Mission Planner latest 1.3.27 quirk

APM 2.5
Frame QUAD
FW V3.2.1 (36b405fb)

 After loading latest MP 1.3.27,  I had a flight this morning that nearly ended in disaster.  Upon selecting RTL it went out of control with excessive yaw, pitching and rolling. Quickly recovered by switching to stabilize.  Downloaded dataflash log and viewed that I had suffered my first GPS glitch with a momentary hDOP spike to 4 and sat count drop to 6, it was a first for this platform which has performed countless loiter and RTL activity with great precision (sat counts of 14 & hDOP avg 1.7).  Anyways, while troubleshooting I was unable to carry out Automatic Analysis of the data flash log on neither of my three systems ( XP, 7 and Windows 8.1) which were updated with the latest 1.3.27 so I reverted back to the one I've been utilizing for awhile 1.3.25 and now the Automatic Analysis works.   With 1.3.27 an error window states "global name sqrt is not defined " then you hit OK  and the next pop-up is " bad input file "  Have not tried 1.3.26 yet.

Mission planner does not affect outcome of RTL.
It’s entirely done onboard and limited by your sensor data, tuning, and the hardwares ability to fly.

[quote=“Andre-K”]Mission planner does not affect outcome of RTL.
It’s entirely done onboard and limited by your sensor data, tuning, and the hardwares ability to fly.[/quote]

Missing the point there, the problem he is having is the latest version of MP in not letting him do an auto analysis of the log.

I don’t use the auto analysis so can’t really help.

my bad, subject and two of the first lines misled me.

Ya agree my sbj is misleading, tried editing but was unable to figure a way to do so…

while am not having any RTL flight problem, I am having the same issue with MP after the latest update when doing an auto analysis I get two errors, “global name sqrt is not defined” then you hit OK and the next pop-up is " bad input file "…am running 1.3.27 build 1.1.5620.38044

Just a follow up to mention that I tried MP 1.3.26 and dataflash auto analysis works :slight_smile: moved up to 1.3.27 and the error pop-up re-appears.

for a temp fix

goto LogAnalyzer\tests\TestCompass.py

and replace
def vec_len(x):
return sqrt(x[0]**2+x[1]**2+x[2]**2)
with
def vec_len(x):
return math.sqrt(x[0]**2+x[1]**2+x[2]**2)

I am having the same problem after updating MP this morning - yesterday Auto Analysis worked fine, today I get the same error message as above, after updating to 1.3.27.

I can’t help but noticing that the error message by the name seems to be related with the compass - and I am currently fiddling around with the compass, trying to get the external compass and GPS working correctly (I have cut SJ10, but seem to continue receiving compass data from the internal compass, while the GPS works, but I am getting “GPS_GLITCH” errors).

What could be the relation here - has one thing to do with each other or is this just a coincidence? Will I continue to get correct Auto Analysis data after changing the line in the file?

Thanks for your help!
Anna

meee1 your fix work fine! THX :wink:

I just noticed I had the same problem. Replaced the line, as advised above, but still got error messages. Checked for BETA updates, installed, and everything works! :smiley:

Thanks meee1 :slight_smile: I implemented your fix. Great programs ( MP & APM2 ) , I utilize them in all three operating system Win, OS and Ubuntu LTS