phillxnet
(Philip Guyton)
March 27, 2017, 7:29pm
5
@MvL OK, that was it then. Good to get to the bottom of it.
It’s down to how docker stores it’s images when using a btrfs backend . Since docker images are composed of layers built on top of one another the docker system can use the btrfs snapshots mechanism (or OverlayFS ) to store and share these image snapshots between a number of docker containers. It’s not a waste of space but an efficient use of it.
From the above link: https://docs.docker.com/engine/userguide/storagedriver/btrfs-driver/#the-future-of-btrfs
we have:
“Docker leverages Btrfs subvolumes and snapshots for managing the on-disk components of image and container layers.”
Linking to a related thread on this in the forum:
Hey guys, love rockstor, and appreciate all the features the developers put in. I noticed rocks-ons take a ton of snapshots. After deleting the snapshots, it seems to break the rock-ons. Is this by design or a bug? If I delete the snapshots, it totally breaks all the rock-ons I’ve installed. I have to re-create a new rock-ons share to fix.
I like managing my own snapshots. I wish there was a way to disable the auto-snapshots on the rock-ons, or have it not break all the rock-ons when deleting t…
and
it is possible that I am just not searching with the proper keywords which is causing me to fail in finding info on this but… in the guide it does talk about how the rock-ons are regularly snapped… I am wondering if we can delete these snaps without causing harm to the system. as of now I am about 50% used on the RAID1 array I use for the share that holds my rock-ons but I want to keep it under control if I can… I will do more search later and perhaps make a VM to test on.
-Nick
Given @suman 's response in the later I suspect that there is some improvements to be made, ie I think of late docker has added some ‘clean up’ features that we should be able to enable in the background in time, ie:
docker system df
docker system prune
And a more recently forum thread:
Hi,
I was wondering what is the logic of Rockstor when it comes to snapshots if there is no scheduled snapshots/tasks in place?
It seems as a kind of standard Rockstor performs (irregular?) snapshots for the “rockstor ons root” share, see below
[image]
126 snapshots are listed back - any need to define a task in deleting them after x month?
My NAS is used only for private usage, but I guess I should define a scheduled snapshot for my “main share” (1TB of data) or?
If so, any recommendatio…
I though we had a current issue open on this but can’t find it so I have created one in the docs repo as a start:
opened 07:25PM - 27 Mar 17 UTC
Thanks to knucklehead101, mrseth1, glenngould, MvL, and anborn in the following … forum threads for highlighting this issue. Since docker (Rockons) use snapshots to manage image layers their deletion will break installed Rock-ons. An explanation of their purpose and use will help to clarify their appearance in the Rock-ons-root share.
This explanation could be placed in the following doc section:
http://rockstor.com/docs/docker-based-rock-ons/overview.html#rockons-root
Referenced forum threads:
https://forum.rockstor.com/t/rock-ons-snapshot-removal/637
https://forum.rockstor.com/t/deleting-snapshots-breaks-rock-ons/2431
https://forum.rockstor.com/t/what-is-the-logic-of-rockstor-if-there-are-no-scheduled-snapshots/2928
https://forum.rockstor.com/t/solved-headphones-rock-on/3034
https://forum.rockstor.com/t/some-rock-ons-not-reachable-after-creating-new-rock-on-share-and-new-installation-of-rock-ons/8029
Possible background docker reference:
https://docs.docker.com/engine/userguide/storagedriver/btrfs-driver/
Hope that helps.