influxdata/influxdb · error · CatalogError

node ' ' has compact mode and cannot be removed

Error message

node '{node_id}' has compact mode and cannot be removed

What it means

This error is thrown when an operator attempts to remove a node from the cluster catalog while the node is still configured with a compactor mode. Removing a node that owns compaction work would leave compacted generations unproduced, so the catalog refuses the operation. It is raised by node lifecycle/registration APIs in the catalog.

Solutions

1. Reconfigure the node so its mode no longer includes compaction (remove compact role), then retry removal.
2. Verify the node has been fully stopped and drained of compactor responsibilities with the node state/inspect API.
3. If the node id is wrong, re-run the removal targeting the intended node_id.
4. If compaction has been migrated elsewhere (e.g. another compactor node), confirm the catalog reflects the new assignment before removing.

Example fix

// before
catalog.remove_node("compactor-01").await?;
// after
let node = catalog.node("compactor-01")?;
assert!(!node.mode.is_compactor(), "strip compact role before removal");
catalog.set_node_mode("compactor-01", NodeMode::QueryOnly)?;
catalog.remove_node("compactor-01").await?;

let node = catalog.node(node_id)?;
if node.mode.is_compactor() {
    return Err(format!("node {node_id} still has compact mode; strip role before removal"));
}

fn removable(node: &NodeMetadata) -> bool { !matches!(node.mode, NodeMode::Compact | NodeMode::All) }

When it happens

Trigger: Calling the catalog's node removal/unregister API (Catalog::remove_node / node state machine transitions) with a node_id whose current NodeMode/role includes compact duties.

Common situations: Decommissioning a node during cluster resize or failover; automated scripts draining nodes that happen to be compactor-enabled; operator removes a node via admin CLI/HTTP API without first stripping its compact role.

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

• idempotent no-op
• invalid stop ack for node
• invalid unregister for node
• {0}
• Adding a new database would exceed limit of

AI-assisted analysis of influxdata/influxdb@06200ef96b (2026-09-19). Data as JSON: /api/errors/cd73903132e05fb0. Report an issue: GitHub.

Appendix: source

Thrown at influxdb3_catalog/src/error.rs:308

    #[error("cannot delete internal db")]
    CannotDeleteInternalDatabase,

    #[error("cannot modify internal db")]
    CannotModifyInternalDatabase,

    #[error("tried to stop a node ({node_id}) that is already stopped")]
    NodeAlreadyStopped { node_id: Arc<str> },

    #[error(
        "node '{node_id}' is not fully stopped (current state: {current_state}); run \"stop node\" first"
    )]
    NodeNotFullyStopped {
        node_id: Arc<str>,
        current_state: &'static str,
    },

    #[error("node '{node_id}' has compact mode and cannot be removed")]
    NodeModeNotRemovable { node_id: Arc<str> },

    #[error("invalid stop ack for node '{node_id}' (current state: {current_state})")]
    InvalidStopAck {
        node_id: Arc<str>,
        current_state: &'static str,
    },

    #[error("invalid unregister for node '{node_id}' (current state: {current_state})")]
    InvalidUnregister {
        node_id: Arc<str>,
        current_state: &'static str,
    },

    #[error("idempotent no-op")]
    IdempotentNoOp,

    #[error("cannot delete operator token")]

View on GitHub (pinned to 06200ef96b)