Unsolved
This post is more than 5 years old
20 Posts
1
6534
View total backup size protected by Networker
Hello all.
There are some ways to get the full backup size that is protected by Networker through reports and command line but when you use VMware backup protection with Data Domain you have to make a lot of reports and at the end you cannot say what is the total backup size. Can anyone point out an easy way to get that info so we can review the total backup size protected and check it with capacity based license?
Thank you in advanced
ble1
14.3K Posts
0
January 12th, 2016 10:00
Are you referring to mminfo -k stuff or something else?
bingo.1
2.4K Posts
0
January 12th, 2016 11:00
A very easy way is to run "mminfo -X" ... but it does not tell you how NW will get these numbers.
The other method is to query the MDB for all fulls in the backup cycle like ...
mminfo -q "savetime>=start_savetime,savetime
... and summarize the values in the column "totalsize" using a script. You can do this very easily with PowerShell.
Do not forget that you need to devide this number as follows:
size[TB] = total_bytes/1024/1000/1000/1000 not
size[TB] = total_bytes/1024/1024/1024/1024 as this would result in "tibibytes"
Kleinenbroich
86 Posts
1
January 14th, 2016 08:00
Maybe Alexandros is looking for a solution for the same problem as I have: How to calculate the amount of used capacity license. According to EMC definition: "The capacity is measured as the largest aggregate of full backups or synthetic full backups (which is the combination of full backups plus incremental backups) that are performed for all protected data by the NetWorker software over a two-month period (60 days)." My understanding of this sentence is how to get this aggregation? Use mminfo to get all full backups during the last 60 days, for each save set select the highest amount and sum those values?
If this is correct:
- How can I do?
- How do I have to deal with servers which are protected by both, NetWorker client and VMware Backup Appliance(VBA)? Use only the higher value or both?
ble1
14.3K Posts
0
January 14th, 2016 14:00
I don't use capacity license, but I wonder about following statement: "...over a two-month period (60 days).". Let me explain this further. What does this mean if my retention exceeds 60 days. What does it mean if I do full backup on daily basis? What does it mean if I use smaller retention? I would argue that operational backup retentions in majority cases are below 60 days. Surely, older standard by the end of last century was to keep data for 1 month which may result in 60 days logic, but I think many people decreased their data retention since (at least for operational backups). So, if my retention is 2 weeks, with full once per week, my backup cycle is around 3 weeks - which is 21 day and that is way below 60 days. So, I'm not sure I understand how this is calculated, but I assume 60 days is max and otherwise would need to take into account smaller backup cycle, This goes to show, or at least opens possibilities, to have pros and cons with capacity licensing depending on your retention and backup level (in combination with incr delta) in combination.
bingo.1
2.4K Posts
0
January 14th, 2016 17:00
Hi Hrvoje,
as often it is not easy to understand. That's why you need to install a monitoring server which will do that.
The issue is that you have to monitor the 'activity over time' (statistics) which of course has not anything to do with the retention.
Let's assume a backup volume of 50TB (fulls only) per week. So over a period of 2 months NW will have backed up about 8 x 50TB (fulls). No matter which retention one will use, the backup statistics per backup cycle (usually a week) will not change. Correct?
The monitoring period of 2 months is only used to collect the numbers over a longer period to make sure that peaks will not affect the overall statistical values.
PapadopoulosA
20 Posts
0
January 14th, 2016 22:00
This is what i mean,Kleinenbroich.
ble1
14.3K Posts
1
January 15th, 2016 03:00
Sure, but is that fair? If you use 2 weeks retention and vast majority of capacity is taken by database fulls which happen daily (so they are retained two weeks), it turns out you pay for capacity as if you use 2 months retention. Unless price difference between legacy model and capacity one is such that it takes into account that, then legacy mode of licensing would be still preferred by those who use smaller retention. My view of capacity license, in theory, is that it should be calculated on retained data and capacity used during the period which is retained. Of course, some upper limit might have to exist for environments that have long-term backups as otherwise they would have that as penalty, but for operational backups I would rather see is correctly calculated.