This repository has been archived by the owner on Aug 2, 2022. It is now read-only.
Fix that Trigger page might freeze for high cardinality detectors #230
Conversation
This file contains 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
We can define triggers for an AD monitor using AD preview results. The trigger page runs fine for a single-stream detector where the preview results are few. For a high-cardinality detector, the trigger page is likely freezing simply because React needs to draw too many preview results on the page. This PR fixes the issue by holding a worst-case bound on the preview results to show. Specifically, when the number of preview results exceeded the bound, we split the preview time range into small chunks and only kept the maximum anomaly grade results within each chunk. The reduction can keep important results (i.e., the anomalies) intact while speeding up the trigger page rendering. We have also seen null pointer exceptions during trigger evaluation when the anomaly result index does not exist. The exception can arise when anomaly result indices are deleted by index rollover, and there is no new result index generated. Monitors will send out alerts for the exception. This PR fixes the issue by installing extra null checks. Testing done: 1. Added/modified unit tests for the above 2 fixes. 2. Manually verified the above 2 issues are fixed.
ftianli-amzn
previously approved these changes
Dec 24, 2020
yizheliu-amazon
previously approved these changes
Dec 24, 2020
| }); | ||
|
|
||
| let mid = startTime + (endTime - startTime) / 2; | ||
| console.log(mid); |
Comment on lines
37
to
42
| if ( | ||
| !!anomalyData && | ||
| !!anomalyData.anomalyResult && | ||
| !!anomalyData.anomalyResult.anomalies && | ||
| anomalyData.anomalyResult.anomalies.length > 0 | ||
| ) { |
There was a problem hiding this comment.
you can simplify it to use get(anomalyData, 'anomalyData.anomalyResult.anomalies', []).length > 0
There was a problem hiding this comment.
yes, changed to "get(anomalyData, 'anomalyResult.anomalies', []).length > 0"
c9d4914
yizheliu-amazon
approved these changes
Dec 24, 2020
ftianli-amzn
approved these changes
Dec 24, 2020
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
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.
Issue #, if available:
#220
Description of changes:
We can define triggers for an AD monitor using AD preview results. The trigger page runs fine for a single-stream detector where the preview results are few. For a high-cardinality detector, the trigger page is likely freezing simply because React needs to draw too many preview results on the page. This PR fixes the issue by holding a worst-case bound on the preview results to show. Specifically, when the number of preview results exceeded the bound, we split the preview time range into small chunks and only kept the maximum anomaly grade results within each chunk. The reduction can keep important results (i.e., the anomalies) intact while speeding up the trigger page rendering.
We have also seen null pointer exceptions during trigger evaluation when the anomaly result index does not exist. The exception can arise when anomaly result indices are deleted by index rollover, and there is no new result index generated. Monitors will send out alerts for the exception. This PR fixes the issue by installing extra null checks.
Testing done:
null pointer exception before the change:
after the change:
By submitting this pull request, I confirm that my contribution is made under the terms of the Apache 2.0 license.