Open
Conversation
2607e4e to
4f982e0
Compare
17afd13 to
bbb18ac
Compare
bbb18ac to
6eeb206
Compare
…average queue time
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Describe your changes
https://financialtimes.atlassian.net/browse/DATA-10316 and the accompanying Slack thread describe a very old workaround in
o-trackingto avoid a 'queue bug', whereby if more than 200 events are in the local storage queue (i.e. because the device has been offline or unable to send events for some reason), they are summarised in a singleo-tracking:queue-bugevent. This was presumably to avoid an enormous flood of events due to a buggy release ofo-trackingmany years ago. Importantly, summarising these events rather than sending them properly means the Spoor pipeline is missing key BI events in a significant volume (e.g. 2.6mnpage:viewevents for FT App in a 90-day period).However, the current implementation throws away some useful data, which will be required in order to assess whether the bug still exists or if these queued events can now all be sent safely.
This PR makes a few changes:
add()andrun()from thesend.jsAPI: these events were only used in unit tests, and should be preserved as internal helper methods onlyo-tracking:queue-bugpayload, if the queued events contain an oldero-tracking:queue-bugevent, merge the data in to the current context (rather than throw away the data from any events prior to the last 200)o-tracking:queue-bug:isOfflineCounts(how many of the events were sent withis_offline: true),maxOfflineLagandaverageOfflineLag.Additional context
Checklist before requesting a review
percylabel for o-[COMPONENT] orchromaticlabel for o3-[COMPONENT] on my PR before merging and after review. Find more details in CONTRIBUTING.md