Getting total amount of restorable data

Ah ok, I read title and though you wish to know how much has been restored.  I do something similar for my own reporting... each day around 4am I create so called raw report like:

$MMINFO -avot -q 'sscreate>=yesterday 00:00:00,sscreate<=yesterday 23:59:59' -xc, -r 'totalsize,pool,group,client,sscreate(20),sscomp(20),level,name' | grep -v 'ss-created' |sort -u

From there I process output into archive and transaction logs, database backups, file system backups, NDMP and long-term backups (aka archives). I do this grouped for each tier we have (test, education, development, acceptance and production) from that single raw file and mail that to myself each morning with totals for volume and number of sessions. After summer vacation, I plan to load all raw files into reports with QlikView ti build historical overview and trends - but that will take some time due to other priorities.

This is tricky one as actions like staging and data cloning are seen as reads thus they can easily sneak in into restored data.

Your mminfo query is not coorected. What you do is reporting client,poll to which it wrote and summary size of what has been backed up and not read. To get how much has been read, add read like:

# mminfo -avot -xc"," -r"client,pool,sumsize(20),read(20),name" -q client=FOO -t '1 day ago' | tail -1

FOO,index,253932404,318 GB,bootstrap

Now, above already says this is not good too. Why? Well, sumsize above is shows in bytes so how can read be bigger than write?  The reason why is simple: read give aggregate value for reads from volume. And if you try to nail it down, it does not work.

So, at the end of the day, your best tool to get values per backup set are contained within logs where you can search for read actions and extract the value. I believe NMC in never versions contains these reports and I think I have seen mentioned this has been made better in NetWorker 8, but I didn't test it for that so far.

You also wish to ad- -av too... without it, only browsable save set will be shown (eg. no index and bootstrap and any other which might have browse time expired, but retention is still valid).

Hrvoje,

I would like to know how much space (total) is occupied in the backup backend; this is described as 'Restorable data' .

Sorry if this confusing, these terms are 'invented' by some commercial guys......

Ahh...

Think I got something:

Without  any  options,  mminfo displays information about the save sets

       that completed properly since the  previous  day's  midnight,  and  are

       still  contained  in  an on-line file index (browsable save sets).

So I need to specify a -t with the mminfo, e.g. -t"1 year ago" to be sure everything is included in the query. (Environment is operational for 8 months).

Hrvoje,

Thanks for your quick reply.

For my 'report quest' I am only interested in the amount of written data, not read data. I'd like to know at a certain point in time (say NOW) which client 'consumes' how much GB in the backup storage (before dedup&compression).

I guess the time (-t) is not relevant if I only need to know the amount of space occupied.