Closed dfitchett closed 2 months ago
116 tests ±0 116 :white_check_mark: ±0 36s :stopwatch: -1s 34 suites ±0 0 :zzz: ±0 34 files ±0 0 :x: ±0
Results for commit 23a6f640. ± Comparison against base commit a8a375fe.
:recycle: This comment has been updated with latest results.
Overall Project | 67.56% | :x: |
---|
There is no coverage information present for the Files changed
What was the problem?
During a recent BGS outage, 3 requests for two failed jobs (6 total) were sent it with no response. Currently, there is no queue wide message TTL policy, and no per message TTL was being set by the publisher (AsyncHoppyClient) when sending the requests, and although the service was down, the messages sent to add notes to a claim were still in queue. When the service came back up, it resulted in duplicate notes being added to the queue.
Associated tickets or Slack threads:
3175
How does this fix it?[^1]
Adds a configurable per-message TTL to the message upon publishing. The default is set to zero. From the documentation:
There is currently no dead lettering happening, and is unsupported by VRO and
How to test this PR
cd abd-vro/shared/lib-hoppy
pip install .
pytest
[^1]: Pull-Requests guidelines. If PR is significant, update Current Software State wiki page. [^secrel]: To check if a PR will succeed in the SecRel workflow, test PRs in the SecRel pipeline.