Hi,
I've configured my ceph cluster to use cluster and public networks, but
looks like this configuration doesn't work correctly because when rados
bench returns the following test's result:
Maintaining 16 concurrent writes of 4194304 bytes for at least 10 seconds.
Object prefix: benchmark_data_srv64_4661
sec Cur ops started finished avg MB/s cur MB/s last lat avg lat
0 0 0 0 0 0 - 0
1 16 16 0 0 0 - 0
2 16 17 1 1.99975 2 1.17015 1.17015
3 16 17 1 1.33317 0 - 1.17015
4 16 17 1 0.999894 0 - 1.17015
5 16 17 1 0.799913 0 - 1.17015
------------------------------------------------------------------------------------
eth0 - public network
eth1 - cluster network
ceph.conf:
[global]
public network = 10.8.0.0/24
cluster network = 192.168.102.0/24
[osd.0]
host = srv95
cluster addr = 192.168.102.95
public addr = 10.8.0.95
devs = /dev/sdb
[osd.1]
host = srv115
cluster addr = 192.168.102.115
public addr = 10.8.0.115
devs = /dev/sda6
[osd.2]
host = srv6
cluster addr = 192.168.102.6
public addr = 10.8.0.6
devs = /dev/sda6
[osd.3]
host = srv140
cluster addr = 192.168.102.140
public addr = 10.8.0.140
devs = /dev/sda6
[osd.4]
host = srv140
cluster addr = 192.168.102.140
public addr = 10.8.0.140
devs = /dev/sdb6
[osd.5]
host = srv6
cluster addr = 192.168.102.6
public addr = 10.8.0.6
devs = /dev/sdb6
[osd.6]
host = srv115
cluster addr = 192.168.102.115
public addr = 10.8.0.115
devs = /dev/sdb6
# id weight type name up/down reweight
-1 7 root default
-3 7 rack unknownrack
-2 1 host srv95
0 1 osd.0 up 1
-4 2 host srv115
1 1 osd.1 up 1
6 1 osd.6 up 1
-5 2 host srv6
2 1 osd.2 up 1
5 1 osd.5 up 1
-6 2 host srv140
3 1 osd.3 up 1
4 1 osd.4 up 1
------------------------------------------------------------------------------------
My question is: How to set up cluster network correctly?
Thanks
--
Kind regards,
R. Alekseev
--
To unsubscribe from this list: send the line "unsubscribe ceph-devel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
