I am trying to create a ceph cluster.
I've installed ceph on 3 computers and added their names to hosts files.
Created passwordless ssh login.
Installed and configured ntp server and checked, that all the computers show the same time/time zone.
Installed ceph and created mons on all hosts, that are part of ceph cluster with ceph-deploy.
zapped disks, allocated to ceph on each comp and created osds on them.
Checked, the osds were created on those disks.
ceph status shows health_warn, even after reboots of all of the computers and several days have passed. Also, it shows that I have several osds available, but none of them are up. pgmap shows 0 bytes used/available/total.
rados lspools shows only one pool: rbd.
rados -p rbd ls: freezes the output. The cursor just hangs there forever. I've tried to wait even for a couple of hours. Didn't list anything and didn't exited the command. Reboot didn't help.
Tried to zap and osd create once again. It didn't help, only in ceph status, it now shows twice more osds, then were before.
What did I do wrong. I did everything step-by-step from the book: "Learning Ceph - 2015" (Karan Singh), except, that he uses VMs, but I am using metal and I've changed all the radhat based commands/configs, from his book to debian(Ubuntu 15.10). My ceph version is the standard, that comes from the Ubuntu server repository v0.94.5
My current ceph version is based