Temporal Intro Workshop

This repository contains example code for my Temporal Intro Workshop.

Prerequisites

  1. Git, Make, etc.
  2. Make sure you have the latest Go and Docker installed

Usage

  1. Checkout this repository
  2. Run make up
  3. Wait for Temporal to start
  4. Check if Temporal is running with make ps
  5. Start a new shell with make shell

Alternatively, you can use the following alias for the tctl commands instead of opening a new shell:

alias tctl='docker compose exec temporal-admin-tools tctl'

Running a workflow using the CLI

You can run a workflow using the CLI with the following command:

tctl workflow run --taskqueue workshop --execution_timeout 60 --workflow_type WORKFLOW_TYPE -i 'arg1 arg2...'

For example, running the first example looks like this:

tctl workflow run --taskqueue workshop --execution_timeout 60 --workflow_type example01 -i '1' -i '3'

As a best practice, workflows generally have a single input struct (to remain compatible with other languages). By default, Temporal uses JSON encoding, so such workflow execution looks like this:

tctl workflow run --taskqueue workshop --execution_timeout 60 --workflow_type example02 -i '{"A": 1, "B": 2}'

You can shorten the command a lot by using shorthands for commands and options:

tctl wf run --tq workshop --et 60 --wt example01 -i '1' -i '3'

Last, but not least, if you want to start a workflow without waiting for its result, you can do so by using the start command instead of run:

tctl wf start --tq workshop --et 60 --wt example01 -i '1' -i '3'

Quering workflow state using the CLI

Workflows can register query handlers to expose state about themselves. You can query that state using the following command:

tctl workflow query --workflow_id 72daa600-3cac-49b0-9e86-277a47c80a87 --query_type current_number

Or using a shorter version:

tctl wf query --wid 72daa600-3cac-49b0-9e86-277a47c80a87 --qt current_number

There is a special query type called __stack_trace that gives you the current stack trace of the workflow. Useful if a workflow is stuck for a long time and you want to check where it stopped.

Signaling a workflow using the CLI

Workflows can register query handlers to expose state about themselves. You can query that state using the following command:

tctl workflow signal --workflow_id 72daa600-3cac-49b0-9e86-277a47c80a87 --name set_number --input '2'

Or using a shorter version:

tctl wf signal --wid 72daa600-3cac-49b0-9e86-277a47c80a87 -n set_number -i '2'

There is a special query type called __stack_trace that gives you the current stack trace of the workflow. Useful if a workflow is stuck for a long time and you want to check where it stopped.

Cleaning up

Once you are finished with the workshop, you can clean up all resources (containers) by running the following command:

make down

License

The MIT License (MIT). Please see License File for more information.

GitHub

https://github.com/sagikazarmark/temporal-intro-workshop