Hey all,
We are implementing a modified version of draft-ietf-sip-mib-09.txt,
where we've stripped out the use of the NETWORK-SERVICES-MIB and are
adding X.731 state information (similar to what is covered in RFC 4268)
and additional information we feel we need to capture to successfully
manage an openSER (http://openser.org/) instance. For example, tracking
active phone calls is something we need that isn't supported by the
SIP-*-MIBs.
Most of the data we need to expose is pretty straightforward and doesn't
require a lot of ingenuity. However, there are two tables in the
SIP-SERVER-MIB that are potential problems: sipRegUserTable and
sipContactTable. The sipRegUserTable has a list of all registrations on
the server (think one per SIP phone to make life easier). That list is
potentially unbounded, limited only by available memory and storage.
The sipContactTable can potentially be worse, as it represents contact
information per registration. A single registration can have multiple
contact rows.
The proposed communication mechanism with openSER is to use the official
FIFO interface, where essentially we shove some text into the FIFO and
we get a text response back. At the moment, we've decided to not add
AgentX support directly into openSER. So all of the data is going to
live in the openSER instance and we will be poking it via the FIFO to
expose it in the SNMP agent. No problem so far. Now I need to manage
the sipRegUserTable, where I don't have the list of current
registrations in memory. Querying for that list (just the index
information, not the entire table row) may be prohibitive if it's
several thousand entries. I don't know at this time if the module that
tracks reservations will be our own, or one of the stock modules from
openSER. If it's our own, we can likely keep the list of registrations
in a sorted order; if it's not, we can't depend on the registrations
being properly sorted.
Any suggestions on what approach I should take? While the FAQ entries
for mib2c are helpful, and point me to the iterate or mfd config files
to deal with tables that have external data, I was hoping someone could
point out in advance whether or not the standard approaches will scale
to the 10,000+ row case.
I'll poke away at it myself and report back whatever progress I make.
Thanks,
Glenn McAllister
SOMA Networks, Inc.
-------------------------------------------------------
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-coders mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/net-snmp-coders
