1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
|
/**
* Copyright (C) 2021-present MongoDB, Inc.
*
* This program is free software: you can redistribute it and/or modify
* it under the terms of the Server Side Public License, version 1,
* as published by MongoDB, Inc.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* Server Side Public License for more details.
*
* You should have received a copy of the Server Side Public License
* along with this program. If not, see
* <http://www.mongodb.com/licensing/server-side-public-license>.
*
* As a special exception, the copyright holders give permission to link the
* code of portions of this program with the OpenSSL library under certain
* conditions as described in each individual source file and distribute
* linked combinations including the program with the OpenSSL library. You
* must comply with the Server Side Public License in all respects for
* all of the code used other than as permitted herein. If you modify file(s)
* with this exception, you may extend this exception to your version of the
* file(s), but you are not obligated to do so. If you do not wish to do so,
* delete this exception statement from your version. If you delete this
* exception statement from all source files in the program, then also delete
* it in the license file.
*/
#pragma once
#include <memory>
#include "mongo/db/process_health/fault.h"
#include "mongo/db/process_health/fault_facet.h"
#include "mongo/db/process_health/fault_facet_container.h"
#include "mongo/db/process_health/fault_manager_config.h"
#include "mongo/db/process_health/health_monitoring_server_parameters_gen.h"
#include "mongo/db/process_health/health_observer.h"
#include "mongo/db/process_health/progress_monitor.h"
#include "mongo/db/process_health/state_machine.h"
#include "mongo/db/service_context.h"
#include "mongo/executor/task_executor.h"
#include "mongo/platform/atomic_word.h"
#include "mongo/platform/mutex.h"
namespace mongo {
namespace process_health {
/**
* FaultManager is a singleton constantly monitoring the health of the
* system.
*
* It supports pluggable 'HealthObservers', which are responsible to check various
* aspects of the server health. The aggregate outcome of all periodic checks
* is accesible as FaultState.
*
* If an active fault state persists, FaultManager will terminate the server process.
*/
class FaultManager : protected StateMachine<HealthCheckStatus, FaultState>,
protected FaultFacetsContainerFactory {
FaultManager(const FaultManager&) = delete;
FaultManager& operator=(const FaultManager&) = delete;
public:
// The taskExecutor provided should not be already started.
FaultManager(ServiceContext* svcCtx,
std::shared_ptr<executor::TaskExecutor> taskExecutor,
std::unique_ptr<FaultManagerConfig> config);
// Variant with explicit crash callback for tests.
FaultManager(ServiceContext* svcCtx,
std::shared_ptr<executor::TaskExecutor> taskExecutor,
std::unique_ptr<FaultManagerConfig> config,
std::function<void(std::string cause)> crashCb);
virtual ~FaultManager();
void setupStateMachine();
boost::optional<FaultState> handleStartupCheck(const OptionalMessageType& message);
boost::optional<FaultState> handleOk(const OptionalMessageType& message);
boost::optional<FaultState> handleTransientFault(const OptionalMessageType& message);
boost::optional<FaultState> handleActiveFault(const OptionalMessageType& message);
void setInitialHealthCheckComplete(FaultState, FaultState, const OptionalMessageType&);
void logCurrentState(FaultState, FaultState, const OptionalMessageType&);
void logMessageReceived(FaultState state, const HealthCheckStatus& status);
void setTransientFaultDeadline(FaultState, FaultState, const OptionalMessageType&);
void clearTransientFaultDeadline(FaultState, FaultState, const OptionalMessageType&);
// Start periodic health checks, invoke it once during server startup.
// It is unsafe to start health checks immediately during ServiceContext creation
// because some ServiceContext fields might not be initialized yet.
// Health checks cannot be stopped but could be effectively disabled with health-checker
// specific flags.
SharedSemiFuture<void> startPeriodicHealthChecks();
bool isInitialized();
static FaultManager* get(ServiceContext* svcCtx);
// Replace the FaultManager for the 'svcCtx'. This functionality
// is exposed for testing and initial bootstrap.
static void set(ServiceContext* svcCtx, std::unique_ptr<FaultManager> newFaultManager);
// Signals that the intensity for a health observer has been updated.
static void healthMonitoringIntensitiesUpdated(HealthObserverIntensities oldValue,
HealthObserverIntensities newValue);
// Returns the current fault state for the server.
FaultState getFaultState() const;
// Returns the current fault, if any. Otherwise returns an empty pointer.
FaultConstPtr currentFault() const;
// All observers remain valid for the manager lifetime, thus returning
// just pointers is safe, as long as they are used while manager exists.
std::vector<HealthObserver*> getHealthObservers() const;
// Gets the aggregate configuration for all process health environment.
FaultManagerConfig getConfig() const;
// Gets the timestamp of the last transition
Date_t getLastTransitionTime() const;
protected:
// Returns all health observers not configured as Off
std::vector<HealthObserver*> getActiveHealthObservers() const;
HealthObserver* getHealthObserver(FaultFacetType type) const;
// Runs a particular health observer. Then attempts to transition states. Then schedules next
// run.
virtual void healthCheck(HealthObserver* observer, CancellationToken token);
// Protected interface FaultFacetsContainerFactory implementation.
// The interface FaultFacetsContainerFactory is implemented by the member '_fault'.
FaultFacetsContainerPtr getFaultFacetsContainer() const override;
FaultFacetsContainerPtr getOrCreateFaultFacetsContainer() override;
void updateWithCheckStatus(HealthCheckStatus&& checkStatus) override;
void schedulePeriodicHealthCheckThread();
// TODO: move this into fault class; refactor to remove FaultInternal
bool hasCriticalFacet(const FaultInternal* fault) const;
void progressMonitorCheckForTests(std::function<void(std::string cause)> crashCb);
private:
// One time init.
void _init();
std::unique_ptr<FaultManagerConfig> _config;
ServiceContext* const _svcCtx;
std::shared_ptr<executor::TaskExecutor> _taskExecutor;
// Callback used to crash the server.
std::function<void(std::string cause)> _crashCb;
mutable Mutex _mutex =
MONGO_MAKE_LATCH(HierarchicalAcquisitionLevel(5), "FaultManager::_mutex");
std::shared_ptr<FaultInternal> _fault;
// This source is canceled before the _taskExecutor shutdown(). It
// can be used to check for the start of the shutdown sequence.
CancellationSource _managerShuttingDownCancellationSource;
// Manager owns all observer instances.
std::vector<std::unique_ptr<HealthObserver>> _observers;
SharedPromise<void> _initialHealthCheckCompletedPromise;
// Protects the state below.
mutable Mutex _stateMutex =
MONGO_MAKE_LATCH(HierarchicalAcquisitionLevel(0), "FaultManager::_stateMutex");
bool _initialized = false;
Date_t _lastTransitionTime;
// Responsible for transitioning the state of FaultManager to ActiveFault after a
// timeout while in the TransientFault state. The timer is canceled when the instance is
// destroyed.
struct TransientFaultDeadline {
TransientFaultDeadline() = delete;
TransientFaultDeadline(FaultManager* faultManager,
std::shared_ptr<executor::TaskExecutor> executor,
Milliseconds timeout);
virtual ~TransientFaultDeadline();
protected:
// Cancel timer for transitioning to ActiveFault
CancellationSource cancelActiveFaultTransition;
// Fufilled when we should transition to ActiveFault
ExecutorFuture<void> activeFaultTransition;
};
std::unique_ptr<TransientFaultDeadline> _transientFaultDeadline;
std::unique_ptr<ProgressMonitor> _progressMonitor;
stdx::unordered_set<FaultFacetType> _healthyObservations;
struct HealthCheckContext {
std::unique_ptr<ExecutorFuture<HealthCheckStatus>> result;
boost::optional<executor::TaskExecutor::CallbackHandle> resultStatus;
HealthCheckContext(std::unique_ptr<ExecutorFuture<HealthCheckStatus>> future,
boost::optional<executor::TaskExecutor::CallbackHandle> cbHandle)
: result(std::move(future)), resultStatus(cbHandle){};
};
stdx::unordered_map<FaultFacetType, HealthCheckContext> _healthCheckContexts;
};
} // namespace process_health
} // namespace mongo
|
