Understanding the graph display of Duplicacy webUI

UX
#1

Firstly, thank you to the devs for creating a good software that just works.

Can someone explain to me what the below image means. Why do I have 2 diverging lines in the graph from one point?. I have other backup storages with single IDs that do not have this diverging lines they seem to be on top of each other.

But for some reason here they are diverging in this one. I suspect this is due to prune settings or something. But I do not understand it completely. My prune is set to run weekly on Wednesdays with this settings “-keep 0:360 -keep 90:28 -keep 14:1 -a”. My backups occur on the Monday before every week. I should also note that I had recently paused the scheduled backup but not the check and prune for about a month while I was migrating some data.

01
011081×699 39.7 KB

Thank you

#2

At a guess, this might be caused by unreferenced chunks?

You may need to run a prune job with the -exhaustive flag to clean those up. These may only get complete removal after a subsequent round of backups and a further (normal) prune.

Edit: to do that, just create a new schedule, untick all the days - so you can run it manually - then add a Prune job. Just OK the settings and edit the option, remove all the -keep flags and add only -exhaustive. Press the play button to run it.

#3

Thank you for your reply @Droolio. Sorry for the late reply as I had to run all prune and backups… I have tried what you suggested by setting up a manual prune with -exhaustive flag. Then I let an additional cycle of backup prune and check run. Unfortunately it seems like I still have the two lines showing up.

Any ideas as to why this is happening. I can provide some logs if you can find out what is happening. Would love any input from others as well if they know what is happening here.

1
11239×742 42 KB

Also if anyone could tell me what does this -hash mean in this backup. It is not present when I go to restore any of the other backups. This backup also has two lines in the graph even though it only has one id backing up to it.

2

#4 
OPTIONS:
   -hash                        detect file differences by hash (rather than size and timestamp)

-hash flag is systematic during the first backup and first snapshot.

#5

Thanks, I missed that. The other backups must have pruned that revision as they are revision 3 mostly