Mirantis OpenStack

OSTF test "Create volume and boot instance from it" failed using Ceph volumes/images

Bug #1385828 reported by Dennis Dmitriev on 2014-10-26

This bug affects 2 people

	Status	Importance	Assigned to	Milestone
Mirantis OpenStack	Invalid	High	Andrey Kurilin	Mirantis OpenStack 6.0-updates
6.0.x	Invalid	High	Andrey Kurilin	Mirantis OpenStack 6.0-updates
6.1.x	Invalid	High	Andrey Kurilin	Mirantis OpenStack 6.1

Bug Description

        Scenario:
            1. Create cluster (CentOS simple, Ceph for volumes, images and RadosGW for objects; nova-network flat-dhcp)
            2. Add 1 node with controller role
            3. Add 1 node with compute role
            4. Add 3 nodes with ceph-osd role
            5. Deploy the cluster
            6. Check ceph status
            7. Run OSTF tests
            8. Check the radosqw daemon is started

OSTF test "Create volume and boot instance from it" failed with error in nova-compute log (http://paste.openstack.org/show/124829/):

-------------------
2014-10-24 14:50:53.834 17714 TRACE oslo.messaging.rpc.dispatcher DeviceIsBusy: The supplied device (vdb) is busy.
-------------------

Failed CI job:
http://jenkins-product.srt.mirantis.net:8080/view/6.0_swarm/job/6.0_fuelmain.system_test.centos.thread_4/6/

Tags:

Revision history for this message

Dennis Dmitriev (ddmitriev) wrote on 2014-10-26:

fail_error_ceph_rados_gw-2014_10_24__14_59_37.tar.gz Edit (3.5 MiB, application/x-tar)

Dennis Dmitriev (ddmitriev) on 2014-10-26

affects:	mos → fuel
Changed in fuel:
milestone:	6.0 → none
milestone:	none → 6.0

Revision history for this message

Dmitry Mescheryakov (dmitrymex) wrote on 2014-10-27:

The exception occurs inside Nova, so assigning to Nova team for now.

no longer affects:	fuel
Changed in mos:
milestone:	none → 6.0
assignee:	nobody → MOS Nova (mos-nova)
status:	New → Confirmed
importance:	Undecided → High

Dmitry Mescheryakov (dmitrymex) on 2014-11-27

tags:

added: nova

Revision history for this message

Roman Podoliaka (rpodolyaka) wrote on 2014-11-27:

Are you guys sure Ceph works correct and is accessible from the compute nodes?

Andrey Kurilin (andreykurilin) on 2014-11-27

Changed in mos:
assignee:	MOS Nova (mos-nova) → Andrey Kurilin (akurilin)

Revision history for this message

Roman Podoliaka (rpodolyaka) wrote on 2014-12-01:

works like a charm for me on the latest ISO

Changed in mos:
status:	Confirmed → Incomplete

Revision history for this message

Roman Podoliaka (rpodolyaka) wrote on 2014-12-01:

Shouldn't these two steps in the test actually be swapped?

7. Run OSTF tests
8. Check the radosqw daemon is started

Revision history for this message

Alexander Gubanov (ogubanov) wrote on 2014-12-09:

I checked it on mos 6.0 (build_number: '43')
OSTF test "Create volume and boot instance from it" passed

and radosqw daemon is running

[root@node-1 ~]# /etc/init.d/ceph-radosgw status
/usr/bin/radosgw is running.
[root@node-1 ~]# netstat -antup | grep rad
tcp 0 0 192.168.0.1:33667 192.168.0.4:6805 ESTABLISHED 30040/radosgw
tcp 0 0 192.168.0.1:56273 192.168.0.3:6803 ESTABLISHED 30040/radosgw
tcp 0 0 192.168.0.1:41745 192.168.0.4:6800 ESTABLISHED 30040/radosgw
tcp 0 0 192.168.0.1:55715 192.168.0.5:6803 ESTABLISHED 30040/radosgw
tcp 0 0 192.168.0.1:50138 192.168.0.1:6789 ESTABLISHED 30040/radosgw
tcp 0 0 192.168.0.1:34944 192.168.0.3:6800 ESTABLISHED 30040/radosgw
tcp 0 0 192.168.0.1:39235 192.168.0.5:6800 ESTABLISHED 30040/radosgw