Re: ✘Draft GPSD-MIB

[Gary E. Miller]

Yo Greg!

On Thu, 07 Jul 2022 18:52:49 -0400
Greg Troxel <gdt@lexort.com> wrote:

> "Gary E. Miller" <gem@rellim.com> writes:
> 
> >> I am not following at all.   Having management status available is
> >> separate from making the actual position available.  
> >
> > Is it?  Not in the Draft GPSD-MIB.  
> 
> It should be separate, or rather position not included.  That's
> management plane vs data plane.  For example, a mail server's MIB
> would include counts of messages but not their contents.

IMHO everything gpsd has to say is "contents".  What is the point of
talking to gpsd if you can't get its position?

Currently gpssnmp only supports nSat, uSat and a few similar status
variables, but no reason to limit it to that.  gpssnmp can only report
what is in the gpsd JSON and anyone locally can read the gpsd JSON.

> >>   I would expect
> >> normal practice is to have gpsd make the gpsd socket data
> >> available on localhost only.  
> >
> > Yes.  Ditto for gpssnmp.  
> 
> Normal practice in SNMP is to expose it to some amount of network,

Not my normal practice.  I have snmpd configured to only answer to localhost
so that Zabbix Agent2 can access it.  Agent2 then communicates with the
Zabbix server.

> more so than gpsd.  Or at least it used to be.  People may have local
> agents and then move to some other inter-machine protocol.

I can see gpsd allowing both.  gpssnmp could use the AgentX protocol
with snmp and let snmpd do all the access control, rate limiting,
network filtering, etc.

But that is a long way down this road.

RGDS
GARY
---------------------------------------------------------------------------
Gary E. Miller Rellim 109 NW Wilmington Ave., Suite E, Bend, OR 97703
        gem@rellim.com  Tel:+1 541 382 8588

            Veritas liberabit vos. -- Quid est veritas?
    "If you can't measure it, you can't improve it." - Lord Kelvin

pgp3RSJTX7k9M.pgp
Description: OpenPGP digital signature