Best Practices for Cloning and Restoring (Backup My Org)

Owned by Kenny Tran
Last updated: Sept 20, 2024
3 min read

Note: Formatting of the spreadsheet data is normally completed through Microsoft Excel. However, it can also be completed via Google Sheets, by uploading/importing the backup/restore .csv file from the local device where it is stored. An example for the restore planning template can be found here:

  1. Located in the Backup Logs, select the ‘View’ button for the restore that you wish to export, and click the ‘Export’ button which will download the backup .csv file to your local device.

  2. Begin formatting the csv. file by removing the following columns:

    1. localItemInfoID,

    2. localItemInfoSize,

    3. localItemDataID,

    4. localItemDataSize,

    5. localItemFGDBID, and

    6. localItemFGDBDataSize,

    7. and rename the status column as backup status.

  3. In your spreadsheet editor, select the freeze tool and lock in the column headers, or first row.

  4. Add the following columns:

    1. restore step (priority),

    2. restore substep (more relevant for larger restores of 50 or more),

    3. restore operator (if multiple people are going to be doing restores),

    4. restore status,

    5. restore errors,

    6. restore id,

    7. restore url,

    8. testing operator (if multiple people are going to be testing restored content),

    9. testing status, and

    10. testing errors

  5. Assign numeric values to the restore steps column of all content you want to restore. Each row is assigned a numeric value that’s based on the level of dependency among the collection of items, groups, and users in the log. Entities classified with 1 has very little to no dependencies to other items in your log collection and increasing the value shows the more dependent an item is to other entities. In the example list below, we’ll assume that there is a collection of ArcGIS types present in a backup log. They are assigned values to distinguish the level of dependency.

 

1 = data items (users and groups will proceed this if present in backup log), 2= layers (hosted), 3 = layers & views (registered), 4 = maps, scenes, etc. 5 and above = applications and projects

 

Based on your collection of backups, the assignment of values may differ from the example above. If ArcGIS users and groups are included in the backup log, they would be classified as 1 for users, 2 for groups, and 3 and up for items based on their level of dependency. When planning your template each item type should ideally have its own restore step based on the priority. Further example is shown in the workflow spreadsheet (found in the attachment link above or below the article).

  1. Under the type columns, further categorization of what the item’s type is expanded in this column.

  2. If restoring more than 50 of any given content type, it is ideal to break out the collection with the substeps column with each collection holding 50 at a time. The amount of content in each substep can be more or less depending on your comfort level and item type restore complexity with Backup My Org. However, we would not recommend results more than 100 at a time in the restore log. Splitting the restore items into smaller substeps helps for better manageability.

  3. Postprocessing Rewire:

    1. Find an item with an old id in source org (or planning .csv file),

    2. Find corresponding item in destination org (or planning document restore id column),

    3. Repeat step for URLs.

  4. Rewire relationships:

    1. Find an item with an old id in source org (or planning .csv),

    2. Find corresponding item in destination org (or planning document restore id column),

    3. If the item does not yet exist, this can be left blank.

  5. Under the Testing Errors column, be sure to record any errors while completing a spot checking test, preferably for more complex items, OR individual items.

Attachment: