34 Posts
0
1153
index save set level and flag
Hello,
I have a tape storing some save sets which clone retention time has expired but it is still in "recoverable" state. so the tape is still not recyclable and re-usable yet. These save sets are index save sets:
Client SaveSet SSID SaveTime CloneRetentionTime Level Status Size Flags
hvbra02b index:nasbra01 2072880661 5/1/21 8:00:35 PM 8/1/21 full recoverable 13 GB cr
hvbra02b index:dbbra07b 2089657877 5/1/21 8:00:34 PM 8/1/21 full recoverable 267 MB cr
hvbra02b index:dbbra05b 2106435093 5/1/21 8:00:33 PM 8/1/21 full recoverable 262 MB cr
hvbra02b index:prnbra02 2123212309 5/1/21 8:00:32 PM 8/1/21 full recoverable 309 MB cr
tsebra02 DISASTER_RECOVERY:\ 178093172 5/13/21 8:22:25 PM 8/13/21 incr recyclable 105 KB cE
tsebra02 WINDOWS ROLES AND FEATURES:\ 194870383 5/13/21 8:22:20 PM 8/13/21 incr recyclable 5723 KB cE
Can someone clarify to me when these index save sets will become recyclable. It seems it depends on the level because when I looked for this specific save sets I found that level "1" ones are expired:
mminfo -avot -q name=index:issbra02 -r "volume(20), client(30), name(8), ssretent, clretent, level, ssid, sumflags"
360273L6 hvbra02b index:issbra02 8/1/2021 8/1/2021 full 2022549013 cr
360273L6 hvbra02b index:issbra02 8/2/2021 8/2/2021 1 2861496249 cE
360273L6 hvbra02b index:issbra02 8/3/2021 8/3/2021 1 1217415421 cE
360273L6 hvbra02b index:issbra02 8/4/2021 8/4/2021 1 2274466411 cE
360273L6 hvbra02b index:issbra02 8/5/2021 8/5/2021 1 2828200890 cE
Will these full index save sets ever become recyclable also? if so when?
Thank you.
crazyrov
4 Operator
4 Operator
•
1.3K Posts
1
August 26th, 2021 23:00
@hend_bs, A saveset (index or otherwise) goes into an expired state when it crosses its retention period and when there are no other savesets depend on it. For example: A full saveset does not expire even though its retention has crossed if there are any other level or incremental savesets that are depended on it.
hend_bs
34 Posts
0
August 27th, 2021 01:00
Aha got it. Thank you
bingo.1
2.4K Posts
0
August 27th, 2021 03:00
Of course, crazyrov's answer is true and valid. But let me make another comment.
As you can easily verify, the index backups following the full are set to level 1 (which is correct) but they already expired . This is wrong unless you changed their status manually. Usually, their correct status should also be recoverable. So if you did not manipulate their status' this could point to a bug.
barry_beckers
393 Posts
0
September 5th, 2021 09:00
If there is a chain still depending on the full, it should indeed show for all ssid's depending the same state, not that all inc. made after it would have already expired as then there would no longer be a dependency on an already expired full.
Same for regular filesystem backups or NW NMM backup that knows about a backup chain and its dependencies, unlike many other module backups.
You did not use the option, introduced in nw19 of O recall correctly to always keep the last full?
This as your query does not state any index backups expiring after aug. 5th? Or wasn't it a current query? So were there backups of this client after may 5th?
Haven't looked into that always-keep-last-full-backup feature and its ins and outs yet to see if that also applies therefor to the corresponding index entry?