Skip to content
GitLab
Projects Groups Snippets
  • /
  • Help
    • Help
    • Support
    • Community forum
    • Submit feedback
    • Contribute to GitLab
  • Sign in / Register
  • Kea Kea
  • Project information
    • Project information
    • Activity
    • Labels
    • Members
  • Repository
    • Repository
    • Files
    • Commits
    • Branches
    • Tags
    • Contributors
    • Graph
    • Compare
  • Issues 557
    • Issues 557
    • List
    • Boards
    • Service Desk
    • Milestones
  • Merge requests 65
    • Merge requests 65
  • Deployments
    • Deployments
    • Releases
  • Packages and registries
    • Packages and registries
    • Container Registry
  • Monitor
    • Monitor
    • Incidents
  • Analytics
    • Analytics
    • Value stream
    • Repository
  • Wiki
    • Wiki
  • Snippets
    • Snippets
  • Activity
  • Graph
  • Create a new issue
  • Commits
  • Issue Boards
Collapse sidebar
  • ISC Open Source ProjectsISC Open Source Projects
  • KeaKea
  • Issues
  • #1959
Closed
Open
Issue created Jun 30, 2021 by Marcin Siodelski@marcinDeveloper

Possible skipped standby server synchronization after temporary communication interruption

Consider the following scenario. There are two HA partners in the hot-standby mode. Communication between them is interrupted but they still run. The primary transitions to the partner-down state quickly because it does not monitor the DHCP traffic directed to standby. The standby is not supposed to respond to any traffic. The primary gets to the partner-down state and the standby is still in hot-standby state. When the communication is re-established, it is possible that the primary first spots that the standby is in the hot-standby state. In that case, the primary may transition to the hot-standby state. The standby would see the primary in the hot-standby state and would assume that they operate normally. As a result, the standby won't synchronize its lease database, and will lose the leases allocated by the primary in the partner-down state.

The proposed solution is to modify the state machine, so that the primary remains in the partner-down state when he spots that the standby is in the hot-standby state. This should force the standby to transition to the waiting state and synchronize.

To upload designs, you'll need to enable LFS and have an admin enable hashed storage. More information
Assignee
Assign to
Time tracking