Commit Graph

1078 Commits

Author SHA1 Message Date
Bob Van Landuyt 2e2a63c866 Rename child namespaces in migrationhelpers 2017-05-01 11:14:24 +02:00
Bob Van Landuyt b85f2fa4c7 Clear html cache for a projects milestones 2017-05-01 11:14:24 +02:00
Bob Van Landuyt 99a03fd6e9 Move ReservedPathsMigration into V1 namespace 2017-05-01 11:14:24 +02:00
Bob Van Landuyt 39efd0c030 Clear cached markdown after renaming projects 2017-05-01 11:14:24 +02:00
Bob Van Landuyt 389057f001 Rename Projects & Namespaces based on entire paths 2017-05-01 11:14:24 +02:00
Bob Van Landuyt 27f54bebb2 Use objects for renaming namespaces and projects 2017-05-01 11:14:24 +02:00
Bob Van Landuyt 0369ef1452 Add a WIP spec for clearing the cache 2017-05-01 11:14:24 +02:00
Bob Van Landuyt 579d8891d5 Rename projects in a migrationhelper 2017-05-01 11:14:24 +02:00
Bob Van Landuyt 7508ee5667 Make renaming records in the database reusable
So we can use it for projects
2017-05-01 11:14:24 +02:00
Bob Van Landuyt 58bc628d30 Rename namespace-paths in a migration helper 2017-05-01 11:14:24 +02:00
Bob Van Landuyt 2f95e6a0c4 Move `replace_sql` into `Database::MigrationHelpers` 2017-05-01 11:14:24 +02:00
Douwe Maan 3584f7dd52 Merge branch 'zero-downtime-migrations' into 'master'
Prepare for zero downtime migrations

See merge request !9976
2017-04-12 20:43:12 +00:00
Yorick Peterse 223d8a3d26
Prepare for zero downtime migrations
Starting with GitLab 9.1.0 we will no longer allow downtime migrations
unless absolutely necessary. This commit updates the various developer
guides and adds code that is necessary to make zero downtime migrations
less painful.
2017-04-12 18:15:19 +02:00
Tiago Botelho c41df67676 removes redundant code from database.rb 2017-04-11 20:20:19 +01:00
blackst0ne 9997c58fc0 Add remove_concurrent_index to database helper 2017-04-06 09:53:57 +11:00
Yorick Peterse 79696f5b7a
Hash concurrent foreign key names similar to Rails
This was initially not implemented simply because I forgot about the
size limit of constraint names in PostgreSQL (63 bytes). Using the old
technique we can't add foreign keys for certain tables. For example,
adding a foreign key on
protected_branch_merge_access_levels.protected_branch_id would lead to
the following key name:

    fk_protected_branch_merge_access_levels_protected_branches_protected_branch_id

This key is 78 bytes long, thus violating the PostgreSQL size
requirements.

The hashing strategy is copied from Rails' foreign_key_name() method,
which unfortunately is private and subject to change without notice.
2017-02-21 15:07:02 +01:00
Yorick Peterse a97dcc077c
Add method for creating foreign keys concurrently
This method allows one to create foreign keys without blocking access to
the source table, but only on PostgreSQL.

When creating a regular foreign key the "ALTER TABLE" statement used for
this won't return until all data has been validated. This statement in
turn will acquire a lock on the source table. As a result this lock can
be held for quite a long amount of time, depending on the number of rows
and system load.

By breaking up the foreign key creation process in two steps (creation,
and validation) we can reduce the amount of locking to a minimum.
Locking is still necessary for the "ALTER TABLE" statement that adds the
constraint, but this is a fast process and so will only block access for
a few milliseconds.
2017-02-10 21:51:09 +01:00
Drew Blessing 1f399fe437 fix 2016-09-15 21:59:59 -05:00
Drew Blessing e000f02bd3 Add support for column limits in add_column_with_default 2016-09-15 21:59:58 -05:00
Stan Hu 3fd304c1a5 Fix spec Don't attempt to disable statement timeout on a MySQL DB 2016-07-14 20:55:33 -07:00
Yorick Peterse 816c453558
Don't update columns in batches in a transaction
This ensures that whatever locks are acquired aren't held onto until the
end of the transaction (= after _all_ rows have been updated). Timing
wise there's also no difference between using a transaction and not
using one.
2016-06-15 17:04:07 +02:00
Yorick Peterse 8966263e0c
Customizing of update_column_in_batches queries
By passing a block to update_column_in_batches() one can now customize
the queries executed. This in turn can be used to only update a specific
set of rows instead of simply all the rows in the table.
2016-06-15 16:44:44 +02:00
Yorick Peterse b33b7be53e
Handle NULL migration errors in migration helpers
This ensures that whenever changing the NULL constraint of a column
fails we still drop the column.
2016-06-13 11:22:58 +02:00
Felipe Artur 9264203103 change add_concurrent_index function arguments 2016-06-06 13:06:21 -03:00
Grzegorz Bizon 9e92b0eead Add specs for migration helpers and boolean values 2016-05-22 22:41:14 +02:00
Grzegorz Bizon 13bf612600 Test migration helpers using a migration class 2016-05-22 22:40:43 +02:00
Yorick Peterse 36083b4d0e
Fixed styling per Rubocop pedantics 2016-05-12 13:03:55 +02:00
Yorick Peterse e867099091
Added helper methods for database migrations
These helpers can be used to perform migrations without taking down the
entire application.

For example, the method "add_column_with_default" can be used to add a
new column with a default value without locking the entire table.
2016-05-12 12:59:40 +02:00