Monitoring API

[Kaiede]

It would be useful to be able to monitor the backup service directly using a tool like Dashy or Homepage. It would also provide a way for docker health checks to directly query the service rather than looking for a state file that the service writes to disk. For example, pulling health information could look like this:

GET http://backup-service/live

RESPONSE: either 200 or 503
{}

GET http://backup-service/status

RESPONSE: 200
{
  lastBackup: ‘<lastBackupDate>’,
  lastBackupState ‘<success/failed>’,
  lastBackupSizes: {
    ‘yosemite’: 12345,
    ‘colorado’: 56789
  },
  lastBackupTotalSize: 69134,
  storageTotalUsed: <size of all backups>
}

This could also be potentially used for other things, such as registering for notifications via HTTP callbacks. For example, if I create a container that has a python microservice, part of it’s startup would be to PUT a request to the backup service:

PUT http://backup-service/listen
{
  callback: ‘http://python-service/backupEvent’,
  events: [‘backup-failed’, ‘backup-complete’]
}

And the callback would look like

POST http://python-service/backupEvent
{
  event: ‘backup-complete’
}