I've think i've just hit a DataONTAP bug. Netapp 7.3.2 here. Snap list displays 0% everywhere, after several SMSAP(think SMO) snap restores from secondary (not Volume Restores but LUN restores directly in the FlexVol area), df reports this:
Filesystem total used avail capacity Mounted on /vol/sap_XXX_sapdata/ 53GB 43GB 9GB 83% /vol/sap_XXX_sapdata/ /vol/sap_XXX_sapdata/.snapshot 0GB 19GB 0GB ---% /vol/sap_XXX_sapdata/.snapshot
From Snapshot To KB changed Time Rate (KB/hour) --------------- -------------------- ----------- ------------ ---------------
smsap_XYZ_host1_XYZ_f_h_1_8a4a65ed2ddaa881012ddaa88ad50001_0 Active File System 21060124 1d 05:42 709088.452
If you take 10315488 kB and do math it is 9.837 GB since last snapshots (compared to AFS). The LUN size is 15GB, fractional_reserve=100%, LUN space reservation = enabled, snap reserve = 0%. There is 20.084 GB in snapshots, which is pretty close to what df .snapshot reports.
At least for me this is : LUN+LUN*fractional_reserve+space_used_by_snapshots= 15GB + 15GB + 20.084GB =~ 50 GB. Not sure how much of it is in fractional_reserver(BTW: how to check this?), i guess that at least: 19GB - (53GB - 43GB) = 19GB - 10GB = 9GB is in fractional_reserve, and this leaves me with 50GB-9GB = 41GB (pretty close value to the used space for Vol , as reported by df).
So why the hell snap list reports 0% everywhere? It looks like it doesn't support LUN snap restores "overwrite semantics"? Is this known "feature" ? In which version has this been patched?
Where does 7.3.5 come from? Three of these bugs should be fixed in version you have and fourth (237378) is not fixed in any (released) version. Looks like you really hit it; even more, looks like everyone doing SFSR should see this bug. I have to check ☺
Re: snap list returns 0% for every snapshot, after LUN secondary(primary too?) restores from SMO/SMSAP