>>>>> On Fri, 21 Oct 2005 10:42:00 +0100, Dave Shield <[EMAIL PROTECTED]> said:

>> setaccess limitedgroup "" v2c noAuthNoPriv prefix log anything

Dave> What is this access view being applied to?
Dave> The OID of the trap?
Dave> The OIDs of the payload varbinds?
Dave> Or both?

Sorry.  That would be in the missing docs:  currently just the OID of
the trap.


>> # wrapper scripts to let xyzzy community log and execute
>> ipv4logcommunity -v execute xyzzy

Dave> I don't understand this one.
Dave> I thought you were proposing something along the lines of:

Dave> logcommunity   xyzzy
Dave> execcommunity  xyzzy

Dave> A little more explanation would be useful.

Well, it's not that trivial it turns out.  In the same way that
rocommunity and rwcommunity can't both be specified with the same
community, it's true for this as well.  The problem comes from the way
these wrappers always assume generation of a new group as well as just
the access line.

  ipv4logcommunity mycommunity

Adds a view for allowing traps from 'mycommunity' to be logged.  But
adding this next:

  ipv4executecommunity mycommunity

Will not work.  This is true for snmpd rocommunity/rwcommunity as well
where you can't specify rwcommunity twice with the same community and
two different OIDS to merge them together.  If you want to do
something that complex you need to use the real VACM tokens instead of
the easier wrappers.

The above example is a bad choice, my bad.  But all the wrapper tokens
have an additional flag that lets you add "extra" views into the
current view list.  IE this:

  ipv4logcommunity -v execute xyzzy

Means add xyzzy as a community string that is good for both log (in
the token name) and execute (the -v flag).  A better example would
have been:

  ipv4authcommunity -v log -v execute xyzzy

which is the exact same thing.  I realized the confusion of the other
one wasn't worth it, so the last example I think is more clear.  I'm
not sure I wanted the flag syntax (-v) or to use something like
log|execute or something...  All of the tokens are parsed by the same
thing, so the -v flag actually works across them all.  We probably
shouldn't advertise them for some of them though.  This is the one
thing I'm not happy with generally, but the options are all bad in
different ways.

There are ipv4 and ipv6 prefixes to indicate what the community is to
be used under for which transport, in the same way that we have
rocommunity and rocommunity6 today.  I think these names are nicer
than the snmpd equivalents though since they're more explicit.


-- 
Wes Hardaker
Sparta, Inc.


-------------------------------------------------------
This SF.Net email is sponsored by:
Power Architecture Resource Center: Free content, downloads, discussions,
and more. http://solutions.newsforge.com/ibmarch.tmpl
_______________________________________________
Net-snmp-coders mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/net-snmp-coders

Reply via email to