Scripts to upload, download, restore, or reconfigure arbor instances based on Kitware's girder & Flow technologies
One script (arbor_full_collection_backup.py) allows an Arbor user to create a local backup copy of all the datasets and analyses uploaded in Arbor instance. A second script (arbor_restore_collections.py) uploads all data and analyses stored in a local backup onto an Arbor instance.
Since Arbor includes a password-authenticated datastore, a valid user login is required to execute backup or restore jobs. During a backup operation, only collections visible to the chosen login credentials will be included in the backup. Therefore, administrative level user authentication is recommended to perform complete backups and re-initializations of Arbor instances.
python arbor_full_collections_backup.py -a "http://localhost:9080" -u user -p password
The above command will backup the Arbor available at the URL (http://localhost:9080) using the authentication credentials (user, password). A backup will be written in the default locaton ($HOME/arbor/backups). The backup will contain the URL of the Arbor instance andthe date and time of the backup embedded in the name to give unique names to each backup.
python arbor_restore_collections.py -a "http://localhost:9080" -u user -p password -d "/home/username/arbor/backups/arbor_backup_52.204.9.236_2016-11-30_19:43:49"
The above command will estore all the collections and datasets from a previously saved collection archived (stored at the /home/username/arbor.... URL). This does not delete existing collections, the script only adds any new collections contained in the backup and/or adds datasets and analyses to an already existing collection which were present in the backup.
python arbor_full_collections_backup.py --help
python arbor_restore_collections.py --help
Explain the command line options available for either script.