In July, I reported a wm(4) interface had stopped receiving after a successful mpt(4) timeout recovery. ifconfig down/up solved the problem.
Yesterday, on the same machine, the interface again stopped working (I didn't check whether it did work in one direction) and again, ifconfig up/down helped. However, this time, there was no mpt(4) timeout recovery involved. So maybe the mpt timeout recovery is not causing the wm lockup, but they both have some common cause (lost interrupts or the like)?
