Hi Kenneth,
You should check for drive or XFS related errors in /var/log/message
files on all nodes. We've had a similar issue in the past with a bad
block on a hard drive.
We've had to :
1. Stop the OSD associated to the drive that had a bad block, flush its
journal (ceph-osd -i $osd --flus
This says it's actually missing one object, and a repair won't fix that (if
it could, the object wouldn't be missing!). There should be more details
somewhere in the logs about which object.
On Wed, Oct 4, 2017 at 5:03 AM Kenneth Waegeman
wrote:
> Hi,
>
> We have some inconsistency / scrub error
