• State
  • Locked Locked
  • Replies 4 replies
  • Subscribers 55 subscribers
  • Views 502 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

TI OTA Upgrade server finding wrong PAN IDs

Jordan Reynolds1
Prodigy 60 points
Other Parts Discussed in Thread: CC2531, CC2538, Z-STACK, CC2530

Hello:


We are currently working with the TI OTA cluster + OTA server application.  We have built a project adding USB UART capabilities to the CC2531 dongle and all seems well, most of the time.  We did most of our testing with the dongle in co-coordinator mode, and have now switched to router mode.


What we have noticed is somewhat erratic behavior in the PANs found when the TI OTA server is used to scan for networks. On occasion the scan will identify the correct networks, but more often then not, it seems to shift the data related to the network.  For example (from the debug log output):

[07/28/2014 09:43:37.050] <Debug  > Pan 0x0202
Logical Channel = 0x03
Router Capacity = 0x67
Device Capacity = 0x10
Version = 0x0F
Stack Profile = 0x0F
 {OtaServer}
[07/28/2014 09:43:37.129] <Info   > New Pan Discovered -> Pan 0x0202 {MtOtaMsgs}

In this case, it has discovered a PAN it believes is 0x0202, however the correct PAN ID is 0x6703 (these values are parsed as the router capacity and logical channel values).  Of course the tool cannot connect to a PAN ID of 0x0202 and therefore is not able to upgrade any devices.


Any thoughts / suggestions what may be going wrong here?

Thanks

-JR

I

  • 0
    TI__Genius 12415 points

    Hi Jordan,

    In order for me to be able to help you with this problem I will need some more information. 

    - What function are you using to get the information or how are you retrieving this information.

    - If possible could you provide the block of code where you are parsing the data and also the sniffer log might be useful. 

    Hector

  • 0 in reply to Hector_r
    Prodigy 60 points

    Hi Hector:


    Thanks for the reply.

    I'll try and clarify our setup.

    - The network co-coordinator is a commercial gateway device.  This is currently configured to produce a network with PAN ID of 0x6702 on Channel 19 (tho I have tried a number of other channels with similar behavior).

    - Several target CC2538 based target devices runing TI Z-Stack with the OTA upgrade cluster implemented.

    - TI CC2531USB PC based dongle.  This is running the CC2530 OTA firmware provided as part of the Z-Stack modified to include the USB Uart driver.

    - PC Running windows 7 and the TI OTA upgrade server v2.0.3


    The basic process:

    - Form a network with the co-ordinator and a single target devices.  This network is left open for the duration.

    - Plug in the CC2531USB dongle

    -Launch the TI OTA upgrade server tool

    - Enable "Use operations log" (note the previous log output was from this tool).

    - Select "Scan for PANs" in the OTA upgrade server tool.

    The TI OTA upgrade tool operations log shows :

    [07/28/2014 09:43:37.050] <Debug  > Pan 0x0202
    Logical Channel = 0x02
    Router Capacity = 0x67
    Device Capacity = 0x10
    Version = 0x0F
    Stack Profile = 0x0F
     {OtaServer}
    [07/28/2014 09:43:37.129] <Info   > New Pan Discovered -> Pan 0x0202 {MtOtaMsgs}

    The following is a wire shark capture of the activity on the given channel:

    2    6.503391000        Broadcast    IEEE 802.15.4    84    Beacon Request

    3    6.592168000    0x0000        ZigBee    102    Beacon, Src: 0x0000, EPID: 5a:42:432d:42:5041:43

    The first line seems to be the scan request from the CC2531USB dongle, the second is the response from the gateway :

    IEEE 802.15.4 Beacon, Src: 0x0000
        Frame Control Field: Beacon (0x8000)
            .... .... .... .000 = Frame Type: Beacon (0x0000)
            .... .... .... 0... = Security Enabled: False
            .... .... ...0 .... = Frame Pending: False
            .... .... ..0. .... = Acknowledge Request: False
            .... .... .0.. .... = Intra-PAN: False
            .... 00.. .... .... = Destination Addressing Mode: None (0x0000)
            ..00 .... .... .... = Frame Version: 0
            10.. .... .... .... = Source Addressing Mode: Short/16-bit (0x0002)
        Sequence Number: 206
        Source PAN: 0x6702
        Source: 0x0000
        Superframe Specification
            .... .... .... 1111 = Beacon Interval: 15
            .... .... 1111 .... = Superframe Interval: 15
            .... 1111 .... .... = Final CAP Slot: 15
            ...0 .... .... .... = Battery Extension: False
            .1.. .... .... .... = PAN Coordinator: True
            1... .... .... .... = Association Permit: True
        GTS
            GTS Descriptor Count: 0
            GTS Permit: False
        Pending Addresses: 0 Short and 0 Long
        Frame Check Sequence (TI CC24xx format): FCS OK
            RSSI: 28 dB
            FCS Valid: True
            LQI Correlation Value: 108

  • 0 in reply to Jordan Reynolds1
    Prodigy 60 points

    Here is a sniff of the CC2531USB -> PC response over the USB bus.  It seems the first device in the list parses ok, and the remaining are shifted by a number of bytes (I suspect this is in the OTA upgrade server side, which I do not have the source for).  Also note the bottom item in the display shows the correct PAN ID, the second is shifted by 2 bytes, and the third is shifted by 4 bytes.

    1B 00 30 75 2C 10 80 FA FF FF 00 00 00 00 09 00
    01 03 00 02 00 84 03 48 00 00 00 FE 43 43 89 03
    02 67 13 0F 0F 01 01 02 02 AC CB 17 0F 0F 01 01
    02 02 87 0D 14 0F 0F 00 00 02 02 07 01 05 00 00
    B5 00 00 04 77 0A 00 C7 02 0D 00 10 05 FE 8D 11
    3C 00 23 84 C2 11 00 9E 11 0F E3 11 00 00 FE 5F
    A6 80 DF

    This is with 3 open networks in the office :

    0x6702, 0xCBAC  0x0D87 (the first of which is the first 2 bytes in the third line of data).

  • 0 in reply to Jordan Reynolds1
    Prodigy 60 points

    We have found a very old version of OTA console from a much older project, and this does not seem to exhibit the same bug as the OTA server, tho we cannot locate the installer for the OTA Console (a developer happened to have it on their machine already).

    Thanks

    -Jordan