nsrjobd
extends the capabilities previously offered by nsrexec.
An example of where
nsrjobd's
capabilities are taken advantage of is when
savegrp(1m)
requires execution of
save(1m)
and
savefs(1m).
In addition to the remote spawning,
nsrjobd
collects run-time information to be used by the
NetWorker GUI to report and monitor on both actively running as well as completed jobs.
For storing job related information, nsrjobd maintains its own RAP database in /nsr/res/jobsdb. To prevent this database from constantly growing, there is an upper limit of size and a retention period placed on the data in the database. Data pertaining to completed jobs will be migrated to an SQL database maintained by the NetWorker GUI, then deemed eligible for purging from the RAP database. The values for the retention period and database size are stored in the NSR Resource in nsrd's RAP database. They are configurable by the administrator. In contrast to NetWorker's RAP database, nsrjobd's database is considered an opaque data store for nsrjobd's private use, and thus no tools are provided for viewing or manipulating its contents.
nsrjobd is started and stopped automatically when nsrd(1m) is started and shut down respectively. It is not meant to be started manually.
nsrjobd uses the client side nsrexecd(1m) for remote execution, so the NetWorker server requesting command execution must be in the client's servers file.