The situation is resolved

We're monitoring the situation. Errors appear to have stopped and metrics look good our end, though this could coincide with work day ending in Europe and in turn less opportunities for errors to occur. Further work will continue tomorrow unless the situation drastically changes later today.

Another change has been to made to mitigate the issue to further reduce error frequency.

We have also identified a way to properly fix the underlying issue, and discovered why jsrunner is affected by an unrelated service. Both of these things will be implemented/patched.

We have taken two steps to try to mitigate the underlying issue. This should reduce, but not remove, the frequency of errors. We are working on a more permanent fix.

The underlying issue only affects our internal observability and does not negatively impact any customer.