Skip to content

Synchronizing records from device dictionary

LAVA has the ability to synchronize database records with the device dictionary. It currenty supports these models:

  • Device
  • Device type
  • Tag
  • Alias

This can make administration less cumbersome and can help with Ansible and similar setups. By using this feature administrators are able to keep the list of these records inside the version control.

For device records, a flag named is_synced is used to recognize devices which are synced to/from the device dictionary. This option can be updated via usual channels (web UI admin, APIs).

Once the new device dictionary is added to the filesystem it needs to include the following dictionary variable which will control which records should be created:

{% set sync_to_lava = {
   "device_type": "qemu",
   "worker": "worker-01"
   "tags": ["tag1", "tag2"],
   "aliases": ["alias1", "alias2"],
   }
%}

A management command is used to perform a sync to the database:

shell tab="debian" lava-server manage sync

You can for example set this to run when the contents of the device dictionary folder change using a watch utility.

In the above example, a system will create:

  • device named after the device dictionary file
  • device type named qemu (if it doesn't exist already) and relate this device to it
  • worker named worker-01 (if it doesn't exist already) and relate this device to it
  • aliases in the list; and relate them to the specified device type
  • tags in the list; and relate them to the device

If at any point the device dictionary is removed (and once the sync command is executed) the system will move the device that was created to it to Retired health. If all the devices of the synced device type are Retired that particular device type will be marked as invisible in the system.