-
Bug
-
Resolution: Not a Bug
-
Undefined
-
None
-
rhos-18.0.7
-
None
-
False
-
-
False
-
?
-
None
-
-
-
Important
Related to OSPCIX-768
In uni07eta jobs, octavia tempest tests fail to create an amphora LB because that requires to create two VMs on different computes simultaneously (because octavia is configured with ACTIVE_STANDBY). After some investigation, we found an error like this in glance logs:
cinderclient.exceptions.BadRequest: Invalid volume: Volume addef80b-74b5-46ed-856a-8420a337f856 status must be available or downloading to reserve, but the current status is in-use. (HTTP 400) (Request-ID: req-293ad09e-11a4-4b48-9711-f82632586dd2)
We have reproduced the issue without using octavia, but simply creating two VMs simultaneously on different computes:
sh-5.1$ openstack image list --name custom-fedora2 --long
+--------------------------------------+----------------+-------------+------------------+-----------+----------------------------------+--------+------------+-----------+----------------------------------+------+
| ID | Name | Disk Format | Container Format | Size | Checksum | Status | Visibility | Protected | Project | Tags |
+--------------------------------------+----------------+-------------+------------------+-----------+----------------------------------+--------+------------+-----------+----------------------------------+------+
| 00dd401f-5133-40fc-83e1-582ea701f8f1 | custom-fedora2 | qcow2 | bare | 700055552 | c5417b9bdbcb2076aae6ea7771aab878 | active | shared | False | 140fb987283e4d25afc5066a5deb08e7 | |
+--------------------------------------+----------------+-------------+------------------+-----------+----------------------------------+--------+------------+-----------+----------------------------------+------+
sh-5.1$ cat script.sh
openstack server create --network n --flavor custom_neutron_guest_rhel_8.4 --image custom-fedora2 --hypervisor-hostname compute-v2br35gl-$1.ctlplane.example.com --os-compute-api-version 2.74 vmX$1&
sh-5.1$ for i in {0..1}; do ./script.sh $i ; done
One of the VM creation fails (status is ERROR after some seconds).
I am attaching some files with the status of the VMs and the relevant glance logs.
