[HDDS-1649] On installSnapshot notification from OM leader, download checkpoint and reload OM state

• Type: Sub-task
• Status: Resolved
• Priority: Major
• Resolution: Fixed
• Fix Version/s: 0.5.0
• Component/s: OM HA, Ozone Manager
• Labels: pull-request-available
• Assignee: Hanisha Koneru
• Reporter: Hanisha Koneru
• Created: 05/Jun/19 18:22
• Updated: 06/Dec/20 22:55
• Resolved: 22/Jul/19 19:07

Description

Installing a DB checkpoint on the OM involves following steps:
1. When an OM follower receives installSnapshot notification from OM leader, it should initiate a new checkpoint on the OM leader and download that checkpoint through Http.
2. After downloading the checkpoint, the StateMachine must be paused so that the old OM DB can be replaced with the new downloaded checkpoint.
3. The OM should be reloaded with the new state . All the services having a dependency on the OM DB (such as MetadataManager, KeyManager etc.) must be re-initialized/ restarted.
4. Once the OM is ready with the new state, the state machine must be unpaused to resume participating in the Ratis ring.

Issue Links

• is related to: HDDS-3208 Implement Ratis Snapshots on SCM
• is required by: HDDS-1829 On OM reload/restart OmMetrics#numKeys should be updated
• is required by: HDDS-1831 Use annotation based configuration for OZONE_OM_RATIS_LOG_PURGE_GAP
• links to: GitHub Pull Request #948