Backup to Google Drive

Logo

Overview

This add-on will upload files from your hass[]().io backup folder (typically .tar files created by the hass[]().io SnapShot feature) to your Google Drive. A few key things to note:

  1. This add-on asks only for permission to add new files to your Google Drive, and to manage the files that it adds. It will not have permission to view any files on your Google Drive that it did not create itself. This is to protect the contents of your Google Drive.
  2. This add-on exposes a Web User Interface for obtaining your authorization to upload files to your Google Drive. Once that authorization is established, the Web User Interface is no longer needed - it exists only for this initial setup.
  3. Backups are performed by executing a REST service exposed on the same port as the Web User Interface. Find more information on the doBackup service operation below. You can use Home Assistant's RESTful Command to integrate this add-on's REST service opertion into your own scripts and automations. You may want to use a REST testing tool like Postman to perform initial testing.
  4. Optionally, you may configure this add-on to purge older files from your hass[]().io backup folder.
  5. Optionally, you may also configure this add-on to purge older files from the Google Drive folder that your are archiving your snapshots to.

Installation

You can install this hass[]().io add-on using my add-on repository following these Home Assistant Add-on Instructions.

Configuration Options

Example configuration:

{"fromPattern" : "/backup/*.tar",
 "backupDirID" : "1CvPDzNz1v-OuOUqKq3jjoKQt020hKK7R",
 "purge" : {"enabled" : true, "preserve" : 3},
 "purge_google" : {"enabled" : true, "preserve" : 12},
 "debug" : false}

fromPattern

Use this to identify the files on your hass[]().io host that you wish to backup.

This pattern is used to identify a list of files to backup. That list is then pared down by checking Google Drive to see if any files in the list have already been backed up by this add-on. This check is performed to avoid backing up the same file twice (Google Drive allows duplicate files with the same name).

Note that this add-on can only see files on your Google Drive that it created itself. Therefore, if you have backed up some of your snapshots on your own to Google Drive, this add-on will not be aware of those and it will back them up anyway.

backupDirID

This identifies the Google Drive folder in which you want to place your backed up files. Because this add-on does not have permission to browse any files or directories on your Google Drive that it does not itself create, you cannot simply provide the folder name. You must instead provide Google Drive's unique opaque ID of the folder. Google Drive doesn't make it easy to get this value. But, here's how you can get it:

  1. In your favorite web browser, navigate to the Google Drive folder to which you plan to upload your files (create a new folder in Google Drive if you wish). Be sure that you have the perferred folder open so that its contents (even if it's empty) are displayed.
  2. From the address bar of your browser, copy the last portion of the URL. That value is Google Drive's unique opaque ID for the folder. Paste that value in for the backupDirID value in the configuration.

purge

This configures the option to purge (delete) older files from your source location (e.g. your /backup folder on hass[]().io). There two sub-elements:

purge_google

Contrary to its ominous sounding name, this does not purge every file from your Google Drive ;-). This add-on can see only files that it creates on your Google Drive. It can also see the folder that you identify for it to place backup files into, but it cannot see files inside that folder unless this add-on created them itself. This option configures the feature to purge (delete) older files from your Google Drive folder (the one you identify in the backupDirID option). There two sub-elements:

Important notes about this option:

  1. This permanently deletes the selected files from your Google Drive, bypassing your Google Drive Trash. The idea is to free up available storage on your Google Drive.
  2. It only considers files in the currently configured backupDirID folder on Google Drive.
  3. It only considers files created by this add-on (because those are the only files it can see).
  4. It does not consider the fromPattern setting at all.

debug

Defaults to false if not present. Set this to true to enable debug-level logging.

Authorizing this Add-On to Upload to Google Drive

This add-on requires you to authorize it to a limited scope of access to your Google Drive. The specific scope it requires is https://www.googleapis.com/auth/drive.file. You can read information about what that scope entails in Google's Guide to OAuth 2.0 Scopes. Essentially, it allows this add-on to view and manage Google Drive files and folders that you have opened or created with this add-on.

Before using the doBackup service operation, you need to follow these steps to grant this add-on the required authorization:

  1. Start this add-on and make sure that it is set to Start on boot.
  2. Open the Web User Interface for this add-on using the Open Web UI link on the add-on details page. Alternatively, you can open your own browser window and navigate to: http://<Your_Hassio_Host>:<Host_Port>/gb, substituting your hass[]().io host name and the Host Port number you've configured for this add-on.
  3. Following the instructions in the Web User Interface, do the following:
    1. Click on the AUTHORIZE button to launch a separate browser tab to the Google Authorization Server.
    2. If required, login to Google and confirm your Google user ID.
    3. Google will then tell you what application (this add-on) is requesting authority and what scope of authority is being requested. Click Google's Allow button.
    4. Google will then show you an authorization code. Copy this code so that you can paste into this add-on's Web User Interface (next step).
    5. Return to the browser tab containing this add-on's Web User Interface and paste the copied value (from the previous step) into the provided `Authorization Code' field.
    6. Click the INGEST CODE button.
    7. You should be presented with a message indicating authorization received.
  4. This completes the authorization step! You're now ready to begin using the doBackup Service as described below.

Calling the doBackup Operation

Backups are performed by calling the doBackup service operation exposed by this add-on. When you start this add-on, the service becomes available on the Host Port that you've configured this add-on to use.

The doBackup service operation does not require any arguments. It gets the information it needs from the Configuration Options and from your having completed the authorization process described above.

You call the doBackup service operation by simply performing a GET against this URI (in fact, you can just click this link):

    http://<YOUR_HASSIO_HOST>:<HOST_PORT>/gb/doBackup

Substitute in your hass[]().io host name (usually hassio.local) and the Host Port number you've configured for this add-on.

The doBackup service operation will respond with JSON reminding you of the configuration settings that it used and indicating:

Sample JSON Response

{
    "backupTimestamp": "2018-11-26T09:57:36.206259",
    "fromPattern": "/backup/*.tar",
    "backupDirID": "1CvPDzNz1v-OuOUqKq3jjoKQt020hKK7R",
    "fileCount": 5,
    "alreadyCount": 2,
    "backedUpCount": 3,
    "deletedCount": 2,
    "deletedFromGoogle": 1
}

Unexpected errors will return an HTTP Status Code of some value other than the normal 200 Success Code.

Calling the adhocBackup Operation

As a completely separate feature, adhoc backups may be performed. These backups do not use the configured options fromPattern, backupDirID, purge and purge_google. Instead, an adhoc backup request identifies which files are to be backed up and which Google Drive folder is to be targeted each time it is placed (hence the name, "adhoc"). The concept of "purging" of older files while preserving more recent files does not apply at all to adhoc backups. Adhoc backups simply copy each identified file from your hass[]().io host to your Google Drive, replacing the target file on Google Drive if it already exists. Adhoc backups are performed by calling the adhocBackup service operation exposed by this add-on. When you start this add-on, the service becomes available on the Host Port that you've configured this add-on to use.

The adhocBackup service operation has two required arguments that must be provided in the form of a JSON body on the HTTP POST.

WARNING: It is recommended that you do NOT specify the same target Google Drive folder for adhoc backups that you have specified in the add-on options configuration for the normal backup operation, especially if you are using the purge_google feature.

You call the adhocBackup service operation by performing a POST against this URI:

    http://<YOUR_HASSIO_HOST>:<HOST_PORT>/gb/adhocBackup

Substitute in your hass[]().io host name (usually hassio.local) and the Host Port number you've configured for this add-on.

Sample JSON Request

{"fromPatterns" : ["/config/configuration.yaml", "/config/automations.yaml"],
 "backupDirID" : "4FtLMzNz1v-OuOUqKq3jjoKQt020hLL9P"}

The adhocBackup service operation will respond with JSON indicating:

Sample JSON Response

{
    "adhocBackupTimestamp": "2019-03-15T07:59:56.205540",
    "fromPatterns": [
        "/config/configuration.yaml",
        "/config/automations.yaml"
    ],
    "backupDirID": "4FtLMzNz1v-OuOUqKq3jjoKQt020hLL9P",
    "copyCount": 2,
    "newCount": 0,
    "replacedCount": 2
}

Unexpected errors will return an HTTP Status Code of some value other than the normal 200 Success Code.

Integrating into Home Assistant

Automating Backups with Home Assistant

You can easily integrate this add-on's doBackup REST service operation into Home Assistant using Home Assistant's RESTful Command. You'll probably need to use localhost instead of hassio.local in this configuration. You'll also want to specify an adequate timeout value. Here's how I setup mine:

rest_command:
  google_backup:
    url: 'http://localhost:8055/gb/doBackup'
    timeout: '300'

With this REST Command created, you'll see your Google Backup Service available as rest_command.google_backup in Home Assistant's Services Development Tool, and you'll also be able to call it as part of Home Assistant Automations.

Because the adhocBackup operation requires a JSON array, I've not been able to figure out how to configure a REST Command for it in Home Assistant. If you figure it out, let me know. For now, you can call the adhocBackup operation from tools like Postman. You'll want to setup your tool do a POST method, passing in JSON (Content-Type of appliction/json). Here's an example JSON request: {"fromPatterns" : ["/config/configuration.yaml", "/config/automations.yaml"], "backupDirID" : "4FtLMzNz1v-OuOUqKq3jjoKQt020hLL9P"}.

MQTT Event

Everytime the doBackup operation, or the adhocBackup operation, is executed, an event is published to the MQTT broker that you have configured in Home Assistant. The payload of the event is a copy of the JSON response described above. The event is published to either the googlebackup/result or the googlebackup/adhocresult topic, depending on which operation was called.

Here is an example Home Assistant Sensor configuration that monitors for these events:

  - platform: mqtt
    name: "Google Backup Results"
    state_topic: "googlebackup/result"
    json_attributes_topic: "googlebackup/result"
  - platform: mqtt
    name: "Google Adhoc Backup Results"
    state_topic: "googlebackup/adhocresult"
    json_attributes_topic: "googlebackup/adhocresult"