3

我尝试使用“cap deploy:migrations”进行部署，因为我有 3 个需要在实时服务器上执行的新迁移。结果是：

executing "cd /home/martin/public_html/project/releases/20110905131238; bundle exec rake RAILS_ENV=production  db:migrate"
    servers: ["50.56.82.190"]
    [50.56.82.190] executing command
 ** [out :: 50.56.82.190] ==  CreateUsers: migrating ====================================================
 ** [out :: 50.56.82.190] -- create_table(:users)
 ** [out :: 50.56.82.190] rake aborted!
 ** [out :: 50.56.82.190] An error has occurred, all later migrations canceled:
 ** [out :: 50.56.82.190] 
 ** [out :: 50.56.82.190] Mysql2::Error: Table 'users' already exists: CREATE TABLE `users` (`id` int(11) DEFAULT NULL auto_increment PRIMARY KEY, `name` varchar(255), `email` varchar(255), `created_at` datetime, `updated_at` datetime) ENGINE=InnoDB
 ** [out :: 50.56.82.190] 
 ** [out :: 50.56.82.190] Tasks: TOP => db:migrate
 ** [out :: 50.56.82.190] (See full trace by running task with --trace)
    command finished in 5816ms
failed: "sh -c 'cd /home/martin/public_html/project/releases/20110905131238; bundle exec rake RAILS_ENV=production  db:migrate'" on 50.56.82.190

通过以前的部署，我在实时服务器上创建了一个工作数据库。我创建了 3 个新的迁移（由于使用新的 gemscarrierwave 和 rmagick 上传图像），我现在想部署它们。显然 cap deploy:migrations 尝试从第一个迁移运行所有迁移，而不仅仅是待处理的迁移。

我的 deploy.rb 是：

require 'bundler/capistrano'
set :application, "otg.in"
set :domain, "otg.in"
set :user, "martin"
set :sudo_use, false
set :repository, "git@github.com:Martin118/otg.in.git"
set :local_repository,  '~/rails_projects/otg.in/.git'
set :port, 48000
set :deploy_to, "/home/martin/public_html/#{application}"
set :scm, :git
set :branch, "master"
default_run_options[:pty] = true
ssh_options[:forward_agent] = true

server "50.56.82.190", :app, :web, :db, :primary => true

after "deploy", "deploy:bundle_gems"
after "deploy:bundle_gems", "deploy:restart"
after "deploy:update_code", "deploy:migrate"
after "deploy", "deploy:cleanup"

# Passenger
namespace :deploy do
     task :bundle_gems do
    run "cd #{deploy_to}/current && bundle install vendor/gems"
  end
  task :start do ; end
  task :stop do ; end
  task :restart, :roles => :app, :except => { :no_release => true } do
    run "#{try_sudo} touch #{File.join(current_path,'tmp','restart.txt')}"
  end
end

require 'whenever/capistrano'

我运行 rails 3.0.8，cap deploy:check 给我“你似乎已经安装了所有必要的依赖项”。

有任何想法吗？或者您需要更多信息？

谢谢你的帮助！

现场直播的 schema.rb 是：

ActiveRecord::Schema.define(:version => 20110130000344) do

  create_table "businesses", :force => true do |t|
    t.string   "business_name"
    t.string   "postal_code"
    t.string   "business_email"
    t.string   "phone"
    t.string   "fax"
    t.string   "web"
    t.text     "description"
    t.datetime "created_at"
    t.datetime "updated_at"
    t.integer  "user_id"
    t.text     "address"
    t.integer  "city_id"
    t.integer  "state_id"
  end

  create_table "cities", :force => true do |t|
    t.string   "name"
    t.datetime "created_at"
    t.datetime "updated_at"
    t.integer  "state_id"
  end

  create_table "states", :force => true do |t|
    t.datetime "created_at"
    t.datetime "updated_at"
    t.string   "name"
  end

  create_table "users", :force => true do |t|
    t.string   "name"
    t.string   "email"
    t.datetime "created_at"
    t.datetime "updated_at"
    t.string   "encrypted_password"
    t.string   "salt"
    t.boolean  "admin",              :default => false
  end

  add_index "users", ["email"], :name => "index_users_on_email", :unique => true

end

另一个更新：select * from schema_migrations; 在我的开发数据库上给了我：

+----------------+ | 版本 | +----------------+ | 20110112163009 | | 20110113141953 | | 20110113192958 | | 20110114214158 | | 20110115002206 | | 20110119100832 | | 20110120134443 | | 20110127171331 | | 20110127171427 | | 20110127171921 | | 20110127172903 | | 20110127183252 | | 20110129201949 | | 20110129204159 | | 20110129205833 | | 20110130000344 | | 20110808142844 | | 20110809133339 | | 20110809142303 | | 20110809154349 | | 20110810092306 | | 20110810093531 | | 20110812085010 | +----------------+

但是用我得到的 MySQL 工作台检查这个（仍然是开发数据库）：

错误：project_development。schema_migrations: 表数据不可编辑，因为没有为表定义主键

此外，生产数据库上的 select * from schema_migrations 只给了我一次迁移（我尝试更新之前的最后一次）：

+----------------+ | 版本 | +----------------+ | 0 | | 20110130000344 | +----------------+

有谁知道这里发生了什么？

谢谢！

4

1 回答 1

3

解决方案是手动将缺少的迁移添加到 schema_migrations 中。不知何故，这张表似乎丢失了一些数据。无论如何，通过添加已经使用“INSERT into schema_migrations (version) values ('2011 ....');”直接完成的迁移，然后 cap deploy:migrations 就从新的迁移开始顺利运行。

于 2011-09-07T14:02:10.317 回答