apache/hadoop · error · IOException

No live nodes contain current block

Error message

No live nodes contain current block

What it means

Error "No live nodes contain current block" thrown in apache/hadoop.

Source

Thrown at hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/namenode/NamenodeFsck.java:1180

    }
    return bytesRead;
  }

  @Override
  public DataEncryptionKey newDataEncryptionKey() throws IOException {
    return namenode.getRpcServer().getDataEncryptionKey();
  }

  /*
   * XXX (ab) See comment above for copyBlock().
   *
   * Pick the best node from which to stream the data.
   * That's the local one, if available.
   */
  private DatanodeInfo bestNode(DFSClient dfs, DatanodeInfo[] nodes,
      Set<DatanodeInfo> deadNodes) throws IOException {
    if ((nodes == null) || (nodes.length - deadNodes.size() < 1)) {
      throw new IOException("No live nodes contain current block");
    }
    DatanodeInfo chosenNode;
    do {
      chosenNode = nodes[ThreadLocalRandom.current().nextInt(nodes.length)];
    } while (deadNodes.contains(chosenNode));
    return chosenNode;
  }

  private void lostFoundInit(DFSClient dfs) {
    lfInited = true;
    try {
      String lfName = "/lost+found";

      final HdfsFileStatus lfStatus = dfs.getFileInfo(lfName);
      if (lfStatus == null) { // not exists
        lfInitedOk = dfs.mkdirs(lfName, null, true);
        lostFound = lfName;
      } else if (!lfStatus.isDirectory()) { // exists but not a directory

View on GitHub (pinned to 2add963021)

Solutions

1. Bring back DataNodes holding the block or restore from backup; the block is unavailable on all live nodes.

When it happens

Trigger: fsck finds a block whose replicas exist only on dead or decommissioned DataNodes.

Common situations: Multiple DataNode failures or disks lost, leaving no live replica of the block.

AI-assisted analysis of apache/hadoop@2add963021 (2026-08-22). Data as JSON: /api/errors/226b97f771ba157c. Report an issue: GitHub.