dohertj2/mxaccessgw
1
0
Fork 0
Code Issues Pull Requests 2 Actions 29 Packages Projects Releases Wiki Activity

metrics(alarms): expose provider-switch count in snapshot, bound the reason tag

Browse Source 
B1: add AlarmProviderSwitchCount to GatewayMetricsSnapshot so the switch total is
readable without scraping the OTEL counter.
B2: replace the free-text reason tag on mxgateway.alarms.provider_switches with a
bounded AlarmProviderSwitchReason enum (failover/failback/unknown); the human-readable
reason stays in the structured log.
This commit is contained in:
Joseph Doherty
2026-06-14 02:33:02 -04:00
parent 5b31e99ab6
commit 56abd64c6c
5 changed files with 47 additions and 8 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/ZB.MOM.WW.MxGateway.Server/Metrics/GatewayMetrics.cs
+15 -4
View File
@@ -50,6 +50,7 @@ public sealed class GatewayMetrics : IDisposable
private long _heartbeatFailures;
private long _streamDisconnects;
private long _retryAttempts;
private long _alarmProviderSwitches;
private bool _disposed;
/// <summary>
@@ -383,25 +384,34 @@ public sealed class GatewayMetrics : IDisposable
}
/// <summary>
/// Records that the alarm provider switched modes and updates the current provider mode gauge.
/// Records that the alarm provider switched modes, increments the switch count, and updates the
/// current provider mode gauge.
/// </summary>
/// <param name="fromMode">Provider mode before the switch (1=alarmmgr, 2=subtag, 0=unknown).</param>
/// <param name="toMode">Provider mode after the switch (1=alarmmgr, 2=subtag, 0=unknown).</param>
/// <param name="reason">Human-readable reason for the switch.</param>
public void AlarmProviderSwitched(int fromMode, int toMode, string reason)
/// <param name="reason">Bounded switch classification used as the counter's <c>reason</c> tag.</param>
public void AlarmProviderSwitched(int fromMode, int toMode, AlarmProviderSwitchReason reason)
{
lock (_syncRoot)
{
_alarmProviderMode = toMode;
_alarmProviderSwitches++;
}
_alarmProviderSwitchesCounter.Add(
1,
new KeyValuePair<string, object?>("from", fromMode.ToString(CultureInfo.InvariantCulture)),
new KeyValuePair<string, object?>("to", toMode.ToString(CultureInfo.InvariantCulture)),
new KeyValuePair<string, object?>("reason", reason));
new KeyValuePair<string, object?>("reason", ReasonTag(reason)));
}
private static string ReasonTag(AlarmProviderSwitchReason reason) => reason switch
{
AlarmProviderSwitchReason.Failover => "failover",
AlarmProviderSwitchReason.Failback => "failback",
_ => "unknown",
};
/// <summary>Sets the current alarm provider-mode gauge without recording a switch (e.g. startup baseline).</summary>
public void SetAlarmProviderMode(int mode)
{
@@ -433,6 +443,7 @@ public sealed class GatewayMetrics : IDisposable
HeartbeatFailures: _heartbeatFailures,
StreamDisconnects: _streamDisconnects,
RetryAttempts: _retryAttempts,
AlarmProviderSwitchCount: _alarmProviderSwitches,
CommandFailuresByMethod: new Dictionary<string, long>(_commandFailuresByMethod, StringComparer.OrdinalIgnoreCase),
EventsByFamily: new Dictionary<string, long>(_eventsByFamily, StringComparer.OrdinalIgnoreCase),
EventsBySession: new Dictionary<string, long>(_eventsBySession, StringComparer.Ordinal),