Important update regarding LTE modem updates

[Prawnhead]

Very glad to hear you've replicated the issue and have a fix underway. The icing on the cake would be if you can come up with an "unbrick procedure" too :) Thanks for all you do. We love Pycom stuff!

[sh4nnongoh]

@xykon Thank you, looking forward to this. Yes, please have an "unbrick procedure" as well. What are the 2 issues that cause the brick? I am not sure if mine is bricked or not.

[le_wildfox]

@xykon Thank you for updating us.
Does the "unresponsive" -ness symptoms include intermittent successful register & attach attempts?

I did connectivity tests at Vodafone UK nb-iot lab (Huawei equipments) using GPy with 37781 NB-IoT FW version. Registrations to nb-iot network were failed using the example sequence as in example neither via REPL nor putting it in the script.

Slight modification to the example as below snippet worked several times:

print("Trying to connect to nb-IoT network...")
send_at_cmd_pretty('AT+CFUN=0')
time.sleep(3)
send_at_cmd_pretty('AT+CFUN=0')
time.sleep(5)
send_at_cmd_pretty('AT+CFUN=1')
time.sleep(2)
send_at_cmd_pretty('AT!="clearscanconfig"')
time.sleep(2)
send_at_cmd_pretty('AT!="addscanfreq band=8 dl-earfcn=3730"') #band & earfcn
time.sleep(2)
send_at_cmd_pretty('AT!="zsp0:npc 1"')
time.sleep(2)
send_at_cmd_pretty('AT+CEREG=2')
time.sleep(1)
send_at_cmd_pretty('AT+CGDCONT=1,"IP","vdf"') #apn
time.sleep(1)

However, register & attach attempts were only intermittently successfull, either success to register or failed with AT+CEREG? returned 2,4 (4 --> unknown cause).

At some point, the device flash memory seemed to be wiped by itself, all the scripts were gone. After this event, connection attempts were always failed with AT+CEREG? returned 2,4.

Any lights to this issue?

Cheers!

[dp]

@xykon As others have said, thank you for the update. Waiting patiently.

I'll also add my +1 to the above comments regarding "unbricking" or, at least, some other resolution for the devices which have been bricked.

Lastly, it would be helpful to publish a "confirm the device is bricked" guidance. For instance, my device still responds to AT commands such as:

print(lte.send_at_cmd('ATI1'))
UE5.0.0.0d
LR5.1.1.0-36417
OK

Is that an indication that it is still functional? Or is it possible to be unable to perform its cellular functions ("bricked") while still responding to AT commands?

Thanx again,
dp

[tuftec]

Can we please have an indication of when an updated solution for NB-IoT will be available for testing.
Peter.

[RWall]

@Xykon, add my +1 to sh4nnongoh's request for a description of the two issues. Is one of them power supply compliance? Thanks!

[tuftec]

Any further progress on this issue?
I have developments dependant on nb-iot working.
It has been 3 weeks since pulling the problematic firmware release.
Thanks
Peter.

[RWall]

tuftec beat me to the request for an update. "A few more days" has now extended out to a few weeks even though Pycom had " received two corrections from Sequans." Like others, I have Pycom hardware gathering dust while critical time is being lost. This needs to be made a top priority.

Specific details on the 'two issues that cause the modem to become "bricked"' should also be shared with the community. The lack of details is concerning. Roger

[IOAg]

Not sure when you guys will have an ETA on this issue, yet do you mind providing us an alternate download link for the firmware...

[Ben]

@xykon Any ETA on when the firmware update will be available again? Currently I have a stalled project where I require CAT-M1 availability.

[Jiemde]

@xykon I'm also waiting the firmware update for an NB-iot project, is it possible to have at least news of the progress of the solving issues.

[dan]

Hi everyone,

We are finalising the update process and will be releasing the instructions next week. We spent the past few weeks working on this issue and making sure the update will fix the problems.

Thank you for waiting, check back next week for more details!

[arneme]

@dan Will NB-IoT support for Huawei eNodeB based networks be part of the update? If not, what is the roadmap for that?

[tuftec]

Ok.

The update using the new firmware 1.18.1r1 appears to work. At least the update part anyway.
Thanks

Peter.

[IOAg]

@tuftec I did not see any announcement for the patch. Is that safe to install? Have changed been loaded in github?.
Thanks.
Philippe

[robert-hh]

@ioag The patch was published yesterday, and the github repository was updated. The new version is 1.18.1.r1.
As a statistical irrelevant info: I have build the image from the repository and updated two FiPy's.

[IOAg]

@robert-hh I did succeed in upgrading a couple of fipys. Thanks for your help.

[tuftec]

I also managed to update ok. But this is my first attempt to get NB-IoT working.
I get a message back from the modem indicating that it does not understand the zsp0:npc command.

Anyone else seen this? What does it mean?

Peter.

[dan]

@arneme Huawei eNodeB is supported, but the network operators could have specific protocol requirements.

[dan]

We have released an update (v1.18.1.r1) with updated scripts for modem upgrade. You can find the forum post here with more info.

For the bricked modems, we are still working on the process. We'll make a post here on the forum when it's published.