When you add new nodes to existing cluster, docker will restart
all anyway and that will break a lot, including mariadb.
Change-Id: Ie46f99a141f99480a87218ead4b76ba65f2edae9
Closes-Bug: #1699335
Haproxy fails to deploy if outward group is not
in inventory.
This change adds an optional outward setting for
haproxy
Change-Id: I351578582b1057bb48ac69859583a5db13f0bfb8
Implement an ansible role that adds Hyper-V as a compute node for
OpenStack using Kolla.
This will install and configure the Nova Compute service, the
Hyper-V Neutron agent and FreeRDP-WebConnect.
https://docs.openstack.org/ocata/config-reference/compute/hypervisor-hyper-v.html
Change-Id: I601835b0769c5ff173a980a05a752391ae8cc82f
Implements: blueprint hyperv-ansible-role
Co-Authored-By: Alessandro Pilotti <apilotti@cloudbasesolutions.com>
Certain services such as Murano and trove require access to a rabbitmq
instance from tenant networks. [0]
Exposing the internal rabbitmq to end users is a security hole, hence
there are two options, 1) use vhosts in the existing rabbitmq, or two a
separate rabbitmq instances. Given the importance of rabbitmq to the
OpenStack deployment, we have decided to go with a separate instance.
Refer to [1] for more detail on the various options.
This change makes the rabbitmq role generic so that it can be reused, in
this case to start 'outward_rabbitmq'. It needs to be exposed via
haproxy both for network isolation and also because this is what Murano
configuration requires.
Follow on patches will be added to add a vhost in this outward instance
for Murano and other services which require access.
Based on the original work by bdaca[2]
[0] http://murano.readthedocs.io/en/stable-liberty/intro/architecture.html
[1] http://lists.openstack.org/pipermail/openstack-dev/2016-December/109091.html
[2] https://review.openstack.org/#/c/374525
Change-Id: Ib2bcc7ed4bf4f883a7cd1dfad3db89201e3cfd8d
Partial-Bug: #1620374
Depends-On: I020eb6219f89a310451becde41f6f1c7f54baadd
Co-Authored-By: Bartłomiej Daca <bartek.daca@gmail.com>
always_run is deprecated and removed in Ansible 2.4
check_mode is introduced in Ansible 2.2 and Kolla-ansible bump Ansible to
2.2.0 so it's safe to replace always_run by check_mode now.
Change-Id: Id1028d38b7bde30a6afe17b319dcdc77907914ab
Closes-Bug: #1643633
Implements: blueprint migrate-to-ansible-2-2-0
On many systems IPv6 related modules are not loaded by default.
Usually when one runs ip6tables-* commands required modules are
probed. In neutron_openvswitch_agent container /lib/modules
does not exist. The commands fail to process ip6 firewall
rules as consequence.
Change-Id: Ic4e72eb4f5304f013b7a09ddd31794cfafa67e0b
Closes-Bug: #1615715
Different Y stream version in Ansible is not compatible. Since ansible
2.2.0 is released for a while. It is time for kolla-ansible to bump the
min Ansible version. Then we can remove the annoying WARNING message
in deploying logs and use the new features.
Change-Id: I99f3c5678f6d2d1f93d61c660fbd166184ff6422
Cron is always restarting due dummy environment
variable is None.
This change adds a useless env value.
Change-Id: Iad435bb0671a25bc12c6a6d0d988d555faba76c3
Closes-Bug: #1697706
keystone-paste.ini file is introduced by
I3a3ca2e74c0ae341105d3481f97956c6da473046 for a security risk of
admin_token_auth middleware. Now this middleware is removed by
I57586ccfa0ad1309cc806d95377dc1ecad015914. So it is safe to use upstream
keystone-paste.ini file.
This patch also keep custom paste file feature. Just put the file to
/etc/kolla/config/keystone/keyston-paste.ini path.
Closes-Bug: #1695023
Partially-Implements: blueprint custom-paste
Change-Id: Ieb983b6a9edb6a156928f6b56a4bd2dbed4281e2
Since whole issue was related to check whether user wants to wipe
device, loopbacks can be opt out from this warnings
Change-Id: Idd823b282e3055457ed041a98c848deb8509cc30
Closes-Bug: #1667074
Sometimes the "Enable influxdb datasource" task failed with "503 Service
Unavailable" error. Waiting for grafana port UP seems not always enough,
it requires to wait for application ready.
Change-Id: I3a27b3086698b89a80a3f585beb810831289ae7a
Closes-Bug: #1697213
With the following configuration in globals.yml:
enable_ceilometer="no"
enable_designate="no"
enable_searchlight="yes"
neutron.conf is generated like following:
[oslo_messaging_notifications]
driver = messagingv2
topics =
=> topics value is missing.
This patch fix it.
Closes-Bug: #1671940
Change-Id: I28ab60c61882caaba823bab84f30f77e270f29b4
With the following configuration in globals.yml
enable_ceilometer="no"
enable_designate="no"
enable_searchlight="yes"
nova.conf is generated like following:
[oslo_messaging_notifications]
driver = messagingv2
topics =
topics value is missing.
Change-Id: I27145c0da8b864b2614091933c33d83bdec8b9be
Closes-Bug: #1671935
Co-Authored-By: Jeffrey Zhang <jeffrey.zhang@99cloud.net>
Currently TCMalloc's default tc size is 32MB.
This causes poor performance in ceph storage.
A new ceph_tcmalloc_tc_bytes option has been added
with a default of 128MB.
128 MB is default TC size at above jewel version.
and if we don't set this config,
osd daemon will running with 32 MB.
because 32MB is default size in TCmalloc 2.4 version.
32MB and 128MB are twice the performance difference.
- reference : https://www.slideshare.net/Red_Hat_Storage/
ceph-performance-projects-leading-up-to-jewel-61050682
Closes-Bug: #1693692
Change-Id: I0d25c92917b11a29bcfd18f9c129cae328fa2d3e
Signed-off-by: jangseon ryu <jangseon.ryu@navercorp.com>
