Docker will change the name volume folder permission to image already
set. /var/log/kolla is not created and well configured in image.
This patch set creates and configures proper permission /var/log/kolla
in base image.
This patch set also fixed the ceph image by adding ceph user to kolla
group.
TrivialFix
Change-Id: Ib5ef0187e90de2699b3cda31e819b4babb07e0af
Since initial setup is arguably the hardest part of kolla deployment,
let's show people our new plabybook to lower the barrier of entry.
TrivialFix
Change-Id: I6edf355772b5705207cdecfbe8c45672a63badf0
By having serial in site.yml for these roles will mean (particularly
during an upgrade) that tasks can be executed out of the intended
order across hosts, leading to unexpected results.
Change-Id: I44f0c66c445d853a2e4af69119562f13b4424936
Closes-Bug: #1626096
Currently Cinder enables iscsid service only when LVM is choosen as
backend. This fix will allow to specify external iSCSI storage via
configuration add-on and to run iscsid independently of LVM settings.
Change-Id: Iaae0f8025534e5ab1c5f9d5e812a7d41129915ef
Closes-Bug: 1614125
Currently Kolla operators are restricted to configuring one physical
network (physnet1).
This change along with ml2_conf.ini augmentation can be used to setup
multiple physical networks in openvswitch.
E.g. To configure two physical networks, physnet1 and physnet2, with
ports eth1 and eth2 associated respectively:
In /etc/kolla/globals.yml, set
neutron_bridge_name: "br-ex,br-ex2"
neutron_external_interface: "eth1,eth2"
In /etc/kolla/config/neutron/ml2_conf.ini
[ovs]
bridge_mappings = physnet1:br-ex,physnet2:br-ex2
Co-Authored-By: Mick Thompson <michael.a.thompson@oracle.com>
Closes-Bug: #1625700
DocImpact
Change-Id: I9454ca98d9b058368129123109ccc56f95519874
This solves the following issue:
oslo_config.cfg.ConfigFilesNotFoundError: Failed to find some config
files: /etc/neutron/fwaas_driver.ini
Change-Id: I606e591057574ca113f2cba18618f455f7c36bc6
Closes-bug: #1625672
Fix typo in variable name ("yup" instead of "yum"), which means we now
import the repo keys we intended to. Furthermore we no longer try to
import EPEL key in that very spot, as it should be imported after
epel-release is installed, which already happens later during the base
image build.
CentOS/EPEL keys in the image before this patch:
()[root@6843e7e136cf ~]# rpm -q gpg-pubkey --qf '%{summary}\n' | grep -i -E '(centos)|(epel)'
gpg(CentOS-7 Key (CentOS 7 Official Signing Key) <security@centos.org>)
gpg(Fedora EPEL (7) <epel@fedoraproject.org>)
With this patch:
()[root@64a6ae7d0347 ~]# rpm -q gpg-pubkey --qf '%{summary}\n' | grep -i -E '(centos)|(epel)'
gpg(CentOS-7 Key (CentOS 7 Official Signing Key) <security@centos.org>)
gpg(CentOS Storage SIG (http://wiki.centos.org/SpecialInterestGroup/Storage) <security@centos.org>)
gpg(CentOS Virtualization SIG (http://wiki.centos.org/SpecialInterestGroup/Virtualization) <security@centos.org>)
gpg(Fedora EPEL (7) <epel@fedoraproject.org>)
Change-Id: I5cd22a887282e5de199a809b0798a439ce1a99e7
Closes-Bug: #1625585
Currently __pycache__ directory, which also holds
python bytecode is not deleted when running tox.
TrivialFix
Change-Id: I0fdcf9a3f9b01cee813eefdbfe29d208dd67f826
