apache/seatunnel · error · ClickhouseConnectorException

Failed to close ssh session

Error message

Failed to close ssh session

What it means

RsyncFileTransfer.close() wraps any IOException thrown while closing the Apache SSHD clientSession into a ClickhouseConnectorException with SSH_OPERATION_FAILED. This fires during sink cleanup when the SSH connection to the remote host is already broken or the session teardown fails. It signals that resources may not have been released cleanly.

Source

Thrown at seatunnel-connectors-v2/connector-clickhouse/src/main/java/org/apache/seatunnel/connectors/seatunnel/clickhouse/sink/file/RsyncFileTransfer.java:180

        }
    }

    @Override
    public void transferAndChown(List<String> sourcePaths, String targetPath) {
        if (sourcePaths == null) {
            throw new ClickhouseConnectorException(
                    CommonErrorCodeDeprecated.ILLEGAL_ARGUMENT, "sourcePath is null");
        }
        sourcePaths.forEach(sourcePath -> transferAndChown(sourcePath, targetPath));
    }

    @Override
    public void close() {
        if (clientSession != null && clientSession.isOpen()) {
            try {
                clientSession.close();
            } catch (IOException e) {
                throw new ClickhouseConnectorException(
                        ClickhouseConnectorErrorCode.SSH_OPERATION_FAILED,
                        "Failed to close ssh session",
                        e);
            }
        }
        if (sshClient != null && sshClient.isOpen()) {
            sshClient.stop();
            try {
                sshClient.close();
            } catch (IOException e) {
                throw new ClickhouseConnectorException(
                        ClickhouseConnectorErrorCode.SSH_OPERATION_FAILED,
                        "Failed to close ssh client",
                        e);
            }
        }
    }
}

View on GitHub (pinned to cf67b549a7)

Solutions

  1. Verify network stability and SSH server KeepAlive/timeout settings between SeaTunnel worker and the ClickHouse host
  2. Check sshd logs on the remote host for connection resets around the failure time
  3. Retry the job; if intermittent, this is a transient network issue
  4. Upgrade connector/sshd-common dependency if the close-failure stems from a known Apache SSHD bug

Example fix

// before: no keepalive configured
SshClient sshClient = SshClient.setUpDefaultClient();
sshClient.start();
// after: enable heartbeats so idle sessions are not reaped
ClientFactoryManager manager = sshClient;
manager.setKeepAliveProvider(KeepAliveProvider.KEEP_ALIVE);
manager.setProperty(ClientFactoryManager.HEARTBEAT_INTERVAL, 30000);
Defensive patterns

Strategy: try-catch

Validate before calling

// before job: verify SSH reachability
ssh -o ConnectTimeout=5 -o BatchMode=yes user@host true && echo OK

Try / catch

try {
    sink.close();
} catch (ClickhouseConnectorException e) {
    if (e.getErrorCode() == ClickhouseConnectorErrorCode.SSH_OPERATION_FAILED) {
        log.warn("SSH session close failed (possibly already dead connection), continuing", e);
    } else { throw e; }
}

Prevention

When it happens

Trigger: Sink close() is called and clientSession.close() throws IOException — typically because the network connection dropped mid-job, the SSH server closed the connection abruptly, or the session was already half-closed by a timeout.

Common situations: Long-running ClickHouse rsync-based file transfers where an idle SSH session was reaped by a firewall/NAT, remote ClickHouse host restarted during the job, or SSH server IdleTimeout terminated the session before close().

Related errors

AI-assisted analysis of apache/seatunnel@cf67b549a7 (2026-09-10). Data as JSON: /api/errors/686ba32f5db41460. Report an issue: GitHub.