After trying to restart the mds master, it also failed. Now the cluster state is :
# ceph status
cluster:
id: dd024fe1-4996-4fed-ba57-03090e53724d
health: HEALTH_WARN
1 filesystem is degraded
insufficient standby MDS daemons available
29 daemons have recently crashed
services:
mon: 3 daemons, quorum 2,0,1 (age 2d)
mgr: mgr.0(active, since 10d), standbys: mgr.2, mgr.1
mds: cephfs:1/1 {0=node1=up:replay(laggy or crashed)}
osd: 3 osds: 3 up (since 32h), 3 in (since 10d)
data:
pools: 3 pools, 49 pgs
objects: 29 objects, 203 KiB
usage: 3.0 GiB used, 41 TiB / 41 TiB avail
pgs: 49 active+clean
PS: The cluster in on Centos 7
Thanks,