Something is buggy with iburst...

[Achim Gratz]

Achim Gratz via devel writes:
> It's likely that removing iburst justs makes the probability of this
> error occuring much lower on startup.  I've seen one system pick up the
> dreaded symptoms overnight to the tune of an effective poll interval
> just short of one hour, the peerstats file shows one entry _exactly_
> every 3072 seconds.  Anybody have an idea what could produce such a poll
> interval?

OK, it just happened again:

     remote                       refid      st t when poll reach   delay   offset   jitter
===========================================================================================
 NMEA(0)                     .uBx6.           0 u    9   16  377   0.0000   3.1188   4.0313
*raspberrypi2                .NavS.           1 u    4   16  377   0.5921   0.0064   0.0319
+raspberrypi3                .NavS.           1 u  38m   16  252   0.6008  -0.1239   0.0727

Note that some reach bits are zero, but not nearly enough to account for
the amount of time that has elapsed.  Note also that the time offset
seems quite wrong (both rasPi are running on PPS and are very close to
each other).

Anything I can look at?  I didn't activate rawstats so far.


Regards,
Achim.
-- 
+<[Q+ Matrix-12 WAVE#46+305 Neuron microQkb Andromeda XTk Blofeld]>+

Factory and User Sound Singles for Waldorf rackAttack:
http://Synth.Stromeko.net/Downloads.html#WaldorfSounds