https://github.com/trusteddomainproject/OpenDKIM/issues unsure if ed25519-sha256 works with openssl or not :/ no patch ready imho using beta2 is untrusted unstable, but it would be nice to see if it builds at all https://github.com/trusteddomainproject/OpenDKIM/releases/tag/rel-opendkim-2-11-0-Beta2
(In reply to Benny Pedersen from comment #0) > https://github.com/trusteddomainproject/OpenDKIM/issues I guess you mean [1] which currently lists 5 issues. Or [2] (which the HOMEPAGE links to for bug tracking) which lists many more, probably because of a misinterpretation of the dash in "ed25519-sha256". > unsure if ed25519-sha256 works with openssl or not :/ Do you mean you are not sure of opendkim (somehow) configured for "ed25519-sha256" might work as expected with a specific version of dev-libs/openssl? > no patch ready imho Patch for what problem? [1] https://github.com/trusteddomainproject/OpenDKIM/issues?utf8=%E2%9C%93&q=is%3Aissue+is%3Aopen+ed25519 [2] https://sourceforge.net/p/opendkim/bugs/search/?q=ed25519-sha256&limit=50
Perhaps maintainers can fix the Summary once you figure out what the problem is.
Upstream development has pretty much stopped. The referenced beta (!) release is almost a year old now. Also, I opened pull request https://github.com/trusteddomainproject/OpenDKIM/pull/41 on April 2, 2019 and we have not even received a comment yet. My conclusion as of today is that spending time on OpenDKIM to support new signature types is not worth it as long as upstream development is paused. The combination of pymilter/dkimpy-milter seems more promising to me. Both authors have accepted my recent contributions, and the library/milter are actively maintained by them.
If neither Michael nor Francisco have an objection, I will close this issue.
(In reply to Ralph Seichter from comment #4) > If neither Michael nor Francisco have an objection, I will close this issue. None here.
I may try to fix this one a lot later on. Consider reopening this in December as I may have more time to try to backport the beta patches.
i have used opendkim 2.11 in last 24 hours, no problem with it, so beta2 is more or less ready for gentoo i have disabled poll, it makes QA atps its depricated vbr experimental not used widely, this could be added to use flag if demend for it my unstable ebuild is in ::fidonet overlay if there is some that would like to try it i will make systemd work again, just to laysy not make it right now since i dont use systemd also tarball have imho no working initd for openrc, so used the one from 2.10 that still works https://tools.ietf.org/html/rfc6376#section-4.1 sha-1024 is planned in ietf hope the best from here
Benny, what part of "Consider reopening this in December" do you find confusing?
