swarmexec › Docs › Install & quick start
Install & quick start_
One static binary, then one command to provision the agents.
On this page
Install
Requires Docker Engine 19.03 or newer (API 1.40) on the manager and on every node. swarmexec refuses an older daemon at connect with a message naming both versions, rather than connecting and then failing one view at a time. Two features want a little more from the node's daemon — live resource usage and the per-node image view need Docker 23.0 (API 1.41/1.42); below that they stay empty and say so instead of failing.
Quickest: the install script picks the build for your machine,
checks it against the release's SHA256SUMS and refuses to install
anything it could not verify. It never uses sudo and never edits your
shell profile — read it first, it is short.
With a package manager — then brew upgrade or
scoop update brings each new release, which keeps the client in step
with the agents:
Or by hand:
The client is a single static binary. Grab the latest release build for your platform:
Builds: linux-amd64,
linux-arm64,
darwin-arm64,
darwin-amd64,
windows-amd64.
The agent image is public on Docker Hub as
logleio/swarmexec-agent.
Updates: when a newer release is out, the TUI's status line and
swarmexec doctor say so; update the way you installed (the script,
brew upgrade, scoop update or a new download). The
client learns of it from this site at most once a day, sending nothing about you
or your cluster; ui.check_updates: false
turns that off.
Shell completion
swarmexec completion bash|zsh|fish|powershell prints a completion
script. It completes more than flags: exec <TAB> offers the services
that exist, logs web.<TAB> their running slots, --stack
<TAB> the deployed stacks and --context <TAB> your Docker
contexts. Only the manager API is asked, never an agent; a completion gives up after
two seconds and over ssh:// it is never allowed to prompt, so a context
that would ask for a password offers nothing instead of freezing the shell.
Quick start
Point your Docker context at a Swarm manager, then provision the agents once.
init creates a shared secret, deploys the agent on every node in
self-signed mode, and writes a matching client config — so the other commands
work immediately.
Tear the agents down again with swarmexec down (it leaves your
client config untouched).