quote:

Originally posted by SirBeep:
I believe FreeNAS is zfs so you probably have snapshot issues. I don't use clicky drooly interfaces, I build the systems that do the real work behind those clicky droolies, so I can't tell you where to click to find what you need to click, but..

first you check and clean the samba .recycle folder at your share root, the equivalent to trash can? whatever winderz calls the undelete hole... turn on See Hidden Files in your File Explorer Options to see it, then clean it out.

After that is there should be a Snapshots section somewhere in the management UI of your NAS. This is a feature similar to Apples Time Machine. go there and clean up any snapshot timestamps you want to get rid of..

and while you're there at snapshots there may be some settings you should tweak. ZFS can operate in an immutable mode.. erm... think a quantum physicists wet dream where everything that could happen did happen and every decision creates a fork and an entirely new universe. this, but for data. nothing deletes it just gets hidden, updating a file makes a copy of the file and keeps both updated and non updated data.. etc.. as you can imagine this will eat data.

quote:

Originally posted by architect:
FreeNAS is a FreeBSD distribution, not Linux. It uses the Z file system (ZFS) pioneered by Sun. a ZVol is a pre-allocated volume scaled to a specified size (comparable to a file or a disk of the allocated size). A "regular" ZFS volume is not bound by size, or more accurately, bound only by the size of the underlying media (a disk or zpool of disks). There is a ton of information on-line about ZFS. Be aware that anyone you ask for help on this will assume that you have read the relevant documentation beforehand.

One of the processes you are running wants more space than you are giving it. Sometimes this is cache, or data, but more often it is logs. Find out what this is, and stop running it, or change its configuration to be less needy, or write its logs to a volume with more space. I haven't run pi-hole, but this seems unlikely to be the problem.

I also think it unlikely that ZFS snapshots are responsible for filling the volume(s), but it wouldn't hurt to take a look at these too "zfs list -t all" if you have a lot of snaps (name has an @ in it) in the problematic file system, more then thirty or so, might not hurt to remove a few, and take snaps less frequently.

If you don't mind working at the command line, the following FreeBSD commands can help home in on the problem:

% df # shows free disk space, i.e. which volume is getting filled)
# systat -vmstat # shows the general state of the virtual memory sub-system, including disk activity)
# ps auxww # process status, what is using CPU and other resources
% top # like ps, but updated frequently

The first character of the previous four lines is typed by the system as a prompt, the "%" indicates that it can be run as an ordinary user, the "#" means that becoming the superuser (root) is advised to run this.

There are other commands to inspect the file systems, zfs and zpool, these have enough options that reading the man pages is advised before executing them, actually, reading the man page is a good idea before running any command. In a Unix-like system, man pages can be read by running "man COMMAND" e.g. "man man" or "man zfs" You can also find most man pages on-line via a web search. There is also the FreeBSD Handbook that is a very useful resource. The FreeBSD support forums can also be very helpful, but the FreeNAS ones are probably a little better resource for FreeNAS-specific questions these days.

quote:

Originally posted by Bytes:
Ensigmatic is another good resource. He's managed tons of Linux servers.