Segment metadata #456
Merged
Segment metadata #456
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
These are typically where a metric type has changed between segments. The column type defaults to string, but provides a non-null errorMessage. Previously we would assume this was a dimension since it is reported as string. This will now omit it from visibility.
Requires using lenient aggregator merge if we want to accommodate types changing
brharrington
approved these changes
May 19, 2023
manolama
pushed a commit
to manolama/iep-apps
that referenced
this pull request
Oct 25, 2023
There are 2 issues with Druid segment metadata 1. If a column type is changed, the entry for the column default to a string type and an errorMessage is populated. These column entries cannot be used. Since we treat all string columns as dimension, a metric that changes type gets listed as a dimension instead of a metric and as such cannot be queried. 2. The HLLSketchBuilder ingestion aggregator initially reports column type of HLLSketchBuilder but as soon as that is persisted into an actual segment it becomes HLLSketch. This results in the same symptom as (1) since the type appears to have changed. We encounter problem (1) when someone accidentally adds a new metric as the wrong type and needs to fix it. e.g. as a spectatorHistogramTimer when it was supposed to be a spectatorHistogramDistribution. Today we have to run a hadoop job to restate that column under the correct type. We encounter problem (2) when ingesting HLLSketches from raw values using realtime ingestion. This does not appear when using batch ingestion since the segment and metadata is already finalized once the cluster loads the segment. Changes: 1. Filter out columns that report an error. These are typically where a metric type has changed between segments. The column type defaults to string, but provides a non-null errorMessage. Previously we would assume this was a dimension since it is reported as string. This will now omit it from visibility. 2. Use aggregators to define metrics rather than columns. Requires using lenient aggregator merge if we want to accommodate types changing.
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.
There are 2 issues with Druid segment metadata
HLLSketchBuilderbut as soon as that is persisted into an actual segment it becomesHLLSketch. This results in the same symptom as (1) since the type appears to have changed.We encounter problem (1) when someone accidentally adds a new metric as the wrong type and needs to fix it. e.g. as a
spectatorHistogramTimerwhen it was supposed to be aspectatorHistogramDistribution. Today we have to run a hadoop job to restate that column under the correct type.We encounter problem (2) when ingesting HLLSketches from raw values using realtime ingestion. This does not appear when using batch ingestion since the segment and metadata is already finalized once the cluster loads the segment.