Ondrej Moris
Ondrej Moris
I am wondering why `#2633` eventually gets deleted while `#885` does not, I would expect the same behavior...?
Oh, that explains it. Sounds like we are in a pretty pickle. If I get it right the problem is: > `#2633` can't send a delete as it has IKE_AUTH...
@cagney @ueno I tried to get back to this and I am puzzled by this TEMPORARY_FAILURE. Suppose that left side of the connection receives IKE_AUTH request from the other side...
(edited previous message in order to keep relevant information, the original message was actually a rather confusing) I managed to reproduce this, finally. **Configuration** ``` conn test type=transport authby=secret auto=add...
XFRM actually has a policy to always choose the latest state so I am wondering why it wasn't the case when we hit the problem for the first time (see...
Good point, I'll add one.
Hm, it seem tricky to keep both IKE SA in the test. I am able to make it work on my VMs but for some reason the same procedure does...
Right, that sounds better IMO. Let me fix it quickly.
@mamtagambhir IMO the current approach (targeting permanent connections only) would not fix your issue since in your case connection are not permanent (but IMO it can be extend to cover...
@mamtagambhir Sorry for the delay, is the connection actually broken when you encounter this problem?