index As shown in the article below, I have summarized as a personal reminder what I stumbled upon when deploying a Rails app to AWS.
(Preparation) The world's most polite AWS commentary. Until you bring your Rails app to AWS using EC2 https://qiita.com/naoki_mochizuki/items/f795fe3e661a3349a7ce https://qiita.com/naoki_mochizuki/items/22cfbf4bf7ec95f6ac1c https://qiita.com/naoki_mochizuki/items/814e0979217b1a25aa3e https://qiita.com/naoki_mochizuki/items/5a1757d222806cbe0cd1
When I set up RDS and click "Create Database", this error occurs.
DB Subnet Group doesn't meet availability zone coverage requirement. Please add subnets to cover at least 2 availability zones. Current coverage: 1 (Service: AmazonRDS; Status Code: 400; Error Code: DBSubnetGroupDoesNotCoverEnoughAZs; Request ID: 3e87202c-e6b3-46dc-8396-47c64a2f0dd6)
I referred to this article. https://www.wantanblog.com/entry/2019/09/24/225020 Apparently, it seems that the problem was that only one subnet group was created, so if you create another subnet with the following settings from "VPC" → "Subnet" → "Create subnet", you can create an RDS instance. It became so. ・ VPC ・ ・ ・ Select the created VPC ・ Availability zone ・ ・ ・ Specify a location different from the subnet you have already created. ・ IPv4 CIDR block ・ ・ ・ 10.0.1.0/24
*[ .ssh ] $: ssh -i mumu.pem [email protected]
When I tried to log in to EC2 with, I got a connection error after waiting for a while. I referred to this article. https://xn--o9j8h1c9hb5756dt0ua226amc1a.com/?p=3583 When I checked the route table of the subnet where EC2 is located, the route to the outside (destination 0.0.0.0/0, target igw -...) was not set (inadvertently). After adding the following in "VPC"-> "Route table"-> "Check the corresponding route table"-> "Route table"-> "Edit route table"-> "Add route", SSH connection to EC2 becomes possible. I did. ・ Destination ・ ・ ・ 0.0.0.0/0 ・ Target ・ ・ ・ igw -... (created internet gateway)
It was a typo. It was ssh-keygen instead of ssh-key gem.
The following error occurs
$ rake secret
You must use Bundler 2 or greater with this lockfile.
I referred to this article. https://programming-beginner-zeroichi.jp/articles/169
$ gem install bundler
$ bundle install
$ bundle exec rake secret
I solved it with. (* By the way, since I set the database of the application to sqlite, an error occurred in "bundle install", and after following the steps in the next article, I did "bundle install".
I'm planning to use MySQL as a database, so I'll change the rails app database from sqlite to MySQL. I referred to this article. https://note.com/itoa06/n/n31fe4f9cd6b9
Click here for the difference /Gemfile
-gem 'sqlite3', '~> 1.4'
+gem 'mysql2', '>= 0.4.4'
/config/database.yml
default: &default
- adapter: sqlite3
+ adapter: mysql2
+ encoding: utf8mb4
pool: <%= ENV.fetch("RAILS_MAX_THREADS") { 5 } %>
- timeout: 5000
+ username: root
+ password:
+ host: localhost
development:
<<: *default
- database: db/development.sqlite3
+ database: hello_rails_development
test:
<<: *default
- database: db/test.sqlite3
+ database: hello_rails_test
production:
<<: *default
- database: db/production.sqlite3
+ database: hello_rails_production
+ username: hello_rails
+ password: <%= ENV['HELLO_RAILS_DATABASE_PASSWORD'] %>
Failed to start mysqld.service: Unit not found. I got an error when I tried to launch MySQL
sudo service mysqld start
Redirecting to /bin/systemctl start mysqld.service
Failed to start mysqld.service: Unit not found.
I referred to this article. https://qiita.com/hamham/items/fd77bb0bb167a150dc8e#mysql57%E3%81%AE%E5%B0%8E%E5%85%A5
As @MurakamiKazutaka wrote in the comment, it seems that Amazon Linux 2 will try to install MariaDB when trying to install mysql with yum.
$ yum -y install http://dev.mysql.com/get/mysql57-community-release-el7-11.noarch.rpm
$ yum -y install mysql mysql-community-server
$ mysqld --version
mysqld Ver 5.7.23 for Linux on x86_64 (MySQL Community Server (GPL))
$ cd /var/www/rails/app name
$ sudo service mysqld start
It was solved safely.
I thought it wasn't safe. In this case, MySQL will create the root user password on its own, so you need to put the password in database.yml, as @hat_log said. https://qiita.com/Dough/items/7493ad374a51b24abb58
$ sudo cat /var/log/mysqld.log | grep 'temporary password'
[Note] A temporary password is generated for root@localhost: XXXXXX
$ mysql -u root -p
Enter password: XXXXXX
mysql> set password for root@localhost=password('passwordPASSWORD@999');
Next, enter the password in database.yml
production:
<<: *default
database: mumu_production
username: root
password: passwordPASSWORD@999
An error occurred while creating the database.
$ rake db:create RAILS_ENV=production
...
NoMethodError (undefined method `deep_symbolize_keys' for...
I wondered if the .yml file was malformed and probably didn't have two indented spaces. Then I didn't write the key name secret_key_base: in config / secrets.yml (inadvertently).
Wrong ↓
production:
(Generated secret key)
Correct one ↓
production:
secret_key_base:(Generated secret key)
I get an error when trying to start Nginx
$ sudo service nginx start
Redirecting to /bin/systemctl start nginx.service
Job for nginx.service failed because the control process exited with error code. See "systemctl status nginx.service" and "journalctl -xe" for details.
I referred to this article. https://qiita.com/shota0701nemoto/items/a6929ef6f396cf3bede4
$ sudo nginx -t
nginx: the configuration file /etc/nginx/nginx.conf syntax is ok
nginx: [emerg] open() "/var/www/rails/[Wrong app name]/log/nginx.error.log" failed (2: No such file or directory)
nginx: configuration file /etc/nginx/nginx.conf test failed
I think the content of the error varies from person to person. I was told that "/var/www/rails/[wrong app name] /log/nginx.error.log" does not exist, so when I moved to the file,
$ cd /var/www/rails/[Wrong app name]/log
-bash: cd: /var/www/rails/[Wrong app name]/log: No such file or directory
When I looked closely, I noticed that I made a mistake in writing the hyphen (-) and underscore (_) in the app name.
$ vim config/unicorn.conf.rb
$ cd /etc/nginx/conf.d/
$ sudo vim mumu.conf
Corrected the wrong part in.
$ cd /var/www/rails/[app name]/
$ sudo service nginx start
It started up safely.
After starting Nginx, when I try to access the IP address of EC2 with Chrome,
We're sorry, but something went wrong.
It became. I thought it was because I stopped RDS once,
$ sudo service mysqld start
I started MySQL, but there is no change.
$ less log/production.log
I checked the Rails log in (the bottom is the latest information),
ActionView::Template::Error (The asset "application.css" is not present in the asset pipeline.
Since the error was displayed, rewrite config / envitonments / production.rb referring to this article. https://kanoe.studio/archives/791
# Do not fallback to assets pipeline if a precompiled asset is missed.
config.assets.compile = true
Restart the application server, (I referred to this article https://qiita.com/takuyanagai0213/items/259ca105e35f6eb066d6 ）
$ ps -ef | grep unicorn | grep -v grep
takuya 2460 1 0 March 11? 00:00:04 unicorn_rails master -c /var/www/rails/myapp/config/unicorn.conf.rb -D -E production
takuya 2465 2460 0 March 11? 00:00:05 unicorn_rails worker[0] -c /var/www/rails/myapp/config/unicorn.conf.rb -D -E production
takuya 2467 2460 0 March 11? 00:00:04 unicorn_rails worker[1] -c /var/www/rails/myapp/config/unicorn.conf.rb -D -E production
$ kill 2460
$ unicorn_rails -c /var/www/rails/myapp(My app name)/config/unicorn.conf.rb -D -E production
I accessed the EC2 IP address again, but as usual, "We're sorry, but something went wrong." Continued (Is it still an error ...)
When I checked the log again,
$ less log/production.log
ActionView::Template::Error (Webpacker can't find application in /var/www/rails/hello-rails/public/packs/manifest.json. Possible causes:
1. You want to set webpacker.yml value of compile to true for your environment
unless you are using the `webpack -w` or the webpack-dev-server.
2. webpack has not yet re-run to reflect updates.
3. You have misconfigured Webpacker's config/webpacker.yml file.
4. Your webpack configuration is not creating a manifest.
Error was occurring.
I don't know what you're talking about
I referred to this article. https://qiita.com/natecotus/items/a2bd9f3ebd5b1866d48e
$ rm -rf bin/webpack*
$ rails webpacker:install
Webpacker requires Node.js >= 8.16.0 and you are using 6.17.1
Please upgrade Node.js https://nodejs.org/en/download/
An error occurred again. It seems that the version of Node.js is old. I referred to this article. https://qiita.com/paranishian/items/bddaed7c3aacedb11967
$ git clone git://github.com/creationix/nvm.git .nvm
$ . ~/.nvm/nvm.sh
$ nvm install
$ curl -o- -L https://yarnpkg.com/install.sh | bash
$ source ~/.bashrc
$ yarn -v
Well then, let's change it ~
$ rails webpacker:install
[Ynaqdhm] Y
[Ynaqdhm] Y
$ RAILS_ENV=production bundle exec rails webpacker:compile
It looks like it's compiled, so restart Unicorn and Nginx.
$ ps -ef | grep unicorn | grep -v grep
$ kill [Process ID]
$ unicorn_rails -c /var/www/rails/[app name]/config/unicorn.conf.rb -D -E production
$ sudo nginx -s reload
Now, try connecting with an IP address. .. .. It was displayed!
Recommended Posts