What causes "%s has already fully joined the CMS" in apache/cassandra?

Query current CMS membership first; if the node already has full membership, skip the join step entirely

apache/cassandra · warning · IllegalStateException

%s has already fully joined the CMS

Error message

%s has already fully joined the CMS

What it means

In CMSMembership.startJoining, after checking the joining set, it verifies the node is not already in fullMembers; if it is, IllegalStateException is thrown because the node has already fully joined the CMS and starting a join for it is meaningless and would corrupt membership state.

Source

Thrown at src/java/org/apache/cassandra/tcm/CMSMembership.java:142

        return lastModified;
    }

    public Set<NodeId> joiningMembers()
    {
        return joiningMembers;
    }

    public Set<NodeId> fullMembers()
    {
        return fullMembers;
    }

    public CMSMembership startJoining(NodeId id)
    {
        if (joiningMembers.contains(id))
            throw new IllegalStateException(id + " is already joining the CMS");
        if (fullMembers.contains(id))
            throw new IllegalStateException(id + " has already fully joined the CMS");

        return new CMSMembership(lastModified, fullMembers, joiningMembers.with(id));
    }

    public CMSMembership cancelJoining(NodeId id)
    {
        if (!joiningMembers.contains(id))
            throw new IllegalStateException(id + " is not currently joining the CMS");
        if (fullMembers.contains(id))
            throw new IllegalStateException(id + " has already fully joined the CMS");

        return new CMSMembership(lastModified, fullMembers, joiningMembers.without(id));
    }

    public CMSMembership finishJoining(NodeId id)
    {
        if (!joiningMembers.contains(id))
            throw new IllegalStateException(id + " is not currently joining the CMS");

View on GitHub (pinned to 88fd0f6a0e)

Solutions

Query current CMS membership first; if the node already has full membership, skip the join step entirely
Make the join orchestration idempotent: treat 'already fully joined' as a no-op success
Update stale automation state to reflect the node's completed join before issuing further CMS operations
Only cancel/rejoin if you genuinely intend to remove and re-add the node, using cancelJoining/finishJoining transitions in the correct order

Example fix

// before
membership = cms.startJoining(nodeId);
// after
if (!cms.fullMembers().contains(nodeId))
    membership = cms.startJoining(nodeId);

Defensive patterns

Strategy: validation

Validate before calling

if (membership.fullMembers().contains(nodeId)) return; // already joined; skip startJoining

Prevention

Always read current CMS membership from the latest epoch before joining
Treat completed joins as success in retry logic
Update orchestration state after join completion to avoid replays

When it happens

Trigger: Applying a startJoining transformation for a NodeId present in fullMembers — a join request replayed after successful completion, or an operator/automation re-running the join step after the node already joined.

Common situations: Retry logic that does not check whether the previous join succeeded; re-running bootstrap/CMS-join tooling on an already-joined node; stale orchestration state following an ambiguous command result.

Understand the failure class

Background: "Invalid state transition" errors: "status must be X, actually Y", "already rejected/charging/uninstalled", "cannot ... while running" — what they mean when a library rejects your call — this error's family across 31 libraries.

Related errors

AI-assisted analysis of apache/cassandra@88fd0f6a0e (2026-09-10). Data as JSON: /api/errors/6b6eca81d05f1cf9. Report an issue: GitHub.