On Mon, 20 Mar 2023 at 21:41, Paul Wouters <[email protected]> wrote: > > > On Mar 21, 2023, at 09:33, Andrew Cagney <[email protected]> wrote: > > > > New commits: > > commit 98ebb28b89895d14aac2a4dab71edc54f4277220 > > Author: Andrew Cagney <[email protected]> > > Date: Mon Mar 20 18:19:40 2023 -0400 > > > > opportunistic: in orphan_holdpass() always replace the kernel policy > > > > Drop code skipping a replace when negotiation==failure. > > This way all failure shunts have 0 priority. > > This will also fix the text in the shunt table, eg “oe-trying” vs > “oe-failing”. But I’m not sure how atomic this operation is. Is there a > chance of packet leak?
The negotiation->bare and bare->prospective operations each use a single syscall. We should be good. > > > > Note: #1042 #1041 ... > > > > commit 210a0f6e5e73ce3fef2bf4e11299f79f85c246ff > > Author: Andrew Cagney <[email protected]> > > Date: Mon Mar 20 18:58:16 2023 -0400 > > > > testing: expect bare shunts with priority 0 > > > > _______________________________________________ > > Swan-commit mailing list > > [email protected] > > https://lists.libreswan.org/mailman/listinfo/swan-commit > _______________________________________________ > Swan-dev mailing list > [email protected] > https://lists.libreswan.org/mailman/listinfo/swan-dev _______________________________________________ Swan-dev mailing list [email protected] https://lists.libreswan.org/mailman/listinfo/swan-dev
