Limiting number of jobs that run or rather number of drives - Veritas Net Backup

This is a discussion on Limiting number of jobs that run or rather number of drives - Veritas Net Backup ; Hello All My aim is to switch netbackup configs on the fly such that I re-arrange the storage units : the true aim is so I can morph my backup infrastructure from a weekend mode to weekday mode: weekend is ...

+ Reply to Thread
Results 1 to 2 of 2

Thread: Limiting number of jobs that run or rather number of drives

  1. Limiting number of jobs that run or rather number of drives


    Hello All

    My aim is to switch netbackup configs on the fly such that I re-arrange
    the storage units : the true aim is so I can morph my backup infrastructure
    from a weekend mode to weekday mode: weekend is all drives up for DR backups
    , weekday is to use a reduced subset of drives for DR backups , leaving say
    3 drives for adhoc backups, online oracle backups, restores etc that run
    through the working day.

    Having read the doc, Im supposed to use bpsturep -cj to reduce the number
    of concurrent jobs associated with my storage unit. Ive done this and then
    used the storage unit and netbackup ignores my limit. That is, Ive set -cj=4,
    checked its changed, then I kick of a cron job which runs 18 backups: and
    all my tape drives in the storage unit get used. Ive not restarted any daemons.
    I was expecting 4 drives to get used, the remainder of the jobs queueing.

    The situation is quite complex: Ive got one robot, 9 drives, SSO across 5
    media servers. The job in question is initiated on the master server, but
    the jobs themselves run on floating addresses that exist on one of two (clsutered)
    servers.
    Netbackup 5.1mp5, L180, LTO2, all solaris.

    Thanks in advance for your thoughts,Jim

  2. Re: Limiting number of jobs that run or rather number of drives


    I experimented with

    ./bin/admincmd/bpsturep -label xxx-L180 -cj 4 ; ./bin/admincmd/bprdreq -rereadconfig


    and it worked perfectly. One hit...nothing else was running at the time and
    true enough four drives used, five free.

    This seems to be the way it ought to work but others have had headaches (
    as in "repeat this in a loop fifty times") getting it to be successful .
    I will see how reliable it is: assume it works perfectly unless I post a
    followup.

    Jim


    "jim dalton" wrote:
    >
    >Hello All
    >
    >My aim is to switch netbackup configs on the fly such that I re-arrange
    >the storage units : the true aim is so I can morph my backup infrastructure
    >from a weekend mode to weekday mode: weekend is all drives up for DR backups
    >, weekday is to use a reduced subset of drives for DR backups , leaving

    say
    >3 drives for adhoc backups, online oracle backups, restores etc that run
    >through the working day.
    >
    >Having read the doc, Im supposed to use bpsturep -cj to reduce the number
    >of concurrent jobs associated with my storage unit. Ive done this and then
    >used the storage unit and netbackup ignores my limit. That is, Ive set -cj=4,
    >checked its changed, then I kick of a cron job which runs 18 backups: and
    >all my tape drives in the storage unit get used. Ive not restarted any daemons.
    >I was expecting 4 drives to get used, the remainder of the jobs queueing.
    >
    >The situation is quite complex: Ive got one robot, 9 drives, SSO across

    5
    >media servers. The job in question is initiated on the master server, but
    >the jobs themselves run on floating addresses that exist on one of two (clsutered)
    >servers.
    >Netbackup 5.1mp5, L180, LTO2, all solaris.
    >
    >Thanks in advance for your thoughts,Jim



+ Reply to Thread