This PS adds support for testing fqdn over-rides in zuul gates.
When enabled it will direct requests to a configurable domain to
the default ip of the primary node.
Change-Id: I3d9a4a0bf06532caf0f544d44027493622f4ae5b
Signed-off-by: Pete Birley <pete@port.direct>
Upgrades the kubernetes version to v1.10.5 from v1.10.4.
Change-Id: Ic2a1f73c935136135e587945180e67ac928f8178
Signed-off-by: Tin Lam <tin@irrational.io>
This reverts commit ad5d4259c45940bf4ddd150aad37c5b8d1aabba6.
We need to revert this - as until the charts are updated OSH is broken.
Change-Id: I58db4c0bf7bdccd8ba7cd1e63af00ff1f01c343a
This modifies the manifest files to include volume mounts for
the logging configuration file, which is required for the jobs
in the charts to function
This also makes the keystone-webhook job nonvting, as the htk
changes will break the osh-charts required for the keystone
webhook job. The change to add the required fixes can be found
here: https://review.openstack.org/#/c/576001/.
Once that change is merged, we can move the keystone-webhook job
back to a voting job
Change-Id: I6ae59e2736624fff5b072e89b6043b23bc8b7f5d
This PS updates the openrc functions to use the internal interface by
default for keystone actions performed within the cluster.
Change-Id: I491618d9fd473917e2034a315f292db746f0d7cc
Signed-off-by: Pete Birley <pete@port.direct>
This PS updates the rally test runner script to allow the keystone
endpoint interface to be defined.
Change-Id: I88d7446c6bbb85090929be1728a308886cb41a74
Signed-off-by: Pete Birley <pete@port.direct>
This PS removes the use of the `quote and truncate` approach to
suppress output from gotpl actions in templates and replaces it
with the recommended practice of defining `$_` instead.
Change-Id: I5fedc3471dcbecef37d2fe1302bf9760b3163467
Signed-off-by: Pete Birley <pete@port.direct>
This PS udpate the dependency mixin function to permit cases where
the mixin is disabled by passing a null value as the key to use
for resolution.
Change-Id: Idcade7eebed317852b70392431ed02a352241c9b
Signed-off-by: Pete Birley <pete@port.direct>
This changes the default image for kube-state-metrics to use the
bitnami image instead of the coreos image. This allows us to
override the image entrypoint, as the Alpine based image used
previously did not easily allow us to do so. Adding this also
makes creating a common prometheus exporter deployment template
easier, as it reduces the functional differences between exporter
charts and templates
Change-Id: I6c4aac36f563fcb15f52640bc6f9913b45b4358a
This PS moves to use the current ga version for kubernetes daemonsets,
additionally any remaining deployments that were using the
`extensions/v1beta1` have been updated to `apps/v1`.
Story: 2002205
Task: 21735
Change-Id: If9703162dc472af1e6096bf2b9062802fd5ce8ab
Signed-off-by: Pete Birley <pete@port.direct>
This moves the Fedora job in openstack-helm-infra to an
experimental check until the issues with Fedora can be resolved
Change-Id: I5080351e3e12f2759b7fc9d73f361918d19041f6
This updates the node-exporter chart to provide the mechanism
for overriding the default image's entrypoint, which brings it in
line with the other charts
Change-Id: Ia8f6a306a6f72d7dba37e5c5736e0f5a11ad1bf0
This PS moves to use the current API version for kubernetes rcs'
that were previously using `apps/v1beta1`.
Story: 2002205
Task: 21735
Change-Id: Icb4e7aa2392da6867427a58926be2da6f424bd56
Signed-off-by: Pete Birley <pete@port.direct>
This updates the resources and the apigroups in the clusterrole
for kube-state-metrics to reflect the additional collectors that
are included in the image we use
Change-Id: I4b1c1779598e6488e4e1c8def18ad767d5d5fab4
This patch set will allow the docker setting to take effect even if
docker is already installed. The proxy setting will take effect as
long as the proxy variables are in place.
Change-Id: I1df812001d37d094fc3a31bea9e13d1488277a67
Signed-off-by: Tin Lam <tin@irrational.io>
This adds a basic check for capacity utilization for persistent
volume claims. To accomplish this, it adds a basic alerting rule
to prometheus that triggers after a persistent volume's usage
exceeds 80%, and triggers 5 minutes after that state has been
reached. In addition, there is a service check added to the
nagios chart that will query Prometheus to check if the alarm
for that threshhold is firing for any of the volume claims.
Change-Id: I862c860ac479a715733202f679bb151885d7aa7c
This PS simply moves functions within the chart to their correct location.
Change-Id: Ia3d693713903d226a864dcdcf9884dee67f07d2b
Signed-off-by: Pete Birley <pete@port.direct>
This PS simply moves functions within the chart to their correct location.
Change-Id: Ia5ac02a25a76ff759160cc352404b71b4208b216
Signed-off-by: Pete Birley <pete@port.direct>
This PS updates the gate to not deploy the hwe kernel by default
on ubuntu nodes.
Change-Id: I2ea6ba899ad022d0203874693fd5f16dc76535e2
Signed-off-by: Pete Birley <pete@port.direct>
This ps adds a function for cleaning up the test data used to
verify Elasticsearch is functioning properly. It removes the test
index created and populated with test data to ensure the resulting
elasticsearch cluster is clean and does not contain random data
Change-Id: Ibdeb90e3f3b6307bf16c68469556bef256ed2d78
This removes the megacli collector from the node exporter, as this
collector is no longer supported and supplied as part of the node
exporter image used.
Change-Id: Ic9d7bc906435227337aed7a40f4c25bbb16fd3ba
This PS moves to use the internal keystone endpoint for boostrapping
jobs, to allow osh to be deployed when the support infrastructure may
not have been fully configured (eg - dns for public endpoints)
Change-Id: I5a9eb07b88bdc0ffbcda371bdd4c03d4da4ed78d
Signed-off-by: Pete Birley <pete@port.direct>
For some reason it appears that the task to wait for nodes to come up
requires privilege escalation to work. I suspect this is due to
interation between ansible 2.5.4 and zuuls log streamer, but am not
sure. In the meantime this PS unblocks the gates.
Change-Id: I705f2ddf3facfe56838f606f88cfb15b822d18a5
Signed-off-by: Pete Birley <pete@port.direct>
This updates the fluent-logging chart to support the creation of
an arbitrary number of templates for elasticsearch. This allows
for the definition of multiple index mappings driven via the
chart's values. This provides flexibility in determining specific
structures for indexes that may differ between log types.
This also moves to define these mappings via json instead of XML.
As gotpl can convert yaml directly to json, and elasticsearch can
ingest json directly for index creation, we no longer need an XML
helper function to generate the required configuration. This helps
reduce the number of helper functions we need to maintain
Change-Id: I3c85fb9a1e700eb1592d96f83e632172d0eb2681
This PS updates the version of kubernetes deployed in the gates/dev
env to 1.10.3.
Change-Id: I4916a669ab0cb58760c0497b2264d4a7d0a9bffe
Signed-off-by: Pete Birley <pete@port.direct>
