Flowpipe Server

By default, most Flowpipe commands run in client-only mode. For example, flowpipe pipeline run will parse and load a mod from the current directory (or --mod-location), run the pipeline, print the results, and then exit. This makes it easy to run pipelines on an ad hoc basis - just run the command.

But what if you want to run a pipeline at regular intervals, or in response to an event? Flowpipe triggers provide that capability. To use triggers, you must run Flowpipe in server mode. The flowpipe server command runs Flowpipe in server mode in the foreground:

Once it is running you can connect to from another terminal with the --host argument:

Like all Flowpipe commands, flowpipe server will load the mod from the current directory by default, but you can specify a different directory with --mod-location:

By default, Flowpipe will listen on all network interfaces, but you can pass --listen local if you only want to listen on the loopback addresses:

Flowpipe listens on port 7103 by default, but you can use the --port argument to use a different one:

While the server is running, Flowpipe will watch your mod files for changes and automatically update the server. You can disable this with the --watch argument if you prefer not to update the server instance as the files change:

Often it is simpler to manage all of these settings with a workspace instead. For example, you can add a workspace to your workspaces file (~/.flowpipe/config/workspaces.fpc):

And then use the --workspace flag when you start the server:

and when you run other Flowpipe commands against that server: