Track not saved due to unknown error

myneur shared this problem 5 years ago

Today couple of times and last days several times I got a message when ‚saving actual part of track‘:

a) Track not saved due to unknown error

b) or Track saved but actually no track appers in data

I don`t know which update it was but last weeks it is much more worse than it was and I encounter it sometimes now. There are no problems with SD and track added manually is possible to save every time.

Please, try to find a bug. There is nothing worse than unreliable tracking when you need it.

Thanks, m

PS: I was quite hysteric because I had unsaved 2 hours tracking for OSM and actually I added path along by manual tool not to loose it which quite drains my battery. Fortunately it finally saved the track after more than 10 attempts.

Comments (20)



you have problems with "save part" of recorded track right? Not just whole track. If so, you may check this version, where saving a part of track during a record should be fixed. Anyway if problem is in saving whole track when you want to stop track record, then I`ll need some logs (by catlog program for example) to check where is problem


I had a problem with saving whole track also, but only once: Track not saved due to unknown error. After that all the tracked data disappear.

The problem with saving whole track is worse because there is no second trial - data just disappear with error message.

Therefore I started to Save actual parts because I often do tracking for many hours for OSM and can`t afford to lose the data. Error while saving actual data I encountered multiple times.

I can`t predict when it appears again. Mostly it is OK.

I would appreciate a recovery scenario from the error on saving a whole track: option to Export it without saving or Send it by e-mail or whatever just not to lose it.

I`ll try to use the linked version for a while..


hmm understand. You`re absolutely correct ...

As I see in code, I firstly finished track recording service and then finally saved track. In case this failed, track was lost... weird system

In new version will this be switched. So track will be completely removed just in case, save will be successful ..

hope this will prevent any problems


Hi Menion, sorry but it is back. Track not saved again in actual version 2.12.2.

Fortunately track remained paused.

Saving duration was over 40s.

1. saving actual part was OK (length 2:47, 28km)

2. then saving all track FAILED (length 3h, 33km). It told something like „Saving failed with unknown error“ on green background

3. then saving actual was OK

4. then saving all track again was OK

Can`t there be a problem because of the length of the track?

Do it help you if I attach the tracks?

I would expect red background of the error message and not to disappear until any user action to mention it. Even short tracks last 15s to save; long ones like this over 40s.

It seams to me that it happens on long tracks, but just sometimes.

Isn`t it possible to simulate it e. g. by setting Android device to demo mode with GPS move simulation? It happens occasionally, so I can`t predict when it happens again..


hi myneur

newest testversion:

"Saving failed with unknown error“ on green background is strange :)

Your device?

Your Androidversion?

rooted device?

any mod?

A distance of 200km and a recording time of 6h should be no problem and will be saved in seconds.

How frequently you save track points? After every second? After every meter?

There are other tasks in background?


Record NMEA

Live tracking


GPS sharing


You wrote: Do it help you if I attach the tracks?

Maybe - lets try


Device: Xperia Active (ST17i)

Android: 4.0.4

non-rooted, original ROM

Track: interval 5m, 2s

No other tasks on background except for notification Toggle, Minimalistic Text and those original androidous (Services, Maps, Update Center), nothing you have mentioned.

Get tracks here:

1. - saving actual part without problems

3. - saving actual part after error message

4. - regular save after-that



1. sorry for confusing green background - fixed npw

2. I`m worried that some part of track etc. not help here. When this error happen, it should be mainly due to single problem - storing track itself into database.

I would really suggest to check your SD card in some desktop application if do not contain any errors or bad sectors etc. You may also install any program that allow to handle SQLite database and check if database is OK.

When this problem happen, I suggest to try for example add any manual track, or just edit name of any already stored track. If there will be also a problem, then it definitely will be issue in card/database, nothing different.

I also added some more debug messages into log when such problem happen, so after you update to next Locus version (probably this week), be prepared with CatLog program to create a log message


There is no problem to add track or point manually when saving track error occurs.

I did a test of SD by OS X Disk Utility - no problems there.

So I will wait for next version and CatLog it.

Enough to run the Cat, Record (in Verbose Default) and than save after error occurs? Wouldn`t there be too much data after 3 hours?

Isn`t it possible to set a mock location provider like this for easy test?


Wouldn`t there be too much data after 3 hours?



new Testversion ->


Here is the log:

I have set the test version above to track every second and let it dance by the window in the rhythm of GPS inaccuracy.

The last try of ‚save actual part‘ raised that unknown error (on orange background). Stop and save the track immediately after that saved it OK.




perfect, log helped a lot! There seems to be a problem with saving track that is currently active. Issue should be anyway fixed so you`ll see in next version


I think I also experienced that issue. Here is my log:

It is from a previous version of Locus.

I think it was occurring during saving long tracks (20k points), but that might be only a coincidence. I will observe this issue and let you know if it is also in a new Locus version


hmm this seems to be a different problem then problem of myneur. You`ve got "Database locked" issue ... hard to say how to fix it, as Android should handle correctly multi-thread access to database by it`s own. Hmm...


Hi, error still there and with high probability to occur now: this weekend 50 % of save attempts were errors. Even quite short tracks lasting couple of minutes.

Lots of crashes of application (like when trying to create a point). After couple of restarts finaly saved.


Hi!. Now I`m pretty sure you were right, there are two bugs here. Because of one, Ive already lost two track! This is appalling:( I`m going on a trip to Czech Republic in one hour so I don`t have time to describe them. But I`m sending a lot of logs.

I`m really sorry that I`m just dropping them off, in a raw form, but I believe it is better than don`t share them.

Just quickly, important: "my" bug is destroying tracks, when the widget is used to stop the track!


Menion. Did you look into the logs? Should I describe them?


Ah sorry guys, forget to answer

@myneur: I`m almost sure, that your problem has something to do with problem of SD card. These SQLite errors are quite rare, but mostly happen when there is some corruption on SD card. So I suggest to check it with some tool (like in windows etc.)

@elmuSSo: I was looking into your logs, and seems there is problem with getting a root directory of Locus. Don`t know why this happen, but I added one more check into code, so this should solve your problem


OK, I`ll do an extensive wr test (native OS X and Win test passed) and hope I will not catch any virus by some bloody utility for that - which to use?

But it is strange that any other app works fine. I would expect that card errors would be more prevalent on camera photos and other more data-extensive documents stored on the card. But there are no any errors other than Locus.

Supposing there is a SD card error, the response of app should be: “There is a card error - to avoid them in future exchange a card for a new one".

And app should not crash totally with exception above only when adding a single point.


hi myneur

Bad story and so many tasks...

why you don`t make a simply test with an old sdcard?


Because my old card doesn`t fit a 10GB map I use for navigation. And it also does not prove there is no error. The error occurs sometimes, sometimes very frequently and sometimes not at all. So the only I would prove by using a different card would be just that the error was not there that time, not that it cannot come in future in general. Last days there are not errors on my current card e. g. at all.