t2t2
In case this issue gets active once again, archive.org link for above blog post: https://web.archive.org/web/20210123103641/https://plumbr.io/blog/monitoring/time-in-distributed-systems
Made a sandbox with this patch, most frequently used otel web instrumentations (document-load, user-interaction, xhr and fetch) and an artificial time drift simulator: https://codesandbox.io/s/modest-fermat-4eu09e?file=/src/otel-fixes.ts It doesn't improve this test case...
_Updated codesandbox with pr changes as of now_ >This was the root of the 0-duration spans in your example too Yep fixed on auto start + perf end, however span...
> If the user provides manual start and end times we can't correct drift because they may be intentionally describing a span that happened in the past. There is currently...
like previous pr, [codesandbox with the changes, artificial drift and most common instrumentations](https://codesandbox.io/s/otel-js-pr-3434-jupv0h?file=/src/otel-fixes.ts) (with already fixed the `_performanceOffset` issue in first comment)
> FWIW, Chrome 108 was released 9 hours ago with a fix - from my testing the timings aren't broken anymore. Also possible that as the new version had been...
How about [suppressTracing](https://github.com/open-telemetry/opentelemetry-js/blob/c85fbe6682b9a8fd52e4b99bdbb079833b04e845/packages/opentelemetry-core/src/trace/suppress-tracing.ts) (already used by exporters to avoid creating spans for exporting - [see example](https://github.com/open-telemetry/opentelemetry-js/blob/369b07e1c7483556fa9da952ba27c2e2828ecefb/packages/opentelemetry-core/src/internal/exporter.ts#L35)) or doing the logic in a sampler?
After suggestion from the (prefers to be) anonymous contributor, here's sql to update current movie's boxmojo_id-s [updatemovies.zip](https://github.com/t2t2/draft/files/224325/updatemovies.zip)
Which would people prefer, just links straight to IMDB&mojo or via a "movie details" page that could have related info?
Not atm, feel free to