openstack cinder：volume：create：找不到有效的主机。 没有可用的主机

Question

I am trying to get a working openstack liberty in one of our dev servers that has minimal configuration and has ubuntu 14:04.

All the components looks good and I was able to create instance and view the dashboard too.

Now, I am trying to create a volume and this is where I am facing an issue.

We have a node that has a disk 'sda' with 8GB and we followed the steps listed here to prepare the disk. https://eatpeppershothot.blogspot.sg/2016/04/use-real-block-storage-backend-for.html

​output of lsblk​

NAME FSTYPE LABEL MOUNTPOINT

sda xfs

└─sda1 LVM2_member

​

Then we followed the standard documentation http://docs.openstack.org/liberty/install-guide-ubuntu/cinder-storage-install.html

On Controller node

**root@dev-controller:~/scripts# cinder service-list**
+------------------+----------------+----------+---------+-------+----------------------------+-----------------+
|      Binary      |      Host      |   Zone   |  Status | State |         Updated_at         | Disabled Reason |
+------------------+----------------+----------+---------+-------+----------------------------+-----------------+
| cinder-scheduler | dev-controller |   nova   | enabled |   up  | 2016-08-04T12:03:55.000000 |        -        |
|  cinder-volume   |  dev-osd1@lvm  |   osd1   | enabled |   up  | 2016-08-04T12:03:51.000000 |        -        |
|  cinder-volume   |  dev-osd2@lvm  | dev-osd2 | enabled |   up  | 2016-08-04T12:03:54.000000 |        -        |
|  cinder-volume   |  dev-osd3@lvm  |   nova   | enabled |   up  | 2016-08-04T12:03:55.000000 |        -        |
+------------------+----------------+----------+---------+-------+----------------------------+-----------------+

​By default all the volumes were listed under "nova" zone but we changed it to different node while trying things out.

Command to create volume cinder create --volume_type lvm --display_name testvol 1 --availability-zone dev-osd2 cinder create --volume_type lvm --display_name testb 1

The 'cinder create' command works but 'cinder list' command shows that the volume is in error state. Both of these commands end up in error state.

cinder-scheduler.log

2016-08-04 20:11:46.466 3461 INFO cinder.openstack.common.scheduler.base_filter [req-d1e3c63e-0d15-42cc-af3f-7867689cf4fe db30662c15f44a5e9b4cdd3fb5125762 0d1a9f632e4c4dd3a090c7c377fbcc77 - - -] Filter AvailabilityZoneFilter returned 0 host(s)
2016-08-04 20:11:46.469 3461 ERROR cinder.scheduler.flows.create_volume [req-d1e3c63e-0d15-42cc-af3f-7867689cf4fe db30662c15f44a5e9b4cdd3fb5125762 0d1a9f632e4c4dd3a090c7c377fbcc77 - - -] Failed to run task cinder.scheduler.flows.create_volume.ScheduleCreateVolumeTask;volume:create: No valid host was found. No weighed hosts available
2016-08-04 20:11:46.471 3461 DEBUG cinder.volume.flows.common [req-d1e3c63e-0d15-42cc-af3f-7867689cf4fe db30662c15f44a5e9b4cdd3fb5125762 0d1a9f632e4c4dd3a090c7c377fbcc77 - - -] Updating volume: 89a42d39-f7a9-41ce-84b7-76160d444597 with {'status': 'error'} due to: No valid host was found. No weighed hosts available _update_object /usr/lib/python2.7/dist-packages/cinder/volume/flows/common.py:87

I have checked and made sure that all the database connect string are correct and that the host are able to connect to the controller. RabbitMQ is fine too.

OUTPUT from pvdisplay

root@dev-osd2:~# pvdisplay
  --- Physical volume ---
  PV Name               /dev/vda5
  VG Name               dev-mon01-vg
  PV Size               15.57 GiB / not usable 2.00 MiB
  Allocatable           yes
  PE Size               4.00 MiB
  Total PE              3986
  Free PE               6
  Allocated PE          3980
  PV UUID               gux2KI-5hbS-VSOf-RLaQ-dCk8-ZXCX-oBOocf

  --- Physical volume ---
  PV Name               /dev/sda1
  VG Name               cinder-volumes
  PV Size               8.00 GiB / not usable 2.98 MiB
  Allocatable           yes
  PE Size               4.00 MiB
  Total PE              2047
  Free PE               2047
  Allocated PE          0
  PV UUID               AfjDlL-vSjm-stE4-4jPJ-ze7G-dH1f-GmWLSv

OUTPUT from vgdisplay

 --- Volume group ---
  VG Name               cinder-volumes
  System ID
  Format                lvm2
  Metadata Areas        1
  Metadata Sequence No  1
  VG Access             read/write
  VG Status             resizable
  MAX LV                0
  Cur LV                0
  Open LV               0
  Max PV                0
  Cur PV                1
  Act PV                1
  VG Size               8.00 GiB
  PE Size               4.00 MiB
  Total PE              2047
  Alloc PE / Size       0 / 0
  Free  PE / Size       2047 / 8.00 GiB
  VG UUID               P6t7E7-TQqq-zhEW-HAX8-CFl6-kkH3-q3Aova

Please help me figure this out :)

Regards

Answer 1

We had integrated cinder with ceph , we had to add the ceph secret key to all our compute nodes which we did not do before. Once we did that, it worked fine