BGS5 restart (?) during eclipse OTAP | Telit Cinterion IoT Developer Community
February 10, 2017 - 2:17pm, 3181 views
At first I was using the EHS6 for my program, everything running fine in Eclipse (uploading the program, terminating it through an AT command, re-uploading with the fixes etc..)
I switched to BGS5 because of the lower price, and the its capabilites are just fine for my scenario. The problem I'm facing with the BGS5 is that it seems to restart during the "internal" otap Eclilpse is making (happens around 90% of the *****).
What I'm getting in eclipse:
Using USB port COM89. Connecting to module... Initializing module for debugging... Establishing "IP connection for remote debugging of BGSx"... Registering ip address "192.168.244.1" of remote debugging device... Waiting for debug device registration of "IMP_NG_BGS5_REMOTE"... Passing control to external device emulator... Installing suite from: http://192.168.244.2:50168/MeterData.jad Exception in thread "Object Server Connection - Handle JMX Notification Thread" java.util.ConcurrentModificationException at java.util.ArrayList$Itr.checkForComodification(Unknown Source) at java.util.ArrayList$Itr.next(Unknown Source) at com.sun.jme.toolkit.remoting.client.rmiimpl.ObjectServerConnectionImpl$1.run(Unknown Source) at java.lang.Thread.run(Unknown Source)
And what I receive when running a trace on the USB com port
(the bottom line "input/output error" means that connection was lost with the com port)
So what I assume that is happening is that for some reason the module is restarting.
Things to notice
- The same code/procedure works flawlessly with the EHS6
- With the BGS5 the above error usually happens, so then I have to turn off/turn on the device and try again. The program uploads successfully and can debug in Eclipse around 1 out of 10 ***** that I try
Is there a way fix to that? Is it a common/known issue?
Could you check ATI1 for the detailed firmware version?
Is it working when you install and run the application directly on the module without Eclipse?
What is the application doing?
It's happening with both
Updating the firmware was one of the first things I tried to fix this.
The application is running just as fine as on the EHS6 boards (no weird restarts, nothing). I'm uploading the program through SMS otap. The exact code is deployed on both EHS6 and BGS5 boards to test them.
The application itself, without getting into much detail, is using the ASC0 interface to talk to an external peripheral via RS232, and uses a SIM card to post some results on an HTTP server. Normal stuff.
Another thing to notice, the 1st time I saw this "Exception in thread "Object Server Connection - Handle JMX Notification Thread"" error is when i was using the internal watchdog and forgetting to "kick" it. Also it's tha same error that I would if I would just shut the module off while I'm debugging in Eclipse/Netbeans. So, now it's the same exact error. (Thus my assumption that the board is restarting - along with that I loose the serial connection on my terminal)
Also, sorry that I didn't clarify this, the program doesn't start at all* (I mean having a System.out.println command as the first command on the MIDlet startApp() )
*Unless for the ***** that it actually runs ofcourse (which as I said above might happen in 1 out of every 10 *****). If the program gets uploaded on the device and it starts running, then there are no problem later.
I have found out that somebody has recently already reported such problem and it's being analyzed.
It does not exist in A-REVISION 00.000.10.
The problem only occurs when the MIDlet is run form IDE.
So for now if you need to use this feature during the development you can downgrade to the above version.
Thanks, I will give A-REVISION 00.000.10 a try.
Is there a changelog however available to know the changes since that version? Because if the improvements are for things we don't need we can just as well use that firmware.
There are release notes documents available for every software revision.
Writing to inform that on the "A-REVISION 00.000.15" it works fine as well (netbeans debugging)
Is there however an expected date for this fix on a new firmware version?
I'm sorry but the plans for upcoming firmware releases are not public so I can't share any date.