by

Backblaze Object Storage



Backblaze B2 Cloud Storage vs Google Cloud Storage: What are the differences? Developers describe Backblaze B2 Cloud Storage as 'A Cloud Storage service'. It is a generic cloud storage. You can use it from any computer, any operating system or any cloud storage purpose. You can manage and store large amounts of media files using this service. Innovative Companies Build on B2 Cloud Storage Organizations around the world choose Backblaze to solve for their use cases while improving their cloud OpEx vs. Amazon S3 and others.

This guide will show you how to use Backblaze B2 as a database replica path forLitestream.

Create a bucket#

To begin, you’ll need a bucket to store your data. If you don’t already have oneset up in Backblaze, follow the steps below:

  1. From your Backblaze dashboard, go to My Account > Buckets
  2. Click “Create a Bucket”
  3. Enter a unique name for “Bucket Unique Name” and leave the remaining fields as their defaults.
  4. Click “Create a Bucket”
  5. Copy the bucket details to the appropriate places in your litestream.yml file:

Create a user#

Next, create a user with programmatic access to the bucket you just created.

  1. From your Backblaze dashboard, go to My Account > App Keys
  2. Click “Add Application Key”
  3. In “Name of Key”, enter any unique name
  4. Under “Allow access to Bucket(s)' select the bucket you created above.
  5. Leave the remaining fields as their defaults.
  6. Click “Create new key”
  7. Copy the credentials to the appropriate places in your litestream.yml file
    • The keyID field is your access-key-id
    • The applicationKey field is your secret-access-key

Configuration file#

Your final litestream.yml configuration file should look like this:

Backblaze is an online backup tool that allows Windows and Mac OS X users toback up their data to an offsite data center.

Backblaze B2 is their cloud object storage offering similar to Amazon S3 andother object storage services.

Instantiating a driver¶

To instantiate the driver you need to pass your key id and application key tothe driver constructor as shown below.

To access the credentials, you can login tohttps://secure.backblaze.com/user_signin.htm, then click “App Keys” or goto https://secure.backblaze.com/app_keys.htm directly.

keyID serves as the first and applicationKey as the second argument tothe driver constructor.

API Docs¶

class libcloud.storage.drivers.backblaze_b2.BackblazeB2StorageDriver(key, secret=None, secure=True, host=None, port=None, api_version=None, region=None, **kwargs)[source]
Parameters:
  • key (str) – API key or username to be used (required)
  • secret (str) – Secret password to be used (required)
  • secure (bool) – Whether to use HTTPS or HTTP. Note: Some providersonly support HTTPS, and it is on by default.
  • host (str) – Override hostname used for connections.
  • port (int) – Override port used for connections.
  • api_version (str) – Optional API version. Only used by driverswhich support multiple API versions.
  • region (str) – Optional driver region. Only used by drivers whichsupport multiple regions.
Return type:

None

connectionCls

alias of BackblazeB2Connection

create_container(container_name, ex_type='allPrivate')[source]

Create a new container.

Parameters:container_name (str) – Container name.
Returns:Container instance on success.
Return type:libcloud.storage.base.Container
delete_container(container)[source]

Delete a container.

Parameters:container (libcloud.storage.base.Container) – Container instance
Returns:True on success, False otherwise.
Return type:bool
delete_object(obj)[source]

Delete an object.

Parameters:obj (libcloud.storage.base.Object) – Object instance.
Returns:bool True on success.
Return type:bool
download_object(obj, destination_path, overwrite_existing=False, delete_on_failure=True)[source]

Download an object to the specified destination path.

Parameters:
  • obj (libcloud.storage.base.Object) – Object instance.
  • destination_path (str) – Full path to a file or a directory where theincoming file will be saved.
  • overwrite_existing (bool) – True to overwrite an existing file,defaults to False.
  • delete_on_failure (bool) – True to delete a partially downloaded file ifthe download was not successful (hashmismatch / file size).
Returns:

True if an object has been successfully downloaded, Falseotherwise.

Return type:

bool

download_object_as_stream(obj, chunk_size=None)[source]

Return a iterator which yields object data.

Parameters:
  • obj (libcloud.storage.base.Object) – Object instance
  • chunk_size (int) – Optional chunk size (in bytes).
Return type:

iterator of bytes

download_object_range(obj, destination_path, start_bytes, end_bytes=None, overwrite_existing=False, delete_on_failure=True)
Backblaze

Download part of an object.

Parameters:
  • obj (libcloud.storage.base.Object) – Object instance.
  • destination_path (str) – Full path to a file or a directory where theincoming file will be saved.
  • start_bytes (int) – Start byte offset (inclusive) for the rangedownload. Offset is 0 index based so the firstbyte in file file is “0”.
  • end_bytes (int) – End byte offset (non-inclusive) for the rangedownload. If not provided, it will default to theend of the file.
  • overwrite_existing (bool) – True to overwrite an existing file,defaults to False.
  • delete_on_failure (bool) – True to delete a partially downloaded file ifthe download was not successful (hashmismatch / file size).
Returns:

True if an object has been successfully downloaded, Falseotherwise.

Return type:

bool

download_object_range_as_stream(obj, start_bytes, end_bytes=None, chunk_size=None)

Return a iterator which yields range / part of the object data.

Parameters:
  • obj (libcloud.storage.base.Object) – Object instance
  • start_bytes (int) – Start byte offset (inclusive) for the rangedownload. Offset is 0 index based so the firstbyte in file file is “0”.
  • end_bytes (int) – End byte offset (non-inclusive) for the rangedownload. If not provided, it will default to theend of the file.
  • chunk_size (int) – Optional chunk size (in bytes).
Return type:

iterator of bytes

enable_container_cdn(container)

Enable container CDN.

Parameters:container (libcloud.storage.base.Container) – Container instance
Return type:bool
enable_object_cdn(obj)
Backblaze object storage

Enable object CDN.

Parameters:obj (libcloud.storage.base.Object) – Object instance
Return type:bool
ex_get_upload_data(container_id)[source]

Retrieve information used for uploading files (upload url, auth token,etc).

Rype:dict
ex_get_upload_url(container_id)[source]

Retrieve URL used for file uploads.

Return type:str
What is object storage
get_container(container_name)[source]

Return a container instance.

Parameters:container_name (str) – Container name.
Returns:Container instance.
Return type:libcloud.storage.base.Container
get_container_cdn_url(container)

Backblaze Cloud Storage

Return a container CDN URL.

Parameters:container (libcloud.storage.base.Container) – Container instance
Returns:A CDN URL for this container.
Return type:str
get_object(container_name, object_name)[source]

Return an object instance.

Storage
Parameters:
  • container_name (str) – Container name.
  • object_name (str) – Object name.
Returns:

Object instance.

Return type:
get_object_cdn_url(obj)

Return an object CDN URL.

Parameters:obj (libcloud.storage.base.Object) – Object instance
Returns:A CDN URL for this object.
Return type:str
iterate_container_objects(container, prefix=None, ex_prefix=None)[source]

Return a generator of objects for the given container.

Parameters:
  • container (Container) – Container instance
  • prefix (str) – Filter objects starting with a prefix.Filtering is performed client-side.
  • ex_prefix (str) – (Deprecated.) Filter objects starting with a prefix.Filtering is performed client-side.
Returns:

A generator of Object instances.

Return type:

generator of Object

iterate_containers()[source]

Return a iterator of containers for the given account

Returns:A iterator of Container instances.
Return type:iterator of libcloud.storage.base.Container
list_container_objects(container, prefix=None, ex_prefix=None)

Return a list of objects for the given container.

Parameters:
  • container (libcloud.storage.base.Container) – Container instance.
  • prefix (str) – Filter objects starting with a prefix.
  • ex_prefix (str) – (Deprecated.) Filter objects starting with a prefix.
Returns:

A list of Object instances.

Return type:

list of libcloud.storage.base.Object

list_containers()

Return a list of containers.

Returns:A list of Container instances.
Return type:list of Container
upload_object(file_path, container, object_name, extra=None, verify_hash=True, headers=None)[source]

What Is Object Storage

Upload an object.

Note: This will override file with a same name if it already exists.

upload_object_via_stream(iterator, container, object_name, extra=None, headers=None)[source]

Upload an object.

Backblaze Object Storage Device

Note: Backblaze does not yet support uploading via stream,so this calls upload_object internally requiring the object datato be loaded into memory at once