By default, Ansible injects a variable for every fact, prefixed with
ansible_. This can result in a large number of variables for each host,
which at scale can incur a performance penalty. Ansible provides a
configuration option [0] that can be set to False to prevent this
injection of facts. In this case, facts should be referenced via
ansible_facts.<fact>.
This change updates all references to Ansible facts within Kolla Ansible
from using individual fact variables to using the items in the
ansible_facts dictionary. This allows users to disable fact variable
injection in their Ansible configuration, which may provide some
performance improvement.
This change disables fact variable injection in the ansible
configuration used in CI, to catch any attempts to use the injected
variables.
[0] https://docs.ansible.com/ansible/latest/reference_appendices/config.html#inject-facts-as-vars
Change-Id: I7e9d5c9b8b9164d4aee3abb4e37c8f28d98ff5d1
Partially-Implements: blueprint performance-improvements
The mariadb image was removed in Wallaby, leading to database backup
failures.
Change-Id: I90986e7521779997df2782767bb95efcbd8ef232
Closes-Bug: #1928129
- Replace hardcoded haproxy monitor user with variable.
- Rename mariadb_backup variable to mariadb_backup_possible.
- Drop creation of monitor user in handlers as this is
now handled in register.yml for good reason.
Change-Id: I255a79d36ae18ca42d0befd00b235ca509197db3
Kolla-ansible is currently installing mariadb
cluster on hosts defined in group['mariadb']
and render haproxy configuration for this hosts.
This is not enough if user want to have several
service databases in several mariadb clusters (shards).
Spread service databases to multiple clusters (shards)
is usefull especially for databases with high load
(neutron,nova).
How it works ?
It works exactly same as now, but group reference 'mariadb'
is now used as group where all mariadb clusters (shards)
are located, and mariadb clusters are installed to
dynamic groups created by group_by and host variable
'mariadb_shard_id'.
It also adding special user 'shard_X' which will be used
for creating users and databases, but only if haproxy
is not used as load-balance solution.
This patch will not affect user which has all databases
on same db cluster on hosts in group 'mariadb', host
variable 'mariadb_shard_id' is set to 0 if not defined.
Mariadb's task in loadbalancer.yml (haproxy) is configuring
mariadb default shard hosts as haproxy backends. If mariadb
role is used to install several clusters (shards), only
default one is loadbalanced via haproxy.
Mariadb's backup is working only for default shard (cluster)
when using haproxy as mariadb loadbalancer, if proxysql
is used, all shards are backuped.
After this patch will be merged, there will be way for proxysql
patches which will implement L7 SQL balancing based on
users and schemas.
Example of inventory:
[mariadb]
server1
server2
server3 mariadb_shard_id=1
server4 mariadb_shard_id=1
server5 mariadb_shard_id=2
server6 mariadb_shard_id=3
Extra:
wait_for_loadbalancer is removed instead of modified as its role
is served by check already. The relevant refactor is applied as
well.
Change-Id: I933067f22ecabc03247ea42baf04f19100dffd08
Co-Authored-By: Radosław Piliszek <radoslaw.piliszek@gmail.com>
This trivial patch is just turning off ansible
changed report for group_by tasks as it could
be confusing for user.
Change-Id: I7512af573782359a6f01290a55291ac7eb0de867
Need to consider Negative seqno to compare in some cases,
but the task does not support to do that, we need to make it work.
1.we use mariabackup to restore datas on control1, delete the
mariadb data on control2 and control3, and then use cluster recovery,
as a result that the seqno of the other two nodes will be '-1'.
2. add one more control node into our existing mariadb cluster,
and then use cluster recovery, the seqno of the new node will be '-1'.
Change-Id: Ic1ac8656f28c3835e091637014f075ac5479d390
The handler was firing even when we were only generating config.
This is an issue because the services may not have been deployed.
TrivialFix
Change-Id: I2f832d73138b4c9f29e3c71e2463293eab71483a
This reverts commit 9cae59be51e8d2d798830042a5fd448a4aa5e7dc.
Reason for revert: This patch was found to introduce issues with fluentd customisation. The underlying issue is not currently fully understood, but could be a sign of other obscure issues.
Change-Id: Ia4859c23d85699621a3b734d6cedb70225576dfc
Closes-Bug: #1906288
Mariadb recovery fails if a cluster has previously been deployed, but any of
the mariadb containers do not exist.
Steps to reproduce
==================
* Deploy a mariadb galera cluster
* Remove the mariadb container from at least one host (docker rm -f mariadb)
* Run kolla-ansible mariadb_recovery
Expected results
================
The cluster is recovered, and a new container deployed where necessary.
Actual results
==============
The task 'Stop MariaDB containers' fails on any host where the container does
not exist.
Solution
========
This change fixes the issue by using the 'ignore_missing' flag for kolla_docker
with the stop_container action. This means the task does not fail when the
container does not exist. It is also necessary to swap some 'docker cp'
commands for 'cp' on the host, using the path to the volume.
Closes-Bug: #1907658
Change-Id: Ibd4a6adeb8443e12c45cbab65f501392ffb16fc7
CentOS 8 should work fine without the workaround.
This change adds the missing CentOS 8 IPv6 CI job as well.
Change-Id: I58af7a09b5ae09a10b9efc33c1f30c2efc6613f7
Main plays are action-redirect-stubs, ideal for import_tasks.
This avoids 'include' penalty and makes logs/ara look nicer.
Fixes haproxy and rabbitmq not to check the host group as well.
Change-Id: I46136fc40b815e341befff80b54a91ef431eabc0
Partially-Implements: blueprint performance-improvements
Config plays do not need to check containers. This avoids skipping
tasks during the genconfig action.
Ironic and Glance rolling upgrades are handled specially.
Swift and Bifrost do not use the handlers at all.
Partially-Implements: blueprint performance-improvements
Change-Id: I140bf71d62e8f0932c96270d1f08940a5ba4542a
Trivial: log-error & log-bin are both invalid mariadb config options.
The appropriate options are log_error & log_bin.
Note - this change mostly unnecessary as log_error is provided via cli
and log_bin value is the default.
Change-Id: If7051f7139a68864e599cccffaf17c21855fc4a8
Since change [1] merged we have two mariadb images (mariadb and mariadb-server)
Let's use mariadb-server in kolla-ansible, so we can deprecate mariadb image.
[1]: https://review.opendev.org/#/c/710217/
Change-Id: I4ae2ccaaba8fb516f469f4ce8628e8c61de03f0d
Including tasks has a performance penalty when compared with importing
tasks. If the include has a condition associated with it, then the
overhead of the include may be lower than the overhead of skipping all
imported tasks. For unconditionally included tasks, switching to
import_tasks provides a clear benefit.
Benchmarking of include vs. import is available at [1].
This change switches from include_tasks to import_tasks where there is
no condition applied to the include.
[1] https://github.com/stackhpc/ansible-scaling/blob/master/doc/include-and-import.md#task-include-and-import
Partially-Implements: blueprint performance-improvements
Change-Id: Ia45af4a198e422773d9f009c7f7b2e32ce9e3b97
Previously we mounted /etc/timezone if the kolla_base_distro is debian
or ubuntu. This would fail prechecks if debian or ubuntu images were
deployed on CentOS. While this is not a supported combination, for
correctness we should fix the condition to reference the host OS rather
than the container OS, since that is where the /etc/timezone file is
located.
Change-Id: Ifc252ae793e6974356fcdca810b373f362d24ba5
Closes-Bug: #1882553
Including tasks has a performance penalty when compared with importing
tasks. If the include has a condition associated with it, then the
overhead of the include may be lower than the overhead of skipping all
imported tasks. In the case of the check-containers.yml include, the
included file only has a single task, so the overhead of skipping this
task will not be greater than the overhead of the task import. It
therefore makes sense to switch to use import_tasks there.
Partially-Implements: blueprint performance-improvements
Change-Id: I65d911670649960708b9f6a4c110d1a7df1ad8f7
The common role was previously added as a dependency to all other roles.
It would set a fact after running on a host to avoid running twice. This
had the nice effect that deploying any service would automatically pull
in the common services for that host. When using tags, any services with
matching tags would also run the common role. This could be both
surprising and sometimes useful.
When using Ansible at large scale, there is a penalty associated with
executing a task against a large number of hosts, even if it is skipped.
The common role introduces some overhead, just in determining that it
has already run.
This change extracts the common role into a separate play, and removes
the dependency on it from all other roles. New groups have been added
for cron, fluentd, and kolla-toolbox, similar to other services. This
changes the behaviour in the following ways:
* The common role is now run for all hosts at the beginning, rather than
prior to their first enabled service
* Hosts must be in the necessary group for each of the common services
in order to have that service deployed. This is mostly to avoid
deploying on localhost or the deployment host
* If tags are specified for another service e.g. nova, the common role
will *not* automatically run for matching hosts. The common tag must
be specified explicitly
The last of these is probably the largest behaviour change. While it
would be possible to determine which hosts should automatically run the
common role, it would be quite complex, and would introduce some
overhead that would probably negate the benefit of splitting out the
common role.
Partially-Implements: blueprint performance-improvements
Change-Id: I6a4676bf6efeebc61383ec7a406db07c7a868b2a
Some services look for /etc/timezone on Debian/Ubuntu, so we should
introduce it to the containers.
In addition, added prechecks for /etc/localtime and /etc/timezone.
Closes-Bug: #1821592
Change-Id: I9fef14643d1bcc7eee9547eb87fa1fb436d8a6b3
mariadb container name variable is fixed in some places,
but in the defaults directory, mariadb container_name variable
is variable. If the mariadb container_name variable is changed
during deployment, it will not be assigned to container_name,
but a fixed 'mariadb' name.
Change-Id: Ie8efa509953d5efa5c3073c9b550be051a7f4f9b
Both include_role and import_role expect role's name to be given
via "name" param instead of "role".
This worked but caused errors with ansible-lint.
See: https://review.opendev.org/694779
Change-Id: I388d4ae27111e430d38df1abcb6c6127d90a06e0
We assume that all groups are present in the inventory, and quite obtuse
errors can result if any are not.
This change adds a precheck that checks for the presence of all expected
groups in the inventory for each service. It also introduces a common
service-precheck role that we can use for other common prechecks.
Change-Id: Ia0af1e7df4fff7f07cd6530e5b017db8fba530b3
Partially-Implements: blueprint improve-prechecks
This fixes issues reported by Mark:
- possible failure with 4-node cluster (however unlikely)
- failure to stop all nodes from progressing when conditions are
not valid (due to: "any_errors_fatal: False")
Change-Id: Ib6995bf4c99202c9813859b3d9e2f420448f0445
These affected both deploy (and reconfigure) and upgrade
resulting in WSREP issues, failed deploys or need to
recover the cluster.
This patch makes sure k-a does not abruptly terminate
nodes to break cluster.
This is achieved by cleaner separation between stages
(bootstrap, restart current, deploy new) and 3 phases
for restarts (to keep the quorum).
Upgrade actions, which operate on a healthy cluster,
went to its section.
Service restart was refactored.
We no longer rely on the master/slave distinction as
all nodes are masters in Galera.
Closes-bug: #1857908
Closes-bug: #1859145
Change-Id: I83600c69141714fc412df0976f49019a857655f5
For the CentOS 7 to 8 transition, we will have a period where both
CentOS 7 and 8 images are available. We differentiate these images via a
tag - the CentOS 8 images will have a tag of train-centos8 (or
master-centos8 temporarily).
To achieve this, and maintain backwards compatibility for the
openstack_release variable, we introduce a new 'openstack_tag' variable.
This variable is based on openstack_release, but has a suffix of
'openstack_tag_suffix', which is empty except on CentOS 8 where it has a
value of '-centos8'.
Change-Id: I12ce4661afb3c255136cdc1aabe7cbd25560d625
Partially-Implements: blueprint centos-rhel-8
As part of the effort to implement Ansible code linting in CI
(using ansible-lint) - we need to implement recommendations from
ansible-lint output [1].
One of them is to stop using local_action in favor of delegate_to -
to increase readability and and match the style of typical ansible
tasks.
[1]: https://review.opendev.org/694779/
Partially implements: blueprint ansible-lint
Change-Id: I46c259ddad5a6aaf9c7301e6c44cd8a1d5c457d3
After performing a recovery of MariaDB, the mariadb containers are left
without a restart policy. This leaves them unable to recover from the
crash of a single galera node. There is another issue, in that the
'master' node is left in a bootstrap configuration, with the
--wsrep-new-cluster argument configured as BOOTSTRAP_ARGS.
This change fixes these issues by removing the restart policy of 'no'
from the 'slave' containers, and recreating the master container without
the restart policy or bootstrap arguments.
Change-Id: I36c875611931163ca2c29ae93b71d3af64cb197c
Closes-Bug: #1851594
