Documentation
¶
Index ¶
Constants ¶
This section is empty.
Variables ¶
This section is empty.
Functions ¶
This section is empty.
Types ¶
type Condition ¶
type Condition struct {
// Type is the condition type. It should describe the condition of node in problem. For example
// KernelDeadlock, OutOfResource etc.
Type string `json:"type"`
// Status indicates whether the node is in the condition or not.
Status ConditionStatus `json:"status"`
// Transition is the time when the node transits to this condition.
Transition time.Time `json:"transition"`
// Reason is a short reason of why node goes into this condition.
Reason string `json:"reason"`
// Message is a human readable message of why node goes into this condition.
Message string `json:"message"`
}
Condition is the node condition used internally by problem detector.
type ConditionStatus ¶ added in v0.5.0
type ConditionStatus string
ConditionStatus is the status of the condition.
const ( // True means the condition status is true. True ConditionStatus = "True" // False means the condition status is false. False ConditionStatus = "False" // Unknown means the condition status is unknown. Unknown ConditionStatus = "Unknown" )
type Event ¶
type Event struct {
// Severity is the severity level of the event.
Severity Severity `json:"severity"`
// Timestamp is the time when the event is generated.
Timestamp time.Time `json:"timestamp"`
// Reason is a short reason of why the event is generated.
Reason string `json:"reason"`
// Message is a human readable message of why the event is generated.
Message string `json:"message"`
}
Event is the event used internally by node problem detector.
type Monitor ¶ added in v0.5.0
type Monitor interface {
// Start starts the log monitor.
Start() (<-chan *Status, error)
// Stop stops the log monitor.
Stop()
}
Monitor monitors log and custom plugins and reports node problem condition and event according to the rules.
type Severity ¶
type Severity string
Severity is the severity of the problem event. Now we only have 2 severity levels: Info and Warn, which are corresponding to the current kubernetes event types. We may want to add more severity levels in the future.
type Status ¶
type Status struct {
// Source is the name of the problem daemon.
Source string `json:"source"`
// Events are temporary node problem events. If the status is only a condition update,
// this field could be nil. Notice that the events should be sorted from oldest to newest.
Events []Event `json:"events"`
// Conditions are the permanent node conditions. The problem daemon should always report the
// newest node conditions in this field.
Conditions []Condition `json:"conditions"`
}
Status is the status other problem daemons should report to node problem detector.
Click to show internal directories.
Click to hide internal directories.