[OZAPRS] Are 4 x iGates in Sydney too many?

Ray Wells vk2tv at exemail.com.au
Mon May 26 10:14:01 EST 2014

I'm happy, what's wrong with the rest of you <grin>

A simple question has now branched in three directions; number of 
igates, server load, and paths so lets consider each of them 
individually, because they should not be grouped together.

I think you need to ask each operator why *they* thought it was 
necessary to add their gateway to the network. Is an extra gateway or 
two such really an issue, in the absence of a truly wide area system, 
when greater Sydney covers such a huge area with diverse geographic 
features such as mountains, floodplain, coastal strip, cities with tall 
buildings, etc? Mandating for maximum numbers of igates can only ever be 
a recommendation which is not enforceable, but even if such a scheme 
could be implemented, who would determine which stations would be 
granted the right to operate an igate and on what basis would that 
station be chosen over others. I believe that such decisions are best 
left to the amateurs in a given area to decide their needs, rather than 
a clueless committee with no idea of the "local" situation. The people 
at the coal face know best their local requirements.

Scott Evans, who hosts a T2 server, has already told us that server load 
is a non-issue, with the proviso that rotate servers are used.

Australia has chosen the same path recommendations as most of the rest 
of the world, and uses the WIDEn-N paradigm which was devised by Stephen 
Smith, WA8LMF, as a means of improving channel carrying capacity. The 
original RELAY and WIDE system caused horrendous duplication of digied 
packets. Stephen lives near Los Angeles which has the greatest aprs 
density of any place on earth. The WIDEn-N paradigm limits the 
uncontrolled spread of packets that existed with the old scheme and, 
when properly implemented gives the best possible utilisation of a given 
aprs channel. I recommend you to this page by Stephen Smith for an 
explanation ... http://wa8lmf.net/DigiPaths/

The recommended vhf paths for mobiles in Oz is WIDE1-1, WIDE2-1 which 
gives two hops. ( http://www.aprs.net.au/vhf/national-2m-aprs-network). 
For more remote areas, WIDE1-1, WIDE2-2 which gives three hops, is a 
suggested alternative. Combined with intelligent digi software that 
detects duplicated packets, the system works quite well, given that all 
digis are responding to the same aliases. To further control the spread 
of packets, and to put the brakes on operators who want to try to use 
unrealistic paths, intelligent digi software such as UIDigi can truncate 
long path requests. Paths of WIDE3-3, WIDE4-4, WIDE5-5, WIDE6-6 and 
WIDE7-7 are repeated by a smart digi but in a way that enforces "the 
LAST hop" for such packets. The TAPR aprs mailing list is a good source 
of discussion of such things as paths and packet duplication and Google 
can be your friend.

Ray vk2tv

On 25/05/14 21:44, Justin Albury wrote:
> We are a bloody hard bunch to keep happy ;-)
> So we all love to be able to log onto aprs.fi and see our gated beacons and track others like we do, but then when we hit a black spot with no digi or igate we are all thinking of ways to rectify this.
> Goulburn and Highrange were both "black spots" along the highway until I install the digi's and a igate in Goulburn and everyone was happy, messaging works, weather warnings.... the whole bit.  A few dup packet here and there is the least of or issues...... I log every single packet gated daily, the log looks long but the file size is small and is @ 1200 baud ......Im sure a few extra igates and some dup packets is not going to make the aprsis server melt to the ground ......maybe im wrong i dunno.  What is good is to see the system getting use.....we would all hate to see it fade away like packet did....that would be sad.
> Do we have a AU config standard???  do all the digi's and igates need to be updated???  this is prob a good question.
> love our hobby....dont shoot it down......but lets make it work the best we can with each others help!
> '73
> Justin Albury
> J Albury Communications
> justin at jacomms.com
> 0417246791
> -----Original Message-----
> From: ozaprs-bounces at aprs.net.au [mailto:ozaprs-bounces at aprs.net.au] On Behalf Of vk2ycj at bravo.net.au
> Sent: Sunday, 25 May 2014 8:19 PM
> To: Australian APRS Users
> Subject: Re: [OZAPRS] Are 4 x iGates in Sydney too many?
> Thanks Ray,
> The capture effect of FM and the free for all that happens on a VHF simplex channel is probably difficult to overcome.
> Thanks for the description Ray.
> Jamie VK2YCJ
>> Guys and Gals,
>> Take a look at recommended paths and you'll see why multiple digis are
>> retransmitting a station.
>> A mobile with WIDE1-1 WIDE2-1 will have two hops. Most, if not all,
>> digis respond to WIDE1-1, so most digis will retransmit a WIDE1-1
>> packet for another hop (to any other digis that hears it) and that
>> (those) digis will use the WIDE2-1 part of the path. A path of WIDE1-1
>> WIDE2-2 will have three hops, extending the spread to another ring of
>> digis and potentially more igates.
>> The reason for the closest igate not always being the one that appears
>> on aprs.fi (et al) is that the nearest igate might be clobbered by
>> another station that is stronger than the local digi - think FM
>> capture effect - OR the received packet might be mangled in some way
>> through collisions, and "lost", whilst a distant igate will receive a
>> valid signal via its local digi and deal with it appropriately.
>> APRS is a bit (a lot!) of a free-for-all for channel access compared
>> with connected mode packet where stations hold back transmitting, as
>> defined by the channel access parameters of DWAIT, Slottime and
>> Ppersistence.
>> A situation I see reasonably often is, I can be gating a station
>> that's using vk2rcn-1 just fine, sending numerous packets to aprs-is
>> and then, out of the blue a packet will go via a couple more digis to
>> be gated by a station in Tamworth. I'm in Kempsey. Obviously my rx has
>> been clobbered by vk2rdo-3, vk2rmy-1 or vk2jub-3 and I lost the packet.
>> Ray vk2tv
>> On 25/05/14 17:12, vk2ycj at bravo.net.au wrote:
>>> Gary,
>>> Interesting yes.   I can see a mobile station in Newcastle going to
>>> VK2RTZ-1 then to VK2RTG-5 rather than via our local iGate VK2ZEN-5
>>> VK2ZEN-5 looks like it's okay and working.
>>> Another mobile up the valley is going via VK2RTZ-1 and then to iGate
>>> VK2ZEN-5
>>> Maybe it's a difficult job to get to site and set the routing table
>>> on the
>>> VK2RTZ-5 digipeater if changes are needed.
>>> Can the routing tables be set remotely via the 145.175 access
>>> frequency OR can it only be done with a PC and cable onsite ?
>>> Jamie VK2YCJ
>>>> Maybe the "path police" should be looking at some of the digi's
>>>> around the place.
>>>> I wonder why I see traffic from sugarloaf on highrange being digi'd
>>>> ? I would have thought there are closer igates that the path should
>>>> be set (& honoured) that highrange wouldn't digi this stuff ? Maybe
>>>> its highrange at fault ?
>>>> On 25/05/14 15:20, vk2ycj at bravo.net.au wrote:
>>>>> G'day,
>>>>> I am thinking about using a hand-held 2m rig like a UV-5R as a low
>>>>> power "fill in" digipeater for a small country town ?
>>>>> I drive to Dungog from Newcastle (80Km by road north of Newcastle)
>>>>> 4-5 days per week and the APRS coverage is poor or non-existent.
>>>>> I was considering two options:
>>>>> 1. Putting a UV-5R hand-held radio on a hill near Dungog as a "fill
>>>>> in"
>>>>> digipeater running 1-2 watts and solar powered.  The hill is line
>>>>> of sight to the main digipeater on the Sugarloaf Range in
>>>>> Newcastle.
>>>>> 2. Putting an Alinco DR-135T (with Argent Data card T2-135) running
>>>>> 20 watts and not solar powered.  The boys at Goulburn NSW have
>>>>> installed two of these recently with good results.
>>>>> Can someone please comment on the suitability or stupidity of the
>>>>> above two options ?
>>>>> Thanks,  Jamie VK2YCJ
>>>>>> I have been looking at the load on one server in the rotation,
>>>>>> which is about to move hosting into one of my racks... which is
>>>>>> why previously I said there was no issue.
>>>>>> I do see some issues about the place, with a few WIDE3* and a few
>>>>>> NOGATE. I would have thought a NOGATE should also include NO-DIGI,
>>>>>> but its seems maybe I have the wrong assumption.
>>>>>> Maybe we do need someone to have a look at some of the paths being
>>>>>> used and clean them up.
>>>>>> I would have thought its best to try and keep RF clear so more can
>>>>>> be handle without drama.
>>>>>> Gary
>>>>>> vk2zbb/vk2lv
>>>>>> On 25/05/14 11:50, Scott Evans wrote:
>>>>>>>     From a T2 perspective there's no issue! If you're using
>>>>>>> aunz.aprs2.net rotate the load balancing is automatically taken
>>>>>>> care of, however if you have set up a static server and if
>>>>>>> everyone else had done similar then maybe a problem would exist!
>>>>>>> On May 25, 2014 11:32:59 AM GMT+10:00, Liz VK2XSE
>>>>>>> <edodd at billiau.net>
>>>>>>> wrote:
>>>>>>>        QTR Thu, 22 May 2014 21:47:13 +1000 Ian Mills QTC
>>>>>>>            I've noticed we now have at least 4 x iGates in Sydney,
>>>>>>> I would
>>>>>>>            have
>>>>>>>            thought that's too many and may be adding extra loads
>>>>>>> on the
>>>>>>>            sever as
>>>>>>>            they all gate the same traffic?
>>>>>>>            Ian vk2him
>>>>>>>        The guys in the US, who have more experience in handling
>>>>>>> large traffic
>>>>>>>        volumes, seem to whinge far more about excessive
>>>>>>> digipeating than
>>>>>>>        excessive iGating.
>>>>>>>        I did notice on my last trip to Wollongong that my packets
>>>>>>> were being
>>>>>>>        spread far and wide, and often gated from Newcastle after
>>>>>>> being picked
>>>>>>>        up by VK2AMW-1. (WIDE2-2)
>>>>>>>        The questions can best be answered by experimentation.
>>>>>>>        There is apparently a need for better RF coverage in Sydney.
>>>>>>> How
>>>>>>> much
>>>>>>>        redundancy is required in the iGating system could be explored
>>>>>>>           by
>>>>>>>        deliberately turning off internet connectivity at one or
>>>>>>> more of the
>>>>>>>        Sydney iGates.
>>>>>>>        Sydney users would need to know when these changes were
>>>>>>> made, and then
>>>>>>>        comment on negative and positive aspects of the changes.
>>>>>>>        We would also need to know how many hops people are using
>>>>>>> in their
>>>>>>>        transmissions.
>>>>>>>        And is there any information from the APRS-IS server
>>>>>>> operators about
>>>>>>>        network load?
>>>>>>>        -+-+-+-+-+-+-+-+-+-+-+
>>>>>>>        Liz
>>>>>>>        VK2XSE
>>>>>>>        who hopes not to be using APRS in Sinny, 15 years living
>>>>>>> there was
>>>>>>>        enough.
>>>>>>> -----------------------------------------------------------------
>>>>>>> -------
>>>>>>>        OZAPRS mailing list
>>>>>>>        OZAPRS at aprs.net.au
>>>>>>>        http://lists.aprs.net.au/mailman/listinfo/ozaprs
>>>>>>> --
>>>>>>> Sent from Kaiten Mail. Please excuse my brevity.
>>>>>>> _______________________________________________
>>>>>>> OZAPRS mailing list
>>>>>>> OZAPRS at aprs.net.au
>>>>>>> http://lists.aprs.net.au/mailman/listinfo/ozaprs
>>>>>> _______________________________________________
>>>>>> OZAPRS mailing list
>>>>>> OZAPRS at aprs.net.au
>>>>>> http://lists.aprs.net.au/mailman/listinfo/ozaprs
>>>>> _______________________________________________
>>>>> OZAPRS mailing list
>>>>> OZAPRS at aprs.net.au
>>>>> http://lists.aprs.net.au/mailman/listinfo/ozaprs
>>>> _______________________________________________
>>>> OZAPRS mailing list
>>>> OZAPRS at aprs.net.au
>>>> http://lists.aprs.net.au/mailman/listinfo/ozaprs
>>> _______________________________________________
>>> OZAPRS mailing list
>>> OZAPRS at aprs.net.au
>>> http://lists.aprs.net.au/mailman/listinfo/ozaprs
>> _______________________________________________
>> OZAPRS mailing list
>> OZAPRS at aprs.net.au
>> http://lists.aprs.net.au/mailman/listinfo/ozaprs
> _______________________________________________
> OZAPRS mailing list
> OZAPRS at aprs.net.au
> http://lists.aprs.net.au/mailman/listinfo/ozaprs
> _______________________________________________
> OZAPRS mailing list
> OZAPRS at aprs.net.au
> http://lists.aprs.net.au/mailman/listinfo/ozaprs

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.aprs.net.au/pipermail/ozaprs/attachments/20140526/481b3598/attachment-0001.html>

More information about the OZAPRS mailing list