Noob using GUI version. Do my settings make sense? (and some questions)

Support
#1

Hi everyone, my goal is to first backup my files to my internal backup drive, and then copy those files to Backblaze B2. I tried to Google the best method to do this and attempted to make a schedule, although I’m still a bit confused. So any feedback on how I can improve it would be appreciated.

  1. Every 15 minutes daily.
  • Backup my files to hard drive
  1. Every 2 hours daily
  • Copy files from above to cloud.
  1. Once a day, daily (both running parallel)
  • Check HD
  • Check cloud
  1. Once a day, daily (both running parallel)
  • Prune on HD
  • Prune cloud

Questions:

  1. How do I find out the max number of threads I can use? I’ve used Duplicacy for 2 days so far and haven’t even completed the first copy job yet because it was so slow, until I found out how to change the settings. I changed it to 5 threads and it’s A LOT faster, so regardless I’m happy about the improvement but would like to know how to get the maximum amount of speed.
  2. After changing the threads as described above, I also noticed that there’s less chunks. It went from something like 65000 to 6600. Is this normal? It’s possible I’m remembering incorrectly but I’m pretty sure I’m not.
  3. If my files backup to my internal HD while they’re also being copied to the cloud (due to the different time intervals), could this be an issue? Or can it not get messed up from this?
  4. The computer that I’m using isn’t on all of the time. If I miss the scheduled time for a job, does Duplicacy automatically run it the next time it can? I Googled this and supposedly it does, but (unless I’m mistaken) it didn’t seem like this was happening for me. Is there a setting I have to enable?

Any help is greatly appreciated! Thank you.

#2

It depends on the storage backend and your upload bandwidth. For local disk storage, generally you don’t benefit from multiple uploading threads but for B2 I would suggest 4-8 threads if you have enough upload bandwidth.

This is because chunks that have already been uploaded during the last run may not be counted.

This should not cause any problem.

If a scheduled time is missed due to sleep, Duplicacy will run the schedule once the computer wakes up. This is by default and not setting needs to be enabled. If you’re seeing otherwise, check the log file (~/.duplicacy-web/duplicacy-web.log) and post the relevant messages here.

1 Like
#3

Thank you for your help, I really appreciate it!. That pretty much answers my questions, but I just realized I have another one. Like I said in the original post, the computer I’m using isn’t on all of the time. I’m still struggling to get it to do the initial copy job because even though I increased the threads and it was a lot faster, when I turned off my computer last night and turned it on this morning, it said that the job failed and when I manually ran it, it began from the start again. This was the second time I noticed this happen as well. I would be really helpful if you had an ideas as to why this happens and how I can prevent it. Thank you!

Edit: also, another weird thing I’m noticing right now is that the amount of chunks uploaded for the copy job in progress are constantly increasing and then decreasing slightly. The log says they were skipped at the destination. Is this a glitch or intended feature?

#4

On resume Duplicacy may appear to start from the beginning, but chunks that have been uploaded will be skipped so the upload speed will be much faster.

This is a glitch. It is showing the index of the current chunk being copied. If you use multiple threads then this number may go backwards.

1 Like