I’m installing on a Ubuntu 18 Google Cloud with (i believe) all the prerequisite met. It seemed to have gone well to the end when it threw up these messages below. I will appreciate any directions on what when wrong and how i can resolve it.
Thanks
" **files and directories currently installed**.)", "Preparing to unpack .../resolvconf_1.78ubuntu7_all
.deb ...", "Unpacking resolvconf (1.78ubuntu7) ...", "Processing triggers for systemd (229-4ubuntu17) ...", "Settin
g up resolvconf (1.78ubuntu7) ...", "ln: cannot remove '/etc/resolv.conf': Device or resource busy", "dpkg: error p
rocessing package resolvconf (--configure):", " subprocess installed post-installation script returned error exit s
tatus 1", "Processing triggers for systemd (229-4ubuntu17) ...", "Errors were encountered while processing:", " res
olvconf"]}
**to retry, use: --limit @/edx/app/edx_ansible/edx_ansible/playbooks/demo.retry**
PLAY RECAP ********************************************************************************************************
127.0.0.1 : ok=17 changed=3 unreachable=0 failed=1
INFO:ansible.plugins.callback./edx/app/edx_ansible/edx_ansible/playbooks/callback_plugins/task_timing: Install role
-independent useful system packages --------------------------------- 9.11s
INFO:ansible.plugins.callback./edx/app/edx_ansible/edx_ansible/playbooks/callback_plugins/task_timing: Install ubun
tu system packages -------------------------------------------------- 6.49s
INFO:ansible.plugins.callback./edx/app/edx_ansible/edx_ansible/playbooks/callback_plugins/task_timing: Gathering Fa
cts ----------------------------------------------------------------- 1.07s
INFO:ansible.plugins.callback./edx/app/edx_ansible/edx_ansible/playbooks/callback_plugins/task_timing: Update CA Ce
rtificates ---------------------------------------------------------- 0.93s
INFO:ansible.plugins.callback./edx/app/edx_ansible/edx_ansible/playbooks/callback_plugins/task_timing: Add common_u
sers ---------------------------------------------------------------- 0.79s
INFO:ansible.plugins.callback./edx/app/edx_ansible/edx_ansible/playbooks/callback_plugins/task_timing: Grant full s
udo access to the edxadmin group ------------------------------------ 0.77s
INFO:ansible.plugins.callback./edx/app/edx_ansible/edx_ansible/playbooks/callback_plugins/task_timing: Update expir
ed apt keys --------------------------------------------------------- 0.76s
INFO:ansible.plugins.callback./edx/app/edx_ansible/edx_ansible/playbooks/callback_plugins/task_timing: Add edX PPA
apt key ------------------------------------------------------------- 0.71s
INFO:ansible.plugins.callback./edx/app/edx_ansible/edx_ansible/playbooks/callback_plugins/task_timing: Add git apt
repository ---------------------------------------------------------- 0.63s
INFO:ansible.plugins.callback./edx/app/edx_ansible/edx_ansible/playbooks/callback_plugins/task_timing: Check for ex
pired edx key ------------------------------------------------------- 0.50s
INFO:ansible.plugins.callback./edx/app/edx_ansible/edx_ansible/playbooks/callback_plugins/task_timing:
Playbook Deploy demo course finished: 2019-11-27 20:23:37.425273, 50 total tasks. 0:00:26.166122 elapsed.
Makefile:59: recipe for target 'dev.provision.run' failed
make: *** [dev.provision.run] Error 2"
Hello @startoffs, and welcome to the Open edX community!
Can you provider further information on how did you get to this point? This will help other understand what you are trying to achieve and better help you.
For instance, to provision an ironwood.master devstack, I’d usually:
$ cd devstack
$ git checkout open-release/ironwood.master && git pull
$ export OPENEDX_RELEASE=ironwood.master
$ make dev.clone
$ make pull
$ make dev.provision.run
@startoffs, can you try manually running the command corresponding to the task that failed? I am unable to see the details of the task in the output you have shared.
Also, are you using a vanilla Ubuntu 16.04 image on GCP?
A quick search on the internet for your issue lead me to a StackOverflow post and a Docker bug. Can you check those out and see if any of the possible solutions posted there work for you?
@startoffs The issue was likely introduced in this PR which was merged a few days ago. If you can pin your configuration repo to a revision before that PR was merged, you should be fine.
There was a small issue in one of the docker images, which has already been fixed. You might want to try this process again by running:
$ make pull && make dev.provision
I’m just curious on what’s your use case here. The devstack is commonly used for development and we don’t see a lot of remote deployments. Do you mean to have a native installation?