During my last upload I have run into this problem again. Uploaded overnight. Today morning Uploader has uploaded only 4 sequences out of 39 due to any error. Selected “Retry upload”. First it looked ok, then he suddenly started to upload sequence_idx 0, 'md5sum': 'd21d44509db082f86ed236f037904a4b'
again. Digged into the logs. Here you can see it:
The upload before I retried stopped with “There appear to be 1 leaked semaphore objects to clean up at shutdown”. See here:
2024-11-08 22:17:42,606 - DEBUG - The next offset will be: 989855744
2024-11-08 22:17:42,606 - DEBUG - Sending upload_progress via IPC: {'file_type': 'image', 'sequence_idx': 4, 'total_sequence_count': 39, 'sequence_image_count': 34, 'sequence_uuid': '4', 'entity_size': 2128745744, 'md5sum': '96a16bb5a8899ca18dc5b0c4118f8236', 'upload_start_time': 1731100410.7024598, 'upload_total_time': 0, 'offset': 989855744, 'retries': 0, 'upload_last_restart_time': 1731100410.864608, 'upload_first_offset': 0, 'chunk_size': 16777216}
[2024-11-08 22:17:42.612] [debug] [mapillary:tools] 2024-11-08 22:17:42,612 - DEBUG - POST https://rupload.facebook.com/mapillary_public_uploads/mly_tools_96a16bb5a8899ca18dc5b0c4118f8236.zip HEADERS {"Offset": "989855744", "X-Entity-Length": "2128745744", "X-Entity-Name": "mly_tools_96a16bb5a8899ca18dc5b0c4118f8236.zip", "X-Entity-Type": "application/zip"}
[2024-11-08 22:17:42.916] [debug] [mapillary:tools] multiprocessing/resource_tracker.py:224: UserWarning: resource_tracker: There appear to be 1 leaked semaphore objects to clean up at shutdown
[2024-11-08 22:17:42.928] [debug] [mapillary:tools] process exit code null and signal SIGPIPE
[2024-11-08 22:17:42.933] [error] [store:modules:uploadSession] exit with code null and signal SIGPIPE
[2024-11-08 22:17:42.963] [info] [power-saver-blocker] stop: 0
[2024-11-08 22:17:42.963] [info] [services:observer-manager] Stopping 8ec6fc20-9e12-11ef-9886-85bca27b185a
[2024-11-08 22:17:42.978] [error] [vue] { code: null, message: 'exit with code null and signal SIGPIPE' }
I started a fresh new upload with “Skip uploaded files” switched on, but sequence_idx 0-3 were uploaded anyway:
35 sequences out of 39 later - Boom!
BTW: You should really not mix different log formats in a log file. It makes it hard to use log analyzers:
[2024-11-08 14:28:32.234] [debug] [mapillary:tools] 2024-11-08 14:28:32,232 - DEBUG - HTTP response 206: b'{"debug_info":{"retriable":true,"type":"PartialRequestError","message":"Partial request (did not match length of file)"}}'
2024-11-08 14:28:32,232 - DEBUG - The next offset will be: 1593835520
2024-11-08 14:28:32,232 - DEBUG - Sending upload_progress via IPC: {'file_type': 'image', 'sequence_idx': 1, 'total_sequence_count': 8, 'sequence_image_count': 31, 'sequence_uuid': '1', 'entity_size': 2108923895, 'md5sum': '6b1a0ffdfd64fdf46251ae3c8921e205', 'upload_start_time': 1731072019.616509, 'upload_total_time': 0, 'offset': 1593835520, 'retries': 0, 'upload_last_restart_time': 1731072019.944736, 'upload_first_offset': 0, 'chunk_size': 16777216}
[2024-11-08 22:17:42.933] [error] [store:modules:uploadSession] exit with code null and signal SIGPIPE
DEBUG vs. [debug]
[2024-11-08 14:28:32.234] vs. 2024-11-08 14:28:32,232
…32.234 vs. …32,232
and so on.