Zaza seems to currently wait for 64+32+16+8+4+2+1 seconds (127) for
a guest to boot. This is not quite enough sometimes, so increase
this to 511 seconds.
The current usage of ``datetime.today()`` is at the grace of the
locale settings of the test executor and the instances spun up.
The ``cryptography.x509.CertificateBuilder`` ``not_valud_before``
and ``not_valid_after`` attributes do expect [0] a UTC datetime.
0: https://cryptography.io/en/latest/x509/reference/#cryptography.x509.CertificateBuilder
This patch is the peer change for:
https://review.opendev.org/#/c/679599/
charm-glance currently has following Amulet tests:
https://github.com/openstack/charm-glance/blob/master/tests/basic_deployment.py
The 1xx, 2xx and 5xx ones are mostly covered in zaza keystone tests, or could
be skipped in zaza.
Add the only missing 900 one, to test glance restart on config change.
Test option `debug` instead of `use-syslog` to avoid conversion between
`-` and `_`, which is out the scope of this test.
Signed-off-by: Joe Guo <guoqiao@gmail.com>
After a change to the way application name is set and because
keystone-saml-mellon uses BaseKeystoneTest which sets the application
name to "keystone" which breaks KSM actions.
This change explicitly sets application name to keystone-saml-mellon.
During the series upgrade tests, zaza pauses subordinate services.
However, the cinder-ceph charm has no pause/resume action, and so the
test fails. This PR adds a blacklist that ensures that the cinder-ceph
subordinate is not paused during series upgrade.
If a unit is rebooted while executing a hook Juju will mark it as
errored when it comes back. To work around this in tests, units in
an error state are resolved on reboot. However, this could mask a
genuine error. So, to minimise the risk only resolve units that
went into an error state when running update-status as the system
was in a stable state prior to reboot so that should have been the
only hook running.
Charm going into an error state on reboot applies equally to
subordinates so use the same logic for hacluster.
block_until_all_units_idle can return when a subordinate hook is
still queued so check for update-status errors when manually
queuing update-status run via action.
This is a temporary fix to work around a bug in nova-compute.
The bug causes nova-compute to cache the placement api endpoint
which causes placement api calls to fail when the endpoint switches
from http to https
The existing code lead to intermittent failures as there are a few
Neutron ports tagged with ``device:owner`` ``compute:nova`` that
has nothing to do with our payload instances.
Get IP information by asking Nova to list instances instead.
Liam Young
Andrew McLeod
Eduardo Sousa
David Ames
Joe Guo
Frode Nordahl
Chris MacNaughton
Ryan Beisner
Ryan Beisner
Camille Rodriguez
Alex Kavanagh
lolwww