Timeout occurs when attempting to download the file that comes after a 'large' file

[one-giant-leap]

I am able to successfully download multiple small files without issue, however when a 'larger' file (e.g. 20910 bytes) is encountered, the 'larger' file will be downloaded successfully but the next file will produce a timeout.

For example, given the following Directory:

Index Type FIT Type FIT Number Size Date FIT Flags Flags 12 4 4 0 666 2000-01-01 04:29:42+00:00 0 rwe--- 13 4 4 0 666 2000-01-01 04:29:50+00:00 0 rwe--- 14 4 4 0 666 2000-01-01 04:29:56+00:00 0 rwe--- 15 4 4 0 666 2000-01-01 04:30:02+00:00 0 rwe--- 16 4 4 0 666 2000-01-01 04:30:08+00:00 0 rwe--- 17 4 4 0 666 2000-01-01 04:30:14+00:00 0 rwe--- 18 4 4 0 666 2000-01-01 04:30:20+00:00 0 rwe--- 25 4 4 0 20910 2000-01-01 04:46:12+00:00 0 rwe--- 22 4 4 0 666 2000-01-01 04:40:36+00:00 0 rwe--- 23 4 4 0 666 2000-01-01 04:40:40+00:00 0 rwe--- 24 4 4 0 666 2000-01-01 04:40:44+00:00 0 rwe---

Downloading index 22 will fail and the following error is printed:

File "/Library/Frameworks/Python.framework/Versions/3.10/lib/python3.10/site-packages/antfs_cli-0.4-py3.10.egg/antfs_cli/program.py", line 384, in main g.start() File "/Library/Frameworks/Python.framework/Versions/3.10/lib/python3.10/site-packages/openant-0.4-py3.10.egg/ant/fs/manager.py", line 223, in start self._main() File "/Library/Frameworks/Python.framework/Versions/3.10/lib/python3.10/site-packages/openant-0.4-py3.10.egg/ant/fs/manager.py", line 161, in _main self.on_transport(beacon) File "/Library/Frameworks/Python.framework/Versions/3.10/lib/python3.10/site-packages/antfs_cli-0.4-py3.10.egg/antfs_cli/program.py", line 254, in on_transport self.download_file(fileobject) File "/Library/Frameworks/Python.framework/Versions/3.10/lib/python3.10/site-packages/antfs_cli-0.4-py3.10.egg/antfs_cli/program.py", line 297, in download_file data = self.download(fil.get_index(), AntFSCLI._get_progress_callback()) File "/Library/Frameworks/Python.framework/Versions/3.10/lib/python3.10/site-packages/openant-0.4-py3.10.egg/ant/fs/manager.py", line 325, in download self._send_command(DownloadRequest(index, offset, True, crc)) File "/Library/Frameworks/Python.framework/Versions/3.10/lib/python3.10/site-packages/openant-0.4-py3.10.egg/ant/fs/manager.py", line 201, in _send_command self._channel.send_burst_transfer(data) File "/Library/Frameworks/Python.framework/Versions/3.10/lib/python3.10/site-packages/openant-0.4-py3.10.egg/ant/easy/channel.py", line 133, in send_burst_transfer self.wait_for_event([Message.Code.EVENT_TRANSFER_TX_START]) File "/Library/Frameworks/Python.framework/Versions/3.10/lib/python3.10/site-packages/openant-0.4-py3.10.egg/ant/easy/channel.py", line 50, in wait_for_event return wait_for_event(ok_codes, self._node._events, self._node._event_cond) File "/Library/Frameworks/Python.framework/Versions/3.10/lib/python3.10/site-packages/openant-0.4-py3.10.egg/ant/easy/filter.py", line 73, in wait_for_event return wait_for_message(match, process, queue, condition) File "/Library/Frameworks/Python.framework/Versions/3.10/lib/python3.10/site-packages/openant-0.4-py3.10.egg/ant/easy/filter.py", line 62, in wait_for_message raise AntException("Timed out while waiting for message") ant.easy.exception.AntException: Timed out while waiting for message Interrupted: Timed out while waiting for message

I have visibility into the device where the files originate (I have a debugger set up on the embedded device i.e. fitness watch). The device doesn't appear to receive the 'ANTFS_EVENT_DOWNLOAD_REQUEST' message from antfs-cli when the time comes to download the file in question (the one that follows a larger file). Hence, the device doesn't respond with the message that antfs-cli is waiting for and antfs-cli promptly times out (within 1-2 seconds of having downloaded the 'larger' file).

Any thoughts on what might be occurring? My next test is to see how long antfs-cli typically takes to time out after sending the 'ANTFS_EVENT_DOWNLOAD_REQUEST' message to the device. If it is after quite some time, then that suggests the time taken to download the 'larger' file is possibly starving a e.g. watchdog time or something similar.

[one-giant-leap]

I've managed to get it working. From what I can tell the issue was that multiple 'EVENT_RX_FAIL' messages would pile up when a large file transfer was taking place. When antfs-cli started to process the 'EVENT_RX_FAIL' messages upon having transferred the large file, it would time out prior to sending the next 'DOWNLOAD_REQUEST'.

The 'fix' was to check for any 'EVENT_RX_FAIL' messages and stop them going into the queue (code found in base/ant.py in openant).

  # Channel event
  elif (
      message._id == Message.ID.RESPONSE_CHANNEL
      and message._data[1] == 0x01
  ):
      _logger.debug("Got channel event, %r", message)
      if (
          message._data[2] != Message.Code.EVENT_RX_FAIL
      ):
          self._events.put(
              (
                  "event",
                  (message._data[0], message._data[1], message._data[2:]),
              )
          )
  elif message._id == Message.ID.BROADCAST_DATA:
      self._on_broadcast(message)
  elif message._id == Message.ID.ACKNOWLEDGED_DATA:
      self._on_acknowledge(message)
  elif message._id == Message.ID.BURST_TRANSFER_DATA:
      self._on_burst_data(message)
  else:
      _logger.warning("Got unknown message, %r", message)