Is there any good way to monitor how much space is being consumed by snapshots?I am running on a NAS with somewhat limited space (using about 65% at the moment). And I have snapshots for all my shares.However, I've been re-organizing folders deleteing files, moving files etc...SO, it occurred to me, as I keep seeing my free space drop, that I must be holding on to space with some snapshots... Surely I can go back and start deleting them, but wanted to focus on areas where I can recover most space.I figure there must be some mechanisms for managing snapshots and how much size they are consuming. Any leads before I go off googling and getting into trouble??Thanks in advance!

You could start by deleting the oldest snapshot for each share.I don't think there is an easy way to monitor this.Generally I recommend being selective as to which shares to use snapshots with, but it does depend on the use case, how diligently you backup, what you use the shares for, how important the data in the shares is etc.

From what I've read, you need to enable the btrfs quota features to even see the actual on-disk space used by any share (snapshots or the main folder). CoW makes it rather messy in general, since the the data blocks are referenced in multiple folders (and in the general case this is not neatly structured like it is on the NAS). In cases where a block is multiply referenced, the file system needs to decide which folder it actually belongs to - an arbitrary decision at that level if you think about it.

I did find this, would that not be useful?http://dustymabe.com/2013/09/22/btrfs-h ... snapshots/

Note it begins with With quota and qgroups... we can now see how big each of those snapshots are, including exclusive usage.So it turns on whether the quota features are stable enough to deploy in production environments yet. Not sure on the answer, but I do think Netgear would need to kick the tires a bit before turning them on.

Yea, would be good to get the pros to look at it. But might be worth trying? I assume enabling/disabling those two features would do any *harm* ?

Best way to monitor snapshot disk space?

15 Replies

Replies have been turned off for this discussion

StephenB
Guru - Experienced User
Jan 10, 2015
ScottChapman wrote:
...All this makes me think that snapshot management might be a useful add-on
Totally agree. The automatic pruning that happens now is basically a fail-safe the ensures the NAS can boot, etc. It would be great if there were user-adjustable settings via an add-on (or just built in). I think setting the max age of a snapshot would be very useful, as well as being able to set a snapshot schedule.

ScottChapman wrote:
...All this makes me think that snapshot management might be a useful add-on

ScottChapman

Apprentice

Jan 10, 2015

StephenB wrote:
ScottChapman wrote:
...All this makes me think that snapshot management might be a useful add-on
Totally agree. The automatic pruning that happens now is basically a fail-safe the ensures the NAS can boot, etc. It would be great if there were user-adjustable settings via an add-on (or just built in). I think setting the max age of a snapshot would be very useful, as well as being able to set a snapshot schedule.

Now that I've figured out how to figure things out (and delete snapshots) I might write some PHP to help prune mine before the automatic pruning. Let me know if you are interested and maybe I can incorporate what you want to see.

But yea, a more official add-on would be stellar.

btaroli
Prodigy
Feb 17, 2015
Does your PHP manage to clean up the ROS list of snapshots that is kept apart from the BtRFS structures? I ask because I have a share for which I had done some manual subvolume deletes (really to try and get rid of a massive number of unwanted snapshots, rather than clicking on them one at a time in the web UI). BtRFS happily obliged, but ROS still thinks there are hundreds of them, and won't let me blash this share into oblivion -- it "fails" to delete the snapshots. :( Not precisely the problem you were looking to solve here, I know. But if you worked out how to update that list in your PHP, that could be useful for me in this case. :D
mdgm-ntgr
NETGEAR Employee Retired
Feb 17, 2015
btaroli you need to clean up the entries in the readynasd database using sqlite3. If you deleted them on 6.2.x or 6.3.x you may need to fix the snapper config as well.
btaroli
Prodigy
Feb 17, 2015
I've no doubt... but this is an issue the referenced PHP script would have needed to address as well, yes? :) Hence my inquiry. :D

Forum Discussion

Best way to monitor snapshot disk space?

15 Replies

Related Content

RN316 FW: 6.7.5 - Snapshots - Reclaiming Space

Snapshots taking a lot of space but all snapshots are disabled.

Deleting Files and Snapshots does not free up space - UI shows no more snapshots to delete

ReadyNAS 312 showing space used for snapshots when snapshots set to never

[RNDU6000] How do I recover the space taken up by the 'ghost' snapshots?

NETGEAR Academy

ProSupport for Business