[openstack-dev] [Keystone] SQLite support (migrations, work-arounds, and more), is it worth it?

Morgan Fainberg morgan.fainberg at gmail.com
Mon Apr 6 18:52:24 UTC 2015

> On Apr 6, 2015, at 11:41, Mike Bayer <mbayer at redhat.com> wrote:
>> On 4/6/15 12:49 PM, David Stanek wrote:
>> Exactly. This is the direction I have been going. Functional tests are written using the public APIs using the client.
>> I would also add that I don't like that the Keystone unit tests are so database heavy. I would not want MySQL or ant RDBMS to be a requirement for running the tests.
> is that because you'd prefer that the unit tests were more isolated, or just that an external service is being used?
> I've done some work with extensive mocking of SQL databases; specifically mocking at the ORM level.  It is nice when you get it to run, but it's also a much bigger job to write fine-grained mocks like that, the mocks can be brittle in relation to the code they're targeting, and you also need to come up with some solution to actually functional test your database access code.
> I tend to think that having a mysqld service running is the lesser of two evils and you get a lot more code coverage by going all the way out to the DB.

What David is specifically referencing is that we want our functional tests to only require direct API access. There is almost no reason to need access to the DB backend. We have many ways to perform isolation where needed (tempest does a lot of this today). 

The goal is to allow the functional test suite to run against any keystone deployment (be agnostic to db, non-db, etc driver used). This makes environment setup a separate concern the tests don't need to be involved in/aware of. It makes our functional tests more useful for validating a driver or configuration passes muster. 

If there are legitimately cases we need to test a specific db function in isolation we will make specific efforts to support it. Those are apt to be the exception to the rule. 

Sent via mobile 

>>> On Mon, Apr 6, 2015, 12:42 Morgan Fainberg <morgan.fainberg at gmail.com> wrote:
>>> > On Apr 6, 2015, at 09:20, Mike Bayer <mbayer at redhat.com> wrote:
>>> >
>>> >
>>> >
>>> >> On 4/6/15 12:06 PM, Clint Byrum wrote:
>>> >> Excerpts from Boris Bobrov's message of 2015-04-03 18:29:08 -0700:
>>> >>>> On Saturday 04 April 2015 03:55:59 Morgan Fainberg wrote:
>>> >>>> I am looking forward to the Liberty cycle and seeing the special casing we
>>> >>>> do for SQLite in our migrations (and elsewhere). My inclination is that we
>>> >>>> should (similar to the deprecation of eventlet) deprecate support for
>>> >>>> SQLite in Keystone. In Liberty we will have a full functional test suite
>>> >>>> that can (and will) be used to validate everything against much more real
>>> >>>> environments instead of in-process “eventlet-like” test-keystone-services;
>>> >>>> the “Restful test cases” will no longer be part of the standard unit tests
>>> >>>> (as they are functional testing). With this change I’m inclined to say
>>> >>>> SQLite (being the non-production usable DB) what it is we should look at
>>> >>>> dropping migration support for SQLite and the custom work-arounds.
>>> >>>>
>>> >>>> Most deployers and developers (as far as I know) use devstack and MySQL or
>>> >>>> Postgres to really suss out DB interactions.
>>> >>>>
>>> >>>> I am looking for feedback from the community on the general stance for
>>> >>>> SQLite, and more specifically the benefit (if any) of supporting it in
>>> >>>> Keystone.
>>> >>> +1. Drop it and clean up tons of code used for support of sqlite only.
>>> >>>
>>> >>> Doing tests with mysql is as easy, as with sqlite ("mysqladmin drop -f;
>>> >>> mysqladmin create" for "reset"), and using it by default will finally make
>>> >>> people test their code on real rdbmses.
>>> >> Please please please be careful with that and make sure the database
>>> >> name is _always_ random in tests... or even better, write a fixture to
>>> >> spin up a mysqld inside a private tempdir. That would be a really cool
>>> >> thing for oslo.db to provide actually.
>>> >>
>>> >> I'm just thinking some poor sap runs the test suite with the wrong
>>> >> .my.cnf in the wrong place and <poof> there went keystone's db.
>>> > The standard approach here is that tests based on the oslo.db approach by default connect using a username openstack_citest on localhost, which is then used to create databases of random names. If you base your database tests on oslo.db, you get that right now.   This username/host/etc is also configurable through environment variables.  I don't see how my.cnf is involved in this nor how it would impact someone's keystone database, unless we're talking about tests that happen to load down and/or crash the whole database server.
>>> >
>>> > spinning up a whole mysqld seems really heavy-handed and unnecessary.  Not to mention the tests run on other backends as well such as Postgresql.
>>> >
>>> The reasons outlined by both Clint and Mike are why we won't be attempting this until we are happy with our functional test suite. But once we are happy dropping SQLite is on the table. The way I see it the functional tests should be performed against a real keystone server, even if it is one spun up for testing specifically.
>>> Per test db creation / other such stuff will be part of that discussion.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstack.org/pipermail/openstack-dev/attachments/20150406/77b2ba12/attachment.html>

More information about the OpenStack-dev mailing list