Network Switches Port Roles Appear Incorrect

Please provide the following info (tick the boxes after creating this topic):
Software Version
DRIVE OS 6.0.6
DRIVE OS 6.0.5
DRIVE OS 6.0.4 (rev. 1)
DRIVE OS 6.0.4 SDK
other

Target Operating System
Linux
QNX
other

Hardware Platform
DRIVE AGX Orin Developer Kit (940-63710-0010-D00)
DRIVE AGX Orin Developer Kit (940-63710-0010-C00)
DRIVE AGX Orin Developer Kit (not sure its number)
other

SDK Manager Version
1.9.2.10884
other

Host Machine Version
native Ubuntu Linux 20.04 Host installed with SDK Manager
native Ubuntu Linux 20.04 Host installed with DRIVE OS Docker Containers
native Ubuntu Linux 18.04 Host installed with DRIVE OS Docker Containers
other

Hello NVIDIA Team,

While testing the Orin Drive DevKit we’ve run into an issue with the on board switches (Spruce). The documents here show that all ports should be defaulted as a T1 Primary (Master). We are using the BASE configuration (we have also tried safety but with no success).

In order to validate the port roles we used the NVIDIA provided 1G Media Converter and read out the status messages over a terminal window. All ports on Spruce are showing as a T1 Master when connected. We tested this with a laptop going from port J4 → NV Media Converter → Laptop. We tested that the laptop can be pinged successfully and communication is working. The print out shows that the port role on the Media Converter is a Master which means the Orin port is a Slave. From the documentation this should be a master.

We have tried to re-install the Switch firmware using

sudo /bin/bash /lib/firmware/marvell_ethernet/driveota/P3710_88Q6113.sh --Install <FW>

and we verified the switch firmware is 0.07.1186.01

We have checked on two DevKits and both show the same behaviour.

We have a number of Ethernet Sensors we’re looking to connect to this switch and they are all T1 Slaves and expecting that Orin is the T1 master. In this configuration that we are seeing right now we cannot connect to the sensors successfully.

Thank you for bringing this issue to our attention. Our team will investigate the problem and we will get back to you soon.

Could you please confirm whether the result was obtained from version 6.0.5 or 6.0.6? Thanks.

Hello VickNV,

We are currently running 6.0.5 and testing with it; working on migrating to 6.0.6 but we have not completed that migration.

Can you please provide the Aurix firmware version by running the “version” command in the Aurix console? Is it “DRIVE-V6.0.5-P3710-AFW-Aurix-StepB-5.06.05”?

Hello VickNV,

Checking the firmware on Aurix it looks to be as you mention

1 Like

Hi @VickNV , any update on the above issue. Thanks!

The issue is currently being investigated by our team. Could you provide more details on the urgency and whether this issue will impact your tasks? This will help us prioritize the investigation accordingly. Thank you for your patience.

This issue is completely blocking our development with Orin. We have multiple sensors in slave mode connected to this switch and not able to access any of them on Orin dev kit. Also, we have multiple devkits and all have the same issue.
It would be great if you could expedite resolving the issue.

@VickNV you mentioned above if there are any additional details that we can provide, is there something you’d like for us to check in helping with the debug?

The port is left as the default configuration for MTU (1466 i believe, which is 1500 with MACSec enabled) and we do initialize it to 192.168.1.100 and 192.168.2.100 for Spruce and Oak respectively on the XFI interface. Otherwise we have not made any changes to the Ethernet portion.

We have also, for debug, been using the EQOS port and an external media converter. With this we can configure the media converter to master mode and the appropriate speed and are able to communicate with devices we’re looking to connect in this case.

Thank you for providing these details. Our team is currently investigating the issue and we will let you know if we need any additional information. Please bear with us as we work on resolving the problem.

Hello @VickNV

just to follow-up to see if the team has been able to make any progress towards this issue. If there is anything you’d like us to try to help with this effort please let me know.