Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

install for content-modeing-overhaul branches of everything #66

Merged

Conversation

dannylamb
Copy link
Member

This PR is for installing all the required changes for testing Islandora/documentation#86.

A lot of it is commented out becaues it depends on code being in packagist, so once everything is tested and PRs are merged, we'll comment it back in. In the meantime, I have a gist that bridges the gap until then.

@whikloj
Copy link
Member

whikloj commented May 7, 2018

Not sure if this is due to the vagrant 2.0.1 versus 2.0.3 differences, but vagrant and ansible are not able to login with this PR.

default: SSH address: 127.0.0.1:2200
    default: SSH username: ubuntu
    default: SSH auth method: password
    default: Warning: Connection reset. Retrying...
    default: Warning: Remote connection disconnect. Retrying...
    default: Warning: Connection reset. Retrying...
    default: Warning: Remote connection disconnect. Retrying...
    default: Warning: Connection reset. Retrying...
    default: Warning: Authentication failure. Retrying...
    default: Warning: Connection reset. Retrying...
    default: 

Then it seems to continue building the box, but then once it is Ansible time...

TASK [install python] **********************************************************
Monday 07 May 2018  12:57:29 -0500 (0:00:00.050)       0:00:00.050 ************ 
fatal: [default]: UNREACHABLE! => {"changed": false, "msg": "Failed to connect to the host via ssh: Warning: Permanently added '[127.0.0.1]:2222' (ECDSA) to the list of known hosts.\r\[email protected]: Permission denied (publickey,password).\r\n", "unreachable": true}
	to retry, use: --limit @/sw/var/www/DAM2/claw-playbook/playbook.retry

I can't run the ansible-playbook command manually as it has the same problem.

I note that the port has changed from :2222 to :2200, maybe that is the problem? But I have no idea.

@DiegoPino
Copy link
Contributor

Also 2.1.0 is out, maybe that is a fix version, worth testing out. It is also worth resetting all the environment vars/caches, a.k.a vagrant destroy && vagrant up when you update since many of the plugins and cached sessions could be outdated?

@dannylamb
Copy link
Member Author

@whikloj Definitely should check that port.

I noticed this morning I was getting the same 'Connection reset' loop when vagrant ssh'ing into a box I built last week. I just gave it a while and it eventually sorted itself out. 🤷‍♂️

@whikloj
Copy link
Member

whikloj commented May 7, 2018

Yeah it is the hosts file, need to update the port there. As this seems to be a vagrant version thing, we should make people aware but unless we want to make vagrant 2.0.3 the minimum version I don't know if there is anything to do.

@whikloj
Copy link
Member

whikloj commented Jun 5, 2018

@dannylamb I'm going to hold on to this until packagist is ready and then you can uncomment those necessary lines.

@dannylamb
Copy link
Member Author

ok, i think that's all the PRs associated with this issue. let me update this PR.

@ajs6f
Copy link
Contributor

ajs6f commented Jun 6, 2018

@dannylamb Once this PR is updated, would it be reasonable to try a full install (using this Ansible gear)?

@dannylamb
Copy link
Member Author

@ajs6f Yes, it should work for ubuntu with both mysql and pgsql. I've hit a minor hurdle releasing Alpaca so the artifact is available to the installer, but once that gets sorted out it should work np.

@ajs6f
Copy link
Contributor

ajs6f commented Jun 6, 2018

@dannylamb Okay, cool, I'll take it out for my usual high-speed low -skill test drive on CentOS once I see a release of Alpaca (as a snapshot to Maven, I assume?).

# Comment this back in once code gets merged and is available on packagist
# - "islandora/islandora:dev-8.x-1.x"
# - "islandora/openseadragon:dev-8.x-1.x"
# - "islandora/islandora_image:dev-8.x-1.x"
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@dannylamb I think we are up-to-date in packagist. These can be uncommented now.

# Comment this back in once code gets merged and is available on packagist
# - islandora_feature
# - islandora_openseadragon
# - islandora_image_feature
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Ditto

@whikloj
Copy link
Member

whikloj commented Jun 12, 2018

Haha I think the comment

We can fix #66

Made Github confused.

@whikloj whikloj reopened this Jun 12, 2018
#- name: Import features
# command: "{{ drush_path }} -y fim --bundle=islandora islandora_feature,islandora_openseadragon,islandora_image_feature"
# args:
# chdir: "{{ drupal_core_path }}"
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Should/Can this be re-enabled?

#- name: Index tags
# command: "{{ drush_path }} -y php:eval 'array_walk(\\Drupal::service('entity_type.manager')->getStorage('taxonomy_term')->loadMultiple(), function($term) { $term->save(); });'"
# args:
# chdir: "{{ drupal_core_path }}"
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Ditto re: re-enabling.

@dannylamb
Copy link
Member Author

Sweet. We're end of day over here, but I'll try and push up the actual changes ASAP so we can test.

@dannylamb
Copy link
Member Author

vagrant up 🚀

@dannylamb
Copy link
Member Author

Ok, so I'm updating this to make the migrations go at the very end. Everything needs to be in place so stuff can get indexed appropriately, etc...

@@ -1,5 +1,6 @@
---

apache_listen_ip: localhost
Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Adding this somehow messes with the virtual host file and now root is /var/www/html instead of /var/www/html/drupal/web

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Likely a geerlingguy artefact. Any reason why we want it there in the first place? Not having it in the past hasn't caused any trouble that I'm aware of.

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I think this will come up fine if we change this to:

apache_listen_ip: "*"

which is the default, so we could probably not set it as well.

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I stand corrected, we do need the variable set for this PR. I'm testing @jonathangreen 's suggestion now.

@whikloj
Copy link
Member

whikloj commented Jun 19, 2018 via email

@seth-shaw-unlv
Copy link
Member

seth-shaw-unlv commented Jun 19, 2018

Most recent run using apache_listen_ip set to "*" resulted in the following error:

TASK [Run migrations] **********************************************************
Tuesday 19 June 2018  15:51:41 -0700 (0:00:00.537)       0:11:34.810 ********** 
fatal: [default]: FAILED! => {"changed": true, "cmd": ["/var/www/html/drupal/vendor/drush/drush/drush", "-y", "-l", "*:8000", "mim", "--group=islandora"], "delta": "0:00:00.100176", "end": "2018-06-19 22:49:49.238597", "msg": "non-zero return code", "rc": 1, "start": "2018-06-19 22:49:49.138421", "stderr": "In DrupalKernel.php line 369:[Symfony\\Component\\HttpKernel\\Exception\\BadRequestHttpException]", "stderr_lines": ["", "In DrupalKernel.php line 369:", "", "  [Symfony\\Component\\HttpKernel\\Exception\\BadRequestHttpException]  ", ""], "stdout": "", "stdout_lines": []}

@dannylamb
Copy link
Member Author

Yeah, using * blows up the migrations. That's where I'm stuck. I don't want to hard code localhost or 127.0.0.1 into the task, but can't seem to get the vhost to comply. At least not without postprocessing...

I may just add another task to the post-install, but that just seems layering more crap on top of something already not quite working right.

@seth-shaw-unlv
Copy link
Member

Well, we should be able to use the migrate command without the uri switch, as it is theoretically only necessary for multi-site installations. (Can we make that assumption with the playbook?)

I can't test removing the option this morning (out of the office and won't make today's call) but I can look into it this afternoon if you don't beat me to it.

@seth-shaw-unlv
Copy link
Member

Or, could we use apache_vhosts.servername instead of apache_listen_ip for the migrate command?

I need to stop thinking about this away from work and my dev env...

@jonathangreen
Copy link
Contributor

This is so close. We need 2 changes in the requirements.yml:
https://github.com/Islandora-Devops/claw-playbook/blob/master/requirements.yml#L44-L46
Needs to be

- src: https://github.com/Islandora-Devops/ansible-role-alpaca
  name: Islandora-Devops.alpaca
  version: 0.0.2

and
https://github.com/Islandora-Devops/claw-playbook/blob/master/requirements.yml#L60-L62
needs to be:

- src: https://github.com/Islandora-Devops/ansible-role-crayfish
  name: Islandora-Devops.crayfish
  version: 0.0.2

So we pull in the correct requirements from crayfish and alpaca.

@jonathangreen
Copy link
Contributor

jonathangreen commented Jun 20, 2018

vagrant up 🚀

lets see this thing go!

@seth-shaw-unlv
Copy link
Member

It's alive!

Note: I just ran it using CentOS which requires a few version updates (ansible-role-crayfish: master, ansible-role-grok: 0.0.2, and ansible-role-karaf: master).

@jonathangreen
Copy link
Contributor

I sliced a new version for ansible-role-karaf so we can update to 0.0.2 for cent.

@seth-shaw-unlv
Copy link
Member

@jonathangreen thanks. Could you also merge islandora-deprecated/ansible-role-crayfish/pull/7 and cut a new ansible-role-crayfish version, too, while you're at it?

@Natkeeran
Copy link
Contributor

Natkeeran commented Jun 20, 2018

Fired up into a VM and seems good (i.e loads it at 8000, syncs to feodra and indexes in triplestore). 👍

Copy link
Contributor

@jonathangreen jonathangreen left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@jonathangreen
Copy link
Contributor

@seth-shaw-unlv I merged islandora-deprecated/ansible-role-crayfish#7 and sliced a release for 0.0.3. I figured I'd merge this PR since it was working for me on Ubuntu.

Want to make a new PR to update the roles as we need for Cent and we'll work on getting that merged?

@seth-shaw-unlv
Copy link
Member

Sure. I'm going to do it as part of Islandora/documentation/issues/809. I have a local branch with updated requirements I'm testing right now.

@dannylamb
Copy link
Member Author

Just spun up a fresh box and looks like islandora-connector-houdini is available in the feature but not installed in Karaf.

@dannylamb
Copy link
Member Author

Ok, I see what I need to do. In the meantime, just feature:install islandora-connector-houdini. I'll issue a PR shortly.

@dannylamb
Copy link
Member Author

D'oh... or just run ansible-galaxy install --force --role-file=requirements.yml --roles-path=roles/external beforehand...

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

7 participants