Searching for a Report - Risk

[Gormag]

I am looking for a script (or an idea on how to best create one) to gather and print the output for clients at "risk" based on number of days not backed up at a per file system level.

So lets say /, /usr, /ora1, all backed up, but /ora2 did not, I would be able to run the script, and see Server, Policy, Schedule, Path Not Backed up.

I am amazed that this is not something default part of NBU 5.X.

 

[comtec17]

There should be a command name bpcoverage located in the admincmd directory.

 

[creakyjoe]

in addition to comtec17's reply, i make sure to keep right on top of our backup infrastructure/back failures etc as the primary part of my job. it can be a right royal pain but avoids problems later on. And i have umpteen automated reports that run each morning too that check all manner of things; better still i have automated scripts that advise of things such as drives going down etc etc.

I must admit i'd forgotten about the bpcoverage script so am now building a script around it, as a just in case script.

 

[Gormag]

comtec17/creakyjoe,

bpcoverage is a great command, if you want to list file systems that are uncovered. This is not the effect I am searching for.

I may not have made myself clear, so I will try again.

Policy1 - Sched-Full - Run's Friday's
Policy1 - Sched-Inc - Run's all other days

Client1 in Policy1 failed Friday's Full, Saturday, Sunday and Monday's Inc for only the /db12 file system, all others were successfull.

I want to be able to generate a report that will show the client failed, for this (or any) file system, on this (or any policy and schedule).

This, in my opinion is not an extreme request.

I understand how monitoring jobs and success vs unsuccessfull jobs is "part of the job" but it is not always a managable option. If you take into account that many sites have multiple masters, some in excess of 20, and many of those masters run over 10,000 jobs a day, then simply watching for failed jobs is on the extreme.

I still find it difficult to understand how a report like this is not a default part of the package, and Advanced Reporter is not the answer, it is a joke. Same for Command Central, been there, done that.

I am finding my options limited to gathering data from "/usr/openv/netbackup/bin/admincmd/bpdbjobs -report -all_columns" saving it for the day, and then beating my head aginst the wall trying to learn perl to parse it because awk can not handel long strings.

I was just hoping someone had already created this wheel as I did not want to re-create it.

Call me lazy.. ;-)

Thanks,

 

[comtec17]

This sounds like a job for the Advanced Reporter Option

 

[Gormag]

Maybe you missed my statement above about AR.

Thanks for your input.

 

[creakyjoe]

i don't use AR, and wouldn't do either as we can achieve all monitorint/reporting via our own simple scripts. However for simply monitoring for failures, that's the easiest bit. WE don't tolerate/allow failures; what i mean is, while we can't avoid errors completely, we do have ALL netbackup completion/failure emails passed thru to our email system (Exchange). us Admins have rules such that any completion emails go to our wastebins (this way we can refer to them if need be) and any failures come to our Inbox. we use Blackberry's so provided email is working we can see failures as they happen (we have no staff in the office overnight or out of hours so it's down to us). we assess failures at the time and rerun if necessary/freeze tapes/down drives or whatever is needed. This would be trickier if we had 10,000 jobs or whatever but in that situation we would have people onsite, doing the same things. just my five pence..

 

[comtec17]

There is a product out there from Bocada called "Backup Reporter". This is very useful tool and does not cost much.

http://www.bocada.com/index.asp

 

[Gormag]

comtec17,

That may be the solution needed, thank you!

 

[OHWS]

Keep in mind if you use a third party scheduler that Advanced Reporter still may not buy you very much bang for you buck. We use a Tivoli scheduler for all of our Open Systems backup and the Veritas Scheduler for all of our Windows Server backups. This causes certain reports to be innacurate such as the At-Risk and the Backup Windows reports. Just an FYI.