-
Notifications
You must be signed in to change notification settings - Fork 1.1k
Description
Describe the bug
Every command in dev tools/console does a GET for _mappings or tries to get it multiple times. This makes simple requests for things like _cluster/health take a long time to respond or timeout. All of the nodes response instantly _cluster/health or _mappings. The response body for _mappings in my environment is large, its over 100MB and around 3500 mappings.
To Reproduce
Steps to reproduce the behavior:
- Go to Management > Dev Tools
- Open developer console
- Run
GET _cluster/health
Expected behavior
I have only experienced this issue since updating from 2.9.0 to 2.11.1. I am not sure how it worked on 2.9.0 and prior. At least it seems like this request should be run less often if it cannot be removed entirely.
OpenSearch Version
2.11.1
Dashboards Version
2.11.1
Plugins
alertingDashboards@2.11.1.0
anomalyDetectionDashboards@2.11.1.0
customImportMapDashboards@2.11.1.0
ganttChartDashboards@2.11.1.0
indexManagementDashboards@2.11.1.0
mlCommonsDashboards@2.11.1.0
notificationsDashboards@2.11.1.0
observabilityDashboards@2.11.1.0
queryWorkbenchDashboards@2.11.1.0
reportsDashboards@2.11.1.0
searchRelevanceDashboards@2.11.1.0
securityAnalyticsDashboards@2.11.1.0
securityDashboards@2.11.1.0
Screenshots

I ran GET _cluster/health once and it tried to get the mapping three times.
Host/Environment (please complete the following information):
- OS: Ubuntu 22
- Browser and version: Google Chrome, Version: 119.0.6045.199
Additional context
If one user is using dev tools it makes dashboards run slow for all users.