FTP and OBEX not working

[Carlos Romero]

Hi, I think I've just broken a serial5 device :p
I'm developing a program which uses message exchange with other motes. The program was working fine, but I didn't use any semicolon in it, so it was quite slow.
I got overconfident. I thought that the messages were received when the main code had already finished its execution so I could put semicolons in every codeline, INIT included, and in the message sending part of the code too. I didn't think that maybe the message sending is not working very well if you are exchanging the bluetooth capabilities for speed. I think that it gets stuck sending a message or in a loop while it is trying to send the message and checking that the message was sent, and it never recovers from it. There are four motes running the same code, and three of them are still accessible because as I thought, the main code is running fine and when it ends they can manage FTP and messages, but the other one after the initialization starts to run a subrutine which asks the status of the other motes.The other motes never received the message from the questioning mote and it gets stuck in the first sending message part. It worked great before of putting the semicolons and if the message was not succesfully sent to the other devices it should just end, so it must fail trying to send the message somehow (if it is like this, maybe could be a good idea to ignore semicolons if the function is a bluetooth one...).

This is the nefarious message sending code part:
100 $0=$3;
101 K=800+X-33;
102 A=message $K;
103 E = status;
104 IF E < 1000 THEN 107;
105 WAIT 5;
106 GOTO 103;
107 S=success;
0 REM GUARDAMOS LA TRAZA
108 GOSUB 710;
109 RETURN

Now I can't access its filesystem with BlueSoleil because it starts to execute the code almost in the same very moment the device is on. Any ideas??
Thanks

[Juergen Kienhoefer]

that was too much indeed.

Normally the BASIC code leaves time for the Bluetooth base band processor

to complete wireless communication. If you put ; everywhere then there is no time

whatsoever for anything wireless.

So, easy enough, you see the two solder pads on the top side of the PCB?

shorten those, power up, wait for 10 secs at least. power off, unshorten and

start up again. All without the battery.

After that you should have an empty Serial5X and you can start all over.

BTW, the BASIC interpreter is not the slow part of the system. It is the Bluetooth

spec by itself. The packages are too small too to be efficient. And connection making

takes awfully long. You feel that most when you run on an embedded system.

Check you Android e.g. It takes also about 5 seconds to get a FTP started....

Hope that helps

Juergen

Wireless Cables Inc.

--
You received this message because you are subscribed to the Google Groups "AIRcable(tm)" group.
To unsubscribe from this group and stop receiving emails from it, send an email to aircable+u...@googlegroups.com.
To post to this group, send email to airc...@googlegroups.com.
Visit this group at http://groups.google.com/group/aircable?hl=en.
For more options, visit https://groups.google.com/groups/opt_out.
 
 

[Carlos Romero]

Did you mean this? Because I did what you said and the device is now like dead :p
It's not answering inquiries and not accepting anything. Now it's really broken, right?

[Juergen Kienhoefer]

no, this is the battery connector

since you don't have a battery you don't need to disconnect

it is on the other side above the 4DIL switch

[Carlos Romero]

Can you be more specific? I ruined this device, but I want to know for the next time :p
If we flip the serial5 in the picture and we keep having the rs232 toward the front like in the picture there are 7 pads to the left and what seem to be two pads with a little tin on them. Are these to the right the solder pads you said?
Thank you

[Juergen Kienhoefer]

exactly, the 7 pads are for debugging the processor

you use the 2 with tin on them for shortening.