For a database-per-service architecture, how do you guys manage your static data for each microservice? I want to make it easy for a new developer to jump in and get everything up and running easily on their local machine. I'm thinking of checking the entire database with static data into source control with Docker bind mounts so people can just docker-compose up the database service locally (along with whatever other infrastructure services they might need to run and test their microservice).
I know each microservice might need to handle this in their own way, but I'd like to provide a good default template for people to start with.
Making a standard for how to do this sort of goes against the reason for making microservices, i.e. that you can adapt each microservice to the context it exists in.
That being said, Postgres, Mongo and MySQL all run scripts in /docker-entrypoint-initdb.d when initializing a fresh database instance. The scripts have to fit the database obviously, but it's a fairly standardized way of doing it.
They all have descriptions of how to do it on the image page on docker hub.
You can either get your scripts into the container by making a custom image that contains the scripts or you can map them into the directory using a docker-compose volume mapping.
There are some databases that don't have an easy way to initialize a new database. MSSQL comes to mind. In that case, you might have to handle it programmatically.
I'm using Django in a development process. It is annoying that every time I change a bit in a model I need to delete database and run syncdb. For the purpose of testing, I want to add some initial data into database automatically every time when I run syncdb. I've tried put these sort of code inside one app's __init__.py, but it would run before database created and it's a bit annoying to deal with exceptions. Isn't there a neater way to do this?
Once you have initially populated the database; use the dumpdata command to create a fixture (a copy of data). Save it to a file. Then use the loaddata command to automatically populate the database.
Suppose you have an app called bookstore for which you want to automatically load a series of books, authors, etc.
Once you have added some records in the database:
python django-admin.py dumpdata bookstore > initial.json
Once you have made some changes or want to recreate the database:
python django-admin.py loaddata initial.json
South is nice, but it is overkill for this purpose.
If you need to make changes in database and you want that your data remains safe then SOUTH is the right software for you.
It manages change in database and its data. You wont need to delete database when you make changes if you have South. Also it has backups of previous states of database so that if you want to revert back you can. I recommend you read its documentation. It will surely help.
Just wondering the best way to handle the following....
I want to have a VS2010 database project to keep the schema of my database in the dev, integration test and production environments in sync.
As part of the test and production environments I have a lot of reference data that needs to be loaded into the database.
For dev and test I can just recreate the database and use Post Deployment scripts to load the data. However, I cant really do this for the production environment as obviously it will have live data on it.
So what is the best solution to do this? I dont think I can use Post Deployment scripts to load the datbase, because in the case of an insert statement I would need to wrap each one inside an IF NOT EXISTS... clause and there are 1000's of rows.
Maybe its best to use the VS2010 + MSBuild tools to keep the schema up to date and then have a seperate solution for managing the data?
Or is there a solution to this that uses purely the tools in VS2010 + MSBuild?
The best solution for live production enviroment - not to use automatic updates at all!
Use very well tested hand made update scripts in touch with your backend and frontend applications update
And there is always a good idea to have a fresh backup
How about truncating and rebuilding the reference data table each time? If there are constraints you can remove them and add them back at the end of the post-deployment script. Would that work for you?
Or is there a reason why you can't remove production reference data?
For reference data you can have a script that handles an insert, update or delete depending if the data is already in the table or not.
Check out this link for more details (this also includes a generator to help you generate your scripts).
Use a populated database to generate merge statements that can be applied in Post-Deployment.
It might be a good idea to take out the DELETE clause though.
I'd like to know your approach/experiences when it's time to initially populate the Grails DB that will hold your app data. Assuming you have CSVs with data, is is "safer" to create a script (with whatever tool fits you) that:
1.-Generates the Bootstrap commands with the domain classes, run it in test or dev environment and then use the native db commands to export it to prod?
2.-Create the DB's insert script assuming GORM's version = 0 and incrementing manually the soon-to-be autogenerated IDs ?
My fear is that the second approach may lead to inconsistencies for hibernate will have the responsability for the IDs generation and there may be something else I'm missing.
Thanks in advance.
Take a look at this link. This allows you to run groovy scripts in the normal grails context giving you access to all grails features including GORM. I'm currently importing data from a legacy database and have found that writing a Groovy script using the Groovy SQL interface to pull out the data then putting that data in domain objects appears to be the easiest thing to do. Once you have the data imported you just use the commands specific to your database system to move that data to the production database.
Update:
Apparently the updated entry referenced from the blog entry I link to no longer exists. I was able to get this working using code at the following link which is also referenced in the comments.
http://pastie.org/180868
Finally it seems that the simplest solution is to consider that GORM as of the current release (1.2) uses a single sequence for all auto-generated ids. So considering this when creating whatever scripts you need (in the language of your preference) should suffice. I understand it's planned for 1.3 release that every table has its own sequence.
I've looked around the Wordpress forums about this and didn't find anything so I thought I might try here.
If you have a staging/dev Wordpress setup used for testing new pluging and such, how do you go about migrating the data in the staging database back to the production database? Is there a "Wordpress best practices" way to do this, or am I limited to having to manually migrate tables from one database to the other?
I have a script that mysqldumps a copy of my production Wordpress DB, restores it over my test Wordpress install & then corrects all the "production" settings & urls in the test DB.
Both my production & test databases live on the same server, but you could change the mysqldump settings to dump from a remote mysql server & restore to a local server quite easily.
Here are my scripts:
overwrite_test.coach_db_with_coache_db.sh
#!/bin/bash
dbUser="co*******"
dbPassword="*****"
dbSource="coach_production"
dbDest="coach_test"
tmpDumpFile="/tmp/$dbSource.sql"
mysqldump --add-drop-table --extended-insert --user=$dbUser --password=$dbPassword --routines --result-file=$tmpDumpFile $dbSource
mysql --user=$dbUser --password=$dbPassword $dbDest < $tmpDumpFile
mysql --user=$dbUser --password=$dbPassword $dbDest < /AdminScripts/change_coach_to_test.coach.sql
change_coach_to_test.coach.sql
-- Change all db references from #oldDomain to #newDomain
SET #oldDomain = 'coach.co.za';
SET #newDomain = 'test.coach.co.za';
SET #testUsersPassword = 'password';
UPDATE `wp_1_options` SET `option_value` = REPLACE(`option_value`,#oldDomain,#newDomain) WHERE `option_name` IN ('siteurl','home','fileupload_url');
UPDATE `wp_1_posts` SET `post_content` = REPLACE(`post_content`,#oldDomain,#newDomain);
UPDATE `wp_1_posts` SET `guid` = REPLACE(`guid`,#oldDomain,#newDomain);
UPDATE `wp_blogs` SET `domain` = #newDomain WHERE `domain` = #oldDomain;
UPDATE `wp_users` SET `user_pass` = MD5( #testUsersPassword );
-- Only valid for main wpmu site
UPDATE `wp_site` SET `domain` = #newDomain WHERE `domain` = #oldDomain;
Perhaps you are just looking for the wrong thing. Wouldn't a backup plugin handle this with ease? I know they exist for all the big CMS packages...
The two methods would be using the export/import feature under tools or copying the database. I email myself a copy of my production database weekly using the WordPress Database Backup plugin.
The import feature can be problematic for moving a wordpress blog as you have to configure your php.ini file often because the default value of files you can upload on a hosted php implementation tends to be too small by default.
I wanted to pull the database from my production wordpress website into an offline development copy of it on my desktop machine so I could modify the site and test it with a
full set of the existing blog content and history.
This proved to be problematic, as simply making an offline backup of the database and importing it into the local development database did not work.
Overcoming these problems in moving data from the production to the dev database can probably be used to go the other way as well - so I think you can just use these guidelines for what you want to do as well - just start with dev data and move it to prod.
The problems here were:
the permalink designations for the
blog posts are all stored in the
database as they would be for the
online version, but my offline copy
isn't at the domain address, instead
it is in the localhost directory. So
when I launch the site locally,
although the css formatting and
images are all in place (the image
links being relative), the actual
blog posts don't show up.
many of the links throughout the
site link back out to the internet,
so if you try to navigate to
archives, or comments, or
categories, or the main posts, you
get sent back out to the internet
instead of staying in the database
on the local machine.
To make sure I was doing this right, I blew away the wordpress install I had on my local machine and restarted from scratch.
Once I had a clean, new wordpress install and brand new default freshly created local database for it, I opened up the database in phpMyAdmin and took a look at the wp_posts
table. Inside there, each record (in other words, each post) has a column titled "guid", which shows the location of that post. For example, the first one in a fresh, default
install contains this "guid" value:
http://localhost/wordpress/?p=1
If you look in the wp_posts table of your online version, you'll see instead in this location the url to your site online.
You can't just import the tables wholesale into your local install, because you'll be importing all these outside references. It will make your local version impossible to navigate locally.
So, I created a backup copy of my online site's database and saved it locally as a .sql file. I then opened that file in a text editor (I used notepad++, a great piece of free software, but you could use any text editor). Things I needed to look out for:
For whatever reason, the tables on my
online site aren't just, for example,
"wp_posts" - they are
"wp_something_posts"... there are
some extra letters in there in the
table names.
Any references to http://... that contain my online url instead of localhost/wordpress
To keep it simple let's just do only the posts. In the backup copy of the .sql you've made of your online database, find the beginning of the wp_posts table. It will look something like this:
--
-- Table structure for table `wp_posts`
--
DROP TABLE IF EXISTS `wp_posts`;
CREATE TABLE `wp_posts` (
...and so on. Highlight everything above that up to just below the comment marking the beginning of the database at the top of the file (it will say -- Database: 'your database name') and delete it. Then go to the end of your wp_posts table, and delete everything after then end of it down to the bottom of the file. Now your file only contains your posts, and nothing else.
Save this as a separate document. Call it posts.sql or something like that.
Now, in this posts.sql file, you need to do two find/replaces actions.
Find every instance of the name of
the table wp_something_posts and
replace it with wp_posts. You only
need to do this if your backup copy
of your online database doesn't
match your clean local install as
far as the table names go. You want
whatever the table name is in this
file to match what your locally
installed wordpress database has as
this table name. If you don't make
these names match, you are just
going to end up importing the posts
into a new, differently named table,
which will be of no use to you at
all.
Find every instance of http://...
(replace the elipsis with your url)
and replace it with
http://localhost/wordpress (or
whatever the local url to your dev
version of the site is)
Now save this file again, to make sure you've got these changes set.
Now that you've done that, use phpMyAdmin to get into the wordpress database on your local machine, select the "import" tab and navigate the selector to the posts.sql file you just made, and then import it. This will pull all the data in that file into your local wp_posts table.
When that finishes, browse your local wordpress site. You'll see all your posts in there now. Hooray!
You may need to do something similar for a few other tables if you want to bring in your comments, tags, categories, and static pages you've created, etc.
I realize this is a convoluted process. There is probably a tool out there somewhere that makes this activity easier, and if someone knows of one I'd love to find out about it. If someone knows of a better way to do this manually than what I've described, I'd love to know that as well!
Until then, this is the way I figured out how to do it. Hopefully it helps get you going in the right direction.
You need to handle the serialized objects. Here is a client side HTML5 utility to handle it. Because it is all javascript, it's quite fast.
The alternative would be hooking a bash script into your deployment. So once the site is deployed, the db is backed up and deserialized with the new domain.
This about sums up the problems with the wordpress core architecture... but I wrote a plugin that solves the problems with domain names and absolute urls being stored in the database:
http://wordpress.org/extend/plugins/root-relative-urls/
This will solve the problems outlined by #oddbill. Though don't worry too much about the url being in the GUID column as that field is never used for link generation.
#markratledge provides a couple links to some lengthy documents that basically say this:
//export
mysqldump -u[username] -p[password] [database] > backup.sql
//import
mysql -u[username] -p[password] [database] < backup.sql
You'll want to exclude the comments/comments_meta tables if you push to production from staging so you don't lose all of your comments and trackbacks (#DavidLaing's approach will wipe those out.) And this assumes you only make content changes in your staging environment. If you want to make changes in production and your staging environment, you'll need to write scripts that sync the data instead of wholesale overwrite it... good luck on that task, may I suggest adding create & modified timestamp columns before you invest too much time with the current schema.
And finally, #RussellStuever's approach is suitable in most circumstances, just be sure to know when you are browsing your host mapped site versus your production site. And really be sure about it, because some browsers cache dns lookups for days until you physically close them and start a new process. At which point switching hosts may take some time, and switching frequently may get frustrating. And if you need to test with an iPhone, you'll need to publish the site live first, or use a good router that can remap outbound internet requests to local servers because you cannot modify hosts files on most mobile devices.
My plugin lets you develop and test from http://localhost/ or http://staging.server.local/ or http://www.production.com without any of the usual pitfalls. And then to migrate data, it's as simple as exporting and importing the data, no search & replace step or database setting tweaks necessary.
And don't rely on the import/export tool, it doesn't capture everything in typical wordpress installations, and still requires a needless search & replace step.