Ask Your Question
0

Unable to delete snapshots in ceph from cinder

asked 2018-06-13 03:40:25 -0500

Robby gravatar image

Hi, I am using openstack pike with ceph backend, when I try to delete snapshots or images from dashboard, it says permission Error, How to say cinder to create snapshots unprotected by default? (or) How to ask cinder to unprotect the snapshots and delete when requested?

ERROR oslo_messaging.rpc.server PermissionError: error unprotecting snapshot volume-3d57031e-a467-4dc8-8640-a15d9cbf1f6d@snapshot-0af999d7-1b89-4b10-b3eb-0eccbbd388b9 2018-06-06 20:23:27.136 6185 INFO cinder.volume.manager [req-0c9397da-edea-427c-9fc9-4fd2dc1208e9 k202059 72f5eabc80c44aa0ba990b57b98745be - default default] Detaching volume 3d57031e-a467-4dc8-8640-a15d9cbf1f6d from instance 5a9ba1aa-670d-414b-8d8e-0975b7966cf7.

edit retag flag offensive close merge delete

Comments

I think protected snapshots is correct, but maybe your cinder and glance clients don't have sufficient permissions to perform deleting operations. Can you paste the output of 

node:~ #  ceph auth get client.glance
node:~ #  ceph auth get client.cinder

(assuming your clients are these two)

eblock gravatar imageeblock ( 2018-06-13 09:55:20 -0500 )edit

Please check below the output

Robby gravatar imageRobby ( 2018-06-13 10:43:40 -0500 )edit

I want to know what privilege is it ？

mr-han gravatar imagemr-han ( 2019-10-14 04:33:16 -0500 )edit
add a comment

2 answers

Sort by » oldest newest most voted
0

answered 2018-06-13 10:38:30 -0500

Robby gravatar image

Thanks so much, this is what i see

[client.glance] key = AQArXJ5aM4xaLBAAeE2tN+a3a2N9iyM1sosYyA== caps mon = "allow r" caps osd = "allow class-read object_prefix rbd_childern, allow rwx pool=images" [root@openstack1 ~]# ceph auth get client.cinder exported keyring for client.cinder [client.cinder] key = AQCfXJ5aaxhqFRAAI1vqQGNkzW6jMD/Cp5YngQ== caps mon = "allow r" caps osd = "allow class-read object_prefix rbd_childern, allow rwx pool=volumes, allow rwx pool=vms, allow rx pool=images"

If I edit rbd.py it works nicely, but it is just a quick fix

edit flag offensive delete link more

Comments

Not sure if it's a typo, but I read "rbd_childern" instead of "rbd_children". If this is actually your configuration, it should work after you correct it.

eblock gravatar imageeblock ( 2018-06-14 09:46:30 -0500 )edit

super!! it works now, thanks a ton

Robby gravatar imageRobby ( 2018-06-14 10:16:34 -0500 )edit

It was a good catch, great help, once again thanks a lot

Robby gravatar imageRobby ( 2018-06-14 10:17:35 -0500 )edit

Great, I'm glad I could help :-)

eblock gravatar imageeblock ( 2018-06-15 02:06:10 -0500 )edit
add a comment

Get to know Ask OpenStack

Resources for moderators

Question Tools

subscribe to rss feed

Stats

Asked: 2018-06-13 03:40:25 -0500

Seen: 1,009 times

Last updated: Jun 13 '18

Related questions

Resize doesn't always extend disk on ceph backend

Glance-api configure with rbd not able to start

Ceph Backend Poor Perfomance

Deploying with Fuel, Ceph management?

HA Storage Options for Private OpenStack Cloud

cinder-volume status becomes inactive after few seconds automatically after restart

triple-o deploy hyperconverged ceph on a single nic with vlans?

how ceph RBD data placement works

Cinder + Ceph - Deduplication?

Wrong Ceph osd disk selection