Snapshot metadata is one thing and backup data is another. What goes to data backup pool (and if it goes in the first place) is controlled by your snapshot policy and backup snapshots field. It sounds as you do not have this set (or set to none). If so, change that field to All.
Well, and that's the part that I didn't quite understand in the first place... The snapshot policy that I currently have defined is:
Number of snapshots 4
Retain snapshots 2
Snapshot expiration Day
Backup snapshots: All
When I run the group, it goes through, backups everything up to the snapshot directory on the storage node, and then just sits there (as opposed to finishing like a normal job). It doesn't try to spin off to tape.
Also, on the group, for Advanced I have the interval set to 4 hrs. I assume that means that snapshots happened every 4 hours?
Ok, one step further (maybe) - I just realized that the group is moving the INDEX of the job over to tape, but not the data itself. Is that what is intended? I only have enough room on the backup server for 3 or so snapshots, so I really need those to move to tape as soon as they're done.
I miss the information from where snapshots should move to tape.
Do you mean from your AdvFile device to tape? This is a staging function you have to create separately.
(and I assume this is what you mean)
Or do you mean from Windows server to NetWorker backup media? This function is called snapshot rollover.
With the snapshot policy you pasted here all snapshots should be moved to NW backup media (snapshot rollover).
Is this done correctly?
But they may reside on your AdvFile device now, this is still correct.
If you would like to get them on tape media you will either have to clone them, assuming your AdvFile device is big enough to keep data for whole lifetime till expiration occurs.
Or if they cannot be kept there all time you will have to create a staging resource to move them to tape media instead of copying them by clone job.
Do you have any pending jobs? I find it a bit strange no data is saved since your snapshot policy says it should be saved. Actually, you can check with mminfo what gets saved and where. I do not use NMM, but I use PowerSnap so we both use snapshot policies in pretty same manner (as far as theory goes).
You are basically asking NMM to take 4 snapshots per day, retain 2 and roll over to backup device all.
I don't quite understand the logic behind that.
Do you really need to save to tape (or any other backup device) the 4 snapshots or is it just to have a quick PIT backup to restore?
Also, taking 4 snapshots and retaining 2, doesn't make sense if you are backing up ALL, right?
Maybe you would like to roll over only 2, instead of all, and then you have 2 in tape (let's say first and last), and the other 2 are stored locally in the storage in case of an urgent PIT restore.
Give a think at that policy and let me know if what I'm saying makes sense
I'll be happy to give you a hand selecting the correct snapshot policy (even if we have to create it manually).
ble1
4 Operator
•
14.4K Posts
0
October 23rd, 2013 14:00
Snapshot metadata is one thing and backup data is another. What goes to data backup pool (and if it goes in the first place) is controlled by your snapshot policy and backup snapshots field. It sounds as you do not have this set (or set to none). If so, change that field to All.
jeremyhuber
3 Posts
0
October 24th, 2013 08:00
Well, and that's the part that I didn't quite understand in the first place... The snapshot policy that I currently have defined is:
Number of snapshots 4
Retain snapshots 2
Snapshot expiration Day
Backup snapshots: All
When I run the group, it goes through, backups everything up to the snapshot directory on the storage node, and then just sits there (as opposed to finishing like a normal job). It doesn't try to spin off to tape.
Also, on the group, for Advanced I have the interval set to 4 hrs. I assume that means that snapshots happened every 4 hours?
jeremyhuber
3 Posts
0
October 24th, 2013 08:00
Ok, one step further (maybe) - I just realized that the group is moving the INDEX of the job over to tape, but not the data itself. Is that what is intended? I only have enough room on the backup server for 3 or so snapshots, so I really need those to move to tape as soon as they're done.
guenterH1
117 Posts
0
October 25th, 2013 08:00
I miss the information from where snapshots should move to tape.
Do you mean from your AdvFile device to tape? This is a staging function you have to create separately.
(and I assume this is what you mean)
Or do you mean from Windows server to NetWorker backup media? This function is called snapshot rollover.
With the snapshot policy you pasted here all snapshots should be moved to NW backup media (snapshot rollover).
Is this done correctly?
But they may reside on your AdvFile device now, this is still correct.
If you would like to get them on tape media you will either have to clone them, assuming your AdvFile device is big enough to keep data for whole lifetime till expiration occurs.
Or if they cannot be kept there all time you will have to create a staging resource to move them to tape media instead of copying them by clone job.
hth.
ble1
4 Operator
•
14.4K Posts
0
October 28th, 2013 17:00
Do you have any pending jobs? I find it a bit strange no data is saved since your snapshot policy says it should be saved. Actually, you can check with mminfo what gets saved and where. I do not use NMM, but I use PowerSnap so we both use snapshot policies in pretty same manner (as far as theory goes).
CarlosRojas
1.7K Posts
0
October 29th, 2013 07:00
Hi Jeremy,
I believe this is because of the snapshot policy.
You are basically asking NMM to take 4 snapshots per day, retain 2 and roll over to backup device all.
I don't quite understand the logic behind that.
Do you really need to save to tape (or any other backup device) the 4 snapshots or is it just to have a quick PIT backup to restore?
Also, taking 4 snapshots and retaining 2, doesn't make sense if you are backing up ALL, right?
Maybe you would like to roll over only 2, instead of all, and then you have 2 in tape (let's say first and last), and the other 2 are stored locally in the storage in case of an urgent PIT restore.
Give a think at that policy and let me know if what I'm saying makes sense
I'll be happy to give you a hand selecting the correct snapshot policy (even if we have to create it manually).
Thank you,
Carlos