it seems our unbound resolvers running on rhel 9 were upgraded during the night and can't start anymore. Downgrading to unbound-1.16.2-8.el9_5.1.x86_64 (from unbound-1.16.2-17.el9.x86_64 which seems to be part of a 9.6 batch that was pushed to CDN) fixes the problem but that means we'll get another issue on next update, so investigating
unbound-1.16.2-8.el9_5.1.x86_64
unbound-1.16.2-17.el9.x86_64
Metadata Update from @arrfab: - Issue assigned to arrfab
Metadata Update from @arrfab: - Issue tagged with: centos-common-infra, high-gain, high-trouble, investigation
Seems related to these commits :
https://gitlab.com/redhat/centos-stream/rpms/unbound/-/commit/0112876df6d5667ae6d469e54efaff57a52041ca
https://gitlab.com/redhat/centos-stream/rpms/unbound/-/commit/67a0f749d9a509d8c35248787b5f66f3956b5c52
investigating why the change and if that's the root cause but clearly related to new error messsage
So indeed due to these changes to systemd and how unbound is started, new config change is needed as by default unbound would try to chroot into a directory and so now would fail to find the path to notify :
chroot
unbound[936911]: [936911:0] fatal error: sd_notify failed /run/systemd/notify: No such file or directory. Make sure that unbound has access/permission to use the socket presented by systemd.
So a new chroot: "" (to disable chroot in fact) was pushed to our ansible role to take that into account. All applied and working now with new unbound pkg
chroot: ""
Metadata Update from @arrfab: - Issue close_status updated to: Fixed with Explanation - Issue status updated to: Closed (was: Open)
Log in to comment on this ticket.