| Age | Commit message (Collapse) | Author | Files | Lines |
|
Change-Id: I81413ffcaa4efb881d13fef62823d60b676e871d
Implements: blueprint docker-named-volumes
|
|
This change let Swift detect and use physical disk for storage. The
old named volume for storage isn't really useful for any serious setup.
Also updated swift-guide.rst accordingly.
Change-Id: I4f577b7b69d8bcd8b3961500946241c65a16db22
Partially-Implements: blueprint swift-physical-disk
|
|
|
|
Add bootstrap label to all bootstrap containers to ensure that when
the a new container is launched a difference is seen between it and
the bootstrap container since we cannot rely on ENV variables for
this. This only affects mariadb at this stage, but it is needed to
ensure rabbitmq works when we switch to named volumes.
Change-Id: Ia022af26212d2e5445c06149848831037a508407
Closes-Bug: #1538136
|
|
With the switch to named volumes we run into a few situations where
we cannot bootstrap volumes like we used to. This labels param will
fix that as the next patchset shows.
Change-Id: Ia93166dd204c5c0d1a0eb9ffeb6d0aba486e269a
Partially-Implements: blueprint docker-named-volumes
|
|
There is no reason to have a hostname-unique pidfile in the container
as we currently have. This posed problems with kolla-mesos reusing
the same script. Since there is no reason for this pidfile to be
configurable in path _at_ _all_, we hardcode the path.
Additionally, we adjust the file perm change to only update the perms
on the folder if it is not already properly set.
This also incorperates a kolla-ansible file in the bootstrap process
which follows our other container techniques of using the idempotent
creation of a volume in the bootstrap process (see nova)
TrivialFix
Related-Bug: #1538136
Change-Id: I2380529fc7146a9603145cdc31e649cb8841f7dd
|
|
Since the fetch script fetched _all_ keyrings from the ceph-mon
container, the ceph-mon container must contain all keyrings. This
setup works AIO but was broken on multinode because the ceph-mon
container did not have the radosgw keyring. This issue affects every
multinode install regardless of using the radosgw or not.
TrivialFix
Change-Id: Ie416de1a5275862da6d77ef0dd174e85e499fc0f
|
|
Change-Id: Ie91b3498e9f9d0d74a502ba6945db531ef967516
Closes-Bug: #1492736
|
|
|
|
Co-Authored-By: Sam Yaple <sam@yaple.net>
Change-Id: I066b2d419abb3ef2d4827ce686865c0dc68a5e97
Closes-bug: #1537763
|
|
We use tcp connection rather than socket so we can remove the config
options related to it.
Additionally adjust the _extremely_ verbose logging from INFO to
WARNING.
TrivialFix
Change-Id: I88bf660134192f11732d012985df5c4f688419ba
|
|
After introduction of pull action and turing every main.yml into
{{action}}.yml we lost ability to perform upgrade.
Change-Id: Id6b5921bd1e3e7b196c4b3223920e51ae5e0b840
Closes-Bug: #1538210
|
|
Change-Id: Ia6ac371845d1f2b545406d096b35a6d5f68be5f6
Implements: blueprint ansible-mongodb
|
|
|
|
|
|
|
|
|
|
|
|
Update Swift config playbook to use correct name of Swift ring
builer file. Those are the names generated if one follows document in
doc/swift-guide.rst
TrivialFix
Change-Id: If1d713e3cccc851ef6f2cc3cb5a736b43fa26957
|
|
After introduction of pull action and turing every main.yml into
{{action}}.yml we lost ability to perform upgrade
Change-Id: Ie9fa2cd083b061033abc733fba53d54f9c55e393
Fixes-Bug: #1538210
|
|
Part of ELK stack. Includes Dockerfiles for both Centos and Ubuntu.
Change-Id: I1d955a5c51e416cc572eb2c9b4c57982a1d6ab67
Partially-implements: blueprint central-logging-service
|
|
PXE files should be copied after volume mount to avoid overwrite.
Change-Id: I023cc316e6873abd658478dfa05be839e44a80c7
Closes-bug: #1538164
|
|
map-file should not be overwriten by volume
TrivialFix
Change-Id: I57bf963fd4af8c010fda0c4504e83dc33f6e9162
|
|
Correct the leftover image name.
Change-Id: I54c46bfabc2afbdd2ada697decec1f67bf70b6ed
Related-Id: I929689f93b56396a41b19fda46e4679c4de84ca1
Closes-Bug: #1537993
|
|
|
|
This change fixes the problem that `swift_hash_path_suffix` and
`swift_hash_path_prefix` not picked up because of the relative path
merge_configs can not recognize.
Change-Id: I4d93269babe71e51ba7949b719f753c1e1b34e3c
Closes-Bug: #1537331
|
|
|
|
rename auth_type to auth_plugin
Change-Id: Idc20cc6e61fe15e64ba5b49e874f1290b9ee685d
closes-Bug: #1536820
|
|
|
|
|
|
|
|
|
|
Change-Id: I688ac5cba128e7eae9df9f51c48273e39fdb09a9
Partially-implements: blueprint precheck-tasks
|
|
|
|
Change-Id: I3d452f30304a83cdeb65d8cac0ddef77c58cfeb9
Partially-Implements: blueprint precheck-tasks
|
|
Due to changes with the drop-root work, we lost the ability to write
to /var/lib/nova/*. This fixes those permissions and ensures cross
container talk works properly between nova_libvirt and nova_compute
Additionally, this fixes another issue introduced which saw that
nova-compute could not run sudo commands as it did not have a proper
sudoers entry
Testing from previous deploys means you need a fresh environment. You
have to remove all of the named volumes that kolla created in docker.
Check these with `docker volume ls`
Signed-off-by: Hui Kang <kangh@us.ibm.com>
Signed-off-by: Artur Zarzycki <azarzycki@mirantis.com>
Co-Authored-By: Sam Yaple <sam@yaple.net>
Co-Authored-By: Hui Kang <kangh@us.ibm.com>
Closes-Bug: #1533350
Change-Id: I7f864c448a2414e0b5d89f48337be411b891df35
|
|
References: https://review.openstack.org/#/c/269042/
TrivialFix
Change-Id: Ief08781342a06f956fc4cf00ba4383759da8c897
|
|
Config file use *_logging_debug as debug default value
Change-Id: I41102fff9056a82f7307694252adff0aedcf2658
|
|
Remove the docker max version check since we are not pinning
docker version any more.
Partially-Implements: blueprint precheck-tasks
Change-Id: Icfa10e72c56092931aae016afaeed4c6ce8cb337
|
|
TrivialFix
It should be more comfortable to keep consistent:
$ find . -name *.yml | xargs grep -h -A5 start_container | grep name: | sort
name: "ceph_mon"
name: "ceph_rgw"
name: "cinder_api"
name: "cinder_backup"
name: "cinder_scheduler"
name: "cinder_volume"
name: "glance_api"
name: "glance_registry"
name: "haproxy"
name: "heat_api"
name: "heat_api_cfn"
name: "heat_engine"
name: "horizon"
name: "ironic-api"
name: "ironic-conductor"
name: "ironic-discoverd"
name: "ironic-pxe"
name: "keepalived"
name: "keystone"
name: "magnum_api"
name: "magnum_conductor"
name: "mariadb"
name: "memcached"
name: mistral_api
name: mistral_engine
name: mistral_executor
name: "murano_api"
name: "murano_engine"
name: "neutron_agents"
name: "neutron_openvswitch_agent"
Change-Id: Ie946286968c830dcc3c0c45cc9ba87418dd44e8c
|
|
|
|
|
|
TrivialFix
Change-Id: Iafd92e4e962dec5a225582ff058ab3603e5a50bb
|
|
TrivialFix
https://review.openstack.org/#/c/249508/
Change-Id: Ie2732ee116b6b4b3e0210eb1f62fc72878d8c277
|
|
|
|
|
|
Option "verbose" from group "DEFAULT" is deprecated for removal.
Its value may be silently ignored in the future.
If this option is not set explicitly, there is no such warning.
Furthermore, the default value of verbose is true, so there is
no need to set this value in config files.
TrivialFix
Change-Id: I3ec2a8900c984a64bc0645672ef89a63975f7f4e
|
|
Ceph-mon bootstrap container shares the same name of `ceph_mon` with
real ceph-mon service. When bootstrapping, it will sleep forever for the
purpose of distribute rings. Later on, when starting real `ceph_mon`
service, kolla_docker module will take bootstrap container as real one
because bootstrap container has a super set of env variables and
kolla_docker doesn't do strict comparing of env variable items.
This change follows the pattern of other bootstrap containers and rename
to use `bootstrap_ceph_mon` to avoid conflict.
Co-Authored-By: Sam Yaple <sam@yaple.net>
Change-Id: I0082820900928c42f9cc778ffd705980fea9ad1c
Closes-Bug: #1535140
|
|
Related-Id: Icb1e81dc8acabe51740edc5658957022ddd13cd7
TrivialFix
Change-Id: Icf7cca554a972370bb109530cce3a0d451c95ae1
|
|
|