MetricsRegistry Class Reference

Lock-free accumulation of command processing statistics. More...

#include <metrics_registry.h>

Collaboration diagram for MetricsRegistry:

Classes
struct	ActiveTypeStats
struct	GameStartSnapshot
struct	TypeStats

Public Member Functions
void	observeCommand (int typeId, qint64 elapsedMs)
void	observeGameStartDurationMs (qint64 elapsedMs)
qint64	totalCommands () const
	Total number of observed commands across all types.
qint64	totalTimeMs () const
	Cumulative processing milliseconds across all types.
int	activeTypeCount () const
	Number of distinct type slots that have seen at least one sample.
QList< ActiveTypeStats >	collectActiveStats () const
GameStartSnapshot	getGameStartSnapshot () const

Static Public Member Functions
static int	typeIdFor (int kindIndex, int extensionNumber)
	Guard against typeIdFor() overflowing into the neighbouring kind's slots.

Static Public Attributes
static constexpr int	KindStride = 1280
static constexpr int	NumKinds = 6
static constexpr const char *	KindNames [NumKinds] = {"session", "room", "game", "moderator", "admin", "developer"}
static constexpr int	MaxTypes = NumKinds * KindStride
	Upper bound on distinct command type ids (see typeIdFor).

Private Member Functions
TypeStats &	slotFor (int typeId)
const TypeStats &	slotFor (int typeId) const

Private Attributes
std::array< TypeStats, MaxTypes >	typeSlots {}
TypeStats	gameStartStats {}
std::atomic< qint64 >	totalCommandsCounter {0}
std::atomic< qint64 >	totalTimeCounter {0}

Detailed Description

Lock-free accumulation of command processing statistics.

observeCommand() is called once per processed command from whichever socket thread handled it. It uses relaxed atomic adds on preallocated storage only, so it introduces no locks, allocations, or shared cache-line ping-pong beyond the unavoidable counter updates.

Reading happens rarely (metrics scraping), accepts momentary tears between related counters, and therefore also needs no synchronization.

Only counts and totals are retained. An earlier Prometheus-style cumulative histogram (per-type, time-bucketed) was cut because nothing in the server ever wrote it out; it belongs to the future /metrics exporter that needs it.

Member Function Documentation

activeTypeCount()

int MetricsRegistry::activeTypeCount

(

)

const

Number of distinct type slots that have seen at least one sample.

Here is the call graph for this function:

Here is the caller graph for this function:

collectActiveStats()

QList< MetricsRegistry::ActiveTypeStats > MetricsRegistry::collectActiveStats

(

)

const

Returns stats for every type slot that has seen at least one sample. Callers resolve the numeric type id to a human-readable label via typeIdFor()/KindNames as needed.

Here is the call graph for this function:

Here is the caller graph for this function:

getGameStartSnapshot()

MetricsRegistry::GameStartSnapshot MetricsRegistry::getGameStartSnapshot

(

)

const

Here is the caller graph for this function:

observeCommand()

void MetricsRegistry::observeCommand	(	int	typeId,
		qint64	elapsedMs )

Here is the call graph for this function:

Here is the caller graph for this function:

observeGameStartDurationMs()

void MetricsRegistry::observeGameStartDurationMs

(

qint64

elapsedMs

)

Records how long one game start took to bring every player's zones online. Kept separate from command timings because it is triggered by the server itself and can dwarf any single command when decks are huge.

Here is the caller graph for this function:

slotFor() [1/2]

MetricsRegistry::TypeStats & MetricsRegistry::slotFor ( int typeId )

private

Here is the caller graph for this function:

slotFor() [2/2]

const MetricsRegistry::TypeStats & MetricsRegistry::slotFor ( int typeId ) const

private

totalCommands()

qint64 MetricsRegistry::totalCommands ( ) const

inline

Total number of observed commands across all types.

Here is the caller graph for this function:

totalTimeMs()

qint64 MetricsRegistry::totalTimeMs ( ) const

inline

Cumulative processing milliseconds across all types.

Here is the caller graph for this function:

typeIdFor()

int MetricsRegistry::typeIdFor ( int kindIndex, int extensionNumber )

inlinestatic

Guard against typeIdFor() overflowing into the neighbouring kind's slots.

Here is the caller graph for this function:

Member Data Documentation

gameStartStats

TypeStats MetricsRegistry::gameStartStats {}

private

KindNames

const char* MetricsRegistry::KindNames[NumKinds] = {"session", "room", "game", "moderator", "admin", "developer"}

staticconstexpr

KindStride

int MetricsRegistry::KindStride = 1280

staticconstexpr

Extension numbers are only unique per command kind, so recorded ids combine the kind index with the protobuf extension number.

The stride is only as wide as it needs to be: 1280 is the first round number above the largest extension actually in use (ModeratorCommand = 1206) and keeps the preallocated TypeStats array small. Bump it if a new command exceeds it.

MaxTypes

int MetricsRegistry::MaxTypes = NumKinds * KindStride

staticconstexpr

Upper bound on distinct command type ids (see typeIdFor).

NumKinds

int MetricsRegistry::NumKinds = 6

staticconstexpr

totalCommandsCounter

std::atomic<qint64> MetricsRegistry::totalCommandsCounter {0}

private

totalTimeCounter

std::atomic<qint64> MetricsRegistry::totalTimeCounter {0}

private

typeSlots

std::array<TypeStats, MaxTypes> MetricsRegistry::typeSlots {}

private

---

The documentation for this class was generated from the following files:

• servatrice/src/metrics_registry.h
• servatrice/src/metrics_registry.cpp