Uploaded image for project: 'Core Server'
  1. Core Server
  2. SERVER-71278

Remove retry for SnapshotUnavailable error in global indexes code

    • Type: Icon: Task Task
    • Resolution: Fixed
    • Priority: Icon: Major - P3 Major - P3
    • 7.1.0-rc0
    • Affects Version/s: None
    • Component/s: None
    • Storage Execution
    • Fully Compatible
    • Execution Team 2023-05-29

      Remove the sleep in the catalog cache added by SERVER-71168, designed to allow time for the config time to advance beyond the creation time of config.csrs.indexes.

      Context:

      During FCV upgrade when enabling theĀ 
      gGlobalIndexesShardingCatalog feature flag, we are creating a new collection on the config server - config.csrs.indexes. The mongoS is also running an aggregation on this collection with snapshot read concern. Since the mongoS has no concept of FCV, it will start sending the aggregation to the config server as soon as the binary is changed to one with the feature flag enabled, regardless of the FCV of the config server. We have encountered the SnapshotUnavailable error after creating the collection and its indexes but before the journal flush happens.
      Once this project finishes, presumably the aggregation will return NamespaceNotFound at the previous config time sent in the aggregation rather than SnapshotUnavailable, and the sleep to allow for journaling to complete can be removed.

            Assignee:
            gregory.wlodarek@mongodb.com Gregory Wlodarek
            Reporter:
            allison.easton@mongodb.com Allison Easton
            Votes:
            0 Vote for this issue
            Watchers:
            4 Start watching this issue

              Created:
              Updated:
              Resolved: