Resolve extreme scale from large manifest files

Overview

When the CrashPlan app scans for files, synchronizes the archive, and restores files, it refers to a set of manifest files containing a reference of your backed-up file paths and file versions. If these manifest files become too large, they will interfere with your ability to back up and restore files.

This article describes how extreme scale of the manifest files interferes with your backup and restore processes, what causes extreme scale, and what you can do to resolve these issues.

Determine extreme scale

Manifest files are considered extreme scale when they are above a certain healthy size.

CrashPlan stores local copies of the manifest files in its cache folder. The location of the cache folder is set by your operating system. For more information, see CrashPlan File and Folder Hierarchy.

To determine whether your manifest files are unhealthy, follow these steps:

1. Navigate to the CrashPlan cache folder for your operating system.
2. Inside the cache folder, open the folder named a series of numbers.
   This folder name reflects the GUID of your backup archive.
3. Find the cpfmf and cphdf files and make note of their file sizes.
4. If either file is over 500 megabytes, you have an extreme scale issue.

Causes of extreme scale

The amount of files being backed up affects the size of the file manifest, and the amount of versions being retained affects the size of the history data file.

Backup archives can reach extreme scale due to large manifest files as a result of:

• Frequency and versioning settings that back up too many files and versions.
• Selecting the root of your operating system drive for backup.
• Altering or removing our recommended global exclusions.
• Backing up live databases, system and application files, and virtual machines.
• Backing up excessive amounts of data from a file share or RAID.

Troubleshooting extreme scale

You can avoid issues with extreme scale by using our default settings.

Once the file manifest and history data file reach extreme scale, you must alter their size by updating your file selection, updating your frequency and version settings, and following our best practices.

• Refer to our best practices for backup settings and what files to back up.
• Change your file selection to reduce the amount of files backed up.
  • Back up your user folder instead of your entire system drive.
  • Limit the total amount of data you back up to only your most valuable files.
• Update your frequency and version settings to limit the amount of versions kept.
  • Reduce the frequency of backups (from 30 minutes to 1 hour for example.)
  • Reduce the number of versions being kept over time.
• Follow our best practices for backing up known problematic file types:
  • Back up open files and databases
  • Back up virtual machines and separate boot partitions
  • Back up applications

Once you have made appropriate changes to your backup and versioning settings, you must put the archive into maintenance for the changes to take effect. You can do this from the CrashPlan console:

1. Sign in to the CrashPlan console. 
2. Select Administration > Environment > Devices from the menu. 
3. On the Active tab, select the device for which you want to start archive maintenance.
4. From the device details page, select the Archive Options icon .
5. Select Maintain Archive.
   A message informs you that running archive maintenance may cause the archive to become unavailable for restore or backup until maintenance finishes. 
6. Click OK.
   Your job is added to the queue.

When the archive maintenance is complete, your backup will synchronize and resume. Check the size of the cpfmf and cphdf to confirm success. They should each be under 500 megabytes. If they are not, you will have to make stricter changes to your file selection and versions and try again.

If you continue to have issues, contact our technical support.