64 and ARM containers work properly only in privileged mode. If you want your containers to have standard privileges, a custom seccomp profile is required.
Variant with privileged container:
docker run -it --privileged ghcr.io/termux-user-repository/termux-docker:aarch64
Variant with seccomp unconfined profile:
docker run -it --security-opt seccomp:unconfined ghcr.io/termux-user-repository/termux-docker:aarch64
You can run commands in non-interactive mode. Just append them to Docker command line.
Example:
docker run -it --rm ghcr.io/termux-user-repository/termux-docker:latest bash -c "apt update && apt install -yq clang"
By default root shell is disabled in container as Termux doesn't really support usage of package manager under root account. In cases where you really need shell with root privileges, entrypoint should be overridden.
The provided images have 2 entry points:
/entrypoint.sh
- the standard one which drops privileges tosystem
user./entrypoint_root.sh
- alternate entrypoint that does not drop privileges.
Usage example:
docker run -it --entrypoint /entrypoint_root.sh ghcr.io/termux-user-repository/termux-docker:latest
Docker:
./build-all.sh
Podman:
./build-all.sh --podman
There a number of known issues which may not be resolved:
-
ARM containers may require a custom seccomp profile to remove restrictions from
personality()
system call. -
When running certain multi threaded program in 32bit containers, the PIDs can balloon and easily exceed libc's limit. The only way to fix this is to set
/proc/sys/kernel/pid_max
to 65535. See termux/termux-docker#40.