GC issues

[Pier]

Hi everyone.

Today I tested GC with the Elite direto and it seems to work nicely.

However I have a couple of issues:

- The speed recorded is totally wrong (max 180 mph)

- I cannot upload the ride to Ridewithgps. GC simply crashes and dumps a core (which I'm not able to find it somehow).

I tried with a Fedora package built from sources and also the official AppImage, same issues.

Thanks

[Ale Martinez (Please don't email or cc me)]

El sábado, 7 de marzo de 2020, 16:21:55 (UTC-3), Pier escribió:

Hi everyone.

Today I tested GC with the Elite direto and it seems to work nicely.

However I have a couple of issues:

- The speed recorded is totally wrong (max 180 mph)

See https://github.com/GoldenCheetah/GoldenCheetah/issues/2999

 

- I cannot upload the ride to Ridewithgps. GC simply crashes and dumps a core (which I'm not able to find it somehow).

I tried with a Fedora package built from sources and also the official AppImage, same issues.

For the AppImage this may help: https://github.com/GoldenCheetah/GoldenCheetah/wiki/SSL-problems-in-some-Linux-distributions

[Pier]

On Saturday, 7 March 2020 19:52:09 UTC, Ale Martinez (Please don't email or cc me) wrote:

- The speed recorded is totally wrong (max 180 mph)

See https://github.com/GoldenCheetah/GoldenCheetah/issues/2999

Ah thanks! I'll make a note to look into the github issues next time. 

I saw once you add a devices, it's not possible to edit it.
I just changed the circumference in the ini file. Hopefully that would be ok.

 

 

- I cannot upload the ride to Ridewithgps. GC simply crashes and dumps a core (which I'm not able to find it somehow).

I tried with a Fedora package built from sources and also the official AppImage, same issues.

For the AppImage this may help: https://github.com/GoldenCheetah/GoldenCheetah/wiki/SSL-problems-in-some-Linux-distributions

Thanks again. Tried that but it still crashes:

$ env | grep OPENSSL_CONF

OPENSSL_CONF=/home/pier/bin/etc/openssl.cnf

$ Downloads/rpms/GoldenCheetah_v3.5_amd64_Linux.AppImage 

[1]    436878 segmentation fault (core dumped)  Downloads/rpms/GoldenCheetah_v3.5_amd64_Linux.AppImage

 

[Ale Martinez (Please don't email or cc me)]

El sábado, 7 de marzo de 2020, 19:42:49 (UTC-3), Pier escribió:

On Saturday, 7 March 2020 19:52:09 UTC, Ale Martinez (Please don't email or cc me) wrote:

- The speed recorded is totally wrong (max 180 mph)

See https://github.com/GoldenCheetah/GoldenCheetah/issues/2999

Ah thanks! I'll make a note to look into the github issues next time. 

I saw once you add a devices, it's not possible to edit it.
I just changed the circumference in the ini file. Hopefully that would be ok.

 

 

- I cannot upload the ride to Ridewithgps. GC simply crashes and dumps a core (which I'm not able to find it somehow).

I tried with a Fedora package built from sources and also the official AppImage, same issues.

For the AppImage this may help: https://github.com/GoldenCheetah/GoldenCheetah/wiki/SSL-problems-in-some-Linux-distributions

Thanks again. Tried that but it still crashes:

$ env | grep OPENSSL_CONF

OPENSSL_CONF=/home/pier/bin/etc/openssl.cnf

Have you checked the openssl.cnf file is located there?

It may be found in other places depending on distribution and install options s.t. https://stackoverflow.com/questions/21477210/correct-location-of-openssl-cnf-file

[Pier]

On Saturday, 7 March 2020 23:18:49 UTC, Ale Martinez (Please don't email or cc me) wrote:

Have you checked the openssl.cnf file is located there?

It may be found in other places depending on distribution and install options s.t. https://stackoverflow.com/questions/21477210/correct-location-of-openssl-cnf-file

 

$ Downloads/rpms/GoldenCheetah_v3.5_amd64_Linux.AppImage 

[1]    436878 segmentation fault (core dumped)  Downloads/rpms/GoldenCheetah_v3.5_amd64_Linux.AppImage

Indeed. On fedora it's in  /etc/pki/tls/openssl.cnf (this is the only place I find the openssl.cnf) but it still crashes (both compiled and appimage).

I was able to find the core...if it may be helpful:

(gdb) bt                                                                                                                                                                                              
#0  0x00005585eaf12e32 in MVJSONNode::getField (this=0x0, name="error") at ../qtsolutions/json/mvjson.cpp:227                                
#1  0x00005585eaf134a2 in MVJSONNode::getFieldString (this=0x0, name=...) at /usr/include/c++/9/bits/basic_string.h:936
#2  0x00005585eab45aa3 in RideWithGPS::writeFileCompleted (this=0x7f75c8018970) at /usr/include/c++/9/bits/char_traits.h:300
#3  0x00005585eafe5295 in RideWithGPS::qt_static_metacall (_o=<optimized out>, _c=<optimized out>, _id=<optimized out>, _a=<optimized out>) at .moc/moc_RideWithGPS.cpp:82
...


(gdb) f 0
#0  0x00005585eaf12e32 in MVJSONNode::getField (this=0x0, name="error") at ../qtsolutions/json/mvjson.cpp:227
227         if (values.size() == 0) return NULL;
(gdb) l
222         return false;
223     }
224
225     MVJSONValue* MVJSONNode::getField(string name)
226     {
227         if (values.size() == 0) return NULL;
228         for (size_t i = 0; i < values.size(); i++)
229             if (values.at(i)->name == name)
230                 return values.at(i);
231         return NULL;

(gdb) p &values
$2 = (std::vector<MVJSONValue*, std::allocator<MVJSONValue*> > *) 0x8

[Ale Martinez (Please don't email or cc me)]

The problem is MVJSONReader sets root=Null when the response cannot be parsed and the code is not checking that but using an exception handler: https://github.com/GoldenCheetah/GoldenCheetah/blob/a86b0e9b74882ed83af3eb9293d97331ea95a500/src/Cloud/RideWithGPS.cpp#L195

If you can build from source it would be useful to enable debug to see which the response is.

[Ale Martinez (Please don't email or cc me)]

I did and here is the result:

[Cloud/RideWithGPS.cpp:81 open] RideWithGPS::open
[Cloud/RideWithGPS.cpp:103 writeFile] RideWithGPS::writeFile(2020_02_27_16_17_14.json)
[Cloud/RideWithGPS.cpp:178 writeFileCompleted] RideWithGPS::writeFileCompleted()
[Cloud/RideWithGPS.cpp:182 writeFileCompleted] reply:<html>
<head><title>301 Moved Permanently</title></head>
<body>
<center><h1>301 Moved Permanently</h1></center>
<hr><center>openresty/1.15.8.2</center>
</body>
</html>

So, it seems the RideWithGPS entry point has changed, we would need to find out the new one and update.

Additionally we need to fix the error handling to avoid crashes.

[Pier]

On Sunday, 8 March 2020 20:35:54 UTC, Ale Martinez (Please don't email or cc me) wrote:

I did and here is the result:

[Cloud/RideWithGPS.cpp:81 open] RideWithGPS::open
[Cloud/RideWithGPS.cpp:103 writeFile] RideWithGPS::writeFile(2020_02_27_16_17_14.json)
[Cloud/RideWithGPS.cpp:178 writeFileCompleted] RideWithGPS::writeFileCompleted()
[Cloud/RideWithGPS.cpp:182 writeFileCompleted] reply:<html>
<head><title>301 Moved Permanently</title></head>
<body>
<center><h1>301 Moved Permanently</h1></center>
<hr><center>openresty/1.15.8.2</center>
</body>
</html>

So, it seems the RideWithGPS entry point has changed, we would need to find out the new one and update.

Additionally we need to fix the error handling to avoid crashes.

I was just doing it but I didn't get much more logs. I did follow the developer guide to enable debug, but somehow I didn't get any more log (or maybe I didn't see it).

Is there something I missed?

[Ale Martinez (Please don't email or cc me)]

On a debug build the output goes to the console, on a release one to goldencheetah.log parallel to your athlete folder.

I raised an issue (https://github.com/GoldenCheetah/GoldenCheetah/issues/3371), the crash is easy to solve but the upload not so since it seems they have changed the API.

[Pier]

On Sunday, 8 March 2020 21:03:01 UTC, Ale Martinez (Please don't email or cc me) wrote:

El domingo, 8 de marzo de 2020, 17:58:14 (UTC-3), Pier escribió:

On Sunday, 8 March 2020 20:35:54 UTC, Ale Martinez (Please don't email or cc me) wrote:

I did and here is the result:

[Cloud/RideWithGPS.cpp:81 open] RideWithGPS::open
[Cloud/RideWithGPS.cpp:103 writeFile] RideWithGPS::writeFile(2020_02_27_16_17_14.json)
[Cloud/RideWithGPS.cpp:178 writeFileCompleted] RideWithGPS::writeFileCompleted()
[Cloud/RideWithGPS.cpp:182 writeFileCompleted] reply:<html>
<head><title>301 Moved Permanently</title></head>
<body>
<center><h1>301 Moved Permanently</h1></center>
<hr><center>openresty/1.15.8.2</center>
</body>
</html>

So, it seems the RideWithGPS entry point has changed, we would need to find out the new one and update.

Additionally we need to fix the error handling to avoid crashes.

I was just doing it but I didn't get much more logs. I did follow the developer guide to enable debug, but somehow I didn't get any more log (or maybe I didn't see it).

Is there something I missed?

On a debug build the output goes to the console, on a release one to goldencheetah.log parallel to your athlete folder.

Thanks, I then must have done something wrong. No output in console or in the log file.

I'll give it another try just to try to learn.

 

I raised an issue (https://github.com/GoldenCheetah/GoldenCheetah/issues/3371), the crash is easy to solve but the upload not so since it seems they have changed the API.

 

Thanks. I tried to give a quick look at the api, not really sure what has been used within GC and my c++ knowledge is indeed pretty poor.

[TWINGSISTER TWINGSISTER]

I have a Segmentation Fault with nothing in .log.

This happens just after "open" after  the initial biker selection

Version 3.5 works and 3.6 snapshot june and august builds fails.

What could be of general interest in all that is that 3.5 works  on Bodhi Linux 64 Bit on a

Compaq Presario V5000 (15y and counting).

Any idea about the tiniest linux 64 distro for outdated laptops that can run GC 3.6?

Am I the only one that is running in the basement with outdated hardware (i.e. Tacx Fortius)?

[Ale Martinez]

El martes, 5 de octubre de 2021 a la(s) 09:38:54 UTC-3, twing...@gmail.com escribió:

I have a Segmentation Fault with nothing in .log.

This happens just after "open" after  the initial biker selection

Version 3.5 works and 3.6 snapshot june and august builds fails.

On Linux you can run GC using gbd to get a stack trace after the Segmentation Fault to help diagnosis, see https://groups.google.com/g/golden-cheetah-users/c/Xed5ppEmXSw/m/BzcM0d8cAQAJ for an example.

[Ale Martinez]

El martes, 5 de octubre de 2021 a la(s) 09:38:54 UTC-3, twing...@gmail.com escribió:

I have a Segmentation Fault with nothing in .log.

This happens just after "open" after  the initial biker selection

Version 3.5 works and 3.6 snapshot june and august builds fails.

What could be of general interest in all that is that 3.5 works  on Bodhi Linux 64 Bit on a

Compaq Presario V5000 (15y and counting).

As explained in https://github.com/GoldenCheetah/GoldenCheetah/wiki/FAQ-GENERAL#for-which-operating-systems-are-binary-distributions-available, for v3.6 you would need a Bodhi Linux version based on Ubuntu 18.04 or higher, I think Bodhi Linux 5.x or 6.x should work, but I have not tested it. 

[Steve Edmonds]

On 06/10/2021 01:38, TWINGSISTER TWINGSISTER wrote:

I have a Segmentation Fault with nothing in .log.

This happens just after "open" after  the initial biker selection

Version 3.5 works and 3.6 snapshot june and august builds fails.

What could be of general interest in all that is that 3.5 works  on Bodhi Linux 64 Bit on a

Compaq Presario V5000 (15y and counting).

Any idea about the tiniest linux 64 distro for outdated laptops that can run GC 3.6?

Am I the only one that is running in the basement with outdated hardware (i.e. Tacx Fortius)?

I had similar on openSUSE when I started with GC but was trying so many things I don't know what fixed it.
Have you tried adding the --no-r and --no-python command line options, that solved some issues, but may be not yours.
steve

On Sunday, March 8, 2020 at 10:17:55 PM UTC+1 Pier wrote:

On Sunday, 8 March 2020 21:03:01 UTC, Ale Martinez (Please don't email or cc me) wrote:

El domingo, 8 de marzo de 2020, 17:58:14 (UTC-3), Pier escribió:

On Sunday, 8 March 2020 20:35:54 UTC, Ale Martinez (Please don't email or cc me) wrote:

I did and here is the result:

[Cloud/RideWithGPS.cpp:81 open] RideWithGPS::open
[Cloud/RideWithGPS.cpp:103 writeFile] RideWithGPS::writeFile(2020_02_27_16_17_14.json)
[Cloud/RideWithGPS.cpp:178 writeFileCompleted] RideWithGPS::writeFileCompleted()
[Cloud/RideWithGPS.cpp:182 writeFileCompleted] reply:<html>
<head><title>301 Moved Permanently</title></head>
<body>
<center><h1>301 Moved Permanently</h1></center>
<hr><center>openresty/1.15.8.2</center>
</body>
</html>

So, it seems the RideWithGPS entry point has changed, we would need to find out the new one and update.

Additionally we need to fix the error handling to avoid crashes.

I was just doing it but I didn't get much more logs. I did follow the developer guide to enable debug, but somehow I didn't get any more log (or maybe I didn't see it).

Is there something I missed?

On a debug build the output goes to the console, on a release one to goldencheetah.log parallel to your athlete folder.

Thanks, I then must have done something wrong. No output in console or in the log file.

I'll give it another try just to try to learn.

 

I raised an issue (https://github.com/GoldenCheetah/GoldenCheetah/issues/3371), the crash is easy to solve but the upload not so since it seems they have changed the API.

 

Thanks. I tried to give a quick look at the api, not really sure what has been used within GC and my c++ knowledge is indeed pretty poor.

--
_______________________________________________
Golden-Cheetah-Users mailing list
golden-che...@googlegroups.com
http://groups.google.com/group/golden-cheetah-users?hl=en
---
You received this message because you are subscribed to the Google Groups "golden-cheetah-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to golden-cheetah-u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/golden-cheetah-users/12ded5ac-9179-4d36-a003-2b0c8cb17489n%40googlegroups.com.

[Steve Edmonds]

On 06/10/2021 07:15, Steve Edmonds wrote:
>
> On 06/10/2021 01:38, TWINGSISTER TWINGSISTER wrote:
>> I have a Segmentation Fault with nothing in .log.
>> This happens just after "open" after  the initial biker selection
>> Version 3.5 works and 3.6 snapshot june and august builds fails.
>> What could be of general interest in all that is that 3.5 works  on
>> Bodhi Linux 64 Bit on a
>> Compaq Presario V5000 (15y and counting).
>> Any idea about the tiniest linux 64 distro for outdated laptops that
>> can run GC 3.6?
>> Am I the only one that is running in the basement with outdated
>> hardware (i.e. Tacx Fortius)?
> I had similar on openSUSE when I started with GC but was trying so
> many things I don't know what fixed it.
> Have you tried adding the --no-r and --no-python command line options,
> that solved some issues, but may be not yours.

I also renamed ~/.goldencheetah created when I installed 3.5 to
~/.goldencheetahlast  and started from scratch with 3.6
> steve