Some devs use NFS mount in vagrant.
Without NFS software they get the following:

The following SSH command responded with a non-zero exit status.
Vagrant assumes that this means the command failed!

mount -o 'vers=3,udp' 10.11.12.1:'/Users/chloelee/DoSomething.org/phoenix' /var/www/dev.dosomething.org

Stdout from the command:



Stderr from the command:

mount.nfs: requested NFS version or transport protocol is not supported

Running sudo apt-get install nfs-common nfs-kernel-server seems to fix the issue.

So we have THE place to look for drupal errors instead of shared /var/log/syslog.

Here's the docs:
https://www.drupal.org/documentation/modules/syslog

cc @sheyd

https://jenkins.dosomething.org/view/Ansible/job/Ansible%20Job%20Run/98/console

TASK [dosomething.search : Checkout default solr core] *************************
task path: /mnt/data/jobs/Ansible Repo Checkout/workspace/roles/dosomething.search/tasks/solr-install-core.yml:3
fatal: [10.100.65.42]: FAILED! => {
    "failed": true, 
    "msg": "'solr_cores' is undefined"
}

when sharing from https://www.dosomething.org/reportback/80947?fid=110809 a different user's image is the one that's actually getting shared on social

on https://www.dosomething.org/campaigns "LGBT HBD", "Headhunters", and "Pineapple Push-Ups" have broken images even though new images exist on campaign page

on non-AWS webhead only Headhunters was broken

Permissions on /var/www/www.dosomething.org/current and /var/www/www.dosomething.org/shared is dosomething:root and 755, when it should be dosomething:dosomething and 775.

Security improvement as discussed with @sheydari-ds.

See issue, here:

#100

Essentially, files directory has a symlink to /var/www/www.dosomething.org in the assets directory, and will break file pathing if site is altered from that.

At the moment we have to add public key into authorized_keys and rebuild Vagrant box each time new dev joins us. This is not optimal.
This approach with SSH keys has been chosen due to a vagrant bug fixed in hashicorp/vagrant#5765.

We need to revisit this issue and find a way to log into non-default ssh user without provisioning it with the authorized keys.

Discovered testing another change

fatal: [default]: FAILED! => {"changed": false, "failed": true, "msg": "Failed to validate the SSL certificate for deb.nodesource.com:443. Make sure your managed systems have a valid CA certificate installed. If the website serving the url uses SNI you need python >= 2.7.9 on your managed machine or you can install the `urllib3`, `pyopenssl`, `ndg-httpsclient`, and `pyasn1` python modules to perform SNI verification in python >= 2.6. You can use validate_certs=False if you do not need to confirm the servers identity but this is unsafe and not recommended. Paths checked for this platform: /etc/ssl/certs, /etc/pki/ca-trust/extracted/pem, /etc/pki/tls/certs, /usr/share/ca-certificates/cacert.org, /etc/ansible"}

Set expose_php = off.

happened previously on DoSomething/legacy-website#4337

https://www.dosomething.org/reportback/80947?fid=110809

As of 1.7.4, vagrant uses it's own key for all ssh connections, including remote_user ansible statement:
https://docs.vagrantup.com/v2/provisioning/ansible.html

Vagrant 1.7+ automatically inserts a different

insecure keypair for each new VM created. The easiest way

to use the same keypair for all the machines is to disable

this feature and rely on the legacy insecure key.

config.ssh.insert_key = false

Because auhorized_keys doesn't contain vagrant key (which is generated on the fly), first task with remote_user: "{{ app_user }}" fails to connect.

Requested by @DFurnes
Should fix https://jenkins.dosomething.org/job/Deploy-Staging/144/console

dosomething@ip-10-100-65-7:/var/www/staging.beta.dosomething.org/releases/20150723202759/lib/themes/dosomething/paraneue_dosomething$ npm install
npm ERR! 404 404 Not Found: dosomething
npm ERR! 404
npm ERR! 404 'dosomething' is not in the npm registry.
npm ERR! 404 You should bug the author to publish it

See https://cobalt.io/dosomething/reports/55.

Let's DoCument Something.

Gleb,

Please purchase one MacBook Pro power adapter per conference room, and one extra one for Morgan for when he departs to Atlanta.

Please purchase the higher wattage ones here:

http://www.apple.com/shop/product/MD506LL/A/apple-85w-magsafe-2-power-adapter-for-macbook-pro-with-retina-display?fnode=fc9e4c5a93d4ba6be78b2e2c99893b214d7de5fcbb976dc694fa4441642247efd8f2b4a608ac28e7103e66acc86e37ba41dba42751a1bc1fc6c1e6c155131912e0f26846e068a0c89fcee08bcc135f62e703f865f0a245e6c89a639303ec5e88f0488ae3fee7b03ca229e1caac472055

immediately after reporting back the user is taken to the permalink page. image is showing up as broken.

1. Regenerate these certs https://github.com/DoSomething/tower/tree/master/files/ssl
2. Put results to the vault to protect actual certificates

Todo draft:

• Setup shared folder, see https://github.com/DoSomething/dosomething-deploy/blob/master/config/tasks/default.cap#L29
• Fix status.php
• Mount files folder
• Fix login
• Fix search
• Check Redis
• Check message broker
• Setup Varnish access
• Install settings.production.php

CC @blisteringherb @sheydari-ds @deezone

changed: [default] => (item={'version': '1.1.0-beta3', 'name': 'casperjs'}) => {"changed": true, "item": {"name": "casperjs", "version": "1.1.0-beta3"}}

Requires both changes on staging RDS server and in ansible variables.
CC @sheydari-ds

Seems like ds.aliases overrides website url on international.
This leads to issue with solr indexing. I manually removed the file, but wee need permanent solution for this issue.

DS build composer task fails with:

Your github oauth token for github.com contains invalid characters: ""

Seems like new composer doesn't ignore empty auth string in auth.json.

See DoSomething/legacy-website#5667.
The reason is that here the dot char (.) is not escaped and represents ALL characters instead of itself:
location ~* ^.+.(jpg|jpeg|gif|css|png|js|ico|xml)$

https://github.com/dosomething/vision

I can set the application up (instructions are in the README regardless). It needs an HTTP port that it can bind too and a URL (vision.dosomething.org?) that Github can push data too + a Slackbot can make requests too.

It technically needs to run two applications (A slackbot and a web app) but they're very lightweight and won't need a lot of resources.

Let me know what else you guys need to know!