apache/hadoop · error · RuntimeException

configured value {}for dfs.datanode.cache.revocation.polling

Error message

configured value {}for dfs.datanode.cache.revocation.polling.ms is too high.  It must not be more than half of the value of dfs.datanode.cache.revocation.timeout.ms.  Reconfigure this to {}

What it means

Error "configured value {}for dfs.datanode.cache.revocation.polling.ms is too high. It must not be more than half of the value of dfs.datanode.cache.revocation.timeout.ms. Reconfigure this to {}" thrown in apache/hadoop.

Source

Thrown at hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/datanode/fsdataset/impl/FsDatasetCache.java:174

        .setNameFormat("FsDatasetCache-%d-" + dataset.toString())
        .build();
    this.uncachingExecutor = new ThreadPoolExecutor(
            0, 1,
            60, TimeUnit.SECONDS,
            new LinkedBlockingQueue<Runnable>(),
            workerFactory);
    this.uncachingExecutor.allowCoreThreadTimeOut(true);
    this.deferredUncachingExecutor = new ScheduledThreadPoolExecutor(
            1, workerFactory);
    this.revocationMs = dataset.datanode.getConf().getLong(
        DFS_DATANODE_CACHE_REVOCATION_TIMEOUT_MS,
        DFS_DATANODE_CACHE_REVOCATION_TIMEOUT_MS_DEFAULT);
    long confRevocationPollingMs = dataset.datanode.getConf().getLong(
        DFS_DATANODE_CACHE_REVOCATION_POLLING_MS,
        DFS_DATANODE_CACHE_REVOCATION_POLLING_MS_DEFAULT);
    long minRevocationPollingMs = revocationMs / 2;
    if (minRevocationPollingMs < confRevocationPollingMs) {
      throw new RuntimeException("configured value " +
              confRevocationPollingMs + "for " +
              DFS_DATANODE_CACHE_REVOCATION_POLLING_MS +
              " is too high.  It must not be more than half of the " +
              "value of " +  DFS_DATANODE_CACHE_REVOCATION_TIMEOUT_MS +
              ".  Reconfigure this to " + minRevocationPollingMs);
    }
    this.revocationPollingMs = confRevocationPollingMs;

    this.cacheLoader = MappableBlockLoaderFactory.createCacheLoader(
        this.getDnConf());
    // Both lazy writer and read cache are sharing this statistics.
    this.memCacheStats = cacheLoader.initialize(this.getDnConf());
  }

  /**
   * For persistent memory cache, create cache subdirectory specified with
   * blockPoolId to store cache data.
   * Recover the status of cache in persistent memory, if any.

View on GitHub (pinned to 2add963021)

Solutions

1. Set dfs.datanode.cache.revocation.polling.ms to a value no greater than half of dfs.datanode.cache.revocation.timeout.ms.
2. Alternatively increase dfs.datanode.cache.revocation.timeout.ms so the polling interval is proportionally smaller.

When it happens

Trigger: DataNode startup when dfs.datanode.cache.revocation.polling.ms is configured larger than half of dfs.datanode.cache.revocation.timeout.ms.

Common situations: See trigger scenarios.

Understand the failure class

• Timeouts: ETIMEDOUT, deadlines, and hung requests — what actually expires when a request times out.

AI-assisted analysis of apache/hadoop@2add963021 (2026-08-22). Data as JSON: /api/errors/bfe9f60b1b2925e5. Report an issue: GitHub.