Can you explain to me when data eventually gets deleted from the node harddisk?
I am observing say e.g. 160 TBh storage usage on one day, 2 days later it is down to 110 TBh, both daily. This means that 2 TB have been deleted from the node, but the harddisk is not freeing up the trashed files.
Is there this one week delay in actually deleting the trash? If yes, can this time be reduced because full nodes are not able to accept ingress until the trash is deleted.
recommended practice is to keep enough free space on the HDD, say a couple of 100GB; also beware, some manufactures do 1TB=1000GB and some 1TB=1024GB.
But this not really related to my question. Please somebody clarify about trash deletion / storage space usage, thanks
My post in the linked thread was related to your question. Files older than 7 days in trash will get deleted. And no, this time was set by Storj for a reason.
Note: Files end up in trash only, if your node was not able to delete the files at the time the customer deleted them.
thanks @donald.m.motsinger
I looked into the trash folder a bit more closely, it contains a lot of old directories that are empty.
Why are the empty directories not removed after 7 days?
Because they don’t take up space and would likely have to be recreated again in the future anyway. There are only 1024 possible directories per satellite and if you have any reasonable amount of trash at all, all of them will see some use over time. It doesn’t make much sense to keep removing and adding them for no reason.