The iSCSI socket address does not change. But the cma_id at the IB
layer, which is bound to the iSCSI socket, will change. The problem
is that the new cma_id is trying to bind to a socket address that is
still bound to the old cma_id. That is, before you bind a new
cma_id to a socket, you must first delete the old one.
So why is iser trying to rebind a listening socket to the same
address? Isn't that the bug here? Just don't do that.
Hey Json,
Yes it is broken, IIRC it was meant to try and handle cases
where the address was restroed, but that needs to do is a periodic
attempt to repair (as bind is expected to fail). Similar to how
nvmet-rdma does this.
(I also have some vague memory that some bonding events mapped to this
cma event although the address was still available but that was a long
time ago so I'm not sure, and the change log says nothing about it...)
