By using AWS re:Post, you agree to the AWS re:Post Terms of Use
AWS re:Postre:Post

Fuota Session error: FragmentationIndexUnsupported

0

We're facing some issues on the FUOTA session, specifically on the FragSessionSetupReq command which AWS IoT Core issues from the Fragmented Data Block Transportation Specification. Once the multicast setup is ready for a multicast group which include just a single device, we schedule a FUOTA task for that group. Unfortunately, we always get the FragmentationIndexUnsupported error: the FragSession field we get is "0x1F" (FragIndex: 1) Is there any way to set the FragIndex on AWS? Still, how can we set the FragMaxSize, NbFrag and the FragMaxRedundancy? We uploaded several firmware images of different size, and it seems that AWS IoT Core always splits them in fragments of 48 bytes.

Topics
Internet of Things (IoT)
Tags
AWS IoT Core
Language
English
OroN
asked a year ago252 views
2 Answers
0

Hi OroN. What device are you using? Does it support only a single fragmentation session and hence an index of 1 exceeds its capacity (in other words, are you saying AWS IoT Core for LoRaWAN sends an index of 1 but you think it should be 0)?

In regards to the fragment and number of fragments, these settings are not currently exposed to the user. However, they change based on the DR inherited from the multicast group.

UPDATE: AWS IoT Core for LoRaWAN uses fragmentation index 0 for unicast FUOTA and fragmentation index 1 for multicast FUOTA.

profile pictureAWS
EXPERT
Greg_B
answered a year ago
  • OroN
    a year ago

    Yes we use an embedded device with a Lorawan stack library supporting only one multicast session. The multicast index value is zero due to the following requirements in "Lorawan Remote Multicast Setup specification v1.0.0". Is it possible to set the AWS lorawan multicast index to zero to be compliant to the requirement?

  • Greg_B EXPERT
    a year ago

    You said in your question that the error is FragmentationIndexUnsupported. This relates to fragmentation index and session, not multicast. If your device is using LoRaMAC-node, then the error comes from here: https://github.com/Lora-net/LoRaMac-node/blob/master/src/apps/LoRaMac/common/LmHandler/packages/LmhpFragmentation.c#L373 . I re-tested FUOTA. AWS IoT Core for LoRaWAN correctly used multicast index 0. However, the fragmentation index is 1 when I think it should be 0 for the first session. I have reported it. I will keep you posted.

  • Greg_B EXPERT
    a year ago

    Are you able to modify the device firmware? If so, you could increase FRAGMENTATION_MAX_SESSIONS to get un-blocked for the time being.

  • Greg_B EXPERT
    a year ago

    OK, AWS IoT Core for LoRaWAN uses fragmentation index 0 for unicast FUOTA and fragmentation index 1 for multicast FUOTA.

0

Yes we use an embedded device with a Lorawan stack library supporting only one multicast session. The multicast index value is zero due to the following requirements in "Lorawan Remote Multicast Setup specification v1.0.0"

Enter image description here

Is it possible to set the AWS lorawan multicast index to zero to be compliant to the requirement?

OroN
answered a year ago
  • Greg_B EXPERT
    a year ago

    As per my other comment, the issue is the fragmentation session and index, not multicast.

You are not logged in. Log in to post an answer.

A good answer clearly answers the question and provides constructive feedback and encourages professional growth in the question asker.

Guidelines for Answering Questions

Relevant content