Potential race in tempest.api.compute.servers.test_device_tagging.DeviceTaggingTestV2_42.test_device_tagging

[Sean Dague]

Looking at http://logs.openstack.org/42/471042/2/check/gate-tempest-dsvm-neutron-linuxbridge-ubuntu-xenial/4cc66aa/console.html#_2017-06-05_17_23_22_337560

It appears that the mount for sr0 doesn't show up in time. This is mostly likely a race in the test not waiting long enough (though I haven't fully debugged).

Adding this for tracking ER bug.

[Andrea Frittoli (andrea-frittoli)]

Logging says:

2017-06-05 17:23:22.321236 | 2017-06-05 16:59:54,204 31479 ERROR [tempest.api.compute.servers.test_device_tagging] Mounting /dev/sr0 on /mnt failed. Right after the failure 'lsblk' in the guest reported:
2017-06-05 17:23:22.321258 | NAME FSTYPE LABEL MOUNTPOINT
2017-06-05 17:23:22.321274 | vda
2017-06-05 17:23:22.321290 | `-vda1 ext3 cirros-rootfs /
2017-06-05 17:23:22.321306 | vdb
2017-06-05 17:23:22.321322 | vdc
2017-06-05 17:23:22.321338 | sr0 iso9660 config-2
2017-06-05 17:23:22.321346 |

Just to confirm that the disk is not already mounted and that the /mnt mount point is not taken.
syslog does not show anything anomalous.

[Andrea Frittoli (andrea-frittoli)]

The nova log shows a guru meditation shortly after the issue: http://logs.openstack.org/42/471042/2/check/gate-tempest-dsvm-neutron-linuxbridge-ubuntu-xenial/4cc66aa/logs/screen-n-cpu.txt.gz#_Jun_05_17_23_27_003886

Jun 05 17:23:27.003886 ubuntu-xenial-ovh-bhs1-9139494 nova-compute[24314]: ========================================================================
Jun 05 17:23:27.003987 ubuntu-xenial-ovh-bhs1-9139494 nova-compute[24314]: ==== Guru Meditation ====

[Andrea Frittoli (andrea-frittoli)]

Disregard that - that was just the end of the test run.
The timestamp in console log is not correct, the correct time of the issue is:
http://logs.openstack.org/42/471042/2/check/gate-tempest-dsvm-neutron-linuxbridge-ubuntu-xenial/4cc66aa/logs/tempest.txt.gz#_2017-06-05_16_59_53_223