Nah, it was me not running container versions.
The bootstrap keys are used to get daemons up without giving out too much admin access to them, so my guess is that as soon as you have a cluster going, you can always read out or create new bootstrap keys if needed later, but they are not necessary for getting a crashed cluster up, rather to expand or initially create nodes.
Den tis 16 juli 2019 kl 17:58 skrev Oscar Segarra <oscar.segarra@xxxxxxxxx>:
Thanks a lot Janne,Well, maybe I'm missunderstanding how ceph stores keyrings in etcd...
bootstrap="bootstrap${array[1]}Keyring" etcdctl "${ETCDCTL_OPTS[@]}" "${KV_TLS[@]}" set "${CLUSTER_PATH}"/"${bootstrap}" < "$keyring" But I'd like to know what happens if etcd loses the keyrings sotred in it when etcd is used to deploy ceph daemons as containers:With KV backend:
docker run -d --net=host \ --privileged=true \ --pid=host \ -v /dev/:/dev/ \ -e OSD_DEVICE=/dev/vdd \ -e KV_TYPE=etcd \ -e KV_IP=192.168.0.20 \ ceph/daemon osdThanks a lot for your help,ÓscarEl mar., 16 jul. 2019 17:34, Janne Johansson <icepic.dz@xxxxxxxxx> escribió:Den mån 15 juli 2019 kl 23:05 skrev Oscar Segarra <oscar.segarra@xxxxxxxxx>:--Hi Frank,Thanks a lot for your quick response.Yes, the use case that concerns me is the following:1.- I bootstrap a complete cluster mons, osds, mgr, mds, nfs, etc using etcd as a key storeas a key store ... for what? Are you stuffing anything ceph-related there? If so, please tell us what.As previously said, ceph has no etcd concept so unless you somehow pull stuff out of ceph and feed it into etcd, ceph will be completely careless if you lose etcd data.May the most significant bit of your life be positive.
May the most significant bit of your life be positive.
_______________________________________________ ceph-users mailing list ceph-users@xxxxxxxxxxxxxx http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
