Hello Lukas,

It sounds like you did it well overally.

Did you check the load on that machine? It could be that it turned out to be too weak to handle the desired set of services. It sounds like it started to swap crazy. Try something like 4G for starters (it would probably work with 2 but 4 is safe for sure!).

Another reason could be that networking went wrong with that VIP address. But it would likely have occurred earlier than the heat deployment step.

-yoctozepto

On Fri, Sep 4, 2020 at 5:01 PM Lukasluedke@web.de Lukasluedke@web.de wrote:

Hi everyone,

I am new to openstack and just followed the "Quick Start" guide for kolla-ansible [https://docs.openstack.org/kolla-ansible/ussuri/user/quickstart.html] on ussuri release, but now having some problems during deployment. I used virtualbox to create a new vm with ubuntu 18.04.5 [https://releases.ubuntu.com/18.04/ubuntu-18.04.5-live-server-amd64.iso.torre...] and attached one interface (enp0s3, 10.0.2.15) with Nat-Network and the second interface as for the neutron interface (enp0s8, no ip) an internal network and named it "neutron". Then I followed the guide to create a new virtual python environment (logged in over ssh using "user" user). -- sudo apt-get update sudo apt-get install python3-dev libffi-dev gcc libssl-dev sudo apt-get install python3-venv

python3 -m venv $(pwd) source $(pwd)/bin/activate pip install -U pip pip install 'ansible<2.10' --

Everything seems to work as mentioned with no problems so far. (Beside the missing "python-venv" package in the docs) -- pip install kolla-ansible sudo mkdir -p /etc/kolla sudo chown $USER:$USER /etc/kolla cp -r $(pwd)/share/kolla-ansible/etc_examples/kolla/* /etc/kolla cp $(pwd)/share/kolla-ansible/ansible/inventory/* .

mkdir /etc/ansible sudo nano /etc/ansible/ansible.cfg ### content from /etc/ansible/ansible.cfg [defaults] host_key_checking=False pipelining=True forks=100

###

I then modified the "multinode" file to only deploy on localhost (replaced all control01, etc. with localhost) as I first wanted to try openstack on one node. The ping worked, changed the globals.yml file according to the documentation, then bootstrap and the prechecks worked so far with no issue. -- ansible -i multinode all -m ping kolla-genpwd

sudo nano /etc/kolla/globals.yml ### additional content of /etc/kolla/globals.yml kolla_base_distro: "ubuntu" kolla_install_type: "source" network_interface: "enp0s3" neutron_external_interface: "enp0s8" kolla_internal_vip_address: "10.0.2.10" ###

kolla-ansible -i ./multinode bootstrap-servers kolla-ansible -i ./multinode prechecks --

But on the deploy stage I am stuck right now:

kolla-ansible -i ./multinode deploy

The mentioned error is telling me, that the keystone service is not available. I attached the error log file, because the error is longer. As a side node, now when I try to login using tty, the vm is nearly frozen and only prompts very, very slowly (about 10 minutes for prompt). ssh in putty is also frozen and the connection broke after about 30 minutes.

Does anyone know what is wrong or has some hints/insight on how to correctly deploy openstack using kolla-ansible?

Best Regards, Lukas Lüdke