Deleting files, shares & snapshots, space still not free OK

[littauer]

RN102 w/ 3TB, 6.1.9

Over the last few months I've added and deleted shares and turned off snapshot and deleted all the snapshots, usually with the GUI. File deletions were w/ the guest OS facilities (mostly openSuSE Linux).

I have a little over 0.5 TB of live data but the GUI says I'm using 1.4TB.

Using ssh I've hunted around and found a lot of data in /data/.purge/ that I recognize from a deleted share and way more in /data/._share/<sharename>/.snapshot/

1) How may these files be safely deleted?

2) How can I prevent a recurrence?

Thanks!

[EskenderNG]

Hello,

Before working on the system using ssh I would advise to run a balance (found on admin page: System > Settings > Volume Schedule) and see if this helps. This will allow re-allocation of data and metadata chunks previously used by the filesystem.

Bye,
Eskender

[mdgm-ntgr]

littauer wrote:
Using ssh I've hunted around and found a lot of data in /data/.purge/

littauer wrote:
that I recognize from a deleted share and way more in /data/._share/<sharename>/.snapshot/

So these snapshots no longer show in the web admin interface? The process for deleting these is different to deleting files.

littauer wrote:
1) How may these files be safely deleted?

See above. I can help with this if you like.

littauer wrote:
2) How can I prevent a recurrence?

Some logs would be helpful (see the Sending Logs link in my sig).

[littauer]

Thanks very much to both of you for the quick response.

@Eskender: I don't see System > Settings > Volume Schedule on my unit. Is this something added since 6.1.9? Otherwise sounds like a good idea. I *DID* try a reboot without any change. I'm wary of mucking about with someone else's setup without guidance as I have much scar tissue from having to clean up others' bad assumptions.

@mdgm: Yes, I understand that deleting files is different from deleting snapshots and yes the snapshots (with the exception of a few created for an empty share... I missed one) are all gone from the web interface.

I probably have a few very large (>1GB) files like isos or large outlook psts and will use a variation of your nice script to nuke them. I hadn't thought that having to truncate >1GB files before deletion was still an issue...

I expect, though, that the issue is a a huge number of tiny files. When you get to looking at the history in the logs (which I'll ship out when I finish typing this) you'll see I did a number of du -s * commands which I ctrl-Ced out of due to the amount of time being taken. Do you have any good advice for that? I worry about running out of buffer space using something like rm -r /data/._share/<sharename>/.snapshot/* and leaving the box in an iffy state.

[littauer]

All better now. mdgm led me through cleaning up the improperly undeleted snapshot entries. Future firmware updates are expected to make the problem go away.

What seems to have gotten me in trouble is ever-increasing numbers of emails, cookies, cache entries etc. from users using rsync to backup without the --delete option.