paparazzi-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Paparazzi-devel] sdlog file names for apogee


From: Gautier Hattenberger
Subject: Re: [Paparazzi-devel] sdlog file names for apogee
Date: Thu, 07 May 2015 18:33:19 +0200
User-agent: Mozilla/5.0 (X11; Linux x86_64; rv:31.0) Gecko/20100101 Icedove/31.6.0

Hello,

We have found some potential sources of bugs in the naming of sdlog files, and the master branch have been updated. Can you test with this new code and report any errors ?

Thank you

Gautier

Le 26/04/2015 15:11, Andreas Philipp a écrit :

Thanks for your replies, Gautier and Alexandre!

After some struggles (with variables &pprzLogFile and pprzLogFile.fs in the module c-files and compiling paparazzi on Linux Mint17.1 which worked for v5.4 but not immediately for master) I've managed to switch to master version v5.5_devel-555-g3b28ef3-dirty, however file names are still not created as expected yet.

As shown in the attached screenshot (05.png), the flightrecorder works but suffers from similar problems regarding the file names as sdlog does. For the latter the -36 file was overwritten again. Additionally, after unplugging and replugging the USB cable, a pprzlog_000.LOG appears with a remarkable date (which might indicate the final breakthrough for decadal climate prediction ;)).

Thus unfortunately it is not fixed yet in master.
Thank's for your help on this (as well as for the apogee/pprz stuff as a whole, really a great tool)!

Cheers,
Andreas

PS: If someone is interested in the humid_sht and airspeed_ets sdlog function, the updated files for master are attached.

Am 24.04.2015 um 23:19 schrieb Alexandre Bustico:

I think i know what is wrong,  i am far from a computer now,  but should fix it on Monday.  I think the problem is i don't chek that the numering part is really only Digit,  and if it's not,  atoi will fail.
Alexandre
Alexandre

Le 24 avril 2015 23:13:53 Gautier Hattenberger <address@hidden> a écrit :

Hi,

Thank you for reporting the issue, we'll have a look at this. In the mean time, the sdlog in master has changed a bit in order to log into several files, with "flight recorder" feature using the periodic telemetry system (https://github.com/paparazzi/paparazzi/pull/1144, to be documented...).
Maybe the issue is fixed in master now, but we still need to check for both master and v5.4.

Gautier

Le 24/04/2015 09:23, Andreas Philipp a écrit :

Hi,

I'm using the SDLOG function for the "humid_sht" and "airspeed_ets" modules (TEMOD-I2C-R1 would be a next candidate ...) on the APOGEE board (see below at PS for the files I use for SDLOG).

It works fine, however the filenames that are created by SDLOG cause reproducible troubles:
sometimes the mechanism of counting up the file name numbers (depending on the already existing files) does not work.
Instead there are filenames including negative numbers (-36 and -2147483648 (smallest 4byte integer) are frequently appearing) and these files are frequently overwriting old files with the same odd name. We have lost some measurement data (very very early in the morning :)) by this, when we switched the plane off and on after flight.

Please find four directory listing screenshots attached, each captured after switching off and on the APOGEE and after some time of measurement plugin in the USB cable to the computer respectively. First (01.png) everything is fine (pprzlog_001.LOG), next (02.png) there is an additional  file pprzlog_-2147483648.LOG (ugly but containing the new data), at the next measurement (003.png) we get a pprzlog_-36.LOG which is finally (04.png) overwritten (see the size and date of the file).

It seems, the integer which is used for the filename is sometimes uninitialized or so.
I found the function "getNextFileName" in
sw/airborne/subsystems/chibios-libopencm3/sdLog.c
But there the integer maxCurrentIndex is initialized to zero in line 248.

Can this be a hardware or filesystem issue? However, it happens on other APOGEE boards too. The filesystem is 8GB FAT and it happens on other (class 10) micro-SD cards too. I'm running the APOGEE on 3S Lipos and 4S Lipos and don't think it's a power issue.

I'm currently using Paparazzi version v5.4.2_stable-0-g21e8716-dirty

I know that this didn't happen half a year ago or so, however I cannot say exactly with which version ...

Can someone please help in finding this bug?

Thanks!!!

Andreas


PS the modified files for using SDLOG with humid_sht and airspeed_ets were reported formerly here (still not included in git):
http://lists.paparazziuav.org/FAT32-for-the-appogie-tt16359.html#a16360
and
http://lists.paparazziuav.org/FAT32-for-the-appogie-tt16359.html#a16361



_______________________________________________
Paparazzi-devel mailing list
address@hidden
https://lists.nongnu.org/mailman/listinfo/paparazzi-devel

_______________________________________________
Paparazzi-devel mailing list
address@hidden
https://lists.nongnu.org/mailman/listinfo/paparazzi-devel



_______________________________________________
Paparazzi-devel mailing list
address@hidden
https://lists.nongnu.org/mailman/listinfo/paparazzi-devel

-- 
--------------------------------------
PD Dr. Andreas Philipp
Institute of Geography
University of Augsburg
Universitaetsstrasse 10
D - 86135 Augsburg
Germany

Phone:  ++49/821/598-2266
Fax:    ++49/821/598-2264
Email:  address@hidden
http://www.geo.uni-augsburg.de/lehrstuehle/phygeo/personal/philipp/
--------------------------------------


_______________________________________________
Paparazzi-devel mailing list
address@hidden
https://lists.nongnu.org/mailman/listinfo/paparazzi-devel


reply via email to

[Prev in Thread] Current Thread [Next in Thread]