I do a couple of sync-jobs between providers and accounts throughout the day and generally rely on this post-transfer set of stats which is logged after each run:
After each such run I parse and add the last transfer's use together so I have an idea of what amount of upload I have used in a 24h period (for gdrive specifically but also for my monitoring system for upload / # files copied and deleted so I have an idea of the lay of the land).
The introduction of the --drive-stop-on-upload-limit and -max-duration commands are of course welcomed in this scenario, however I notice that a non-zero exit code (as you will run into when using these flags) ensures that the above transfer stats are not logged on exit.
Therefore,
Is it possible / would it be possible to log these transfer stats on non-zero exits?
Longer term, -max-[upload|download]-per-period=<time>,<size> would be super - possibly maintain state in rclone.conf but sounds like a nightmare to implement (pairs of providers, etc).
The scripts I use that does the parsing, etc is already in place and fine. It's merely that a non-zero exit code does not log the transfer stats, whilst a zero-exit does.
That sounds like a sensible request... I had a look at that bit of code and it was a bit of a mess so I've tidied it up - it will always output stats before ending now.
Thanks @ncw I'll give that beta a spin as it'll solve my transfer accounting issue
I appreciate --bwlimit however that assumes the quantity to copy/move/sync is quite static and not very variable (size and bandwidth wise) and can therefore get a bit messy in other scenarios.
The use of --drive-stop-on-upload-limit and the parsing I'm already doing of the transfer stats hopefully now even on failure solves that dilemma, thank you!
