Rgw hammer -> jewel Next method helped me After upgrading newly remake rgw on jewel ceph auth del client.rgw.ceph403 rm /var/lib/ceph/radosgw/ceph-rgw.ceph403/ ceph-deploy --overwrite-conf rgw create ceph403 systemctl stop ceph-radosgw.target systemctl start ceph-radosgw.target systemctl status ceph-radosgw.target
After updating of all rgw if they still fall down Check ceph -s has no errors Ceph osd tree shows all osd are there So lets start treatment of schema Must all rgw be stoped systemctl stop ceph-radosgw.target wget https://raw.githubusercontent.com/yehudasa/ceph/wip-fix-default-zone/src/fix-zone chmod +x fix-zone ./fix-zone It makes backup of rgw.root and repairs zones saving next files to current dir: default-zg.json default-zone-fixed.json Edit them by inserting realm_id (radosgw-admin realm list) Remove symbol "_" at the endings of pool names radosgw-admin zonegroup set --rgw-zonegroup default < default-zg.json radosgw-admin zone set --rgw-zonegroup default --rgw-zone default < default-zone-fixed.json radosgw-admin zonegroup default --rgw-zonegroup default radosgw-admin zone default --rgw-zone default check radosgw-admin zonegroup list radosgw-admin zonegroup get radosgw-admin zone list radosgw-admin zone get radosgw-admin period update no errors, only then commit radosgw-admin period update --commit systemctl start ceph-radosgw.target check that everything is ok radosgw-admin metadata list user radosgw-admin bucket list -- Petr Malkov
_______________________________________________ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
