apache/hadoop · error · IOException

Pipeline recovery for {} is interrupted.

Error message

Pipeline recovery for {} is interrupted.

What it means

Error "Pipeline recovery for {} is interrupted." thrown in apache/hadoop.

Source

Thrown at hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/datanode/DataNode.java:3678

    //set visible length
    b.setNumBytes(visible);

    if (targets.length > 0) {
      if (LOG.isDebugEnabled()) {
        final String xferTargetsString =
            StringUtils.join(" ", Arrays.asList(targets));
        LOG.debug("Transferring a replica to {}", xferTargetsString);
      }

      final DataTransfer dataTransferTask = new DataTransfer(targets,
          targetStorageTypes, targetStorageIds, b, stage, client);

      @SuppressWarnings("unchecked")
      Future<Void> f = (Future<Void>) this.xferService.submit(dataTransferTask);
      try {
        f.get();
      } catch (InterruptedException | ExecutionException e) {
        throw new IOException("Pipeline recovery for " + b + " is interrupted.",
            e);
      }
    }
  }

  /**
   * Finalize a pending upgrade in response to DNA_FINALIZE.
   * @param blockPoolId the block pool to finalize
   */
  void finalizeUpgradeForPool(String blockPoolId) throws IOException {
    storage.finalizeUpgrade(blockPoolId);
  }

  static InetSocketAddress getStreamingAddr(Configuration conf) {
    return NetUtils.createSocketAddr(
        conf.getTrimmed(DFS_DATANODE_ADDRESS_KEY, DFS_DATANODE_ADDRESS_DEFAULT));
  }

View on GitHub (pinned to 2add963021)

Solutions

1. Treat as benign if caused by DataNode shutdown or job cancellation; the write will be retried by the client with a new pipeline.
2. If frequent, investigate what interrupts recovery threads (timeouts, premature shutdown) and review pipeline recovery timeouts.

When it happens

Trigger: The thread performing pipeline recovery for a block is interrupted, typically during shutdown or when the client aborts the write.

Common situations: Pipeline recovery was interrupted, usually by DataNode shutdown or thread interruption. The client will retry; no action needed unless persistent.

AI-assisted analysis of apache/hadoop@2add963021 (2026-08-22). Data as JSON: /api/errors/d76c95d22b4ac84c. Report an issue: GitHub.