R: [teqc] latest development version of teqc on-line (16 Mar 2009)

Lou Estey lou at unavco.org
Wed Mar 18 11:02:11 MDT 2009


Giuseppe,

> I am not able to run "teqc +qc" for a specific file: mate0750.09o
> (you can download it from here ftp://geodaf.mt.asi.it/GEOD/GPSD/RINEX/2009/075/
> if you want to reproduce the problem).
> This new teqc version fails showing "Segmentation fault", while the
> previous one (2008Oct2) works fine.
> 
> I performed a quick check trying +qc on all 2009 MATE files: it fails
> 24 times on 76 total. The problem appears on both linux and windows executable.
> 
> Any idea ?

It's a problem on all OS builds.

Yes, the problem appears to be related to the new gap code (because the
seg fault occurs just as an epoch gap is being reported and then the
qc stuff on the gap starts) _and_ related to too many of the GPS nav
messages being deleted:

[2924] teqc +v mate0750.09n
! Notice ! GPS week in GPS week in RINEX NAV = 1471; (default) GPS week = 1523
SV G 1: magnitude of GPS NAV ToE - ToW too large (3600.0 min) @ ToC= 2008 Mar 19 12:00:00.000
SV G 3: magnitude of GPS NAV ToE - ToW too large (-720.0 min) @ ToC= 2009 Mar 15 12:00:00.000
SV G 6: magnitude of GPS NAV ToE - ToW too large (-720.3 min) @ ToC= 2009 Mar 15 11:59:44.000
SV G 7: magnitude of GPS NAV ToE - ToW too large (-240.0 min) @ ToC= 2009 Mar 15 20:00:00.000
SV G11: magnitude of GPS NAV ToE - ToW too large (-480.0 min) @ ToC= 2009 Mar 15 16:00:00.000
SV G13: magnitude of GPS NAV ToE - ToW too large (-240.0 min) @ ToC= 2009 Mar 15 20:00:00.000
SV G16: magnitude of GPS NAV ToE - ToW too large (-480.3 min) @ ToC= 2009 Mar 15 15:59:44.000
SV G19: magnitude of GPS NAV ToE - ToW too large (-240.0 min) @ ToC= 2009 Mar 15 20:00:00.000
SV G20: magnitude of GPS NAV ToE - ToW too large (-480.0 min) @ ToC= 2009 Mar 15 16:00:00.000
SV G21: magnitude of GPS NAV ToE - ToW too large (-960.0 min) @ ToC= 2009 Mar 15 08:00:00.000
SV G23: magnitude of GPS NAV ToE - ToW too large (-360.0 min) @ ToC= 2009 Mar 15 18:00:00.000
SV G24: magnitude of GPS NAV ToE - ToW too large (-240.0 min) @ ToC= 2009 Mar 15 20:00:00.000
SV G25: magnitude of GPS NAV ToE - ToW too large (-240.0 min) @ ToC= 2009 Mar 15 20:00:00.000
SV G31: magnitude of GPS NAV ToE - ToW too large (-600.0 min) @ ToC= 2009 Mar 15 14:00:00.000
SV G32: magnitude of GPS NAV ToE - ToW too large (-480.3 min) @ ToC= 2009 Mar 15 15:59:44.000
teqc: mate0750.09n readable as RINEX V.2.10 format

If instead I use a combined GPS RINEX nav file (e.g. from CDDIS)
instead of your nav file, then the qc works perfectly and does the
gap qc as we want:

-dn|++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++|-dn
+dn|++++++++++++++++++++++++++++++++++-+++++++++++++++++++++++++++++++++++++|+dn
+10|8aa9a99999999aa988777889988999998887899a9a9a99989a9aaaa787899a9999abbb98|+10
Pos|oooooo    o oo   o            o                  o oooo       o  ooo o  |Pos
Clk|                                  ^                                     |Clk
    +---------|--------|--------|--------|--------|--------|--------|--------+
00:00:00.000                                                        23:59:30.000
2009 Mar 16                                                          2009 Mar 16
...
Poss. # of obs epochs   :   2880
Epochs w/ observations  :   2878
Epochs repeated         :      0  (0.00%)
Possible obs >   0.0 deg:  31195
Possible obs >  10.0 deg:  24557
Complete obs >  10.0 deg:  24238
   Missed obs >  10.0 deg:      0
  Deleted obs >  10.0 deg:      8
...
       first epoch    last epoch    hrs   dt  #expt  #have   %   mp1   mp2 o/slps
SUM 09  3 16 00:00 09  3 16 23:59 23.98  30  24557  24238  99  0.24  0.27    898

The seg fault using the original nav file, mate0750.09n, must be occuring
in the gap code and is expecting a nav message for some SV which has been
eliminated.

Besides fixing that problem, I suppose this also calls into question the
current constraint on eliminating a GPS nav message.  The code (for quite
a while now) has been rejecting any GPS nav message where ToE - ToW is
outside of -130 minutes or +180 minutes.  Your nav file has many message
where ToE - ToW to -720 minutes (== 1/2 day; this excludes the one with
ToE - ToW = -960 minutes == 16 hours).  Any thoughts on this?

--lou


More information about the teqc mailing list