One Policy for Three Different Backup SLAs
Helio Manufacturing migrated 180 clients from old NetWorker groups to a single policy 'Prod-Policy'. The policy has one workflow 'Nightly' containing all clients in group 'All-Prod' and one backup action 'Full-Every-Day' that runs daily at 22:00 to pool 'Default'; a clone action was added later but is disabled. Requirements: 20 database clients need weekly full and daily incremental with 8-week retention; 120 file clients need monthly full and daily incremental with 6-month retention; 40 web clients need weekly full with 4-week retention. All fulls must be cloned to tape and kept for 1 year. Symptoms: database backups overlap the file backup window; file clients never perform a monthly full; clone tapes are not created; 6 clients added last week have no backups.
1. Redesign the policy. Specify workflows, groups, schedules, pools, and actions, and explain why the current single action cannot satisfy the SLAs.
Read the worked answer
A single backup action has one schedule and one retention, so it cannot give DB, file, and web clients different levels and retentions. Create three workflows: WF-DB (group Prod-DB, start 22:00, backup action with Sunday full and Mon-Sat incremental, pool DB-Disk, retention 8 weeks); WF-FS (Prod-FS, start 23:00, first Sunday monthly full and daily incremental, pool FS-Disk, retention 6 months); WF-Web (Prod-Web, start 21:00, weekly full, pool Web-Disk, retention 4 weeks). Give each workflow a clone action to a matching tape pool with retention 1 year and a clone storage node that can read the source and write the tape. Separate start times and pools prevent DB overlap. The current 'Full-Every-Day' action explains why file clients never get a monthly full and why cloning never runs.
2. Explain browse versus retention and how to keep clone tapes for 1 year when the backup copy expires in 8 weeks. What must be set on the clone action and pool?
Read the worked answer
Browse period controls how long save-set entries remain visible for recovery; retention controls when the save set can be recycled. Browse cannot exceed retention. For the 8-week DB copy, set browse and retention to 8 weeks. For clones, set the clone action's retention to 1 year; otherwise the clone inherits the source retention and expires with it. Ensure the clone pool has enough volumes and that volume retention supports the 1-year requirement. The original and clone are separate save sets: the original expiring does not delete the clone. Monitor pool capacity and avoid reusing clone volumes before their retention expires.
3. Six clients added last week have no backups. List the checks in order and the commands or views you would use.
Read the worked answer
Check in order: (1) the client resource exists and Scheduled Backup is enabled; (2) the client belongs to the correct group, using nsradmin and printing type: NSR client; name: ...; (3) the workflow includes that group and is enabled; (4) the workflow start time and server time zone are correct; (5) the action schedule has a level for that day and the pool has writable devices; (6) the job monitor or policy log shows 'not in group' or no save sets. Use nsrwatch for active jobs and the NMC/NWUI policy monitoring view for workflow status. New clients most often fail because group membership is missing or scheduled backup is disabled.