I setup a temporary storage NAS using various hard drives and random components lying about my workshop. It is running CentOS 7, with a software RAID of 8 500 GB SATA drives connected as spare drives through cheap fake-RAID controllers.
I then used SDFS to create a deduplicated volume who's base path resides on the software RAID (that's where the chunks.chk, database files, etc reside). I then mount the SDFS volume and share the mounted SDFS volume's folder via CIFS/Samba to my Windows network.
To test it I then copied 1.6 TB worth of random hard disk image files to the share. Worked like a charm! I check the deduplicated data set, 307gb! Great space savings for my DD image files. I then deleted the image files from the CIFS share from my Windows computer that I had used to copy the files over.
However, the 307gb chunks.chk file still persists and when the share is mapped, still reports as having 307gb used, even though I deleted the files and they no longer exist within the share.
Two questions, firstly, does SDFS have a built in setting I can have enabled to clean up the chunks.chk file when files are deleted from the mounted SDFS volume's folder? And primarily, my question is, is it safe to delete the chunks.chk file knowing that I am no longer storing, or intend to store the data residing within the chunks.chk folder?
Thanks so much for your valuable insights!
