Add helper to cleanup delayed sidekiq timechunks #894
Merged
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
For context:
#869 added the
delayed_sidekiq
strategy, which by design will only create a sidekiq worker to update an index every N seconds (10 seconds by default).The implementation details for that involves redis sorted sets: it creates a sorted set for each index that uses the strategy and adds a new member to it everytime we call reindex (with the weight being calculated from the current timestamp, generating a unique number every N seconds).
The problem:
Since we're writing directly to Redis, if we use this strategy in tests we're subject to state leaking and creating flaky tests. For example:
MyClass.import!(1, strategy: :delayed_sidekiq)
and not flush the Sidekiq workers or the redis databaseMyClass.import!(1, strategy: :delayed_sidekiq)
, flushes Sidekiq workers and expects the index to be updated.On the above scenario, test B will fail as no worker will be enqueued since
import!
was called within 10 seconds in between the tests (also note that Sidekiq runs in memory by default when running tests, so workers enqueued by test A are not present in test B).The fix:
This PR adds a way to workaround the issue by cleaning up the saved timechunks. It can easily be added to e.g. rspec hooks via:
Before submitting the PR make sure the following are checked:
[Fix #issue-number]
(if the related issue exists).master
(if not - rebase it).