Deploying Containers
Kanto-auto-deployer (KAD)
This service has been implemented as a stopgap solution by the Leda team since as of now the Kanto CM-native mechanism for initial container deployment does not suit the needs of the Leda Quickstart image well.
In-depth documentation on its operation and development can be found in Kanto Auto deployer.
In the Leda Distro KAD runs as a service that continuously monitors the directory defined with the BitBake variable KANTO_MANIFESTS_DEV_DIR
in the distro config recipes.
By default, the Leda quickstart images use KANTO_MANIFESTS_DEV_DIR=/data/var/containers/manifests"
.
So, these manifests are again stored in the persistent data-partition and can be modified after the image has been deployed.
Important: KAD supports the Kanto Container Management Manifest Template AND the Container Internal State Representation. For future compatiblity reasons it is recommended that you choose Kanto Container Management Manifest Template.
A standard set of containers is deployed through meta-leda, the manifests for which can be found in meta-leda/meta-leda-components/recipes-sdv/eclipse-leda/kanto-containers/core
and meta-leda/meta-leda-components/recipes-sdv/eclipse-leda/kanto-containers/example
.
The advantage of this service is that it can be restarted very quickly (systemctl restart kanto-auto-deployer
), without having to restart the whole container-management service. The implemented filewatcher (the --daemon
flag) allows you to quickly create/edit (or even touch
them) container manifests on the device that would be on-the-fly deployed after saving. This allows for rapid testing when creating new container manifests.
Kanto Container Management init-dir
This feature of Kanto Container Managament by the Kanto Teams is currently under development and hence unstable. That is why it is not used as of the test-0.0.6 release of Leda Distro and 0.1.0-M1 release of meta-leda.
Feedback
Was this page helpful?
Glad to hear it! Please tell us how we can improve.
Sorry to hear that. Please tell us how we can improve.