I would check to see if the images have an exclusive-lock still held by a force-killed VM. librbd will generally automatically clear this up unless it doesn't have the proper permissions to blacklist a dead client from the Ceph cluster. Verify that your OpenStack Ceph user caps are correct [1][2]. [1] http://docs.ceph.com/docs/master/releases/luminous/#upgrade-from-jewel-or-kraken [2] http://docs.ceph.com/docs/luminous/rbd/rbd-openstack/#setup-ceph-client-authentication On Tue, Dec 4, 2018 at 8:56 AM Simon Ironside <sironside@xxxxxxxxxxxxx> wrote: > > On 04/12/2018 09:37, linghucongsong wrote: > > But it is just in case suddenly power off for all the hosts! > > > I'm surprised you're seeing I/O errors inside the VM once they're restarted. > Is the cluster healthy? What's the output of ceph status? > > Simon > _______________________________________________ > ceph-users mailing list > ceph-users@xxxxxxxxxxxxxx > http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com -- Jason _______________________________________________ ceph-users mailing list ceph-users@xxxxxxxxxxxxxx http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
 |