feat: Reindexing supervisor web console builder workfow#19312
feat: Reindexing supervisor web console builder workfow#19312capistrant wants to merge 4 commits into
Conversation
|
unrelated: have you considered using an LLM to port the web console build process to Vite? This should significantly speed up build times |
Had not considered it cuz I had never heard of Vite until you mentioned it 😁. I can push up some draft proposal for it but I'd also like to hear @vogievetsky thoughts on the topic |
FrankChen021
left a comment
There was a problem hiding this comment.
| Severity | Findings |
|---|---|
| P0 | 0 |
| P1 | 0 |
| P2 | 3 |
| P3 | 0 |
| Total | 3 |
Reviewed 17 of 17 changed files.
This is an automated review by Codex GPT-5.5
| dataSource, | ||
| defaultSegmentGranularity: 'DAY', | ||
| defaultPartitionsSpec: { type: 'dynamic', maxRowsPerSegment: 5000000 }, | ||
| ruleProvider: { type: 'inline' }, |
There was a problem hiding this comment.
[P2] Require at least one rule before submitting
Selecting the cascading template creates ruleProvider: { type: 'inline' }, and the top-level form treats that as satisfying the required ruleProvider field even though all rule lists are empty. The server-side cascading template later throws when no partitioning or non-partitioning rule exists, so the console can save a supervisor that cannot generate jobs. Make the custom field invalid until at least one rule list has an entry, or seed a real rule before enabling Submit.
| info: <>Determines how many rows are in each segment.</>, | ||
| }, | ||
| { | ||
| name: 'defaultPartitionsSpec.maxTotalRows', |
There was a problem hiding this comment.
[P2] Remove unsupported maxTotalRows field
Cascading reindexing is MSQ-only, and the server validation rejects dynamic partition specs that set maxTotalRows. This field exposes defaultPartitionsSpec.maxTotalRows in the cascading form, so users can configure a value that the POST endpoint rejects. Drop this field for reindexCascade dynamic defaults or prevent it from being written.
| }, | ||
| // defaultPartitionsSpec: dynamic | ||
| { | ||
| name: 'defaultPartitionsSpec.maxRowsPerSegment', |
There was a problem hiding this comment.
[P2] Allow range defaults to keep maxRowsPerSegment
The only defaultPartitionsSpec.maxRowsPerSegment field is defined for dynamic partitioning, but the range target field still treats maxRowsPerSegment as a valid alternative. If an existing valid cascading spec uses defaultPartitionsSpec: { type: 'range', maxRowsPerSegment: ... }, AutoForm.issueWithModel marks the field as defined when it should not be and disables Submit. Mirror the existing compaction form by adding a range maxRowsPerSegment field defined when targetRowsPerSegment is absent.
Description
This PR enhances the compaction configuration building workflow in the web console by adding template type selection when using compaction supervisors. This allows folks to configure the new cascading reindexing supervisor template type which allows for rules based reindexing that can apply different rules for different intervals of time. Before this PR the only way to work with the new cascading reindexing spec format was via the API. The goal of this PR is to continue down the path of completing all of the pending bullets in #19092 as we try to make this new feature production worthy.
Images
Example Spec
If you want to seed your local cluster with an example spec so you can tinker with modifying via the form, you can use the following spec. Note that the underlying wikipedia data need not be ingested since this is a suspended supervisor and the form works regardless of the underlying data.
Click to expand code
{ "type": "autocompact", "spec": { "type": "reindexCascade", "dataSource": "wikipedia", "taskPriority": 25, "inputSegmentSizeBytes": 100000000000000, "ruleProvider": { "type": "inline", "deletionRules": [ { "id": "delete_japan", "description": null, "olderThan": "P90D", "deleteWhere": { "type": "equals", "column": "countryName", "matchValueType": "STRING", "matchValue": "Japan" }, "virtualColumns": null } ], "partitioningRules": [], "indexSpecRules": [], "dataSchemaRules": [ { "id": "dataSchema-1", "description": null, "olderThan": "P30D", "dimensionsSpec": { "dimensions": null }, "metricsSpec": [], "queryGranularity": "HOUR", "rollup": true, "projections": [] } ] }, "taskContext": { "useConcurrentLocks": true, "maxNumTasks": 2 }, "skipOffsetFromLatest": null, "skipOffsetFromNow": null, "defaultSegmentGranularity": "DAY", "defaultPartitionsSpec": { "type": "range", "targetRowsPerSegment": 5, "maxRowsPerSegment": null, "partitionDimensions": [ "city_and_country", "isRobot" ], "assumeGrouped": false }, "defaultPartitioningVirtualColumns": [ { "type": "expression", "name": "city_and_country", "expression": "concat(cityName, ':', countryName)", "outputType": "STRING" } ], "tuningConfig": null }, "suspended": true }Release note
Adds the ability to configure a compaction supervisor using the
reindexCascadetemplate type in the web console. This contributes to more production worthy UX for the new template type.Key changed/added classes in this PR
This PR has: