Step 6. Launch ansible playbook to install OpenStack, the clouldet library, and the OpenStack++ extensions

[linuix]

I face the following error while performing step 6.
I have tested the script on Ubuntu 14.04 LTS 64-bit (only a single node) with two NIC card.
edit roles/openstack-controller/vars/main.yml and
roles/openstack-compute/vars/main.yml as follows

rabbit_pw: changeme
mysql_pw: changeme
db_pw: changeme
keystone_admin_pw: changeme
admin_token: changeme
metadata_secret: changeme
openstack_controller_node: cloudlet
controller_ip: 192.168.20.137
pub_iface: eth0
flat_iface: eth1

edit roles/cloudlet/vars/main.yml as follows

local_user: root
local_user_pw: 798

edit openstackpp.yaml as follows

 - hosts: controller
    remote_user: root
    vars:
            one_nic: False
    roles:
           -openstack-controller

- hosts: compute
    remote_user: root
    vars:
        single_node: True
        one_nic: False
    tags: compute
    roles:
        -openstack-compute

Then i run into a fatal error.

fatal: [localhost]: FAILED! => {"changed": true, "cmd": "fab --user root --password 798 install", "delta": "0:03:22.615341",
"end": "2018-06-06 19:50:51.945930", "msg": "non-zero return code", "rc": 1, "start": "2018-06-06 19:47:29.330589", "stderr":
"/usr/lib/python2.7/dist-packages/Crypto/Cipher/blockalgo.py:141:
FutureWarning: CTR mode needs counter parameter, not IV\n self._cipher = factory.new(key, *args, **kwargs)\n\n
Warning: sudo() received nonzero return code 4 while executing 'wget

htt p://mirrors.kernel.org/ubuntu/pool/universe/x/xdelta3/python-xdelta3_3.0.0.dfsg-1build1_amd64.deb
-O python-xdelta3.deb '!\n\n\n

Fatal error: Failed to download python-xdelta3.deb \n\nAborting.", "stderr_lines": ["/usr/lib/python2.7/dist-
packages/Crypto/Cipher/blockalgo.py:141:

FutureWarning: CTR mode needs counter parameter, not IV", " self._cipher = factory.new(key, *args, **kwargs)", "",
"Warning: sudo() received nonzero return code 4 while executing 'wget

http://mirrors.kernel.org/ubuntu/pool/universe/x/xdelta3/python-xdelta3_3.0.0.dfsg-1build1_amd64.deb
-O python-xdelta3.deb'!",

"", "",
"Fatal error: Failed to download python-xdelta3.deb", "", "Aborting."], "stdout": "[localhost] Executing task
'install'\n[localhost] run: egrep '^flags.*(vmx|svm)' /proc/cpuinfo > /dev/null\n[localhost] sudo: sudo apt-get
update\n[localhost] out: \n[localhost] out: Reading package lists... 0%\n[localhost] out: \n[localhost] out: Reading package
lists... 100%\n[localhost] out: \n[localhost] out: Reading package lists... Done\n[localhost] out: \n[localhost] out:
\n[localhost] out: Building dependency tree... 0%\n[localhost] out: \n[localhost] out: Building dependency tree...
0%\n[localhost] out: \n[localhost] out: Building dependency tree... 50%\n[localhost] out: \n[localhost] out: Building
dependency tree... 50%\n[localhost] out: \n[localhost] out: Building dependency tree \n[localhost] out: \n[localhost]
out: \n[localhost] out: Reading state information... 0%\n[localhost] out: \n[localhost] out: Reading state information...
0%\n[localhost] out: \n[localhost] out: Reading state information... Done\n[localhost] out: \n[localhost] out: libglu1-mesa is
already the newest version.\n[localhost] out: python-dev is already the newest version.\n[localhost] out: python-lzma is
already the newest version.\n[localhost] out: libc6-i386 is already the newest version.\n[localhost] out: libvirt-bin is
already the newest version.\n[localhost] out: libxml2-dev is already the newest version.\n[localhost] out: libxslt1-dev is
already the newest version.\n[localhost] out: python-lxml is already the newest version.\n[localhost] out: qemu-kvm is already
the newest version.\n[localhost] out: gvncviewer is already the newest version.\n[localhost] out: apparmor-utils is already the
newest version.\n[localhost] out: python-libvirt is already the newest version.\n[localhost] out: python-pip is already the
newest version.\n[localhost] out: 0 upgraded, 0 newly installed, 0 to remove and 6 not upgraded.\n[localhost] out:
\n[localhost] out: --2018-06-06 19:48:43--

htt p://mirrors.kernel.org/ubuntu/pool/universe/x/xdelta3/python-xdelta3_3.0.0.dfsg-1build1_amd64.deb\n[localhost] out:
Resolving mirrors.kernel.org (mirrors.kernel.org)... 198.145.21.9, 2001:19d0:306:6:0:1994:3:14\n[localhost] out: Connecting to
mirrors.kernel.org (mirrors.kernel.org)|198.145.21.9|:80... connected.\n[localhost] out: HTTP request sent, awaiting
response... 301 Moved Permanently\n[localhost] out:

Location: http://mirrors.edge.kernel.org/ubuntu/pool/universe/x/xdelta3/python-xdelta3_3.0.0.dfsg-1build1_amd64.deb

[following]\n[localhost] out: --2018-06-06 19:48:44-- htt p://mirrors.edge.kernel.org/ubuntu/pool/universe/x/xdelta3/python-
xdelta3_3.0.0.dfsg-1build1_amd64.deb\n[localhost] out: Resolving mirrors.edge.kernel.org (mirrors.edge.kernel.org)...
147.75.101.1, 2604:1380:2001:3900::1\n[localhost] out: Connecting to mirrors.edge.kernel.org
(mirrors.edge.kernel.org)|147.75.101.1|:80...

failed: Connection timed out.\n[localhost] out: Connecting to mirrors.edge.kernel.org
(mirrors.edge.kernel.org)|2604:1380:2001:3900::1|:80...

failed: Network is unreachable.\n[localhost] out: \n
Disconnecting from localhost... done.", "stdout_lines": ["[localhost] Executing task 'install'", "[localhost] run: egrep
'^flags.*(vmx|svm)' /proc/cpuinfo > /dev/null", "[localhost] sudo: sudo apt-get update", "[localhost] out: ", "[localhost] out:
Reading package lists... 0%", "[localhost] out: ", "[localhost] out: Reading package lists... 100%", "[localhost] out: ", "
[localhost] out: Reading package lists... Done", "[localhost] out: ", "[localhost] out: ", "[localhost] out: Building
dependency tree... 0%", "[localhost] out: ", "[localhost] out: Building dependency tree... 0%", "[localhost] out: ", "
[localhost] out: Building dependency tree... 50%", "[localhost] out: ", "[localhost] out: Building dependency tree... 50%", "
[localhost] out: ", "[localhost] out: Building dependency tree ", "[localhost] out: ", "[localhost] out: ", "[localhost]
out: Reading state information... 0%", "[localhost] out: ", "[localhost] out: Reading state information... 0%", "[localhost]
out: ", "[localhost] out: Reading state information... Done", "[localhost] out: ", "[localhost] out: libglu1-mesa is already
the newest version.", "[localhost] out: python-dev is already the newest version.", "[localhost] out: python-lzma is already
the newest version.", "[localhost] out: libc6-i386 is already the newest version.", "[localhost] out: libvirt-bin is already
the newest version.", "[localhost] out: libxml2-dev is already the newest version.", "[localhost] out: libxslt1-dev is already
the newest version.", "[localhost] out: python-lxml is already the newest version.", "[localhost] out: qemu-kvm is already the
newest version.", "[localhost] out: gvncviewer is already the newest version.", "[localhost] out: apparmor-utils is already the
newest version.", "[localhost] out: python-libvirt is already the newest version.", "[localhost] out: python-pip is already the
newest version.", "[localhost] out: 0 upgraded, 0 newly installed, 0 to remove and 6 not upgraded.", "[localhost] out: ", "
[localhost] out: --2018-06-06 19:48:43-- htt p://mirrors.kernel.org/ubuntu/pool/universe/x/xdelta3/python-xdelta3_3.0.0.dfsg-
1build1_amd64.deb", "[localhost] out: Resolving mirrors.kernel.org (mirrors.kernel.org)... 198.145.21.9,
2001:19d0:306:6:0:1994:3:14", "[localhost] out: Connecting to mirrors.kernel.org (mirrors.kernel.org)|198.145.21.9|:80...
connected.", "[localhost] out: HTTP request sent, awaiting response... 301 Moved Permanently", "[localhost] out: Location:
htt p://mirrors.edge.kernel.org/ubuntu/pool/universe/x/xdelta3/python-xdelta3_3.0.0.dfsg-1build1_amd64.deb [following]", "
[localhost] out: --2018-06-06 19:48:44-- htt p://mirrors.edge.kernel.org/ubuntu/pool/universe/x/xdelta3/python-
xdelta3_3.0.0.dfsg-1build1_amd64.deb", "[localhost] out: Resolving mirrors.edge.kernel.org (mirrors.edge.kernel.org)...
147.75.101.1, 2604:1380:2001:3900::1", "[localhost] out: Connecting to mirrors.edge.kernel.org
(mirrors.edge.kernel.org)|147.75.101.1|:80...

failed: Connection timed out.", "[localhost] out: Connecting to mirrors.edge.kernel.org
(mirrors.edge.kernel.org)|2604:1380:2001:3900::1|:80...

failed: Network is unreachable.", "[localhost] out: ", "Disconnecting from localhost... done."]}

question 1

Is my configuration correct which displaying above ?

question 2

I see it run into an error when

wget http://mirrors.kernel.org/ubuntu/pool/universe/x/xdelta3/python-xdelta3_3.0.0.dfsg-1build1_amd64.deb

with the notice failed: Network is unreachable.

But i can download it from the web site( i consider the network is correct)
then i download it manually into the directory ~/elijah-openstack/ansible ( i dont know which direcroty should i download python-xdelta3 into)
but unfortunately it still not working.
I don't know how to solve it.

question 3

I always encounter a problem in the step (cloiulet) install library using fabric

Is there another way to install the cloiulet library ?

I need some help .
Thank you very much

[teiszler]

@linuix,

1. The configuration looks good to me based on what you have said regarding your environment.
2. Can you run the wget on that URL from the host? Does it resolve? It looks like it is being redirected via a 301. When you say you can download it, are you doing so via a browser? I would try to execute exactly what is being done by the script and see if it works manually.
3. This step (installing the cloudlet library via fabric) can take some time and there is unfortunately no feedback given to Ansible. I would wait 10 minutes or so before interrupting it. If you reach 15 minutes then it is possible you have run into the issue where the username/password you specify in roles/cloudlet/vars/main.yml are incorrect. You need to make sure that those credentials are correct as fabric uses them to login and perform the installation. You could install the cloudlet library manually by following the instructions on the elijah-provisioning repository, but I suspect you will run into the same issue that you are seeing via Ansible.

Regards,
Tom

[linuix]

First of all, thank you very much for your time to reply to my question.

I can download python-xdelta3_3.0.0.dfsg-1build1_amd64.deb via a browser，
but when i executed wget on the URL from the host,there was something wrong showing as follows.
You said it was being redirected via a 301,how can i solve this issue ?
I executed wget www.google.com ,it doesn`t work too,while i execute wget www.biadu.com ,it finish soon.

root@cloudlet:~# wget http://mirrors.kernel.org/ubuntu/pool/universe/x/xdelta3/python-xdelta3_3.0.0.dfsg-1build1_amd64.deb
--2018-06-08 09:37:26--  http://mirrors.kernel.org/ubuntu/pool/universe/x/xdelta3/python-xdelta3_3.0.0.dfsg-1build1_amd64.deb
Resolving mirrors.kernel.org (mirrors.kernel.org)... 149.20.37.36, 2001:4f8:4:6f:0:1994:3:14
Connecting to mirrors.kernel.org (mirrors.kernel.org)|149.20.37.36|:80... connected.
HTTP request sent, awaiting response... 301 Moved Permanently
Location: http://mirrors.edge.kernel.org/ubuntu/pool/universe/x/xdelta3/python-xdelta3_3.0.0.dfsg-1build1_amd64.deb [following]
--2018-06-08 09:37:26--  http://mirrors.edge.kernel.org/ubuntu/pool/universe/x/xdelta3/python-xdelta3_3.0.0.dfsg-1build1_amd64.deb
Resolving mirrors.edge.kernel.org (mirrors.edge.kernel.org)... 147.75.101.1, 2604:1380:2001:3900::1
Connecting to mirrors.edge.kernel.org (mirrors.edge.kernel.org)|147.75.101.1|:80... failed: Connection timed out.
Connecting to mirrors.edge.kernel.org (mirrors.edge.kernel.org)|2604:1380:2001:3900::1|:80... failed: Network is unreachable.

root@cloudlet:~# wget www.google.com
--2018-06-08 09:41:32--  http://www.google.com/
Resolving www.google.com (www.google.com)... 31.13.65.18, 31.13.66.1
Connecting to www.google.com (www.google.com)|31.13.65.18|:80... failed: Connection timed out.
Connecting to www.google.com (www.google.com)|31.13.66.1|:80... failed: Connection timed out.
Retrying.

--2018-06-08 09:45:48--  (try: 2)  http://www.google.com/
Connecting to www.google.com (www.google.com)|31.13.65.18|:80... failed: Connection timed out.
Connecting to www.google.com (www.google.com)|31.13.66.1|:80... failed: Connection timed out.
Retrying.

--2018-06-08 09:50:04--  (try: 3)  http://www.google.com/
Connecting to www.google.com (www.google.com)|31.13.65.18|:80... 
root@cloudlet:~# ^C

root@cloudlet:~# wget www.baidu.com 
--2018-06-08 09:51:25--  http://www.baidu.com/
Resolving www.baidu.com (www.baidu.com)... 119.75.213.61, 119.75.216.20
Connecting to www.baidu.com (www.baidu.com)|119.75.213.61|:80... connected.
HTTP request sent, awaiting response... 200 OK
Length: 2381 (2.3K) [text/html]
Saving to: ‘index.html’

100%[=============================================================================>] 2,381       --.-K/s   in 0s      

2018-06-08 09:51:25 (286 MB/s) - ‘index.html’ saved [2381/2381]

root@cloudlet:~# wget --server-response http://mirrors.kernel.org/ubuntu/pool/universe/x/xdelta3/python-xdelta3_3.0.0.dfsg-1build1_amd64.deb
--2018-06-08 11:02:01--  http://mirrors.kernel.org/ubuntu/pool/universe/x/xdelta3/python-xdelta3_3.0.0.dfsg-1build1_amd64.deb
Resolving mirrors.kernel.org (mirrors.kernel.org)... 149.20.37.36, 2001:19d0:306:6:0:1994:3:14
Connecting to mirrors.kernel.org (mirrors.kernel.org)|149.20.37.36|:80... connected.
HTTP request sent, awaiting response... 
  HTTP/1.1 301 Moved Permanently
  Content-length: 0
  Location: http://mirrors.edge.kernel.org/ubuntu/pool/universe/x/xdelta3/python-xdelta3_3.0.0.dfsg-1build1_amd64.deb
  Connection: close
Location: http://mirrors.edge.kernel.org/ubuntu/pool/universe/x/xdelta3/python-xdelta3_3.0.0.dfsg-1build1_amd64.deb [following]
--2018-06-08 11:02:01--  http://mirrors.edge.kernel.org/ubuntu/pool/universe/x/xdelta3/python-xdelta3_3.0.0.dfsg-1build1_amd64.deb
Resolving mirrors.edge.kernel.org (mirrors.edge.kernel.org)... 147.75.101.1, 2604:1380:2001:3900::1
Connecting to mirrors.edge.kernel.org (mirrors.edge.kernel.org)|147.75.101.1|:80... failed: Connection timed out.
Connecting to mirrors.edge.kernel.org (mirrors.edge.kernel.org)|2604:1380:2001:3900::1|:80... failed: Network is unreachable.

I set local_user asroot( i login as root user),and i set local_user_pw as my root user's password in the configuration file
roles/cloudlet/vars/main.yml,i think it is correct, yes?
I will try to install the cloudlet library manually following the instructions
on the elijah-provisioning repository,and see if this problem can be solve.

Regards,
linuix

[zetelight]

@teiszler Thank you for your long-term patience and help for all of us! Appreciate your help so much!
I had an error at TASK [cloudlet : (CLOUDLET) install library using fabric] again, like @linuix had here.

fatal: [localhost]: FAILED! => {"changed": true, "cmd": "fab --user adamchen --password admin install", "delta": "0:00:00.594293", "end": "2018-07-08 22:25:40.898323", "msg": "non-zero return code", "rc": 1, "start": "2018-07-08 22:25:40.304030", "stderr": "/usr/lib/python2.7/dist-packages/Crypto/Cipher/blockalgo.py:141: FutureWarning: CTR mode needs counter parameter, not IV\n self.cipher = factory.new(key, *args, **kwargs)\n\nFatal error: Support only Ubuntu Precise (14.04) or Ubuntu Trusty (16.04)\n\nAborting.", "stderrlines": ["/usr/lib/python2.7/dist-packages/Crypto/Cipher/blockalgo.py:141: FutureWarning: CTR mode needs counter parameter, not IV", " self.cipher = factory.new(key, args, kwargs)", "", "Fatal error: Support only Ubuntu Precise (14.04) or Ubuntu Trusty (16.04)", "", "Aborting."], "stdout": "[localhost] Executing task 'install'\n[localhost] run: egrep '^flags.(vmx|svm)' /proc/cpuinfo > /dev/null\n[localhost] out: /etc/profile.d/cloudlet-gateway.sh: line 2: /root/admin-openrc.sh: Permission denied\n[localhost] out: \n\nDisconnecting from localhost... done.", "stdoutlines": ["[localhost] Executing task 'install'", "[localhost] run: egrep '^flags.*(vmx|svm)' /proc/cpuinfo > /dev/null", "[localhost] out: /etc/profile.d/cloudlet-gateway.sh: line 2: /root/admin-openrc.sh: Permission denied", "[localhost] out: ", "", "Disconnecting from localhost... done."]}

These two lines got my attention, but please allow me to give more info here.

I successfully installed the openstack++ with your help before, but then, After I reboot and log in the desktop, I saw this /etc/profile.d/cloudlet-gateway.sh: line 2: /root/admin-openrc.sh: Permission denied prompted by an "Error" window by Ubuntu before I do anything. I ignored this at first time.Then, as the README suggested, after I reboot each time, I re-run

ansible-playbook -i ./hosts openstackpp.yaml

again to recreate the dummy interface and restart the necessary nova services.
So, the error message above was what happened after I re-run the ansible playbook.
Any hints or help will be appreciated!

one more question

By the way, a tiny and silly question. After I installed, how could I login in the dashboard? I tried the port specified in admin-opensrc.sh and the url shown in the picture from README. None of them would show a dashboard for me.

Some helpful info

1. OS info: Ubuntu 14.04
2. I can type cloudlet in the terminal to trigger it.
3. I can open the port specified in admin-openrc.sh from the browser to see some JSON messages, such as

   {"version": {"status": "stable", "updated": "2015-03-30T00:00:00Z", "media-types": [{"base": "application/json", "type": "application/vnd.openstack.identity-v3+json"}], "id": "v3.4", "links": [{"href": "http://localhost:35357/v3/", "rel": "self"}]}}

Update

I tried to remove /root/admin-openrc.sh and /etc/profile.d/cloudlet-gateway.sh to expect ansible-playbook could generate again for me. At this time, there's no error for TASK [cloudlet : (CLOUDLET) install library using fabric] but I had a new one:

TASK [openstack-controller : (KEYSTONE) populate identity service database] **************************
fatal: [localhost]: FAILED! => {"changed": true, "cmd": "/bin/sh -c 'keystone-manage db_sync' keystone", "delta": "0:00:00.066584", "end": "2018-07-08 23:00:14.861341", "msg": "non-zero return code", "rc": 1, "start": "2018-07-08 23:00:14.794757", "stderr": "Traceback (most recent call last):\n File \"/usr/bin/keystone-manage\", line 30, in \n from keystone import cli\n File \"/usr/lib/python2.7/dist-packages/keystone/cli.py\", line 22, in \n import pbr.version\nImportError: No module named pbr.version", "stderr_lines": ["Traceback (most recent call last):", " File \"/usr/bin/keystone-manage\", line 30, in ", " from keystone import cli", " File \"/usr/lib/python2.7/dist-packages/keystone/cli.py\", line 22, in ", " import pbr.version", "ImportError: No module named pbr.version"], "stdout": "", "stdout_lines": []}

How to resolve this now? Is there anyway to re-install Openstack++ without re-install the OS?

Again, thanks for your patience and help!

[teiszler]

@zetelight,

It seems you may have a partially installed OpenStack++ since the script didn't finish when you ran it after a reboot. When the fabric script for eiljah-provisioning is run, it installs a version of pbr that is incompatible with OpenStack Kilo and so there is a step to remove it later on in the Ansible steps. In general you should be able to repair the installation by rerunning the script, though it is unclear what state things are currently in. Did you install the cloudlet gateway role by explicitly executing the playbook with the --tags gateway option? If not, the cloudlet-gateway.sh should not have been copied to /etc/profile.d. it looks like you are receiving the error because admin-openrc.sh doesn't have execute permissions and so at startup, the cloudlet-gateway.sh script cannot run it. If you do a chmod +x /root/admin-openrc.sh it should fix the error at boot time. I will update the Ansible playbook to set the permissions of admin-openrc.sh properly.

Regarding the dashboard, I assume you mean the OpenStack dashboard. That is at http://localhost/horizon (or alternatively the IP of your host machine if coming at it from outside the machine itself).

Let me know if you have any other questions.
Regards,
Tom

[zetelight]

Thank you for your reply, Tom!
I installed the Openstack++ at first and then I tried to install the cloudlet-gateway.
Okay, so my current case is
1. My admin-openrc.sh and cloudlet-gateway.sh are not here anymore, because I deleted them and expected they could be re-generated by running commands again.
2. if I run ansible-playbook -i ./hosts openstackpp.yaml, I will get the error messages describe from my last post
3. If I run ansible-playbook -i ./hosts openstackpp.yaml --tags gateway, I will get these messages below:

TASK [cloudlet-gateway : (GATEWAY) flask db upgrade] ***************************
fatal: [localhost]: FAILED! => {"changed": true, "cmd": ["./venv/bin/flask", "db", "upgrade"], "delta": "0:00:00.522125", "end": "2018-07-09 20:46:19.110340", "msg": "non-zero return code", "rc": 2, "start": "2018-07-09 20:46:18.588215", "stderr": "Usage: flask db upgrade [OPTIONS] [REVISION]\n\nError: Could not locate Flask application. You did not provide the FLASK_APP environment variable.\n\nFor more information see http://flask.pocoo.org/docs/latest/quickstart/", "stderr_lines": ["Usage: flask db upgrade [OPTIONS] [REVISION]", "", "Error: Could not locate Flask application. You did not provide the FLASK_APP environment variable.", "", "For more information see http://flask.pocoo.org/docs/latest/quickstart/"], "stdout": "", "stdout_lines": []}

I didn't install Flask and I also looked over though the quickstart page of Flask, but there's nothing to help me debug for this issue.

Do I have to re-install the whole OS again to make it clean?
I understand that it is also kinda hard for you to debug for me. I appreciate your patience and help!
Best wish,
Adam

[teiszler]

@zetelight,

You could try just running the playbook with the --tags keystone command which will just attempt the tasks associated with the keystone component (including restoring your admin-openrc.sh). You should also in general be able to try the whole playbook again from the top. This should ensure that pbr is installed (I believe this comes from one of the dependencies installed during the openstack python client in the common Ansible role. If running the entire playbook doesn't successfully repair it, you could try to manually install a version of pbr either with apt-get or pip and then launch the keystone tasks with --tags keystone to restore the database tables.

Regards,
Tom

[zetelight]

thank you, Tom! However, after I re-install pbr, re-run commands for Keystone, I have /cloudlet-gateway.sh again but it still has the previous error which is the one about "Permission denied".
Anyway, I will try re-install OS and hopefully, it can work.
Thank you!
Adam

[zetelight]

BTW, some quick question tho.
Since I installed the openstack++ successfully(Ya, I made it after several fails!), do I have to install elijah-provisioning again? I noticed that I already have that repo along with the installation of elijah-openstack.

That is, elijah-openstack provide all UI things for elijah-provisioning so that I can use the UI do everything rather than executing commands in terminal. Am I right?

My main researching purpose is that examining the power/efficiency of cloudlet(Yes, not too deep but I'd like to know more about cloud computing through this research), so could I basically use these Android projects you provided in elijah-provisioning to evaluate the performance, e.g. latency, VM overlay size?

Again, thank you for your help!
Best wishes,

Adam

[teiszler]

@zetelight,

Regarding the permission denied: you can always set the permissions of cloudlet-gateway.sh or admin-openrc.sh manually with sudo chmod.

In terms of elijah-provisioning, it should have been installed as the first set of steps in the Ansible playbook. There should be no reason you need to install eiljah-provisioning manually. You are essentially correct about elijah-openstack. It is a plugin for OpenStack to expose the functionality of elijah-provisioning in the Horizon UI component of OpenStack. You can do all the same operations in elijah-provisioning from the command-line using the cloudlet command.

Regarding your research interests, I think you have the correct idea though I would suggest that you don't need to constrain yourself to any particular workloads. What you run on the VMs could be any existing application. The VM doesn't even need to be running a Linux distribution; it could be a Windows 7/10 VM that you use as a virtual desktop. If you are interested in attributes such as latency, it might be a good idea to setup the same workloads on the cloudlet and in the cloud, like same Amazon EC2. You can then compare the latency on various workloads when going (presumably one hop) to the cloudlet versus all the way to the nearest Amazon site.

Regards,
Tom

[zetelight]

Thank you! Appreciate your help!