ERROR: "XCOMN0319E Error confirming checkpoint: XCOMN0780E..." with large files
search cancel

ERROR: "XCOMN0319E Error confirming checkpoint: XCOMN0780E..." with large files

book

Article ID: 238010

calendar_today

Updated On:

Products

XCOM Data Transport XCOM Data Transport - Windows XCOM Data Transport - Linux PC

Issue/Introduction

Having problems sending files from XCOM Windows server to remote XCOM UNIX server. Small files can transfer successfully, but larger ones do not. The Windows xcom.log file shows that these errors are reported on failure:
XCOMN0319E Error confirming checkpoint: XCOMN0780E TcpIp  507: TXPI function failed with no error code.

Environment

  • XCOM™ Data Transport® for Windows
  • XCOM™ Data Transport® for Linux PC

Cause

Large file transfers at the remote XCOM UNIX receive side were not completing i.e. the files created are truncated, but no corresponding messages were written to its xcom.log file. 
In a previously reported incident with very similar symptoms, it had been found that an XCOM trace on the remote server showed that there was a problem with an .MBR file in the XCOM Queue directory. The queue on that system had become corrupt and the xcomd service needed to be stopped and the queue directory cleared of files before restart.

Resolution

  1. Run a trace on the remote server by editing the xcom.glb file to have XTRACE=10\
  2. Restart xcomd
  3. Add XTRACE=10 to the local transfer command to gather end-to-end traces
  4. When the remote xcom service is stopped to enable the tracing, the option to delete all files in the $XCOM_HOME/Q directory can also be tried before the restart

In this instance, deleting the Q directory entries did indeed resolve the problem, but if the problem persists the trace files can be given to XCOM Support via a support case for further review.

Additional Information

For future reference the full error message "XCOMN0319E Error confirming checkpoint: XCOMN0780E TcpIp  507: TXPI function failed with no error code", indicates that he problem occurs when a checkpoint is being done. Therefore the option to disable checkpointing by setting CKPT=0 on the send side of the transfer could also be considered. By default CKPT (alias CHECKPOINT_COUNT) is 1000 records which for small files won't be reached but for large files it is probably being reached.