Skip to content

J1939 Driver 'Dies' on TPCM Connection Abort #7

@rpkvt

Description

@rpkvt

Hi Kurt,

My application has another mode which tunnels data packets over the J1939 CAN bus to a specific address using PropA PGN (manufacturer defined). After some time, the driver appears to send a Connection Abort request to the destination. The destination then responds and the J1939 driver stops working. In fact, the entire system restarts.

Here's an example where my device (address 128) is sending packets to my bus monitor (address 248):

image

The sequence in red illustrates a successful packet transfer. The sequence in blue shows the abort request and a response from the bus monitor.

This is the reason for the initial request made by the J1939 driver (already connected):
image

Response reason (resources needed):
image

These two messages repeat several times. Eventually the bus monitor sends an additional abort request (timeout is the reason):
image

image

This caused my linux box to restart. It seems to be the reception of this last abort request that does it.

Do you have any thoughts on this? I'm using sendto and it doesn't seem to be sending any errors back to the application.

Thanks,
Rich

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions