Central coordinator for the alert pipeline. More...

#include <alert_manager.h>

Collaboration diagram for kcenon::monitoring::alert_manager:

Public Types
using	metric_provider_func = std::function<std::optional<double>(const std::string&)>

Public Member Functions
	alert_manager ()
	Default constructor.

	alert_manager (const alert_manager_config &config)
	Construct with configuration.

	~alert_manager ()
	Destructor.

	alert_manager (const alert_manager &)=delete

alert_manager &	operator= (const alert_manager &)=delete

	alert_manager (alert_manager &&)=delete

alert_manager &	operator= (alert_manager &&)=delete

common::VoidResult	start ()
	Start the alert manager.

common::VoidResult	stop ()
	Stop the alert manager.

bool	is_running () const
	Check if manager is running.

common::VoidResult	add_rule (std::shared_ptr< alert_rule > rule)
	Add an alert rule.

common::VoidResult	remove_rule (const std::string &rule_name)
	Remove an alert rule.

std::shared_ptr< alert_rule >	get_rule (const std::string &rule_name) const
	Get a rule by name.

std::vector< std::shared_ptr< alert_rule > >	get_rules () const
	Get all rules.

common::VoidResult	add_rule_group (std::shared_ptr< alert_rule_group > group)
	Add a rule group.

common::VoidResult	process_metric (const std::string &metric_name, double value)
	Process a metric value.

common::VoidResult	process_metrics (const std::unordered_map< std::string, double > &metrics)
	Process a batch of metrics.

std::vector< alert >	get_active_alerts () const
	Get all active alerts.

std::optional< alert >	get_alert (const std::string &fingerprint) const
	Get alert by fingerprint.

common::VoidResult	resolve_alert (const std::string &fingerprint)
	Resolve an alert manually.

common::Result< uint64_t >	create_silence (const alert_silence &silence)
	Create a silence.

common::VoidResult	delete_silence (uint64_t silence_id)
	Delete a silence.

std::vector< alert_silence >	get_silences () const
	Get all active silences.

bool	is_silenced (const alert &a) const
	Check if an alert is silenced.

common::VoidResult	add_notifier (std::shared_ptr< alert_notifier > notifier)
	Add a notifier.

common::VoidResult	remove_notifier (const std::string &notifier_name)
	Remove a notifier.

std::vector< std::shared_ptr< alert_notifier > >	get_notifiers () const
	Get all notifiers.

void	set_metric_provider (metric_provider_func provider)
	Set the metric provider function.

void	set_event_bus (std::shared_ptr< interface_event_bus > event_bus)
	Set event bus for publishing alert events.

alert_manager_metrics	get_metrics () const
	Get manager metrics.

const alert_manager_config &	config () const
	Get configuration.

Private Member Functions
void	evaluation_loop ()
	Main evaluation loop.

void	evaluate_rule (const std::shared_ptr< alert_rule > &rule, double value)
	Evaluate a single rule.

void	update_alert_state (const std::string &fingerprint, bool condition_met, double value, const std::shared_ptr< alert_rule > &rule)
	Update alert state.

void	send_notifications (const alert &a)
	Send notifications for an alert.

void	cleanup_silences ()
	Clean up expired silences.

void	cleanup_resolved_alerts ()
	Clean up resolved alerts.

Private Attributes
alert_manager_config	config_

std::mutex	rules_mutex_

std::unordered_map< std::string, std::shared_ptr< alert_rule > >	rules_

std::vector< std::shared_ptr< alert_rule_group > >	rule_groups_

std::mutex	alerts_mutex_

std::unordered_map< std::string, alert >	alerts_

std::mutex	silences_mutex_

std::unordered_map< uint64_t, alert_silence >	silences_

std::mutex	notifiers_mutex_

std::vector< std::shared_ptr< alert_notifier > >	notifiers_

std::mutex	provider_mutex_

metric_provider_func	metric_provider_

std::shared_ptr< interface_event_bus >	event_bus_

alert_manager_metrics	metrics_

std::atomic< bool >	running_ {false}

std::thread	evaluation_thread_

std::condition_variable	cv_

std::mutex	cv_mutex_

std::unordered_map< std::string, std::chrono::steady_clock::time_point >	last_notification_times_

Detailed Description

Central coordinator for the alert pipeline.

The alert manager is responsible for:

Managing alert rules and their lifecycle
Evaluating rules against incoming metrics
Managing alert state transitions
Routing notifications to configured notifiers
Handling alert silencing and grouping

Thread Safety:: This class is thread-safe. All public methods can be called from multiple threads simultaneously.

Examples: alert_pipeline_example.cpp, and production_monitoring_example.cpp.

Definition at line 122 of file alert_manager.h.

Member Typedef Documentation

◆ metric_provider_func

using kcenon::monitoring::alert_manager::metric_provider_func = std::function<std::optional<double>(const std::string&)>

Examples: /home/runner/work/monitoring_system/monitoring_system/include/kcenon/monitoring/alert/alert_manager.h.

Definition at line 124 of file alert_manager.h.

Constructor & Destructor Documentation

◆ alert_manager() [1/4]

kcenon::monitoring::alert_manager::alert_manager ( )

Default constructor.

Examples: /home/runner/work/monitoring_system/monitoring_system/include/kcenon/monitoring/alert/alert_manager.h.

◆ alert_manager() [2/4]

kcenon::monitoring::alert_manager::alert_manager ( const alert_manager_config & config )

explicit

Construct with configuration.

Parameters

config Manager configuration

◆ ~alert_manager()

kcenon::monitoring::alert_manager::~alert_manager ( )

Destructor.

Examples: /home/runner/work/monitoring_system/monitoring_system/include/kcenon/monitoring/alert/alert_manager.h.

◆ alert_manager() [3/4]

kcenon::monitoring::alert_manager::alert_manager ( const alert_manager & )

delete

◆ alert_manager() [4/4]

kcenon::monitoring::alert_manager::alert_manager ( alert_manager && )

delete

Member Function Documentation

◆ add_notifier()

common::VoidResult kcenon::monitoring::alert_manager::add_notifier ( std::shared_ptr< alert_notifier > notifier )

Add a notifier.

Parameters

notifier Notifier to add

Returns: Result indicating success or failure

Examples: /home/runner/work/monitoring_system/monitoring_system/include/kcenon/monitoring/alert/alert_manager.h, and alert_pipeline_example.cpp.

Referenced by main().

Here is the caller graph for this function:

◆ add_rule()

common::VoidResult kcenon::monitoring::alert_manager::add_rule ( std::shared_ptr< alert_rule > rule )

Add an alert rule.

Parameters

rule	Rule to add

Returns: Result indicating success or failure

Examples: /home/runner/work/monitoring_system/monitoring_system/include/kcenon/monitoring/alert/alert_manager.h, and alert_pipeline_example.cpp.

Referenced by main(), AlertManagerProcessingTest::SetUp(), and TEST().

Here is the caller graph for this function:

◆ add_rule_group()

common::VoidResult kcenon::monitoring::alert_manager::add_rule_group ( std::shared_ptr< alert_rule_group > group )

Add a rule group.

Parameters

group Rule group to add

Returns: Result indicating success or failure

Examples: /home/runner/work/monitoring_system/monitoring_system/include/kcenon/monitoring/alert/alert_manager.h, and alert_pipeline_example.cpp.

Referenced by main(), and TEST().

Here is the caller graph for this function:

◆ cleanup_resolved_alerts()

void kcenon::monitoring::alert_manager::cleanup_resolved_alerts ( )

private

Clean up resolved alerts.

Examples: /home/runner/work/monitoring_system/monitoring_system/include/kcenon/monitoring/alert/alert_manager.h.

◆ cleanup_silences()

void kcenon::monitoring::alert_manager::cleanup_silences ( )

private

Clean up expired silences.

Examples: /home/runner/work/monitoring_system/monitoring_system/include/kcenon/monitoring/alert/alert_manager.h.

◆ config()

const alert_manager_config & kcenon::monitoring::alert_manager::config ( ) const

Get configuration.

Returns: Current configuration

Examples: /home/runner/work/monitoring_system/monitoring_system/include/kcenon/monitoring/alert/alert_manager.h.

Referenced by TEST(), TEST(), and TEST_F().

Here is the caller graph for this function:

◆ create_silence()

common::Result< uint64_t > kcenon::monitoring::alert_manager::create_silence ( const alert_silence & silence )

Create a silence.

Parameters

silence Silence configuration

Returns: Result containing silence ID

Examples: /home/runner/work/monitoring_system/monitoring_system/include/kcenon/monitoring/alert/alert_manager.h.

◆ delete_silence()

common::VoidResult kcenon::monitoring::alert_manager::delete_silence ( uint64_t silence_id )

Delete a silence.

Parameters

silence_id Silence ID

Returns: Result indicating success or failure

Examples: /home/runner/work/monitoring_system/monitoring_system/include/kcenon/monitoring/alert/alert_manager.h.

◆ evaluate_rule()

void kcenon::monitoring::alert_manager::evaluate_rule	(	const std::shared_ptr< alert_rule > &	rule,
		double	value )

private

Evaluate a single rule.

Parameters

rule	Rule to evaluate
value	Metric value

Examples: /home/runner/work/monitoring_system/monitoring_system/include/kcenon/monitoring/alert/alert_manager.h.

◆ evaluation_loop()

void kcenon::monitoring::alert_manager::evaluation_loop ( )

private

Main evaluation loop.

Examples: /home/runner/work/monitoring_system/monitoring_system/include/kcenon/monitoring/alert/alert_manager.h.

◆ get_active_alerts()

std::vector< alert > kcenon::monitoring::alert_manager::get_active_alerts ( ) const

Get all active alerts.

Returns: Vector of active alerts

Examples: /home/runner/work/monitoring_system/monitoring_system/include/kcenon/monitoring/alert/alert_manager.h, and alert_pipeline_example.cpp.

Referenced by main(), and print_active_alerts().

Here is the caller graph for this function:

◆ get_alert()

std::optional< alert > kcenon::monitoring::alert_manager::get_alert ( const std::string & fingerprint ) const

Get alert by fingerprint.

Parameters

fingerprint Alert fingerprint

Returns: Alert if found

Examples: /home/runner/work/monitoring_system/monitoring_system/include/kcenon/monitoring/alert/alert_manager.h.

◆ get_metrics()

alert_manager_metrics kcenon::monitoring::alert_manager::get_metrics ( ) const

Get manager metrics.

Returns: Current metrics

Examples: /home/runner/work/monitoring_system/monitoring_system/include/kcenon/monitoring/alert/alert_manager.h, and alert_pipeline_example.cpp.

Referenced by main(), TEST(), and TEST().

Here is the caller graph for this function:

◆ get_notifiers()

std::vector< std::shared_ptr< alert_notifier > > kcenon::monitoring::alert_manager::get_notifiers ( ) const

Get all notifiers.

Returns: Vector of notifiers

Examples: /home/runner/work/monitoring_system/monitoring_system/include/kcenon/monitoring/alert/alert_manager.h.

◆ get_rule()

std::shared_ptr< alert_rule > kcenon::monitoring::alert_manager::get_rule ( const std::string & rule_name ) const

Get a rule by name.

Parameters

rule_name Name of rule

Returns: Rule if found

Examples: /home/runner/work/monitoring_system/monitoring_system/include/kcenon/monitoring/alert/alert_manager.h.

◆ get_rules()

std::vector< std::shared_ptr< alert_rule > > kcenon::monitoring::alert_manager::get_rules ( ) const

Get all rules.

Returns: Vector of all rules

Examples: /home/runner/work/monitoring_system/monitoring_system/include/kcenon/monitoring/alert/alert_manager.h, and alert_pipeline_example.cpp.

Referenced by main().

Here is the caller graph for this function:

◆ get_silences()

std::vector< alert_silence > kcenon::monitoring::alert_manager::get_silences ( ) const

Get all active silences.

Returns: Vector of active silences

Examples: /home/runner/work/monitoring_system/monitoring_system/include/kcenon/monitoring/alert/alert_manager.h.

◆ is_running()

bool kcenon::monitoring::alert_manager::is_running ( ) const

Check if manager is running.

Returns: True if running

Examples: /home/runner/work/monitoring_system/monitoring_system/include/kcenon/monitoring/alert/alert_manager.h.

Referenced by TEST_F(), TEST_F(), TEST_F(), and TEST_F().

Here is the caller graph for this function:

◆ is_silenced()

bool kcenon::monitoring::alert_manager::is_silenced ( const alert & a ) const

Check if an alert is silenced.

Parameters

a	Alert to check

Returns: True if silenced

Examples: /home/runner/work/monitoring_system/monitoring_system/include/kcenon/monitoring/alert/alert_manager.h.

◆ operator=() [1/2]

alert_manager & kcenon::monitoring::alert_manager::operator= ( alert_manager && )

delete

◆ operator=() [2/2]

alert_manager & kcenon::monitoring::alert_manager::operator= ( const alert_manager & )

delete

Examples: /home/runner/work/monitoring_system/monitoring_system/include/kcenon/monitoring/alert/alert_manager.h.

◆ process_metric()

common::VoidResult kcenon::monitoring::alert_manager::process_metric	(	const std::string &	metric_name,
		double	value )

Process a metric value.

Parameters

metric_name	Name of the metric
value	Metric value

Returns: Result indicating success or failure

Examples: /home/runner/work/monitoring_system/monitoring_system/include/kcenon/monitoring/alert/alert_manager.h, and alert_pipeline_example.cpp.

Referenced by main(), and TEST().

Here is the caller graph for this function:

◆ process_metrics()

common::VoidResult kcenon::monitoring::alert_manager::process_metrics ( const std::unordered_map< std::string, double > & metrics )

Process a batch of metrics.

Parameters

metrics Map of metric names to values

Returns: Result indicating success or failure

Examples: /home/runner/work/monitoring_system/monitoring_system/include/kcenon/monitoring/alert/alert_manager.h.

◆ remove_notifier()

common::VoidResult kcenon::monitoring::alert_manager::remove_notifier ( const std::string & notifier_name )

Remove a notifier.

Parameters

notifier_name Name of notifier

Returns: Result indicating success or failure

Examples: /home/runner/work/monitoring_system/monitoring_system/include/kcenon/monitoring/alert/alert_manager.h.

◆ remove_rule()

common::VoidResult kcenon::monitoring::alert_manager::remove_rule ( const std::string & rule_name )

Remove an alert rule.

Parameters

rule_name Name of rule to remove

Returns: Result indicating success or failure

Examples: /home/runner/work/monitoring_system/monitoring_system/include/kcenon/monitoring/alert/alert_manager.h.

◆ resolve_alert()

common::VoidResult kcenon::monitoring::alert_manager::resolve_alert ( const std::string & fingerprint )

Resolve an alert manually.

Parameters

fingerprint Alert fingerprint

Returns: Result indicating success or failure

Examples: /home/runner/work/monitoring_system/monitoring_system/include/kcenon/monitoring/alert/alert_manager.h.

Referenced by TEST_F().

Here is the caller graph for this function:

◆ send_notifications()

void kcenon::monitoring::alert_manager::send_notifications ( const alert & a )

private

Send notifications for an alert.

Parameters

a	Alert to notify about

Examples: /home/runner/work/monitoring_system/monitoring_system/include/kcenon/monitoring/alert/alert_manager.h.

◆ set_event_bus()

void kcenon::monitoring::alert_manager::set_event_bus ( std::shared_ptr< interface_event_bus > event_bus )

Set event bus for publishing alert events.

Parameters

event_bus Event bus instance

Examples: /home/runner/work/monitoring_system/monitoring_system/include/kcenon/monitoring/alert/alert_manager.h.

Referenced by TEST().

Here is the caller graph for this function:

◆ set_metric_provider()

void kcenon::monitoring::alert_manager::set_metric_provider ( metric_provider_func provider )

Set the metric provider function.

Parameters

provider Function that returns metric values by name

Examples: /home/runner/work/monitoring_system/monitoring_system/include/kcenon/monitoring/alert/alert_manager.h.

Referenced by TEST().

Here is the caller graph for this function:

◆ start()

common::VoidResult kcenon::monitoring::alert_manager::start ( )

Start the alert manager.

Returns: Result indicating success or failure

Examples: /home/runner/work/monitoring_system/monitoring_system/include/kcenon/monitoring/alert/alert_manager.h, and alert_pipeline_example.cpp.

Referenced by main(), TEST_F(), TEST_F(), and TEST_F().

Here is the caller graph for this function:

◆ stop()

common::VoidResult kcenon::monitoring::alert_manager::stop ( )

Stop the alert manager.

Returns: Result indicating success or failure

Examples: /home/runner/work/monitoring_system/monitoring_system/include/kcenon/monitoring/alert/alert_manager.h, and alert_pipeline_example.cpp.

Referenced by main(), TEST_F(), TEST_F(), and TEST_F().

Here is the caller graph for this function:

◆ update_alert_state()

void kcenon::monitoring::alert_manager::update_alert_state	(	const std::string &	fingerprint,
		bool	condition_met,
		double	value,
		const std::shared_ptr< alert_rule > &	rule )

private

Update alert state.

Parameters

fingerprint	Alert fingerprint
condition_met	Whether condition is met
value	Current metric value

Examples: /home/runner/work/monitoring_system/monitoring_system/include/kcenon/monitoring/alert/alert_manager.h.

Member Data Documentation

◆ alerts_

std::unordered_map<std::string, alert> kcenon::monitoring::alert_manager::alerts_

private

Examples: /home/runner/work/monitoring_system/monitoring_system/include/kcenon/monitoring/alert/alert_manager.h.

Definition at line 372 of file alert_manager.h.

◆ alerts_mutex_

std::mutex kcenon::monitoring::alert_manager::alerts_mutex_

mutableprivate

Examples: /home/runner/work/monitoring_system/monitoring_system/include/kcenon/monitoring/alert/alert_manager.h.

Definition at line 371 of file alert_manager.h.

◆ config_

alert_manager_config kcenon::monitoring::alert_manager::config_

private

Examples: /home/runner/work/monitoring_system/monitoring_system/include/kcenon/monitoring/alert/alert_manager.h.

Definition at line 363 of file alert_manager.h.

◆ cv_

std::condition_variable kcenon::monitoring::alert_manager::cv_

private

Examples: /home/runner/work/monitoring_system/monitoring_system/include/kcenon/monitoring/alert/alert_manager.h.

Definition at line 395 of file alert_manager.h.

◆ cv_mutex_

std::mutex kcenon::monitoring::alert_manager::cv_mutex_

private

Examples: /home/runner/work/monitoring_system/monitoring_system/include/kcenon/monitoring/alert/alert_manager.h.

Definition at line 396 of file alert_manager.h.

◆ evaluation_thread_

std::thread kcenon::monitoring::alert_manager::evaluation_thread_

private

Examples: /home/runner/work/monitoring_system/monitoring_system/include/kcenon/monitoring/alert/alert_manager.h.

Definition at line 394 of file alert_manager.h.

◆ event_bus_

std::shared_ptr<interface_event_bus> kcenon::monitoring::alert_manager::event_bus_

private

Examples: /home/runner/work/monitoring_system/monitoring_system/include/kcenon/monitoring/alert/alert_manager.h.

Definition at line 387 of file alert_manager.h.

◆ last_notification_times_

std::unordered_map<std::string, std::chrono::steady_clock::time_point> kcenon::monitoring::alert_manager::last_notification_times_

private

Examples: /home/runner/work/monitoring_system/monitoring_system/include/kcenon/monitoring/alert/alert_manager.h.

Definition at line 399 of file alert_manager.h.

◆ metric_provider_

metric_provider_func kcenon::monitoring::alert_manager::metric_provider_

private

Examples: /home/runner/work/monitoring_system/monitoring_system/include/kcenon/monitoring/alert/alert_manager.h.

Definition at line 384 of file alert_manager.h.

◆ metrics_

alert_manager_metrics kcenon::monitoring::alert_manager::metrics_

mutableprivate

Examples: /home/runner/work/monitoring_system/monitoring_system/include/kcenon/monitoring/alert/alert_manager.h.

Definition at line 390 of file alert_manager.h.

◆ notifiers_

std::vector<std::shared_ptr<alert_notifier> > kcenon::monitoring::alert_manager::notifiers_

private

Examples: /home/runner/work/monitoring_system/monitoring_system/include/kcenon/monitoring/alert/alert_manager.h.

Definition at line 380 of file alert_manager.h.

◆ notifiers_mutex_

std::mutex kcenon::monitoring::alert_manager::notifiers_mutex_

mutableprivate

Examples: /home/runner/work/monitoring_system/monitoring_system/include/kcenon/monitoring/alert/alert_manager.h.

Definition at line 379 of file alert_manager.h.

◆ provider_mutex_

std::mutex kcenon::monitoring::alert_manager::provider_mutex_

private

Examples: /home/runner/work/monitoring_system/monitoring_system/include/kcenon/monitoring/alert/alert_manager.h.

Definition at line 383 of file alert_manager.h.

◆ rule_groups_

std::vector<std::shared_ptr<alert_rule_group> > kcenon::monitoring::alert_manager::rule_groups_

private

Examples: /home/runner/work/monitoring_system/monitoring_system/include/kcenon/monitoring/alert/alert_manager.h.

Definition at line 368 of file alert_manager.h.

◆ rules_

std::unordered_map<std::string, std::shared_ptr<alert_rule> > kcenon::monitoring::alert_manager::rules_

private

Examples: /home/runner/work/monitoring_system/monitoring_system/include/kcenon/monitoring/alert/alert_manager.h.

Definition at line 367 of file alert_manager.h.

◆ rules_mutex_

std::mutex kcenon::monitoring::alert_manager::rules_mutex_

mutableprivate

Examples: /home/runner/work/monitoring_system/monitoring_system/include/kcenon/monitoring/alert/alert_manager.h.

Definition at line 366 of file alert_manager.h.

◆ running_

std::atomic<bool> kcenon::monitoring::alert_manager::running_ {false}

private

Examples: /home/runner/work/monitoring_system/monitoring_system/include/kcenon/monitoring/alert/alert_manager.h.

Definition at line 393 of file alert_manager.h.

393{false};

◆ silences_

std::unordered_map<uint64_t, alert_silence> kcenon::monitoring::alert_manager::silences_

private

Examples: /home/runner/work/monitoring_system/monitoring_system/include/kcenon/monitoring/alert/alert_manager.h.

Definition at line 376 of file alert_manager.h.

◆ silences_mutex_

std::mutex kcenon::monitoring::alert_manager::silences_mutex_

mutableprivate

Examples: /home/runner/work/monitoring_system/monitoring_system/include/kcenon/monitoring/alert/alert_manager.h.

Definition at line 375 of file alert_manager.h.

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

include/kcenon/monitoring/alert/alert_manager.h

Public Types

Public Member Functions

Private Member Functions

Private Attributes

Detailed Description

Member Typedef Documentation

◆ metric_provider_func

Constructor & Destructor Documentation

◆ alert_manager() [1/4]

◆ alert_manager() [2/4]

◆ ~alert_manager()

◆ alert_manager() [3/4]

◆ alert_manager() [4/4]

Member Function Documentation

◆ add_notifier()

◆ add_rule()

◆ add_rule_group()

◆ cleanup_resolved_alerts()

◆ cleanup_silences()

◆ config()

◆ create_silence()

◆ delete_silence()

◆ evaluate_rule()

◆ evaluation_loop()

◆ get_active_alerts()

◆ get_alert()

◆ get_metrics()

◆ get_notifiers()

◆ get_rule()

◆ get_rules()

◆ get_silences()

◆ is_running()

◆ is_silenced()

◆ operator=() [1/2]

◆ operator=() [2/2]

◆ process_metric()

◆ process_metrics()

◆ remove_notifier()

◆ remove_rule()

◆ resolve_alert()

◆ send_notifications()

◆ set_event_bus()

◆ set_metric_provider()

◆ start()

◆ stop()

◆ update_alert_state()

Member Data Documentation

◆ alerts_

◆ alerts_mutex_

◆ config_

◆ cv_

◆ cv_mutex_

◆ evaluation_thread_

◆ event_bus_

◆ last_notification_times_

◆ metric_provider_

◆ metrics_

◆ notifiers_

◆ notifiers_mutex_

◆ provider_mutex_

◆ rule_groups_

◆ rules_

◆ rules_mutex_

◆ running_

◆ silences_

◆ silences_mutex_