Ok... We haven't used cephadm/orch at all - it was a very
straightforward manual install. But looking at the PR it seems that
container-related complexities are now spilling over into traditional
deployments too.
This one is easily worked around. But I hope we don't keep getting these
impacts.
Thanks, Chris
On 22/07/2021 17:09, Ernesto Puerta wrote:
Hey Chris,
Nothing dumb from your end. Cephadm has recently changed the way it
configures the network endpoints favoring IP addresses over hostnames.
I think this was the change: https://github.com/ceph/ceph/pull/41483/
<https://github.com/ceph/ceph/pull/41483/>
I may suggest you to use a SAN with the IP address/es in your certificate?
Kind Regards,
Ernesto
On Thu, Jul 22, 2021 at 3:45 PM Chris Palmer <chris.palmer@xxxxxxxxx
<mailto:chris.palmer@xxxxxxxxx>> wrote:
Since updating from Pacific 16.2.4 -> 16.2.5 I've noticed a behaviour
change in the Dashboard. If I connect to the active MGR, it is fine.
However if I connect to a standby MGR, it redirects to the active
one by
placing the active IP address in the URL, rather than the active
hostname as it used to.
When accessing over HTTPS it "works", except that our certificate
doesn't have the IP address as a valid name, so we get certificate
warnings.
Does anyone else see this too, or have I done something dumb?
Thanks, Chris
_______________________________________________
ceph-users mailing list -- ceph-users@xxxxxxx
<mailto:ceph-users@xxxxxxx>
To unsubscribe send an email to ceph-users-leave@xxxxxxx
<mailto:ceph-users-leave@xxxxxxx>
_______________________________________________
ceph-users mailing list -- ceph-users@xxxxxxx
To unsubscribe send an email to ceph-users-leave@xxxxxxx
