Add _metric_names_hash field to OTel metric mappings #120952
Merged
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
If metrics that have the same timestamp and dimensions aren't grouped into the same document, ES will consider them to be a duplicate. The _metric_names_hash field will be set by the OTel ES exporter. As it's mapped as a time_series_dimensions, it creates a different _tsid for documents with different sets of metrics. The tradeoff is that if the composition of the metrics grouping changes over time, a different _tsid will be created. That has an impact on the rate aggregation for counters.
felixbarny added >bug :Data Management/Data streams 
elasticsearchmachine added the Team:Data Management 
| Pinging @elastic/es-data-management (Team:Data Management) |
elasticsearchmachine added the external-contributor | Hi @felixbarny, I've created a changelog YAML for you. |
carsonip reviewed Jan 27, 2025
| priority: 10 | ||
| # workaround for https://github.com/elastic/elasticsearch/issues/99123 | ||
| _metric_names_hash: | ||
| type: keyword |
There was a problem hiding this comment.
Q: will a number be more lightweight, as you're using a 8 digit hex anyway?
There was a problem hiding this comment.
At the moment, numbers can't leverage run-length encoding. So it's actually lighter to use a keyword here as all dimensions are incorporated into the _tsid, which we sort by. Therefore, all values for the same tsid are equal and can be compressed very efficiently.
carsonip approved these changes Jan 28, 2025
elasticsearchmachine added v8.19.0 v9.1.0 v8.17.3 v8.16.5 and removed v8.18.0 v9.0.0 v8.17.2 v8.16.4 labels 
| I had a discussion with @martijnvg about this last week. The conclusion was that this change makes the consequences of imperfect grouping much less bad and we should therefore move forward with it. It's not a replacement for improving the grouping logic. But it's much better to have a different time series rather than dropping metrics. It's a much less stressful situation having to debug why the rate aggregation isn't working properly in some cases rather than debugging a data loss scenario. Longer-term, it seems like we'll go into the one metric per doc route where grouping of metrics isn't required anymore. |
This was referenced Feb 18, 2025
felixbarny added a commit to felixbarny/elasticsearch that referenced this pull request Feb 18, 2025
If metrics that have the same timestamp and dimensions aren't grouped into the same document, ES will consider them to be a duplicate. The _metric_names_hash field will be set by the OTel ES exporter. As it's mapped as a time_series_dimensions, it creates a different _tsid for documents with different sets of metrics. The tradeoff is that if the composition of the metrics grouping changes over time, a different _tsid will be created. That has an impact on the rate aggregation for counters.
This was referenced Feb 18, 2025
felixbarny added a commit to felixbarny/elasticsearch that referenced this pull request Feb 18, 2025
If metrics that have the same timestamp and dimensions aren't grouped into the same document, ES will consider them to be a duplicate. The _metric_names_hash field will be set by the OTel ES exporter. As it's mapped as a time_series_dimensions, it creates a different _tsid for documents with different sets of metrics. The tradeoff is that if the composition of the metrics grouping changes over time, a different _tsid will be created. That has an impact on the rate aggregation for counters.
felixbarny added a commit to felixbarny/elasticsearch that referenced this pull request Feb 18, 2025
If metrics that have the same timestamp and dimensions aren't grouped into the same document, ES will consider them to be a duplicate. The _metric_names_hash field will be set by the OTel ES exporter. As it's mapped as a time_series_dimensions, it creates a different _tsid for documents with different sets of metrics. The tradeoff is that if the composition of the metrics grouping changes over time, a different _tsid will be created. That has an impact on the rate aggregation for counters.
elasticsearchmachine pushed a commit that referenced this pull request Feb 18, 2025
If metrics that have the same timestamp and dimensions aren't grouped into the same document, ES will consider them to be a duplicate. The _metric_names_hash field will be set by the OTel ES exporter. As it's mapped as a time_series_dimensions, it creates a different _tsid for documents with different sets of metrics. The tradeoff is that if the composition of the metrics grouping changes over time, a different _tsid will be created. That has an impact on the rate aggregation for counters.
elasticsearchmachine pushed a commit that referenced this pull request Feb 18, 2025
If metrics that have the same timestamp and dimensions aren't grouped into the same document, ES will consider them to be a duplicate. The _metric_names_hash field will be set by the OTel ES exporter. As it's mapped as a time_series_dimensions, it creates a different _tsid for documents with different sets of metrics. The tradeoff is that if the composition of the metrics grouping changes over time, a different _tsid will be created. That has an impact on the rate aggregation for counters.
elasticsearchmachine pushed a commit that referenced this pull request Feb 18, 2025
If metrics that have the same timestamp and dimensions aren't grouped into the same document, ES will consider them to be a duplicate. The _metric_names_hash field will be set by the OTel ES exporter. As it's mapped as a time_series_dimensions, it creates a different _tsid for documents with different sets of metrics. The tradeoff is that if the composition of the metrics grouping changes over time, a different _tsid will be created. That has an impact on the rate aggregation for counters.
elasticsearchmachine pushed a commit that referenced this pull request Feb 18, 2025
If metrics that have the same timestamp and dimensions aren't grouped into the same document, ES will consider them to be a duplicate. The _metric_names_hash field will be set by the OTel ES exporter. As it's mapped as a time_series_dimensions, it creates a different _tsid for documents with different sets of metrics. The tradeoff is that if the composition of the metrics grouping changes over time, a different _tsid will be created. That has an impact on the rate aggregation for counters.
andrzej-stencel pushed a commit to open-telemetry/opentelemetry-collector-contrib that referenced this pull request Feb 21, 2025
…tions (#37511) If metrics that have the same timestamp and dimensions aren't grouped into the same document, ES will consider them to be a duplicate. This adds a hash of the metric names that will be mapped as a dimension in Elasticsearch. The tradeoff is that if the composition of the metrics grouping changes over time, a new time series will be created. That has an impact on the rate aggregation for counters. ES mapping changes: elastic/elasticsearch#120952 --------- Co-authored-by: Carson Ip <carsonip@users.noreply.github.com>
💚 All backports created successfully
Questions ?Please refer to the Backport tool documentation |
carsonip pushed a commit to carsonip/elasticsearch that referenced this pull request Apr 15, 2025
If metrics that have the same timestamp and dimensions aren't grouped into the same document, ES will consider them to be a duplicate. The _metric_names_hash field will be set by the OTel ES exporter. As it's mapped as a time_series_dimensions, it creates a different _tsid for documents with different sets of metrics. The tradeoff is that if the composition of the metrics grouping changes over time, a different _tsid will be created. That has an impact on the rate aggregation for counters. (cherry picked from commit 5e8865d)
carsonip added a commit that referenced this pull request Apr 15, 2025
If metrics that have the same timestamp and dimensions aren't grouped into the same document, ES will consider them to be a duplicate. The _metric_names_hash field will be set by the OTel ES exporter. As it's mapped as a time_series_dimensions, it creates a different _tsid for documents with different sets of metrics. The tradeoff is that if the composition of the metrics grouping changes over time, a different _tsid will be created. That has an impact on the rate aggregation for counters. (cherry picked from commit 5e8865d) Co-authored-by: Felix Barnsteiner <felixbarny@users.noreply.github.com>
carsonip added a commit to carsonip/elasticsearch that referenced this pull request Apr 16, 2025
…elastic#126850) Bump otel-data plugin version as elastic#120952 missed the bump.
carsonip added a commit to carsonip/elasticsearch that referenced this pull request Apr 16, 2025
…elastic#126850) Bump otel-data plugin version as elastic#120952 missed the bump.
carsonip added a commit to carsonip/elasticsearch that referenced this pull request Apr 16, 2025
…elastic#126850) Bump otel-data plugin version as elastic#120952 missed the bump.
carsonip added a commit to carsonip/elasticsearch that referenced this pull request Apr 16, 2025
…elastic#126850) Bump otel-data plugin version as elastic#120952 missed the bump. (cherry picked from commit 5860ccb) # Conflicts: # x-pack/plugin/otel-data/src/main/resources/resources.yaml
elasticsearchmachine pushed a commit that referenced this pull request Apr 16, 2025
carsonip added a commit that referenced this pull request Apr 17, 2025
elasticsearchmachine pushed a commit that referenced this pull request Apr 17, 2025
elasticsearchmachine pushed a commit that referenced this pull request Apr 17, 2025
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
4 participants
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.
A short-term workaround for #99123
If metrics that have the same timestamp and dimensions aren't grouped into the same document, ES will consider them to be a duplicate. The
_metric_names_hashfield will be set by the OTel ES exporter (see open-telemetry/opentelemetry-collector-contrib#37511). As it's mapped as a time_series_dimensions, it creates a different _tsid for documents with different sets of metrics. The tradeoff is that if the composition of the metrics grouping changes over time, a different _tsid will be created. That has an impact on the rate aggregation for counters.