scripts/bootstrap-ansible.sh failed due to pip version
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
OpenStack-Ansible |
Fix Released
|
High
|
Unassigned |
Bug Description
Hi Folks,
It seems that there is some kind of incompatibility in the dependancies in CentOS 8.2 and the OpenStack ansible:
I have installed the deployment guide as suggested in https:/
While running the script scripts/
Dmitriy Rabotyagov (noonedeadpunk) wrote : | #2 |
Changed in openstack-ansible: | |
status: | New → Incomplete |
phonix6 (phonix6-deactivatedaccount-deactivatedaccount) wrote : | #3 |
- pip errors Edit (50.7 KiB, text/plain)
Thanks for answering.
Here is the info, in the attachment.
yes that's right, i have upgraded the 8.2 to 8.3.2011
KR,
phonix6 (phonix6-deactivatedaccount-deactivatedaccount) wrote : | #4 |
I believe that it could be related to the python installation.
I have the:
python36-
After cloning the git from:
git clone -b master https:/
and trying to run the integrity checker:
openstack-ansible setup-infrastru
I get the following:
[root@srv-infra playbooks]# openstack-ansible setup-infrastru
-bash: openstack-ansible: command not found
I really cannot understand this, the git from openstack-ansible is defenitly cloned.
So it's either the pip package or the python package.
I have tried to remove the python, but the same version will be installed after.
[root@srv-infra playbooks]# dnf remove python36-
Dependencies resolved.
=======
Package Architecture Version Repository Size
=======
Removing:
python36-devel x86_64 3.6.8-2.
Removing unused dependencies:
platform-
python-rpm-macros noarch 3-39.el8 @appstream 3.3 k
python3-
Transaction Summary
=======
Remove 4 Packages
Freed space: 767 k
Is this ok [y/N]:
[root@srv-infra playbooks]# dnf install python3-devel
Last metadata expiration check: 0:35:29 ago on Tue 26 Jan 2021 11:05:06 AM CET.
Dependencies resolved.
=======
Package ...
Dmitriy Rabotyagov (noonedeadpunk) wrote : | #5 |
At the moment you failed to bootstrap openstack-ansible because of the bug. Patch https:/
dnf install rsync
and after that re-run ./scripts/
openstack-ansible binary is being installed with this script, so until it finish with success you won't have openstack-ansible binary.
You can also join #openstack-ansible IRC channel on Freenode for more help if needed.
Dmitriy Rabotyagov (noonedeadpunk) wrote : | #6 |
If we go futher, openstack-ansible is actually a pretty simple wrapper around ansible-playbook and ansible binaries which source code is https:/
phonix6 (phonix6-deactivatedaccount-deactivatedaccount) wrote : | #7 |
Thanks for the input!
It really was the case, NOTABUG, but definitely a documentation inconsistency.
While looking into the guide from here:
https:/
I do not see any instruction to install rsync package.
# dnf install https:/
# dnf install git chrony openssh-server python3-devel sudo
# dnf group install "Development Tools"
Changed in openstack-ansible: | |
status: | Incomplete → In Progress |
Changed in openstack-ansible: | |
importance: | Undecided → High |
OpenStack Infra (hudson-openstack) wrote : Fix included in openstack/openstack-ansible 22.0.1 | #8 |
This issue was fixed in the openstack/
Changed in openstack-ansible: | |
status: | In Progress → Fix Released |
phonix6 (phonix6-deactivatedaccount-deactivatedaccount) wrote : | #9 |
I am having the issue with 22.0.1 rc release.
I am unable to run the playbook, what is wrong with that?
I am using ubuntu 20.4 now, I am getting:
root@iaas-
openstack-ansible: command not found
root@iaas-
total 328
-rw-r--r-- 1 root root 4452 Feb 13 12:52 ceph-install.yml
-rw-r--r-- 1 root root 1840 Feb 13 12:52 ceph-nfs-
-rw-r--r-- 1 root root 2265 Feb 13 12:52 ceph-rgw-
-rw-r--r-- 1 root root 3724 Feb 13 12:52 ceph-rgw-
drwxr-xr-x 2 root root 4096 Feb 13 12:52 common-playbooks
drwxr-xr-x 2 root root 4096 Feb 13 12:52 common-tasks
-rw-r--r-- 1 root root 778 Feb 13 12:52 containers-
-rw-r--r-- 1 root root 2700 Feb 13 12:52 containers-
-rw-r--r-- 1 root root 2715 Feb 13 12:52 containers-
-rw-r--r-- 1 root root 2210 Feb 13 12:52 containers-
-rw-r--r-- 1 root root 3564 Feb 13 12:52 containers-
-rw-r--r-- 1 root root 3546 Feb 13 12:52 containers-
-rw-r--r-- 1 root root 1056 Feb 13 12:52 containers-
drwxr-xr-x 3 root root 4096 Feb 13 12:52 defaults
-rw-r--r-- 1 root root 1273 Feb 13 12:52 etcd-install.yml
-rw-r--r-- 1 root root 2089 Feb 13 12:52 galera-install.yml
-rw-r--r-- 1 root root 1887 Feb 13 12:52 haproxy-install.yml
-rw-r--r-- 1 root root 3578 Feb 13 12:52 healthcheck-
-rw-r--r-- 1 root root 11350 Feb 13 12:52 healthcheck-
-rw-r--r-- 1 root root 18215 Feb 13 12:52 healthcheck-
-rw-r--r-- 1 root root 4494 Feb 13 12:52 infra-journal-
drwxr-xr-x 2 root root 4096 Feb 13 12:52 library
-rw-r--r-- 1 root root 1160 Feb 13 12:52 listening-
lrwxrwxrwx 1 root root 25 Feb 13 12:52 lxc-containers-
lrwxrwxrwx 1 root root 26 Feb 13 12:52 lxc-containers-
lrwxrwxrwx 1 root root 23 Feb 13 12:52 lxc-hosts-setup.yml -> containers-
-rw-r--r-- 1 root root 1265 Feb 13 12:52 memcached-
-rw-r--r-- 1 root root 2959 Feb 13 12:52 openstack-
-rw-r--r-- 1 root root 2145 Feb 13 12:52 os-adjutant-
-rw-r--r-- 1 root root 1302 Feb 13 12:52 os-aodh-install.yml
-rw-r--r-- 1 root root 1328 Feb 13 12:52 os-barbican-
-rw-r--r-- 1 root root 1381 Feb 13 12:52 os-blazar-
-rw-r--r-- 1 root root 1338 Feb 13 12:52 os-ceilometer-
-rw-r--r-- 1 root root 6840 Feb 13 12:52 os-cinder-
-rw-r--r-- 1 root root 1439 Feb 13 12:52 os-designate-
-rw-r--r-- 1 root root 6345 Feb 13 12:52 os-glance-
-rw-r--r-- 1 root root 1808 Feb 13 12:52 os-gnocchi-
-rw-r--r-- 1 root root 1363 Feb 13 12:52 os-heat-install.yml
-rw-r--r-- 1 root root 1382 Feb 13 12:52 os-horizon-
-rw-r--r-- 1 root root 1385 Feb 13 12:52 os-ironic-
-rw-r--r-- 1 root root 6103 Feb 13 12:52 os-keystone-
-rw-r--r-- 1 root root 1422 Feb 13 12:52 os-magnum-
-rw-r--r-- 1 root root 2628 Feb 13 12:52 os-...
phonix6 (phonix6-deactivatedaccount-deactivatedaccount) wrote : | #10 |
Anyone knows why I’m getting “ openstack-ansible: command not found” even tough the openstack-ansible is cloned to the right location?
Thanks,
Dmitriy Rabotyagov (noonedeadpunk) wrote : | #11 |
It might be only, if run of the ./scripts/
phonix6 (phonix6-deactivatedaccount-deactivatedaccount) wrote : | #12 |
Thx,
But I'm still having the same results:
root@iaas-
bootstrap-
root@iaas-
root@iaas-
+ export HTTP_PROXY=
+ HTTP_PROXY=
+ export HTTPS_PROXY=
+ HTTPS_PROXY=
+ export ANSIBLE_
+ ANSIBLE_
+ export ANSIBLE_
+ ANSIBLE_
+ export ANSIBLE_
+ ANSIBLE_
+ export USER_ROLE_
+ USER_ROLE_
+ export USER_COLLECTION
+ USER_COLLECTION
+ export SSH_DIR=/root/.ssh
+ SSH_DIR=/root/.ssh
+ export DEBIAN_
+ DEBIAN_
+ export SETUP_ARA=false
+ SETUP_ARA=false
+ export PIP_OPTS=
+ PIP_OPTS=
+ export OSA_WRAPPER_
+ OSA_WRAPPER_
++ dirname scripts/
+ cd scripts/..
+ info_block 'Checking for required libraries.'
+ source scripts/
++ LINE=--
++ ANSIBLE_PARAMETERS=
+++ date +%s
++ STARTTIME=
++ COMMAND_
++ PIP_COMMAND=
++ ZUUL_PROJECT=
++ GATE_EXIT_
++ GATE_EXIT_
++ GATE_EXIT_
++ GATE_EXIT_
++ [[ -n '' ]]
++ '[' -z '' ']'
+++ grep -c '^processor' /proc/cpuinfo
++ CPU_NUM=4
++ '[' 4 -lt 10 ']'
++ ANSIBLE_FORKS=4
++ trap 'exit_fail 404 0 '\''Received STOP Signal'\''' SIGHUP SIGINT SIGTERM
++ trap 'exit_fail 405 0' ERR
+++ id -u
++ '[' 0 '!=' 0 ']'
++ '[' '!' -d etc -a '!' -d scripts -a '!' -d playbooks ']'
++ export PATH=/usr/
++ PATH=/usr/
++ export HOME=/root
++ HOME=/root
++ [[ -f /usr/local/
+ info_block 'Bootstrapping System with Ansible'
+ echo -------
-------
+ print_info 'Bootstrapping System with Ansible'
+ PROC_NAME='- [ Bootstrapping System with Ansible ] -'
+ printf '\n%s%s\n' '- [ Bootstrapping System with Ansible ] -' -------
- [ Bootstrapping System with Ansible ] -------
+ echo -------
-------
++ pwd
+ export OSA_CLONE_
+ OSA_CLONE_
++ readlink -f ansible-
+ ANSIBLE_
Dmitriy Rabotyagov (noonedeadpunk) wrote : | #13 |
Eventually this is because of the wrong repo that has been added to the deploy host:
E: The repository 'http://
. We don't install such repo and it's smth that was installed by you for this host. You should remove all non working repos from the host and ideally use clean system from the deploy host.
phonix6 (phonix6-deactivatedaccount-deactivatedaccount) wrote : | #14 |
Weird...
I have used the official guide.
I have run over the test environment and started again with centos 8.
It would be good to move away from centos, as it not going to last after 2022.
Thx,
Dmitriy Rabotyagov (noonedeadpunk) wrote : | #15 |
I'd recommend using Ubuntu/Debian indeed, as the issue you had is totally related to third party ppa that has been added somehow. From what I see in url, it's related to munin monitoring tool.
phonix6 (phonix6-deactivatedaccount-deactivatedaccount) wrote : | #16 |
Unfortunately i am getting the error even in the new environment under Centos 8.
All installed and updated today:
TASK [Append user overridden roles] *******
[WARNING]: Unable to find '/etc/openstack
ok: [localhost]
TASK [Clone git repos (parallel)] *******
An exception occurred during task execution. To see the full traceback, use -vvv. The error was: TypeError: Value of unknown type: <class 'multiprocessin
fatal: [localhost]: FAILED! => {"changed": false, "module_stderr": "Traceback (most recent call last):\n File \"/root/
Dmitriy Rabotyagov (noonedeadpunk) wrote : | #17 |
Eventually this task should have fallback to non-parallel clone process which is not critical and the process should proceed
Dmitriy Rabotyagov (noonedeadpunk) wrote : | #18 |
What version are you checking out? I think we should have covered this issue and released fix with 22.0.1 which is https:/
phonix6 (phonix6-deactivatedaccount-deactivatedaccount) wrote : | #19 |
I cloned the git 22.0.0.0rc1 from the Victoria release.
I follow the guide in:
https:/
Thx
Dmitriy Rabotyagov (noonedeadpunk) wrote : | #20 |
This doc supposed to get latest available tag from the repo but this mechanism is broken at the moment. You should checkout to 22.0.1 as it's a stable release.
phonix6 (phonix6-deactivatedaccount-deactivatedaccount) wrote : | #21 |
I checked 22.0.1, but now the bootstrap is not completing as follows:
git clone https:/
/opt/
cd /opt/openstack-
# ext switch the applicable branch/tag to be deployed from. Note that deploying from the head of a branch may result in an unstable build due to changes in flight and upstream OpenStack changes. For a test (ie not a development) build it is usually best to checkout the latest tagged version.
git tag -l
git checkout stable/victoria
git describe --abbrev=0 --tags
git checkout 20.0.1
[root@iaas-infra openstack-ansible]# scripts/
+ export HTTP_PROXY=
+ HTTP_PROXY=
+ export HTTPS_PROXY=
+ HTTPS_PROXY=
+ export ANSIBLE_
+ ANSIBLE_
+ export ANSIBLE_
+ ANSIBLE_
+ export USER_ROLE_
+ USER_ROLE_
+ export SSH_DIR=/root/.ssh
+ SSH_DIR=/root/.ssh
+ export DEBIAN_
+ DEBIAN_
+ export SETUP_ARA=false
+ SETUP_ARA=false
+ export PIP_OPTS=
+ PIP_OPTS=
+ export OSA_WRAPPER_
+ OSA_WRAPPER_
++ dirname scripts/
+ cd scripts/..
+ info_block 'Checking for required libraries.'
+ source scripts/
++ LINE=--
++ ANSIBLE_PARAMETERS=
+++ date +%s
++ STARTTIME=
++ COMMAND_
++ PIP_COMMAND=
++ ZUUL_PROJECT=
++ GATE_EXIT_
++ GATE_EXIT_
++ GATE_EXIT_
++ GATE_EXIT_
++ [[ -n '' ]]
++ '[' -z '' ']'
+++ grep -c '^processor' /proc/cpuinfo
++ CPU_NUM=4
++ '[' 4 -lt 10 ']'
++ ANSIBLE_FORKS=4
++ trap 'exit_fail 398 0 '\''Received STOP Signal'\''' SIGHUP SIGINT SIGTERM
++ trap 'exit_fail 399 0' ERR
+++ id -u
++ '[' 0 '!=' 0 ']'
++ '[' '!' -d etc -a '!' -d scripts -a '!' -d playbooks ']'
++ export PATH=/usr/
++ PATH=/usr/
++ export HOME=/root
++ HOME=/root
++ [[ -f /usr/local/
+ info_block 'Bootstrapping System with Ansible'
+ echo -------
-------
+ print_info 'Bootstrapping System with Ansible'
+ PROC_NAME='- [ Bootstrapping System with Ansible ] -'
+ printf '\n%s%s\n' '- [ Bootstrapping System with Ansible ] -' -------
- [ Bootstrapping System with Ansible ] -------
+ echo -------
-------
++ pwd
+ export OSA_CLONE_
+ OSA_CLONE_
++ readlink -f ansible-role-re...
phonix6 (phonix6-deactivatedaccount-deactivatedaccount) wrote : | #22 |
[root@iaas-infra openstack-ansible]# scripts/
+ export HTTP_PROXY=
+ HTTP_PROXY=
+ export HTTPS_PROXY=
+ HTTPS_PROXY=
+ export ANSIBLE_
+ ANSIBLE_
+ export ANSIBLE_
+ ANSIBLE_
+ export USER_ROLE_
+ USER_ROLE_
+ export SSH_DIR=/root/.ssh
+ SSH_DIR=/root/.ssh
+ export DEBIAN_
+ DEBIAN_
+ export SETUP_ARA=false
+ SETUP_ARA=false
+ export PIP_OPTS=
+ PIP_OPTS=
+ export OSA_WRAPPER_
+ OSA_WRAPPER_
++ dirname scripts/
+ cd scripts/..
+ info_block 'Checking for required libraries.'
+ source scripts/
++ LINE=--
++ ANSIBLE_PARAMETERS=
+++ date +%s
++ STARTTIME=
++ COMMAND_
++ PIP_COMMAND=
++ ZUUL_PROJECT=
++ GATE_EXIT_
++ GATE_EXIT_
++ GATE_EXIT_
++ GATE_EXIT_
++ [[ -n '' ]]
++ '[' -z '' ']'
+++ grep -c '^processor' /proc/cpuinfo
++ CPU_NUM=4
++ '[' 4 -lt 10 ']'
++ ANSIBLE_FORKS=4
++ trap 'exit_fail 398 0 '\''Received STOP Signal'\''' SIGHUP SIGINT SIGTERM
++ trap 'exit_fail 399 0' ERR
+++ id -u
++ '[' 0 '!=' 0 ']'
++ '[' '!' -d etc -a '!' -d scripts -a '!' -d playbooks ']'
++ export PATH=/usr/
++ PATH=/usr/
++ export HOME=/root
++ HOME=/root
++ [[ -f /usr/local/
+ info_block 'Bootstrapping System with Ansible'
+ echo -------
-------
+ print_info 'Bootstrapping System with Ansible'
+ PROC_NAME='- [ Bootstrapping System with Ansible ] -'
+ printf '\n%s%s\n' '- [ Bootstrapping System with Ansible ] -' -------
- [ Bootstrapping System with Ansible ] -------
+ echo -------
-------
++ pwd
+ export OSA_CLONE_
+ OSA_CLONE_
++ readlink -f ansible-
+ ANSIBLE_
++ readlink -f inventory
+ OSA_INVENTORY_
++ readlink -f playbooks
+ OSA_PLAYBOOK_
+ ssh_key_create
+ key_path=/root/.ssh
+ key_file=
+ '[' '!' -d /root/.ssh ']'
+ '[' '!' -f /root/.ssh/id_rsa -o '!' -f /root/.
++ cat /root/.
+ key_content=
+ grep -q 'ssh-rsa ' /root/.
+ determine_distro
+ source /etc/os-release
+ export DISTRO_ID=centos
+ DIS...
Dmitriy Rabotyagov (noonedeadpunk) wrote : | #23 |
Wait, it seems you're currently on 20.0.1 and not 22.0.1. 20.0.1 is train release which does not have support of the CentOS 8.
You can join us in IRC on Freenode at #openstack-ansible for further community support.
phonix6 (phonix6-deactivatedaccount-deactivatedaccount) wrote : | #24 |
thx,
oops i didn't noticed. i changed to 22.0.1.
atal: [infra1_
fatal: [infra1_
phonix6 (phonix6-deactivatedaccount-deactivatedaccount) wrote : | #25 |
PLAY RECAP *******
compute1 : ok=36 changed=21 unreachable=0 failed=0 skipped=7 rescued=0 ignored=0
infra1 : ok=129 changed=67 unreachable=0 failed=0 skipped=12 rescued=0 ignored=0
infra1_
infra1_
infra1_
infra1_
infra1_
infra1_
infra1_
infra1_
infra1_
infra1_
infra1_
infra1_
infra1_
storage1 : ok=36 changed=21 unreachable=0 failed=0 skipped=7 rescued=0 ignored
phonix6 (phonix6-deactivatedaccount-deactivatedaccount) wrote : | #26 |
I think I understand what is wrong here.
I am running on ESXI hypervisor, the nics I have are as follows:
ens160 mgmt
ens192 vxlan
ens224 storage
I also Setup the bridges like the guide states.
The only difference is that I do not use Vlan interfaces at all e.g ens160.10, ens192.20, and so on.
In my case I have a virtual switch that takes care of the vlan tagging.
so I do VST (virtual switch tagging) and not VGT (virtual guest tagging).
I think strongly believe that the ansible book are failing due to this.
phonix6 (phonix6-deactivatedaccount-deactivatedaccount) wrote : | #27 |
I have setup the interfaces to tagging as guided, but unfortunately still the same error:
fatal: [infra1_
fatal: [infra1_
"rc": 1}
fatal: [infra1_
fatal: [infra1_
phonix6 (phonix6-deactivatedaccount-deactivatedaccount) wrote : | #28 |
here are the logs, it seems that some ova pics are not coming up:
lxc-start infra1_
lxc-start infra1_
lxc-start infra1_
lxc-start infra1_
lxc-start infra1_
lxc-start infra1_
lxc-start infra1_
lxc-start infra1_
lxc-start infra1_
lxc-start infra1_
lxc-start infra1_
lxc-start infra1_
lxc-start infra1_
lxc-start infra1_
lxc-start infra1_
lxc-start infra1_
lxc-start infra1_
lxc-start infra1_
lxc-start infra1_
OpenStack Infra (hudson-openstack) wrote : Fix included in openstack/openstack-ansible 23.0.0.0b1 | #29 |
This issue was fixed in the openstack/
CentOS 8.3 is not actually compatible with 8.2 as they have different repo names and some other things, that made us drop 8.2 support in favor of 8.3. However without actually errors it's hard to understand if it is the case for you.
However if you followed the guide, dnf upgrade should have brought you 8.3 anyway.
So please provide us with the actual errors you've faced during deployment.