fix(timestamps): Deal with sub-microsecond inaccuracies on parse #5002
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Dav1dde
force-pushed
the
dav1d/ts-log-nano-inacc
branch
2 times, most recently
from
f117a67 to
ae0f0cb
Compare
Dav1dde
commented
Jul 30, 2025
| return { | ||
| "sentry.observed_timestamp_nanos": { | ||
| "stringValue": time_within(ts, expect_resolution="ns", precision="s") | ||
| "stringValue": time_within(ts, expect_resolution="ns") |
Member
Author
There was a problem hiding this comment.
This was incorrect, before, the observed_timestamp_nanos is set from the received timestamp, which is derived from the current time in Relay, not an input.
Fixed here because I used these tests to verify the flakiness is gone (which was observable under repeated execution).
![cursor[bot]](https://avatars.githubusercontent.com/in/1210556?s=60&v=4){kind=small}
This comment was marked as outdated.
This comment was marked as outdated.
Sorry, something went wrong.
Dav1dde
force-pushed
the
dav1d/ts-log-nano-inacc
branch
from
7acf98c to
d153981
Compare
Dav1dde
force-pushed
the
dav1d/ts-log-nano-inacc
branch
from
d153981 to
3597ba3
Compare
This comment was marked as outdated.
This comment was marked as outdated.
Sorry, something went wrong.
loewenheim
reviewed
Jul 30, 2025
Dav1dde
force-pushed
the
dav1d/ts-log-nano-inacc
branch
from
12cf442 to
9452280
Compare
This comment was marked as outdated.
This comment was marked as outdated.
Sorry, something went wrong.
Dav1dde
force-pushed
the
dav1d/ts-log-nano-inacc
branch
from
9452280 to
32ba503
Compare
loewenheim
approved these changes
Jul 30, 2025
Comment on lines
+1038
to
+1063
| fn test_timestamp_to_datetime() { | ||
| assert_eq!(timestamp_to_datetime(0.), Utc.timestamp_opt(0, 0)); | ||
| assert_eq!(timestamp_to_datetime(1000.), Utc.timestamp_opt(1000, 0)); | ||
| assert_eq!(timestamp_to_datetime(-1000.), Utc.timestamp_opt(-1000, 0)); | ||
| assert_eq!( | ||
| timestamp_to_datetime(1.234_567), | ||
| Utc.timestamp_opt(1, 234_567_000) | ||
| ); | ||
| assert_eq!(timestamp_to_datetime(2.999_999_51), Utc.timestamp_opt(3, 0)); | ||
| assert_eq!( | ||
| timestamp_to_datetime(2.999_999_45), | ||
| Utc.timestamp_opt(2, 999_999_000) | ||
| ); | ||
| assert_eq!( | ||
| timestamp_to_datetime(-0.000_001), | ||
| Utc.timestamp_opt(-1, 999_999_000) | ||
| ); | ||
| assert_eq!( | ||
| timestamp_to_datetime(-3.000_000_49), | ||
| Utc.timestamp_opt(-3, 0) | ||
| ); | ||
| assert_eq!( | ||
| timestamp_to_datetime(-3.000_000_51), | ||
| Utc.timestamp_opt(-4, 999_999_000) | ||
| ); | ||
| } |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
This also deals with floating point sub-microsecond inaccuracies on parsing, instead of only on serialization.
Logs access the nanoseconds of the extracted timestamp and persist these into the payload, this leads to subtle inaccuracies because it never goes through the floating point serialization code which would deal with these inaccuracies.
As it turns out this also applies to calculated span durations which were off.
Instead account for these inaccuracies when parsing from a float, but also when serializing, as the timestamp may have been constructed from something with 'perfect' accuracy.
Also fixes this:
