On Wed, Mar 14, 2018 at 1:59 PM, Marc Marschall <marc@xxxxxxxxxxxxxx> wrote: > Hello there, > > I am trying to set up Cephfs on and for a testing system which will be > running for a couple of months.It is a single node system so far (of course > a non-ideal setup only for non prouctive usage). I fail to mount my cephfs > with error 5. > > I set up ceph on centos7 using the following script (if I made some mistakes > in it feel free to point out). > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% > #!/bin/bash > HOSTS="supervisor" > OSDS="/dev/sdc /dev/sdd /dev/sde /dev/sdf /dev/sdg /dev/sdh /dev/sdi > /dev/sdj /dev/sdj /dev/sdk /dev/sdl /dev/sdm /dev/sdn" > CEPH_PATH="/root/ceph_setup/ceph-deploy-1.5.39" > > echo "hosts: $HOSTS" > echo "osds: $OSDS" > > read -p "Press enter to install software" > > cd ../ceph_data > > > #install software > $CEPH_PATH/ceph-deploy new $HOSTS > echo "osd crush chooseleaf type = 0" >> ceph.conf > echo "osd_pool_default_size = 2" >> ceph.conf > echo "public_network = 10.0.0.0/24" >> ceph.conf > echo "cluster_network = 10.0.0.0/24" >> ceph.conf > echo "max_open_files = 131072" >> ceph.conf > $CEPH_PATH/ceph-deploy install --release luminous $HOSTS > > read -p "Press enter to create mon" > > # create monitor > $CEPH_PATH/ceph-deploy mon create-initial > read -p "Press enter to copy data" > > #copy keys and configuration > $CEPH_PATH/ceph-deploy admin $HOSTS > read -p "Press enter to create mgr" > > #create manager > $CEPH_PATH/ceph-deploy mgr create $HOSTS > read -p "Press enter tozap ad create osds" > > #create osd > for i in $OSDS; do > $CEPH_PATH/ceph-deploy disk zap $HOSTS:$i > $CEPH_PATH/ceph-deploy osd create $HOSTS:$i > done > read -p "Press enter to create pool" > > #create ceph pool > ceph osd pool create rbd 64 > read -p "Press enter to add cepfs" > #create cephfs > ceph osd pool create cephfs_data 64 > ceph osd pool create cephfs_metadata 64 > ceph fs new ceph_data cephfs_metadata cephfs_data > > cd ../ceph_setup > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% > > I think it is running fine: > > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% > ceph -s > cluster: > id: 9f75f5c6-e2c5-4627-92b8-6ce8935aecf7 > health: HEALTH_OK > > services: > mon: 1 daemons, quorum supervisor > mgr: supervisor(active) > mds: ceph_data-0/0/1 up This indicates you don't have any running MDS, which you need for a filesystem to work. And indeed I don't see you doing that in your prior scripting. > osd: 12 osds: 12 up, 12 in > > data: > pools: 3 pools, 192 pgs > objects: 0 objects, 0 bytes > usage: 12714 MB used, 44698 GB / 44711 GB avail > pgs: 192 active+clean Also at some point in here I think you'd have an "fs" block of output if that had worked, but I could be misremembering, or it only appears if there are MDS reports to go along with it. -Greg _______________________________________________ ceph-users mailing list ceph-users@xxxxxxxxxxxxxx http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
 |