Random unit test inputs are not random across test runs

[Alistair Coles]

Unit tests use the random module in places to randomise test inputs, but once the tests in test_builder.py or test_ring_builder_analyzer.py have been run the random seed is left in a repeatable state because calls are made to RingBuilder.balance with a seed value. Consequently, subsequent calls to random in other tests get repeatable results from one test run to another.

I proposed a fix [1] to reset the seed value, but it revealed some intermittently failing tests in test_builder.py - I suspect those tests are relying on non-random input :/ If someone wants to figure that out then please add to my patch.

[1] https://review.openstack.org/#/c/371564/

[OpenStack Infra (hudson-openstack)]

Reviewed: https://review.openstack.org/371564
Committed: https://git.openstack.org/cgit/openstack/swift/commit/?id=7fa9d6e5b4fb74ad95f059742d8844e7c12fa235
Submitter: Jenkins
Branch: master

commit 7fa9d6e5b4fb74ad95f059742d8844e7c12fa235
Author: Alistair Coles <email address hidden>
Date: Fri Sep 16 14:31:44 2016 +0100

    Unset random seed after rebalancing ring

    Unit tests use the random module in places to randomise
    test inputs, but once the tests in test_builder.py or
    test_ring_builder_analyzer.py have been run the random
    seed is left in a repeatable state because calls are made
    to RingBuilder.balance with a seed value. Consequently,
    subsequent calls to random in other tests get repeatable
    results from one test run to another.

    This patch resets the state of the random module before
    returning from RingBuilder.rebalance.

    Closes-Bug: #1639755
    Change-Id: I4b74030afc654e60452e65b3e0f1b45a189c16e3

[OpenStack Infra (hudson-openstack)]

This issue was fixed in the openstack/swift 2.15.1 release.