fix(monitoring,remoteingress,web): Prune old metrics buckets periodically, clear metrics caches on shutdown, simplify edge disconnect handling, and optimize network view data updates
This commit is contained in:
@@ -1,5 +1,14 @@
|
||||
# Changelog
|
||||
|
||||
## 2026-02-21 - 7.4.2 - fix(monitoring,remoteingress,web)
|
||||
Prune old metrics buckets periodically, clear metrics caches on shutdown, simplify edge disconnect handling, and optimize network view data updates
|
||||
|
||||
- Call pruneOldBuckets() each minute to proactively remove stale time-series buckets in MetricsManager
|
||||
- Clear metricsCache, emailMinuteBuckets and dnsMinuteBuckets when MetricsManager stops to avoid stale state on shutdown
|
||||
- On edgeDisconnected remove the edgeStatuses entry instead of mutating an existing record (more explicit cleanup)
|
||||
- Remove unused traffic-timer variables and move requestsPerSec history updates from render() into updateNetworkData() to avoid unnecessary re-renders
|
||||
- Optimize traffic data array updates by shifting in-place then reassigning arrays to preserve Lit reactivity and reduce intermediate allocations
|
||||
|
||||
## 2026-02-21 - 7.4.1 - fix(dcrouter)
|
||||
replace console logging with structured logger, improve metrics logging, add terminal-ready wait in ops UI, bump dees-catalog patch
|
||||
|
||||
|
||||
Reference in New Issue
Block a user