Uploaded image for project: 'Core Server'
  1. Core Server
  2. SERVER-69813

Get rid of temporary solution for stuck secondaries

    • Type: Icon: Task Task
    • Resolution: Unresolved
    • Priority: Icon: Major - P3 Major - P3
    • None
    • Affects Version/s: None
    • Component/s: None
    • Labels:
      None
    • Catalog and Routing

      Starting from 4.0 version, a temporary solution has been implemented to resolve the problem described in SERVER-42737, i.e., the catalog refresh on secondaries stuck waiting for a notification which not arrive (see this analysis for further details).

      The original idea was to get rid of this solution after implementing the catalog loader writes with transactions (see SERVER-44105, closed as "Won't Do"). Nonetheless, the current plan is to remove the ShardServerCatalogLoader (see PM-2948), which should allow to get rid of the temporary solution mentioned above.

            Assignee:
            backlog-server-catalog-and-routing [DO NOT USE] Backlog - Catalog and Routing
            Reporter:
            antonio.fuschetto@mongodb.com Antonio Fuschetto
            Votes:
            0 Vote for this issue
            Watchers:
            2 Start watching this issue

              Created:
              Updated: