On 08/02/2019 20.54, Ashley Merrick wrote:
> Yes that is all fine, the other 3 OSD's on the node work fine as expected,
> When I did the orginal OSD via ceph-deploy i used the external hostname
> at the end of the command instead of the internal hostname, I then
> deleted the OSD and zap'd the disk and re-added using the internal
> hostname + the other 3 OSD's.
> The other 3 are using the internal IP fine, the first OSD is not.
> The config and everything else is fine as I can reboot any of the other
> 3 OSD's and they work fine, just somewhere the osd.22 is still storing
> the orginal hostname/ip it was given via ceph-deploy even after a rm /
> disk zap

The OSDMap stores the OSD IP, though I *think* it's supposed to update
itself when the OSD's IP changes.

If this is a new OSD and you don't care about the data (or can just let
it rebalance away), just follow the instructions for "add/remove OSDs" here:


Make sure when the OSD is gone it really is gone (nothing in 'ceph osd
tree' or 'ceph osd ls'), e.g. 'ceph osd purge <id>
--yes-i-really-mean-it' and make sure there isn't a spurious entry for
it in ceph.conf, then re-deploy. Once you do that there is no possible
other place for the OSD to somehow remember its old IP.

Hector Martin (hec...@marcansoft.com)
Public Key: https://mrcn.st/pub
ceph-users mailing list

Reply via email to