Limit amount of test failures due to updates of tzdata
It seems that recently, Ruby test suite is broken ~ twice a year due to changes in tzdata. I am not sure how detailed the test suite needs to be to have the right amount of test coverage, but it would be nice if Ruby developers had more control above this changes then this changes being forced upon us via OS update. This is followup to #19187.
BTW with my Fedora / RHEL maintainer hat on, this is a bit annoying, because:
- It tends to hide other issues
- In Fedora, it might block other development
- Especially in RHEL, where there is not so big churn, it is unpleasant to start work on fixing CVE just to discover one needs to fix the tzdata test failures first
No data to display