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
|
// Copyright 2019 The Prometheus Authors
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
// +build !nopressure
package collector
import (
"fmt"
"github.com/go-kit/kit/log"
"github.com/go-kit/kit/log/level"
"github.com/prometheus/client_golang/prometheus"
"github.com/prometheus/procfs"
)
var (
psiResources = []string{"cpu", "io", "memory"}
)
type pressureStatsCollector struct {
cpu *prometheus.Desc
io *prometheus.Desc
ioFull *prometheus.Desc
mem *prometheus.Desc
memFull *prometheus.Desc
fs procfs.FS
logger log.Logger
}
func init() {
registerCollector("pressure", defaultEnabled, NewPressureStatsCollector)
}
// NewPressureStatsCollector returns a Collector exposing pressure stall information
func NewPressureStatsCollector(logger log.Logger) (Collector, error) {
fs, err := procfs.NewFS(*procPath)
if err != nil {
return nil, fmt.Errorf("failed to open procfs: %w", err)
}
return &pressureStatsCollector{
cpu: prometheus.NewDesc(
prometheus.BuildFQName(namespace, "pressure", "cpu_waiting_seconds_total"),
"Total time in seconds that processes have waited for CPU time",
nil, nil,
),
io: prometheus.NewDesc(
prometheus.BuildFQName(namespace, "pressure", "io_waiting_seconds_total"),
"Total time in seconds that processes have waited due to IO congestion",
nil, nil,
),
ioFull: prometheus.NewDesc(
prometheus.BuildFQName(namespace, "pressure", "io_stalled_seconds_total"),
"Total time in seconds no process could make progress due to IO congestion",
nil, nil,
),
mem: prometheus.NewDesc(
prometheus.BuildFQName(namespace, "pressure", "memory_waiting_seconds_total"),
"Total time in seconds that processes have waited for memory",
nil, nil,
),
memFull: prometheus.NewDesc(
prometheus.BuildFQName(namespace, "pressure", "memory_stalled_seconds_total"),
"Total time in seconds no process could make progress due to memory congestion",
nil, nil,
),
fs: fs,
logger: logger,
}, nil
}
// Update calls procfs.NewPSIStatsForResource for the different resources and updates the values
func (c *pressureStatsCollector) Update(ch chan<- prometheus.Metric) error {
for _, res := range psiResources {
level.Debug(c.logger).Log("msg", "collecting statistics for resource", "resource", res)
vals, err := c.fs.PSIStatsForResource(res)
if err != nil {
level.Debug(c.logger).Log("msg", "pressure information is unavailable, you need a Linux kernel >= 4.20 and/or CONFIG_PSI enabled for your kernel")
return nil
}
switch res {
case "cpu":
ch <- prometheus.MustNewConstMetric(c.cpu, prometheus.CounterValue, float64(vals.Some.Total)/1000.0/1000.0)
case "io":
ch <- prometheus.MustNewConstMetric(c.io, prometheus.CounterValue, float64(vals.Some.Total)/1000.0/1000.0)
ch <- prometheus.MustNewConstMetric(c.ioFull, prometheus.CounterValue, float64(vals.Full.Total)/1000.0/1000.0)
case "memory":
ch <- prometheus.MustNewConstMetric(c.mem, prometheus.CounterValue, float64(vals.Some.Total)/1000.0/1000.0)
ch <- prometheus.MustNewConstMetric(c.memFull, prometheus.CounterValue, float64(vals.Full.Total)/1000.0/1000.0)
default:
level.Debug(c.logger).Log("msg", "did not account for resource", "resource", res)
}
}
return nil
}
|
