This workload accomplishes the following:
1. Pull OPNFV Opera code
2. Leverage Opera code to install OPEN-O, deploy Clearwater vIMS by
OPEN-O
3. Pull OPNFV Functest
4. Run the Functest vIMS test cases to ensure vIMS is working
5. Generate calling information
The previous patch[1] was merged by accident and it did not follow
community rule. Thus it was reverted by patch[2]. Resubmit the NFV
workload.
[1] https://review.openstack.org/#/c/439492/
[2] https://review.openstack.org/#/c/459557
Change-Id: I615acc5cc19a3ae71f3d8a5023b6fda6e29e7602
Signed-off-by: yaohelan <yaohelan@huawei.com>
There's a logic bug in the destroy action: ip_flag is assuming string
context when it should be assuming list context. This causes ip_flag to
never get set to true and therefore floating IP's never get destroyed.
This can lead to floating IP exhaustion, particularly in environments
with a low number of allocatable floating IP's.
This patch corrects the problem by fixing the logic to check for a list
length of zero rather than an empty string.
Change-Id: I0205ac406e604365560360fc2953a5e90e1c9be3
Depends-On: Ifa9eeb9d761d9801cab580445e6c43c8cf1dfdaa
This patch does the following:
1. Added the apps role so that k8s app deployment can be placed in the role
2. Added cockroachdb-init-pod deployment inthe post role since this has to
be deployed first.
3. Added cockroachdb-pod deployment in the apps role
4. Allow this workload to stand up a standalone cockroachdb cluster or
join in an existing cockroachdb cluster
5. Added the cockroachdb load generator container so that once the cluster
is started, there will be some load generated.
6. Added a way to pre-allocating floating IP addresses so that the workload
can use pre-allocated floating IPs for VMs.
Change-Id: Ifa9eeb9d761d9801cab580445e6c43c8cf1dfdaa
Revert until a proper agreement on merging has been done.
The previous NFV workload patch was merged without having a quorum on at least
2 cores and was self approved by a self-company core reviewer.
This reverts commit 436d97c6725ff02c223ab52bbaa50da3d6f179d2.
Change-Id: If915d7e958c3413851cd0d7dab344744dc5c3e6e
This workload accomplishes the following:
1. Pull OPNFV Opera code
2. Leverage Opera code to install OPEN-O, deploy Clearwater vIMS by
OPEN-O
3. Pull OPNFV Functest
4. Run the Functest vIMS test cases to ensure vIMS is working
5. Generate calling information
Change-Id: I994503fcb52711fb6cd3222bd151be854665b21c
Signed-off-by: yaohelan <yaohelan@huawei.com>
In the LAMP stack workload, we found that one issue for interoperability
between clouds was that in some combinations of
hypervisor/adapter/storage is is necessary to rescan the scsi bus of a
guest VM in order to detect a newly attached Cinder volume.[1] We
neglected to do that in the K8s workload, so this patch ammends the
Ansible play that formats the volume to scan the SCSI bus first.
Note that the LAMP workload was targetted toward one guest OS platform
(Ubuntu), and worked by installing the scsitools package and running the
rescan-scsi-bus script that it provides. The K8s workload is different
in that it's currently targetted toward multiple guests including Ubuntu
and CoreOS [2]...the latter of which doesn't have a package manager.
Rather than build conditionals into the code to handle this, I've taken
a slightly simpler and more platform-agnostic approach in this patch by
simply looking in /sys/class/scsi_host to see what adapters are present
and then scanning each of them with an echo command. This approach has
a few advantages: 1.) it works on guest OS's in which the name of the
package containing the rescan-scsi-tools script is different than on
Ubuntu, 2.) it works on guest OS's that don't have a package manager at
all, and 3.) since it doesn't require installing any additional
software, it is slightly faster and requires trivially less disk
footprint.
[1] https://review.openstack.org/#/c/383923/
[2] CentOS support is coming too:
https://review.openstack.org/#/c/457199/
Change-Id: Ie7204a23bc7a7d1da9a8cc60e684fc987159a41a
The current workload does not the new servers using the pre-allocated
IPs, in some situation, it will be hard to getting floating IPs.
This patch will allow a runner to pre-allocating some IPs, then assign
IPs to the server they like.
Change-Id: I992edb95d5d641983908f38340bb8ed546fc72aa
If using ubuntu image, cloud-init has to finish before other apt
command can be performed. This patch added logic to make sure
that cloud-init get finished before proceed with other tasks.
This patch also contains changes to make the seeding docker images
more generic so that any docker image can be included in a tar.gz
file, the images in the file will be docker loaded onto each worker
node
Change-Id: I505cdb791673dc978bc21b73f7ad672172340a9d
This workload accomplishes the following:
1. Provision 3 nodes or the number of nodes configured by stack_size
2. Create security group
3. Add security rules to allow ping, ssh, and kubernetes ports
4. Install common software onto each node such as docker
5. Download all the required software onto the master node
6. Setup the master node with kube-apiserver, kube-controller-manager
and kube-scheduler and configur each kubernetes services on the
master node
7. Download software for worker node from the master node.
8. Setup flanneld, docker, kubelet and kube-proxy on each work node.
9. Install kubernetes dashboard and dns services.
10. Install cockroachdb.
Change-Id: I3f7ec234aaa72dd6b2542e53e7eae2673ef7b408
The doc in github shows:
'apt-get update apt-get -y install docker.io ln -sf /usr/bin/docker.io /usr/local/bin/docker'
in one line.
I think it should be displayed as 'ssh-keygen -t rsa ...'
Change-Id: Ia579709d4f1eab8d09788d9aa2a3b753cc3ff6d8
* Disable checking for `scp`, add '-o StrictHostKeyChecking=no';
* Add the process that waiting for ssh connection OK of nodes.
Change-Id: Iec6d0c0a51219c42072d26b0920a1263200c2909
Closes-Bug: #1664790
When a cloud image only has root user, deploy wordpress will fail
without the flag since wordpress by default does not allow root user
to deploy artifacts. Adding this flag will allow root and non root
users to be used to deploy wordpress
Change-Id: Icf0a62bdfc5db5d0e14752908477288516143eb9
This should resolve two problems:
* Ansible cannot find `userdata.j2` due to incorrect path used
* Allow the template to be actually rendered as template instead of
being loaded as plain file without evaluation
There's another solution - simple move of `/roles/apply/templates/`
to `/roles/apply/files/templates/`. I have favored this one though
to make it possible to use Jinja2 expressions in the template file,
which I believe was the original creator's idea.
Change-Id: I268e4967597508e15d7de76051e097a43a0bfeac
1. mysql 5.6+ configuration
2. noting supporting ubuntu 16.xx
3. different platform code mixing
4. tags so that you can run selected tasks by role
5. uses openstack superuser theme and posts
6. haproxy used old configuration format
Closes-Bug #1654240
Closes-Bug #1649949
Change-Id: Ibbde92ed4fc18f85d004d54f00d3547c3df8fcc4
