Re: MAXIMUN SIZE of a Display String

On Tue, 2005-12-06 at 11:35 +0100, Gemma Sanchez wrote:

> I have defined all my display strings with a maximum length of 1492
> bytes and it works ok. Now, after reading rfc 1157 I am very confused
> because I wonder how could a data of 1492 bytes fit in a message of
> 484 bytes.

You are quite correct - a 1492 byte string obviously cannot fit into
a 484 byte packet.  But if you look closely at the text you quote,
it doesn't say that packets can't be bigger than this:

  "An implementation of this protocol need not accept messages whose
   length exceeds 484 octets...."     ^^^^

(i.e. it is allowed to be petty minded, and insist on 484 max)


  " However, it is recommended that implementations support larger
    datagrams whenever feasible."

(i.e. it is petty minded to do so!)

Most SNMP implementations (including Net-SNMP) will accept larger
requests.   But 484 is a lower bound - *every* SNMP system should
accept requests of this size.



> Robert Story [said]
> 1492  should be a safe size that will avoid fragmentation on most
> networks."

That's probabably still a fairly conservative suggestion for maximum
packet size.  I suspect that much larger requests would normally
work quite happily.  They may well be fragmented and re-assembled
en route (which increases the possibility of problems), but most
of the time they should work fine.

  Of course, Network Management is often most needed when things
*aren't* working fine - hence the general trend to keep things
small wherever possible.


>    I am sure the board will answer me reason of the election of 
> this maximum size of 1492 for my display strings. could you give me a
> good answer for this quiestion?

I'm not sure there is a good answer, because it's not actually
a very good choice, IMO.

Note that both of the sizes above are concerned with the size of
a *request*  (including the encoding of the various values, the
OIDs and type information and all the header stuff).  The string
value itself is only one part of this.
   So defining a maximum string size of 1492 in order that it will
fit into a packet of size 1492 bytes, just doesn't hold water.
'Cos it won't!


If you're deciding how big to make the strings, the first thing
to ask yourself is what sort of values will they hold, and how
are they to be used?  That may suggest suitable limits.
  Also, are any of them going to be used as table indexes?
That will also limit the length of such strings (as an OID
cannot be >128 subidentifiers).

  But using maximum packet sizes isn't really very sensible.

Dave



-------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc. Do you grep through log files
for problems?  Stop!  Download the new AJAX search engine that makes
searching your log files as easy as surfing the  web.  DOWNLOAD SPLUNK!
http://ads.osdn.com/?ad_id=7637&alloc_id=16865&op=click
_______________________________________________
Net-snmp-users mailing list
[email protected]
Please see the following page to unsubscribe or change other options:
https://lists.sourceforge.net/lists/listinfo/net-snmp-users