On Mon, Oct 21, 2019 at 07:34:36PM +0530, Gowrishankar Muthukrishnan wrote:
> It might be difficult to trace whether LACP PDU tx (as in
> response) was successful when the pdu was not transmitted by
> egress slave for various reasons (including resource contention
> within NIC) and only way to
Hi Ben,
Idea is to log lacp specific tx failure. A one step further as in
ofproto_dpif_send_packet has to make the debug info more generic, even
though it would cater all of its calling functions.
For example, bundle_send_learning_packets reports such tx failures for
gratuitous packets. I think,
On Mon, Oct 21, 2019 at 07:34:36PM +0530, Gowrishankar Muthukrishnan wrote:
> It might be difficult to trace whether LACP PDU tx (as in
> response) was successful when the pdu was not transmitted by
> egress slave for various reasons (including resource contention
> within NIC) and only way to
It might be difficult to trace whether LACP PDU tx (as in
response) was successful when the pdu was not transmitted by
egress slave for various reasons (including resource contention
within NIC) and only way to trace its fate is by looking at
ofproto->stats.tx_[packets/bytes] and slave port stats.