prestodb/presto · error · IllegalArgumentException

Unrecognized memory pool:

Error message

Unrecognized memory pool: 

What it means

ResourceManagerClusterStateProvider aggregates per-node memory usage by classifying each reported memory pool as GENERAL_POOL or RESERVED_POOL. When a node reports a pool ID matching neither, getClusterMemoryPoolInfoInternal throws IllegalArgumentException. This indicates coordinator/resource-manager configuration mismatch (e.g. a custom pool or outdated node config).

Source

Thrown at presto-main-base/src/main/java/com/facebook/presto/resourcemanager/ResourceManagerClusterStateProvider.java:315

                .collect(toImmutableList());

        int queriesAssignedToGeneralPool = 0;
        int queriesAssignedToReservedPool = 0;
        Query largestGeneralPoolQuery = null;
        for (CoordinatorQueriesState nodeQueryState : nodeQueryStates.values()) {
            for (Query query : nodeQueryState.getActiveQueries()) {
                MemoryPoolId memoryPool = query.getBasicQueryInfo().getMemoryPool();
                if (GENERAL_POOL.equals(memoryPool)) {
                    queriesAssignedToGeneralPool = Math.incrementExact(queriesAssignedToGeneralPool);
                    if (!resourceOvercommit(query.getBasicQueryInfo().getSession().toSession(sessionPropertyManager))) {
                        largestGeneralPoolQuery = getLargestMemoryQuery(Optional.ofNullable(largestGeneralPoolQuery), query);
                    }
                }
                else if (RESERVED_POOL.equals(memoryPool)) {
                    queriesAssignedToReservedPool = Math.incrementExact(queriesAssignedToReservedPool);
                }
                else {
                    throw new IllegalArgumentException("Unrecognized memory pool: " + memoryPool);
                }
            }
        }

        List<QueryId> runningQueries = nodeQueryStates.values().stream()
                .map(CoordinatorQueriesState::getActiveQueries)
                .flatMap(Collection::stream)
                .filter(query -> query.getBasicQueryInfo().getState() == RUNNING)
                .map(Query::getQueryId)
                .collect(toImmutableList());

        ImmutableMap.Builder<MemoryPoolId, ClusterMemoryPoolInfo> memoryPoolInfos = ImmutableMap.builder();
        ClusterMemoryPool pool = new ClusterMemoryPool(GENERAL_POOL);
        pool.update(memoryInfos, queriesAssignedToGeneralPool);
        ClusterMemoryPoolInfo clusterInfo = pool.getClusterInfo(Optional.ofNullable(largestGeneralPoolQuery).map(Query::getQueryId), Optional.ofNullable(runningQueries));
        memoryPoolInfos.put(GENERAL_POOL, clusterInfo);
        if (isReservedPoolEnabled) {
            pool = new ClusterMemoryPool(RESERVED_POOL);

View on GitHub (pinned to 55bb57d202)

Solutions

1. Align memory pool configuration across all coordinators, workers, and the resource manager (query.max-memory-per-node / memory-pool settings).
2. Restart/redeploy mismatched nodes so every worker reports only GENERAL_POOL and RESERVED_POOL.
3. If custom pools are in use, update ResourceManagerClusterStateProvider handling to recognize them.
4. Check for version skew: upgrade all nodes to the same Presto version.

Example fix

// before (worker config.properties, mismatched)
memory.pools=custom
// after
memory.pools=general,reserved

# pre-deploy check: every node's config reports only general/reserved pools
grep -E 'memory' config.properties  # ensure no custom pools unless the build supports them

try {
    poolInfo = provider.getClusterMemoryPoolInfo();
} catch (IllegalArgumentException e) {
    if (e.getMessage().startsWith("Unrecognized memory pool:")) {
        log.error("Node reports unknown pool {}; fix memory pool config cluster-wide", e.getMessage());
    }
    throw e;
}

When it happens

Trigger: A Presto node reports a memory pool whose ID is neither the configured general pool nor the reserved pool — typically after changing memory pool configuration on coordinators/workers inconsistently, or with custom pool plugins.

Common situations: Mixed-version clusters during rolling upgrades; one worker with a custom memory-pool config while the resource manager expects only default pools; stale node metadata after config changes.

Related errors

• ACCUMULO_TABLE_EXISTS
• UNEXPECTED_ACCUMULO_ERROR
• NOT_SUPPORTED
• ARROW_INTERNAL_ERROR
• JDBC_ERROR

AI-assisted analysis of prestodb/presto@55bb57d202 (2026-09-04). Data as JSON: /api/errors/ecd3a40392f95d45. Report an issue: GitHub.