I’ve been involved with several projects related to NetBackup reporting over the years. In each of those projects I have heard customers voice a concern about producing a report that tells them conclusively that their clients are protected.
It sounds like a very simple report to produce. I wish it were always simple.
Let’s look at a few of the gory details. First, which client list is canonical?
- All clients who are currently included in a NetBackup policy?
- All clients who have had a backup in the past have data retained in the NetBackup catalog?
- All clients who have had a job run during a certain time period.
Choosing any of the lists of clients has its issues. In an ideal world the lists above would all be the same. In reality they overlap, but rarely match 100%. For example, the act of de-commissioning a server may result in a ticket to remove the client from a policy, but backup images may remain on tape or disk for long periods of time depending on retentions. A client may be removed from a policy during a maintenance window and might not get added back into a policy.
Then there is another more list entirely outside of NetBackup. This is the list of all systems on your network which should be protected. Getting at that list in a dynamic environment requires careful planning. When a new system is provisioned how is the backup team alerted? I wince when I hear an answer like "When we get a restore request."
So what should you do?
To be continued …
Message Edited by Turlas on 04-15-2008 10:49 AM