a quick respond. I quickly double checked with whack --rekey uncommented It works as expected. So I pushed the change. Lets see output testing.libreswan.org produce, however, pluto.log might be empty.
One odd thing is empty pluto.log in master. That makes it a bit harder to analyze. https://testing.libreswan.org/v3.30-162-g412a6e8d86-master/interop-ikev2-strongswan-35-initiator-rekey/OUTPUT/ notice 35 bytes west.pluto.log.gz If you notice rekey issues, I would like to see the test's full output. including pluto.log and charon.log. Without it I can can't respond or comment. -antony On Fri, Feb 28, 2020 at 12:06:51AM -0500, Paul Wouters wrote: > > It seems interop-ikev2-strongswan-35-initiator-rekey is not working. > > The config file has commented out values for "timer based rekey" and > the shell script has commented out values for "command based rekey". > Due to that, the reference output does not show traces of rekey. > So this test never seems to actually do a rekey against strongswan. > > When I changed it to timer based rekey, because according to comments, > 3.29 rekey command is broken, I still do not see strongswan rejecting > anything bad if west is libreswan 3.29 or 3.30. I was epxecting the > test to succeed for 3.29 and fail for 3.30. > > I was hoping to confirm my patch to be stricter about address family > check in TS TYPE using this test, but I can't seem to make this test > work to ever show a failure during rekey over a bad address family > towards strongswan. > > Paul > _______________________________________________ > Swan-dev mailing list > Swan-dev@lists.libreswan.org > https://lists.libreswan.org/mailman/listinfo/swan-dev _______________________________________________ Swan-dev mailing list Swan-dev@lists.libreswan.org https://lists.libreswan.org/mailman/listinfo/swan-dev