2008/5/8 Pamidipati Suresh-G20238 :
> Regarding the example you mentioned, why do you want to gather all the
> instances of the table in the 1st getnext itself when snmpwalk is given.?

I believe that Valantina is concerned about the data changing while the
walk is taking place.

For example, consider a table containing two columns and (initially)
two columns.
Walking this table would look like:

GETNEXT table --> this.1
GETNEXT this.1 --> this.2
GETNEXT this.2 --> that.1[*]
GETNEXT that.1 --> that.2
GETNEXT that.2 --> endOfTable

The client could then display the two rows quite happily:

this.1 that.1
this.2 that.2

But suppose that a new row (3) was created at exact moment that the
walk had got to the point marked[*]
The remaining sequence of GETNEXT requests would then be

GETNEXT that.1 --> that.2
GETNEXT that.2 --> that.3
GETNEXT that.3 --> endOfTable

leaving the client with the inconsistent table

this.1 that.1
this.2 that.2
(?) that.3

Alternatively, suppose the second row was deleted at[*]
The walk would then finish

GETNEXT that.1 --> endOfTable

giving the client a table:

this.1 that.1
this.2 (?)

Loading all the data at the start of the walk would allow the client
to retrieve a consistent set of information (albeit one that was
slightly out of date).


This SF.net email is sponsored by the 2008 JavaOne(SM) Conference
Don't miss this year's exciting event. There's still time to save $100.
Use priority code J8TL2D2.
Net-snmp-users mailing list
Please see the following page to unsubscribe or change other options: