This page documents known limitations, design constraints, and platform-dependent behaviours within the Squirrel system.

These items reflect the realities of operating within Microsoft 365, SharePoint Online, and Azure, and are provided to support customer administrators, compliance teams, and operational owners in designing, validating, and governing Squirrel deployments.

This page documents expected behaviours, dependencies, and constraints when operating Squirrel.

Squirrel processes SharePoint changes such as file creation, modification, movement, and archival asynchronously.

As a result:

• Changes are typically processed overnight.
• A processing window of up to 24–48 hours should be assumed.
• Some actions (such as stub movement detection) rely on scheduled change processing rather than real-time events.

Administrators should allow sufficient time for SharePoint changes to be detected and reflected in Squirrel before initiating dependent operations such as restores.

Stub files are standard SharePoint files and can be moved or renamed by users. Stub movement is tracked on a best-effort basis.

Important considerations:

• Stub movement tracking is not guaranteed.
• Location updates typically appear within 24 hours, depending on processing windows.
• If a stub is moved and immediately restored before processing completes, the file will restore to its original archived location.
• In this scenario, the moved stub may remain and require manual deletion.

If a stub file is moved between SharePoint sites, Squirrel attempts to detect this and retarget the stub.

Because this behaviour depends on SharePoint change processing:

• Cross-site stub tracking is best-effort.
• Successful retargeting is not guaranteed.
• In some cases, restored files may return to their original site and path.

For environments where restore location must be strictly controlled, moving stubs between sites is not recommended.

Squirrel is designed to preserve and restore version history where supported by SharePoint.

However:

• Files with more than 20,000 versions are not formally supported.
• Operations beyond this threshold may still succeed but are not guaranteed.

Administrators managing environments with extremely high version counts should validate behaviour with representative test files.

Squirrel is designed to capture and restore standard SharePoint metadata and commonly used custom columns where supported.

Field types Squirrel attempts to capture include:

• Single line of text
• Multiple lines of text
• Choice
• Number
• Date and Time
• Person or Group
• Currency

Because SharePoint supports a wide range of column types and custom implementations:

• Not all field types are supported for full-fidelity capture and restore.
• Complex or special-purpose fields (for example images, maps, managed metadata, lookup fields, or highly customised fields) may not be fully captured or rehydrated.

Author and Editor metadata fields must contain valid email addresses corresponding to existing directory accounts.

If these fields use:

• Display names
• Aliases
• Non-email identifiers
• Deleted or unresolved accounts

then SharePoint may not accept the value, and the metadata may not be applied to the file, stub, or restored versions.

This behaviour is governed by Microsoft Entra ID and SharePoint identity resolution.

Squirrel attempts to apply original created and modified timestamps where supported.

Because Microsoft 365 stores timestamps internally in UTC and displays them based on tenant and user locale settings:

• Time values may appear shifted after restore.
• Display formats may differ across SharePoint views.
• System timestamps associated with the restore operation may be visible.

Timestamp behaviour remains subject to Microsoft platform handling.

For files larger than 200 MB, content confirmation cannot be performed using MD5 hashing.

In these scenarios, validation is performed using alternative mechanisms such as file size comparison.

Microsoft Information Protection (MIP) labels can only be reapplied to files with non-protected sensitivity labels.

Files with protected sensitivity labels will be restored to their most recent accessible version, with additional historical versions included in a ZIP file.

Sites with legal hold enabled may still be archived if they are not explicitly excluded from Squirrel processing.

Legal hold status is not automatically enforced by Squirrel and must be managed through site selection and policy configuration.

Subsites cannot be individually selected for inclusion or exclusion.

Only primary site scopes can be configured.

Squirrel employs elliptic-curve cryptography using Curve25519 for secure key exchange, combined with AES-256 symmetric encryption for data protection.

This hybrid approach is designed to provide strong security, high performance, and resistance to common cryptographic attacks.

As with all cryptographic systems, long-term risk is influenced by advances in computing, cryptographic research, and operational practices.

If you require guidance on how these constraints apply to your environment, or need assistance designing operational controls, contact support@smikar.com.