Ask Your Question
1

Error connecting to Ceph Cluster with glance api service

asked 2015-09-05 20:42:56 -0500

slosh_liuxin gravatar image

updated 2015-09-06 03:16:16 -0500

Hi,i am testing in devstack with ceph cluster.Now,when i restart glance.api.service,i create a image,then i have got error like this:

ERROR glance.api.v2.image_data [req-75d06dda-1773-414d-9d0d-c7785b336935 c8cc029cb8ec41ebba917d39323edfd2 89e8c584aaf0495f9a2d5d4b6981766a] Failed to upload image data due to internal error
ERROR glance.common.wsgi [req-75d06dda-1773-414d-9d0d-c7785b336935 c8cc029cb8ec41ebba917d39323edfd2 89e8c584aaf0495f9a2d5d4b6981766a] Caught error: error connecting to the cluster

Howerver,in my glance-api.conf,i def like this:

[glance_store]
default_store = rbd 
stores = rbd
rbd_store_pool = images
rbd_store_user = stack
rbd_store_ceph_conf = /etc/ceph/ceph.conf
rbd_store_chunk_size = 8

Where is my mistake?Any help is very useful,please tell me how to troubleshooting.....

edit retag flag offensive close merge delete

Comments

Im sorry for my mistake, default_store in default section is deprecated, you had it right before my answer

Eduardo Gonzalez gravatar imageEduardo Gonzalez ( 2015-09-06 05:12:06 -0500 )edit

Can you paste the content from /etc/ceph/. Make sure the stack user keyring have +r permissions.

Eduardo Gonzalez gravatar imageEduardo Gonzalez ( 2015-09-06 05:15:42 -0500 )edit

Thanks for your answering!!This is my content. -rw-------. 1 stack stack 137 Aug 31 16:41 ceph.client.admin.keyring -rw-r--r--. 1 stack stack 749 Sep 6 14:38 ceph.conf -rw-r--r--. 1 stack stack 64 Sep 6 13:16 client.images.keyring -rw-r--r--. 1 stack stack 65 Sep 6 13:16 client.volumes.keyring

slosh_liuxin gravatar imageslosh_liuxin ( 2015-09-06 18:26:01 -0500 )edit

And,this is my ceph.conf about keyring: [client.volumes] keyring= /etc/ceph/client.volumes.keyring [client.images] keyring= /etc/ceph/client.images.keyring

slosh_liuxin gravatar imageslosh_liuxin ( 2015-09-06 18:41:06 -0500 )edit

I am a newer of ceph and openstack,my purpose is openstack kilo use ceph cluster as the Unified Storage System,which can provide HA and flexible migrate virtual machine.so far as the status,i should do more testing...

slosh_liuxin gravatar imageslosh_liuxin ( 2015-09-06 18:54:41 -0500 )edit

1 answer

Sort by ยป oldest newest most voted
1

answered 2015-09-06 19:32:04 -0500

slosh_liuxin gravatar image

I solve my problem by myself...It is my mistake , my stupid!!!! It is the error user by def rbd_store_user ! I change rbd_store_user = images ,then it works !! I upload a new image ,then i can see it in the ceph pool images!!

edit flag offensive delete link more

Your Answer

Please start posting anonymously - your entry will be published after you log in or create a new account.

Add Answer

Get to know Ask OpenStack

Resources for moderators

Question Tools

1 follower

Stats

Asked: 2015-09-05 20:42:56 -0500

Seen: 2,425 times

Last updated: Sep 06 '15