Uploaded image for project: 'Core Server'
  1. Core Server
  2. SERVER-78496

Clear shard filtering metadata before releasing the critical section in rename participants

    • Type: Icon: Bug Bug
    • Resolution: Fixed
    • Priority: Icon: Major - P3 Major - P3
    • 7.1.0-rc0, 7.0.0-rc8, 6.0.10
    • Affects Version/s: 6.0.7, 7.0.0-rc5
    • Component/s: Sharding
    • Labels:
    • Fully Compatible
    • ALL
    • v7.0, v6.0
    • Sharding EMEA 2023-07-10, Sharding EMEA 2023-07-24

      In v5.0 there was a routing table refresh+wait for flush right before releasing the critical section in rename participants. That code snippet (or something similar) should be reintroduced because it is a way to ensure that both primary and secondaries will update their filtering info for source and target namespaces, properly reacting when a stale router tries to target them.

      Instead of simply refreshing, a better way would be to force a metadata refresh + setting the new filtering metadata as currently done by the collmod participant.

            sergi.mateo-bellido@mongodb.com Sergi Mateo Bellido
            pierlauro.sciarelli@mongodb.com Pierlauro Sciarelli
            0 Vote for this issue
            4 Start watching this issue