Video Screencast Help
Give us your opinion and win with Symantec! Please help us by taking this survey to tell us about your experience with Symantec Connect, so that we can continue to grow and improve.  Take the survey.

Need Script to collect failed backup report

Created: 07 May 2013 | 8 comments

Hi All,

Can anybody help me to get a script to collect failed backup report on daily basis and last successful backup for the failed clients? Below are the overview of environment :

Master Server : Linux- Netbackup 7.1

Media Server (6 in no.) : AIX- Netbackup 7.1

Clients : Mostly- Netbackup 7.1

Operating Systems:

Comments 8 CommentsJump to latest comment

Nicolai's picture

You should consider to deploy OpsCenter for backup reporting.

http://eval.symantec.com/mktginfo/enterprise/white_papers/b-nbu_7_opscenter_analytics_WP.en-us.pdf

http://www.symantec.com/opscenter-analytics

Opscenter is free if you have Netbackup

OpsCenter analytics is licensed.

Assumption is the mother of all mess ups.

If this post answered your'e qustion -  Please mark as a soloution.

niteshry's picture

Thanks Nicolai for your suggesstion. But In my backup enviornment no OpsCenter is implemented and my client need a script to collect failed jobs.

Can anyone help me in this regard ?

Flako's picture

For some quick daily Reporting you might check these commands :

    bperror -backstat -by_statcode -hoursago 36 -U

    bperror -backstat -hoursago 36 -U|grep -v ' 0 '

    bperror -backstat -hoursago 78 -U|grep <client_name> [|grep <policy_name>] [|grep <schedule_name>]

Have fun,
Ingo

Andy Welburn's picture

We used the following (dirty) UNIX script on a daily basis utilising the bperror command mentioned by Flako (it does not, however, detail the last successful client backup - I'll leave that to you!) (EDIT: but you could incorporate, for each failed client, bpimagelist -U -client $CLIENT -d 12/31/12 | head -3 | tail -1 or something more appropriate to yourself?)

bperror -backstat -U| awk '($1 > 1) {print $0}'|cut -c8-20|grep -v \
CLIENT|sort -u > failed_clients
if [ ! -s /app/site/scripts/failed_clients ]
then
  echo "No clients have had any failed jobs in the last 24 hours"
  rm failed_clients
  exit
fi
echo "The following clients have had one or more failed jobs in the \
last 24 hours"
echo "**************************************************************\
*************"
echo
cat failed_clients|
while read CLIENT
do
  echo
  echo "JOBS RUN FOR CLIENT $CLIENT"
  echo "*********************************"
  bperror -backstat -client $CLIENT| cut -d " " -f6 | sort -u > \
failed_jobid
  cat failed_jobid|
        while read JOB_ID
        do
                echo
                echo "JOB_ID $JOB_ID"
                bperror -backstat -U -jobid $JOB_ID
        done
done
rm failed_clients
rm failed_jobid

It *will* also report if said failed job re-started & completed successfully (using the same job_id)

(***EDIT #2: just realised OP is nearly 2 months old & no return since - hopefully managed to work something out!)

Flako's picture

First i do a "bperror -backstat -hoursago 48 -U |grep -v ' 0 '"

In a script i read the client, the policy and the schedule of the failing job.

For every failing client/policy/schedule i do a

"bperror -backstat -U -hoursago ?? -client <client> | grep <policy>  | grep <Schedule>"

That way i see if and when was another run of the failing job.

i.e.:

root@klnnbus1:/ :-) bperror -backstat -U -hoursago 72 -client dsnklnp740-2b
STATUS CLIENT        POLICY           SCHED      SERVER      TIME COMPLETED
  0    dsnklnp740-2b OS-CUM-NEW       CINC_Daily dsnklnt4-2  07/19/2013 17:08:30
 59    dsnklnp740-2b OS-CUM-NEW       FULL_Sat   dsnklnt4-2  07/20/2013 13:06:18
                 (access to the client was not allowed)
 59    dsnklnp740-2b OS-CUM-NEW       FULL_Sat   dsnklnt4-2  07/20/2013 14:10:54
                 (access to the client was not allowed)
 59    dsnklnp740-2b OS-CUM-NEW       FULL_Sat   dsnklnt4-2  07/20/2013 14:57:33
                 (access to the client was not allowed)
  0    dsnklnp740-2b OS-CUM-NEW       FULL_Sat   dsnklnt4-2  07/22/2013 09:46:55
root@klnnbus1:/ :-)

StefanosM's picture

if someone has a way to get a report - opscenter, UNIX shell, Perl whatever - of failed backups and not failed jobs, please share it with us. l'm begging symantec for this report for years.
To clear it more, I do not want to see at the report a job that failed and the succeeded. This is a succesfull backup

Andy Welburn's picture

Hmmm ..... have seen jobs fail & show as failed in the Actiivity Monitor but subsequently re-start as a different job ids & succeed - that's gonna take scripting beyond my basic skills!

Flako's picture

A report of "failed backup and not failed jobs"  ??

Thats what parameter "-backstat" in the bperror CMD means "back(up)stat(us)" .

You dont have to mess with job ids or restarts or whatever.

Just do the

"bperror -backstat -U -hoursago ?? -client <client> | grep <policy>  | grep <Schedule>"

and you get a Date Sorted List of the Backups for this client/policy/schedule.

But you are right, there may be several ways to get to a nice list.     ;-)