[CLOUDSTACK-10222] Clean previous snaphosts from primary storage when taking new one

• Type: Bug
• Status: Resolved
• Priority: Major
• Resolution: Fixed
• Affects Version/s: 4.10.0.0, 4.11.0.0
• Fix Version/s: 4.11.1.1
• Component/s: Secondary Storage, Snapshot, XenServer
• Security Level: Public (Anyone can view this level - this is the default.)
• Labels: easyfix
• Environment: XenServer, Swift
• Assignee: Unassigned
• Reporter: Khosrow Moossavi
• Created: 09/Jan/18 17:08
• Updated: 09/Feb/18 10:54
• Resolved: 06/Feb/18 19:38

Description

When user creates a snapshot (manual or recurring), snapshot remains on the primary storage, even if the snapshot is transferred successfully to secondary storage. This is causing issues because XenServer can only hold a limited number of snapshots in its VDI chain, preventing the user from creating new snapshots after some time, when too many old snapshots are present on the primary storage.

reason: SR_BACKEND_FAILURE_109 The snapshot chain is too long

Proposed solution:
Keep only one snapshot (last successful) on the primary storage regardless that it was a full or partial snapshot.

Issue Links

• links to: GitHub Pull Request #2398