[BUG][1.3] [TC-RR-1.1] Invalid Command error when executing TC_RR_1_1 after several pair/unpair operations

[hung-hotrong]

Reproduction steps

SDK version: v1.3.0.0
Description:

• After several pair/unpair operation, FabricIndex is increased and can be greater than 10, which causes TC_RR_1_1 to fail. This issue blocks us from test harness

Step 1. Put device in commissionable mode, then pair with chip-tool

chip-tool pairing onnetwork 1 77402238

Step 2. unpair the device

chip-tool pairing unpair 1

Step 3. Repeat step 1 and step 2 10 times
Step 4. Execute TC_RR_1_1

rm -f admin_storage.json && python3 python_testing/scripts/sdk/TC_RR_1_1.py --discriminator 3840 --passcode 77402238 --commissioning-method on-network --paa-trust-store-path /paa_roots --storage-path admin_storage.json

Result. Test case failed with chip.interaction_model.InteractionModelError: InteractionModelError: InvalidCommand (0x85) at step 2a because label length = 64 bytes but matter expects 32 bytes

[MatterTest] 02-13 17:42:57.556 INFO Step 2a: Setting fabric label on fabric 12 to '1212121212121212121212121212121212121212121212121212121212121212' using client RD0A

Recovery:

• a reboot doesn't reset fabric index (even when all fabrics are already removed before) so doesn't solve the issue.
• issue gone after a factory reset (removal of chip_kvs)

Log attached.

TC_RR_1_1_InvalidCommand.log

Bug prevalence

Always

GitHub hash of the SDK that was being used

5bb5c9e

Platform

raspi

Platform Version(s)

No response

Type

Platform Issue

Anything else?

No response