Rails-Heroku在本地计算机上的现有开发Postgres数据库

Question

I have developed a Rails 3.2.13 app with PostgreSQL development database on my Windows7 machine. I was able to push my app to heroku with success. However, now I'm stuck at the process of migrating the existing data to Heroku.

I have tried following along the documentation by Heroku, but I can't seem to get it to work.

Here's what I have done so far:

1. Installed the PG Backups heroku addon
2. Created a dump file for my development database with the command pg_dump -Fc --no-acl --no-owner -h localhost -U user myapp_development > myapp_development.dump
3. Uploaded the dump file to AWS S3

At this point, I am not sure how to proceed, to restore the dump file and turn it to a production database in Heroku. Can anyone please help?

PS I also have tried the heroku db:push command, but it gave the following error:

Failed to connect to database: Sequel::AdapterNotFound -> LoadError: cannot load such file -- pg

Answer 1

a) On S3, you need to set your dump file to be publicly readable.

Within the S3 bucket, right-click the dump file you uploaded, and click 'Make Public'

Now, when you click the 'Properties' button (top-right area) for that file, you will see a 'Link' value. This means the file is now accessible from outside, including by Heroku.

b) You will need to run a command like this to restore this pg dump to your heroku database:

$ heroku pgbackups:restore DATABASE 'http://your.s3.domain.here.s3-us-xxxx-xx.amazonaws.com/myapp_development.dump' --app your-heroku-app-name

Notes:

• To run the 'heroku' command, you'll need to install the heroku toolbelt.

• Note that the http URL in step (b) will be the same public URL as the one you saw in 'Properties' in step (a) above.

• DATABASE is a heroku keyword for this command, and should be typed out as is. It equates to the DATABASE_URL config setting you see when you run:

$ heroku config --app your-heroku-app-name

Answer 2

Now that you have the uploaded dump file on S3, you need to tell pgbackups to load up that dump.

Try something like this:

$ heroku pgbackups:restore DATABASE_URL 'https://s3.amazonaws.com/mybucket/path/to/myapp_development.dump'

I found that command here: https://devcenter.heroku.com/articles/heroku-postgres-import-export

Edit:

DATABASE_URL should work verbatim. But if you would like to specify the database, you can use heroku pg:info to find out the name of your database.

For me, heroku pg:info prints:

=== HEROKU_POSTGRESQL_COPPER_URL (DATABASE_URL)
Plan:        Basic
Status:      available
Connections: 1
PG Version:  9.1.9
Created:     2013-03-06 17:21 UTC
Data Size:   9.5 MB
Tables:      9
Fork/Follow: Unsupported

So, in my case I would use:

$ heroku pgbackups:restore HEROKU_POSTGRESQL_COPPER_URL 'https://s3.amazonaws.com/mybucket/path/to/myapp_development.dump'

Answer 3

Make sure all gems are install on your system.

sudo apt-get install sqlite3 libsqlite3-dev
sudo gem install sqlite3-ruby

gem install heroku taps sequel

sudo gem install tabs

sudo heroku db:push

OR

sudo heroku db:push –config herokuServerName

If still problem. Use below command

heroku db:pull postgres://root:@localhost/db_name

Else

sudo heroku db:pull --confirm herokuDomanName postgres://userName:@localhost/Locadatabase

root will be database user name and db_name is database name.

Good Luck.........!!