On 10/10/06, Bruce Shaw <[EMAIL PROTECTED]> wrote:
> >3) hrStorageIndexes

> Locking the beginning of the disk entries into index 30 assumes nobody would
> 'be stupid'/'is insane'/'has a server overworked' enough to have more than
> 29 swap entries.  That's not a good assumption.

Agreed.

One approach would be for the disk entries to default to starting at
30, but allow extra swap entries to push this starting point onwards.

An alternative would be to use 10-29 for swap, then 30+ for disks,
followed by any "overflow" swap.   That would probably be my
preference, though it's harder to code.


>  There's also nothing in the MIB about it.

There can't be.  HostRes indexes are completely arbitrary (and don't
even need to be consistent from one run of the agent to the next).
Having things come in blocks is simply a convenience for the
programmer (and the N/W admin)

>  Why can't they just be contiguous?

The downside of that is that it makes disk indexes unhelpfully changeable.
Adding a new swap file or partition would affect all the disk settings
- either immediately (which probably isn't allowed) or when the agent
next starts up (which is confusing).

Having a fixed starting point makes the common situation much easier
to deal with.

I propose that this is something to think about for future releases
(either 5.4.x or 5.5, I'm not immediately sure which).  But it's not
worth delaying the 5.4 release for.
(Now where have you heard that comment before! :-))

Dave

-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys -- and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
_______________________________________________
Net-snmp-coders mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/net-snmp-coders

Reply via email to