Very large node caches can cause telemetry adapters to fail on Sentinel

Description

dataSourceSync: initialized list of managed IP addresses with 334146 members

While initializing this cache, this code block takes roughly two minutes, and the ElasticFlowRepository is taking approximately three minutes for the bundles to fully start. This is long enough that bundles which depend on these will time out:

 

The current workaround for this issue is to restart the bundles:

 

While moving the telemetry configs away and back again.

Environment

29.0.4

Acceptance / Success Criteria

Large node caches no longer block startup

Lucidchart Diagrams

Activity

Show:

Chandra Gorantla January 7, 2022 at 5:40 PM

Fixed

Details

Assignee

Reporter

FD#

Components

Sprint

Fix versions

Affects versions

Priority

PagerDuty

Created December 15, 2021 at 3:38 PM
Updated February 2, 2022 at 2:29 PM
Resolved January 7, 2022 at 5:40 PM