|Age||Commit message (Collapse)||Author||Files||Lines|
Until bug #1635409 is fixed we'll have to keep the default list
of services deployed by hyperconverged-ceph.yaml in sync with the
ServicesDefault list provided in roles_data.yaml
This change adds some logic in the templates validation script to
ensure that is preserved with future updates.
NOTE: the cherry-pick is not clean because two of the services
which are available in the master branch are not available in the
ocata branch, specifically:
(cherry picked from commit 6ff979aa75bc202fe571d0b2885daa4d7c56d655)
When fixing LP#1643487 we added ?bind_address to all DB URIs.
Since this clashes with Cellsv2 due to the URIs becoming host
dependent, we need a new approach to pass bind_address to pymysql
that leaves the DB URIs host-independent.
In change Iff8bd2d9ee85f7bb1445aa2e1b3cfbff1f397b18 we first create a
/etc/my.cnf.d/tripleo.cnf file with a [tripleo] section with the correct
In this change we make sure that the DB URIs will point to the added
file and to the specific section containing the necessary bind-address
option. We do introduce a new MySQLClient profile which will hold all
this more client-specific configuration so that this change can fit
better in the composable roles work. Also, in the future it might
contain the necessary configuration for SSL for example.
Note that in case the /etc/my.cnf.d/tripleo.cnf file does not exist
(because it is created via the mysqlclient profile), things keep on
working as usual and the bind-address option simply won't be set, which
has no impact on hosts where there are no VIPs.
Co-Authored-By: Damien Ciabrini <email@example.com>
(cherry picked from commit 90431683b5927abb066d7964d513828b5488001c)
This reverts the changes in https://review.openstack.org/414629 for nova as
they are incompatible with cell_v2.
This is a temporary fix for HA while a long-term solution is developed.
Currently we are applying this validation for the services templates, this
submission moves it to run with all templates.
Also fixed those templates not using the alias name.
This replicates the behavior of the custom Jinja2 loader from tripleo-common to
allow template validation on the local filesystem using tox.
Partially-Implements: blueprint overcloud-upgrades-per-service
this attempts to make the error message more useful. This error message
happens if the environment files containing endpoint map overrides
haven't been updated to match the base endpoint map (or the defaults).
Allow for passing the output_dir in the process-templates scripts so
that it doesn't overwrites the templates in the src dir. This is a
desired feature when running the script from a t-h-t installed
Don't walk through hidden files. This avoids going through the .git,
.tox and other hidden directories that we don't care about.
Quick verification to check that the release name
is used instead the date.
Im also adding here all the updated templates required
to pass the check and merge this check as soon as possible.
This validation checks that the TLS-related environment files contain
all of the services defined in the base endpoint map. This will
hopefully help to keep them updated.
Add ReNo support to manage release notes.
The error path for this fails because we don't import six
or install it in the templates tox venv
When a service connects to the database VIP from the node hosting this
VIP, the resulting TCP socket has a src address which is by default
bound to the VIP as well. If the VIP is failed over to another node
while the socket's Send-Q is not empty, TCP keepalive won't engage and
the service will become unavailable for a very long time (by default
more than 10m).
To prevent failover issues, DB connections should have the src address
of their TCP socket bound to the IP of the network interface used for
MySQL traffic. This is achieved by passing a new option to the
database connection URIs. This option is available starting from
We use a new intermediate variable in hiera to hold the IP to be used
as a source address for all DB connections. All services adapt their
database URI accordingly.
Moreover, a new YAML validation check is added to guarantee that new
services will construct their database URI appropriately.
This patch adds a local version of our template processing
routine so that developers can more quickly view the templates
that are actually getting generated. I've noticed multiple developers
now do a full deployment with 'overcloud deploy' only to download
the swift container with the generated templates. This simple task
avoids that step by allowing developers to generate it locally.
It also aims to preserve the ability to use t-h-t templates directly
with Heat (instead of going through Mistral) should users wish to do that.
The new undercloud heat installer requires the ability to generate
templates without requiring Mistral and Swift to do so.
Ideally the Mistral API workflow would use this same code
so perhaps in the future we might modify that routine to:
-download swift tarball containing the templates
-run this local routine that lives in t-h-t
-re-upload the tarball of templates to the swift container
Wire in os-net-config via a normal script heat deployment, which has the
1. Improved error path, currently o-a-c deployments don't report any
errors, thus hang and eventually the deployment times out
2. It's far more hackable from a deployer perspective, e.g it's
much easier to change the os-net-config options or include a
3. Reduces our dependencies on o-a-c (it's only os-net-config and hiera
which requires it), although the script does currently still use oac to
get the metadata IP.
4. May enable passing os-net-config yaml via a json parameter in future,
reducing the need for resource_registry mappings (although we'll have to
support that for backwards compatibility)
The script used is based directly on 20-os-net-config (from t-i-e
at cf94c5e, we can probably improve this now that we have an error path,
but for this initial commit it's a straight copy other than the changes to
replace o-a-c for rendering the json config file.
Co-Authored-By: Steven Hardy <firstname.lastname@example.org>
The first step of generating the Service chain resources via j2,
we'll then incrementally convert other resources to be created
in a similar way.
Partially-Implements: blueprint custom-roles
Skips the unused parameter warning for required parameters.
This patch adds a new DefaultPasswords parameter to
composable services. This is needed to help provide
access to top level password resources that overcloud.yaml
currently manages (passwords for Rabbit, Mysql, etc.).
Moving the RandomString resources into composable services
would cause them to regenerate within the stack. With this
approach we can leave them where they are while we deprecate
the top level mechanism and move the code that uses the
passwords into the composable services.
This will be needed to pick the network where the service has
to bind to from within the service template.
Currently we use hyphens, e.g cinder-api, but in overcloud.yaml
we have a lot of references to services (e.g for AllNodesConfig)
by underscore, e.g cinder_api. To enable dynamic generation of
this data, we need the service name in underscore format.
This patch adds a new service_name section to each composable
service. We now have an explicit unit test check to ensure that
service_name exists in tools/yaml-validate.py.
This patch also wires service_names into hieradata on each
of the roles so that tools can access the deployed services locally
during deployment and upgrades.
Output a warning for parameters which look unused, this should help
developers clean up the template a bit, and eventually could maybe
be developed further into something we can use for gating.
For developer usage it's helpful to have the choice to provide either
an individual files, list of files, or some mix of files and directories
as you don't necessarily want to walk everything all the time.
Adds a "validate" tox env for basic sanity checking of templates.
Currently it just validates that all of the .yaml files are in fact
valid YAML. In the future we might want to add more, but this
seemed like a reasonable start.