Relevant links:
- User guides: https://www.brewblox.com/
- Discord server: https://discord.gg/WaFYD2jaaT
- Previous release notes: https://www.brewblox.com/user/release_notes.html
- Project board: https://github.com/orgs/Brewblox/projects/1
- Code repositories: https://github.com/Brewblox
Brewblox release 2024/03/26
firmware release 2024/03/26
WARNING: Python 3.7 and Debian Buster are no longer supported. If you’re using Debian Buster, see https://www.brewblox.com/user/system_upgrades.html
For this release, we have two new features: Sequences can use variables,
and brewblox configuration files are now generated based on settings in the new brewblox.yml file.
Variables
Before, if you wanted to change a temperature value or a block name in a Sequence, you always had to update the sequence itself.
If you had multiple instructions targeting or waiting on the same block, you had to copy the block name in every instruction.
Now, instruction arguments can either be a value (as before), or a reference to a variable.
Variables are stored in the new Variables block. Multiple Sequence blocks can use the same Variables block.
In the Variables block, you create a new variable, and give it a name and value.
In the sequence instruction, you use a $ + the variable name instead of an argument value.
The variable type (block link, temperature, duration, etc) must match the argument type.
An example use case for variables is to have a generic sequence that you don’t change often, and a variable block that you update for each batch, with variables for setpoint settings and durations.
Example syntax:
SET_SETPOINT target=$setpoint_block, setting=$beer_temp
The bad news is that the variables block is not available on the Spark 2 and 3.
We were unable to make the implementation small enough to fit the firmware partition.
brewblox.yml
In Brewblox, we want to maintain a balance between configurability and ease of use.
brewblox-ctl install creates configuration files, and if you want to, you can override them.
Environment variables were used to declare simple settings, such as the published HTTP / HTTPS ports.
This has its drawbacks. Some configuration could not be partially overridden, or would be reset during updates.
To make it easier to apply specific and common configuration changes, we reworked how configuration files are handled.
brewblox-ctl now stores its own settings in a dedicated file: brewblox.yml.
Other configuration files are no longer static, but are generated based on the settings in brewblox.yml.
For example, you can now disable TLS termination without having to copy all traefik service settings in your own docker-compose.yml.
Configuration files are generated during updates, and when using the brewblox-ctl config apply command.
This includes the .env file. Any manual edits to .env will be lost when configuration is generated.
If you want to include custom environment variables in .env, you can use the environment section in brewblox.yml.
When updating to this release, existing values in .env will not be discarded, but placed in the newly created brewblox.yml.
To see all available settings, run brewblox-ctl config inspect.
Changes
- (feature) Added the Variables block.
- (feature) Sequence instructions can reference named variables defined in a linked Variables block.
- (feature) The Sequence block behavior on startup can now be configured to start/pause/resume at first/current instruction.
- (feature) Brewblox-ctl now uses a central
brewblox.ymlconfig file to generate other configuration files. - (feature) TLS termination in traefik can now be disabled.
- (improve) The Digital Input block now has better detection for Open Load errors.
- (improve) Actuator Soft Start now resets on overcurrent error.
- (fix) Fixed the Spark service page not updating the document title.
- (deprecate) Brewblox no longer supports Python 3.7 / Debian Buster.
