[aprssig] APRS IS Issue ?
Gerry Creager
gerry.creager at tamu.edu
Sun Mar 2 17:41:20 EST 2008
Great in theory, poor in practice. We have worked on identifying, with
Pete, some of the problems and limitations, and to develop work-arounds.
At this time, there are thre (3) Core servers in operation, which I
believe will prove sufficient as we migrate the non-Ham CWOP stations to
their onw, dedicated server pool. IF not, we'll recruit another.
If we continue to see the flood of CWOP weather data on a mixed-service
server core, we'll continue to see problems. Addint the Tier2 servers
does not mitigate the amount of filter hashing that goes on (Scott,
there's more to it than meets the eye). A tiered system should be an
improvement but for this particular problem, it is not.
The solution, however, is simple. For several years, APRS-IS, by
agreement of the volunteer system administrators, has hosted service for
CWOP. Today, because of the rapid, and honestly unanticipated, growth
of CWOP, hosting as a mixed service on one Core infrastructure (with or
without tiering) is no longer practical. As a committed member of both
activities, I've taken some lead in creating and designing a
CWOP-dedicated server pool with different software configurations from
APRS-IS and different configurations, to support this. I'm seeking to
work with the National Weather Service and the CWOP management structure
to see a relatively straightforward transition to that server pool.
That should, without additional software changes, give CWOP some 3-5
years at the current growth rate to evaluate and make changes to support
its operations. At the same time, it buys breathing room for APRS-IS
to evaluate software ant techniques, make changes to management and
operational recommendations and to continue to grow. Simply, hosting
two services was no longer a good idea. Splitting them and using
resources available (javAPRSSrvr code) and expertise derived from our
experiences made good sense. AE5PL and K4HG have been valuable partners
in this endeavor.
I'm glad to entertain questions on the changes and transition, but I
think we need to let some of the rhetoric settle and see how things will
run.
Gerry N5JXS
first.aprs.net
cwop.tamu.edu/cwop1.tamu.edu (CWOP-1)
firenet.us
Earl Needham wrote:
> At 20:46 3/1/2008, Keith VE7GDH wrote:
>
>> Rather than just keep increasing the maximum number of inbound
>> connections, it might be a good idea if the two remaining core servers
>> were to only accept connections from tier 2 servers, and that the end
>> users should connect to the tier 2 servers and spread the load out
>> instead of trying to connect to the core servers. That will take a
>> significant load off the core servers and let them get on with the
>> task of sharing the data to and from the tier 2 servers instead of
>> getting bogged down trying to serve the needs of the individual end
>> users.
>
> Going along with the above, would it be possible to add another
> "rotate" type address, using different servers? I like to use rotate
> when I'm driving, so if I can't connect to a particular server, I can
> try another without having to stop the truck to get in the sleeper to do
> so. The other option is to try and reach the laptop from the drivers
> seat, while at 70 MPH!
>
> 7 3
> Earl
>
> KD5XB -- Earl Needham
> Clovis, New Mexico DM84jk
> http://groups.yahoo.com/group/cw_bugs
> Quoting from the Coast Guard: ZUT
>
>
> _______________________________________________
> aprssig mailing list
> aprssig at lists.tapr.org
> https://lists.tapr.org/cgi-bin/mailman/listinfo/aprssig
--
Gerry Creager -- gerry.creager at tamu.edu
Texas Mesonet -- AATLT, Texas A&M University
Cell: 979.229.5301 Office: 979.862.3982 FAX: 979.862.3983
Office: 1700 Research Parkway Ste 160, TAMU, College Station, TX 77843
More information about the aprssig
mailing list