Aias00 opened a new issue, #7695: URL: https://github.com/apache/shenyu/issues/7695
## Current behavior `collectStateData` aggregates all instances into a `stateHistoryMap` keyed only by state. `getInstanceDataVisual(namespaceId)` filters the pie-chart counts by namespace but reads the global history map for line-chart data. Selecting namespace A therefore shows A-only current counts beside a history containing B and other namespaces. Reproduction: add distinct populations in namespaces A and B, collect one scheduled sample, then request `/instance/analysis/A` and `/instance/analysis/B`. Historical counts should follow the selected namespace but currently share the same data. ## Acceptance criteria - [ ] Keep bounded history per namespace/state, with explicit behavior for any supported aggregate view. - [ ] Sample once per check cycle, preserving the fix for #6497. - [ ] Test A/B independence, empty namespaces and consistent line/pie populations. - [ ] Bound history retention and namespace cleanup. Related: #6497 fixed sampling frequency; this is a separate namespace-scoping defect. Namespace visibility implications have not been tested as a security exploit. ## Source evidence [Reviewed source](https://github.com/apache/shenyu/blob/b4e115367fd57adc8aba332d655b1009932ba394/shenyu-admin/src/main/java/org/apache/shenyu/admin/service/impl/InstanceCheckService.java#L213). Baseline: `b4e115367fd57adc8aba332d655b1009932ba394`. Report based on source inspection; no runtime reproduction was executed during issue preparation. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
