How to report what is being backed up and included or excluded?

ldmwndletsm

ADSM.ORG Senior Member
Joined
Oct 30, 2019
Messages
232
Reaction score
5
Points
0
Running TSM 8.1.11 on Linux.

Let's suppose that you want to generate a report of the nodes and coresponsing file spaces that are on the backups, but you also want to report what is being included or excluded. How would you generate such a report that would cover both? For example, a Perl script could easily run the SQL on the DB2 filespaces and nodes tables to fetch a lot of the information, and it could parse the output, but this doesn't tell you if the given file space is still on the backups for that node, only the information for the last time it was backed up. Moreover, it gives you no information about what is being excluded on the given node. Seems you'd have to parse the client system options file (dsm.sys) on each node. That could get tricky.

You could either 1. remotely run some script on each client and collect the output somewhere to then parse or 2. have each client run a daily cron job to report this to some central location wherein some process then parses that information.

BUT is there instead a way to pull this information from the backup server without having to fiddle around with separarely accessing each client or having each client report this?

We do have the Command Center installed, but I've not played with it much. I find the command line easier for SQL, particularly with parenthetical expressions, not to mention the fact that other Unix commands (e.g. grep, sed, awk, pipes, etc.) are often necessary to filter the output. Is there a built-in report that could do this wherein we could provide the output to the users so they can see what is being backed up and/or excluded?
 
Let's suppose that you want to generate a report of the nodes and coresponsing file spaces that are on the backups, but you also want to report what is being included or excluded. How would you generate such a report that would cover both?
First half is easy, you can query the filespaces table to see which filespaces have been backed and when was the last time it was attempted and last time it was successful. But you can't report the include and excludes from the server as that info is not stored in the database. You would have to run "dsmc query inclexcl" on each individual client to report what's included and excluded.
 
Thank you. That's what I suspected. However, I had another idea, but it appears to have encountered a dead end.

If I run `dsmc -virtualnode=client_node_name` on the backup server then I can run commands like 'query filespace' and 'query backup', natch, and what gets reported does apper to only apply to the given node, not the backup server. Good. However, when I enter 'query inclexcl', it reports only the information for the backup server, not the named node. So that doesn't appear to work.

Am I missing something here ?

Anyway, I was thinking that if it did work then you could do all this on the command line, avoiding interactive, and fetch the information that way. But even if that was possible, it would be clunky to retrieve that for a large number of nodes, unless maybe it was a cron script where it would run once a day with no immediate completion timeframe required, but a quick script, no, whereas retrieving information from the filespaces table is very fast -- only takes a few seconds to report the desired information for all the nodes.

Do you know if the Operations Center is capable of reporting "excludes"? I doubt it.

It would be handy if IBM allowed the "excludes" to be configured on the server itself so everything is centralized, thus permitting such queries to be run from the server where you could report all of that and parse as necessary.
 
When you run "dsmc query inclexcl", it applies to the machine you run it on and for that node. That's because include/excludes come from 3 places:
- client option file
- operating system
- server client option set

The first 2 are unique to the machine you are running it on, the last one follows the node.

It would be handy if IBM allowed the "excludes" to be configured on the server itself so everything is centralized, thus permitting such queries to be run from the server where you could report all of that and parse as necessary.
You can with client option set, but that doesn't prevent other include/exclude from being added to the option file, and you can't override the operating system include/excludes. So that means you could only query 1 of the 3 sources from the server.

More info on client option set:
 
Back
Top