Is it possible to reinitialize the cluster

[10 minus <t10tennn@xxxxxxxxx>]

Hi ,

i have an issue with my ceph cluster were  two nodes wereby accident  and have been recreated.


ceph osd tree
# id    weight  type name       up/down reweight
-1      14.56   root default
-6      14.56           datacenter dc1
-7      14.56                   row row1
-9      14.56                         

  rack rack2
-2      3.64                                    host ceph01
0       1.82                                            osd.0   up      1
1       1.82                                            osd.1   up      1
2       1.82                                            osd.2   up      1
-3      3.64                                    host ceph02
3       1.82                                            osd.3   up      1
8       1.82                                            osd.8   up      1
9       1.82                                            osd.9   up      1
-4      3.64                                    host ceph03
4       1.82                                            osd.4   up      1
5       1.82                                            osd.5   up      1
-5      3.64                                    host ceph04
6       1.82                                            osd.6   up      1
7       1.82                                            osd.7   up      1

For the past four hours the recovery process is running . I 'm not sure if it will comeback.

what I have noticed is that some osd  are going down during recovery and coming back.

--snip--

ceph -s
    cluster 23d53990-4458-4faf-a598-9c60036a51f3
     health HEALTH_WARN 18 pgs down; 1814 pgs peering; 1948 pgs stuck inactive; 1948 pgs stuck unclean; 1 requests are blocked > 32 sec; 1/8 in osds are down
     monmap e1: 3 mons at {mon01=172.16.101.5:6789/0,mon02=172.16.101.6:6789/0,mon03=172.16.101.7:6789/0}, election epoch 46, quorum 0,1,2 mon01,mon02,mon03
     osdmap e9077: 10 osds: 7 up, 8 in
      pgmap v127105: 2240 pgs, 7 pools, 0 bytes data, 0 objects
            400M used, 14896 GB / 14896 GB avail
                 134 creating
                1316 peering
                 394 creating+peering
                 292 active+clean
                  86 remapped+peering
                  18 down+peering

--snip--

 

Should I wait  or should I just zap and start from scratch,  I dont have any data on my ceph cluster ?

_______________________________________________
ceph-users mailing list
ceph-users@xxxxxxxxxxxxxx
http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com