apache/cassandra · warning

Locations differ: {} != {}

Error message

Locations differ: {} != {}

What it means

WARN log from Directory.dumpDiff(Directory) indicating the locations maps (NodeId to DataCenter/location placement) differ between two Directory snapshots. Placement data divergence means the two cluster views would compute different token ownership and replication placement. Diagnostic only.

Source

Thrown at src/java/org/apache/cassandra/tcm/membership/Directory.java:912

    public void dumpDiff(Directory other)
    {
        if (nextId != other.nextId)
        {
            logger.warn("nextId differ: {} != {}", nextId, other.nextId);
        }
        if (!Objects.equals(lastModified, other.lastModified))
        {
            logger.warn("Last modified differ: {} != {}", lastModified, other.lastModified);
        }
        if (!Objects.equals(peers, other.peers))
        {
            logger.warn("Peers differ: {} != {}", peers, other.peers);
            dumpDiff(logger, peers, other.peers);
        }
        if (!Objects.equals(locations, other.locations))
        {
            logger.warn("Locations differ: {} != {}", locations, other.locations);
        }
        if (!Objects.equals(states, other.states))
        {
            logger.warn("States differ: {} != {}", states, other.states);
            dumpDiff(logger, states, other.states);
        }
        if (!Objects.equals(endpointsByDC, other.endpointsByDC))
        {
            logger.warn("Endpoints by dc differ: {} != {}", endpointsByDC, other.endpointsByDC);
            dumpDiff(logger, endpointsByDC.asMap(), other.endpointsByDC.asMap());
        }
        if (!Objects.equals(racksByDC, other.racksByDC))
        {
            logger.warn("Racks by dc differ: {} != {}", racksByDC, other.racksByDC);
        }
        if (!Objects.equals(versions, other.versions))
        {
            logger.warn("Versions differ: {} != {}", versions, other.versions);

View on GitHub (pinned to 88fd0f6a0e)

Solutions

1. Compare which node IDs have differing locations and re-apply the intended topology change as a committed TCM transformation
2. Replay the cluster metadata log on the diverging node until locations converge
3. Ensure snitch-based dc/rack assignments are consistent across the cluster before committing topology changes
4. If a snapshot is simply stale, re-fetch the current Directory from the CMS before diffing

// Validate intended topology is committed cluster-wide before diffing
ClusterMetadata.sync(nodes, targetEpoch);
for (Node node : nodes)
    if (!node.directory.locations().equals(expectedLocations))
        System.out.println(node + " has stale locations");

When it happens

Trigger: dumpDiff(other) called when the locations maps are unequal, typically after a node moved datacenters in one view (e.g. via a Move/Locate transformation) or a node is present in one map but missing in the other.

Common situations: Operator corrected a node's dc/rack on one node only; directory snapshots compared across a topology migration; stale node replayed older metadata where placements had not yet been updated.

Understand the failure class

Background: "This is a bug, please report it": internal invariant violations, unreachable panics, and SNH errors explained — this error's family across 47 libraries.

Related errors

• Endpoints by dc differ: {} != {}
• Racks by dc differ: {} != {}
• nextId differ: {} != {}
• Last modified differ: {} != {}
• Peers differ: {} != {}

AI-assisted analysis of apache/cassandra@88fd0f6a0e (2026-09-10). Data as JSON: /api/errors/bca89d2ebcf0b8c0. Report an issue: GitHub.