CC1310: Sensor went unnormal and switch between Jdllc_states_orphan and Jdllc_states_joined after hours of test.

dennis yang

Intellectual 945 points

Part Number: CC1310

Hi TI friend:

Now we have met a problem of stability.

After 10 hours of test,sensor went to abnormal state and can't recover.

from log we can see:

1.jdllc_State switched between Jdllc_states_orphan and Jdllc_states_joined.

2.pollcnf can't be received after issue time,but CoordRealign can be received after sending orphanScanReq

3.after resetting sensor,it went normal,so it seems to be error of sensor side.

1.SDK as bellow:

simplelink:simplelink_cc13x0_sdk_1_60

2.MAC security : closed; non beacon mode,polling interval 1 second

3.Sensor board: TI lauchpad

[Fri May 18 06:57:19.051 2018] enter pollCnfCb,pollCnt/60 = 642 //we print pollcnf cnt every minute
[Fri May 18 06:58:01.096 2018] enter pollCnfCb,pollCnt high = 0x0
[Fri May 18 06:58:01.111 2018] pollCnt low = 0x1
[Fri May 18 06:58:01.118 2018] status = 0x0
[Fri May 18 06:58:01.118 2018] dataIndCB
[Fri May 18 06:58:01.118 2018] cmdId = 3
[Fri May 18 06:58:01.118 2018] Sensor_sendMsg type = 4
[Fri May 18 06:58:01.118 2018] enter dataCnfCb,status = 0x0
[Fri May 18 06:58:10.166 2018] enter pollCnfCb,pollCnt high = 0x0
[Fri May 18 06:58:10.180 2018] pollCnt low = 0x1
[Fri May 18 06:58:10.180 2018] status = 0xE9 //no ack of poll
[Fri May 18 06:58:10.767 2018] enter pollCnfCb,pollCnt high = 0x0
[Fri May 18 06:58:10.771 2018] pollCnt low = 0x1
[Fri May 18 06:58:10.773 2018] status = 0xE9 //no ack of poll
[Fri May 18 06:58:11.380 2018] enter pollCnfCb,pollCnt high = 0x0
[Fri May 18 06:58:11.393 2018] pollCnt low = 0x1
[Fri May 18 06:58:11.398 2018] status = 0xE9 //no ack of poll
[Fri May 18 06:58:11.398 2018] State Changed: 5
[Fri May 18 06:58:11.398 2018] Board_Led_control,led = 0
[Fri May 18 06:58:11.398 2018] state = 0 //led state ,please ignore
[Fri May 18 06:58:11.398 2018] Board_Led_control,led = 1
[Fri May 18 06:58:11.439 2018] state = 0
[Fri May 18 06:58:11.439 2018] enter processState,state = 3
[Fri May 18 06:58:11.439 2018] send orphan scan request //orphan scan req send
[Fri May 18 06:58:11.458 2018] sendScanReq,type = 3
[Fri May 18 06:58:11.458 2018] enter scanCnfCb,status = 0x0
[Fri May 18 06:58:11.458 2018] devInfoBlock.panID = 4403
[Fri May 18 06:58:11.458 2018] CoordRealign //got CoordRealign

[Fri May 18 06:58:11.458 2018] State Changed: 3
[Fri May 18 06:58:13.763 2018] State Changed: 5
[Fri May 18 06:58:13.814 2018] State Changed: 3
[Fri May 18 06:58:16.182 2018] State Changed: 5
[Fri May 18 06:58:16.216 2018] State Changed: 3
[Fri May 18 06:58:18.585 2018] State Changed: 5
[Fri May 18 06:58:18.605 2018] State Changed: 3
[Fri May 18 06:58:20.966 2018] State Changed: 5
[Fri May 18 06:58:20.989 2018] State Changed: 3
[Fri May 18 06:58:23.323 2018] State Changed: 5
[Fri May 18 06:58:23.359 2018] State Changed: 3
[Fri May 18 06:58:25.670 2018] State Changed: 5
[Fri May 18 06:58:25.710 2018] State Changed: 3
[Fri May 18 06:58:28.037 2018] State Changed: 5
[Fri May 18 06:58:28.088 2018] State Changed: 3

we have tested for days and this issue have been found only one time.

As our product is 7*24 product.

Can you help to give some advice to find the root cause?

over 5 years ago

0 IUsedToBeARealPerson (M-W) over 5 years ago

TI__Guru 58425 points

Hi Yang,

It is hard to give any advice on the issue simply based on this print put. Have you since posting this found a way to recreate this in a controlled way? If you could find a way to recreate this I could setup a test myself and have a look.

Also, what SDK version are you currently using?

0 dennis yang over 5 years ago in reply to IUsedToBeARealPerson (M-W)

Intellectual 945 points

Hi M-W:
Thanks for replying.
we only met this problem for only one time.
Still have no way of reproducing this issue.

BR

0 IUsedToBeARealPerson (M-W) over 5 years ago in reply to dennis yang

TI__Guru 58425 points

Hi Yang,

Any updates on this, have you seen this issue any more or found a way of reproducing it?

0 dennis yang over 5 years ago in reply to IUsedToBeARealPerson (M-W)

Intellectual 945 points

Hi M-W:

still trying to find way of reproducing this issue.

it should have something to do with the environment as we have lots of collector working at the same channel.

Thanks & BR

0 IUsedToBeARealPerson (M-W) over 5 years ago in reply to dennis yang

TI__Guru 58425 points

Hi Yang,

I will close this thread due to inactivity. Feel free to open up a new thread (ask related question to link to this one) if you find the problem or a way to reproduce it.

0 Liuwei over 5 years ago in reply to IUsedToBeARealPerson (M-W)

Prodigy 30 points

Hi M-W,
I think I have the same issue.

SDK:simplelink_cc13x0_sdk_2_10_00_36
Beacon-Mode
CONFIG_MAC_BEACON_ORDER 5
CONFIG_MAC_SUPERFRAME_ORDER 5
MAC Security Enable
Channel: Fix at CH0 (CONFIG_CHANNEL_MASK { 0x01, 0x00, 0x00, 0x00, 0x00, 0x00, \
0x00, 0x00, 0x00, 0x00, 0x00, 0x00, \
0x00, 0x00, 0x00, 0x00, 0x00 })
Devices: TI LaunchPad

FW:
TI example, Collector and Sensor
Collector keep send pkt to Sensor, the interval is 0.6s, run about several hours, the sensor reconnect to collector over and over again.
Sensor state switch between Jdllc_states_orphan and Jdllc_states_joined.

Please help me to fix this issue.
Thanks

0 dennis yang over 5 years ago in reply to Liuwei

Intellectual 945 points

Hi M-W:

We have already reproduced this issue for several times but still cannot find the rule of reproducing it.

From log of collector,the collector send realiagn after getting orphan-indication from sensor,and sensor went to Jdllc_states_joined state but cannot get poll resp from collector and went to Jdllc_states_orphan, then over and over again.

But the collector can still connect with other sensors normally.

So there should be bug I'm not sure coming from collector or sensor.

Thanks&BR

0 IUsedToBeARealPerson (M-W) over 5 years ago in reply to dennis yang

TI__Guru 58425 points

I will try to put up a longer test scenario on this and see if I can catch the same issue you are seeing. I will let you now what I find out.

0 dennis yang over 5 years ago in reply to IUsedToBeARealPerson (M-W)

Intellectual 945 points

Hi M-W:

Thanks for your help.

You can test with interference to see if this issue can be reproduced.

0 IUsedToBeARealPerson (M-W) over 5 years ago in reply to IUsedToBeARealPerson (M-W)

TI__Guru 58425 points

Hi,

I tested the this for an extended time using both the standard examples and using the settings provided by Liuwei, however I was not able to reproduce the issue you are seeing.

Could you maybe provide a sniffer log of the traffic between the sensor and collector when this happens, preferably from Wirehark? It could help in figuring out how to reproduce the issue.

0 IUsedToBeARealPerson (M-W) over 5 years ago in reply to IUsedToBeARealPerson (M-W)

TI__Guru 58425 points

Hi,

Do you got any more feedback on my last post?

0 dennis yang over 5 years ago in reply to IUsedToBeARealPerson (M-W)

Intellectual 945 points

Hi M-W:
We are still trying to reproduce this issue and get OTA log.

But it is not easy to get log of normal to abnormal as it is really hard to reproduce.

Thanks

Sub-1 GHz

Sub-1 GHz forum

CC1310: Sensor went unnormal and switch between Jdllc_states_orphan and Jdllc_states_joined after hours of test.