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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224
|
//
// Copyright 2022 The Sigstore 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.
package ntpmonitor
import (
"errors"
"fmt"
"math/rand"
"sync/atomic"
"time"
"github.com/beevik/ntp"
pkgapi "github.com/sigstore/timestamp-authority/pkg/api"
"github.com/sigstore/timestamp-authority/pkg/log"
)
var (
// ErrInvTime indicates that the local time has drifted too much
// from the monitored NTP servers.
ErrInvTime = errors.New("local time differs from observed")
// ErrTooFewServers means that the number of trusted servers are
// smaller then the selected num servers to query.
ErrTooFewServers = errors.New("too few ntp servers configured")
// ErrNoResponse indicates that there is an error to communicate with
// the remote NTP servers
ErrNoResponse = errors.New("no ntp response")
// ErrThreshold means that there is no positive threshold value
ErrThreshold = errors.New("no valid server threshold set")
// ErrDeltaTooSmall is referring to when the max delta time is
// smaller than the request timeout which can give unstable behaviour.
ErrDeltaTooSmall = errors.New("delta is too small")
)
type serverResponses struct {
tooFewServerResponses bool
tooManyInvalidResponses bool
}
type NTPClient interface {
QueryWithOptions(srv string, opts ntp.QueryOptions) (*ntp.Response, error)
}
type LiveNTPClient struct{}
func (c LiveNTPClient) QueryWithOptions(srv string, opts ntp.QueryOptions) (*ntp.Response, error) {
return ntp.QueryWithOptions(srv, opts)
}
// NTPMonitor compares the local time with a set of trusted NTP servers.
type NTPMonitor struct {
cfg *Config
run atomic.Bool
ntpClient NTPClient
}
// New creates a NTPMonitor, reading the configuration from the provided
// path.
func New(configFile string) (*NTPMonitor, error) {
cfg, err := LoadConfig(configFile)
if err != nil {
return nil, err
}
return NewFromConfig(cfg)
}
// NewFromConfig creates a NTPMonitor from an instantiated configuration.
func NewFromConfig(cfg *Config) (*NTPMonitor, error) {
// default to using a live NTP client
liveNTPClient := LiveNTPClient{}
return NewFromConfigWithClient(cfg, liveNTPClient)
}
func NewFromConfigWithClient(cfg *Config, client NTPClient) (*NTPMonitor, error) {
if len(cfg.Servers) == 0 || len(cfg.Servers) < cfg.NumServers {
return nil, ErrTooFewServers
}
if cfg.ServerThreshold < 1 {
return nil, ErrThreshold
}
if cfg.ServerThreshold > cfg.NumServers {
return nil, ErrTooFewServers
}
if cfg.RequestTimeout < 1 || cfg.MaxTimeDelta < cfg.RequestTimeout {
return nil, ErrDeltaTooSmall
}
return &NTPMonitor{cfg: cfg, ntpClient: client}, nil
}
func (n *NTPMonitor) queryServers(delta time.Duration, servers []string) serverResponses {
validResponses := 0
noResponse := 0
for _, srv := range servers {
// Create a time interval from 'now' with the max
// time delta added/removed
// Make sure the time from the remote NTP server lies
// within this interval.
resp, err := n.queryNTPServer(srv)
if err != nil {
log.Logger.Errorf("ntp response timeout from %s",
srv)
noResponse++
continue
}
// ClockOffset is the estimated difference from
// local time to NTP server's time.
// The estimate assumes latency is similar for both
// sending and receiving data.
// The estimated offset does not depend on the value
// of the latency.
if resp.ClockOffset.Abs() > delta {
log.Logger.Warnf("local time is different from %s: %s",
srv, resp.Time)
} else {
validResponses++
}
}
// Did enough NTP servers respond?
return serverResponses{
tooFewServerResponses: n.cfg.ServerThreshold > n.cfg.NumServers-noResponse,
tooManyInvalidResponses: n.cfg.ServerThreshold > validResponses,
}
}
// Start the periodic monitor. Once started, it runs until Stop() is called,
func (n *NTPMonitor) Start() {
n.run.Store(true)
if n.cfg.RequestTimeout < 1 {
log.Logger.Warnf("NTP request timeout not set, default to 1s")
n.cfg.RequestTimeout = 1
}
delta := time.Duration(n.cfg.MaxTimeDelta) * time.Second
log.Logger.Info("ntp monitoring starting")
//nolint:gosec
r := rand.New(rand.NewSource(time.Now().UTC().UnixNano())) // initialize local pseudorandom generator //nolint:gosec
for n.run.Load() {
servers := RandomChoice(n.cfg.Servers, n.cfg.NumServers, r)
responses := n.queryServers(delta, servers)
// Did enough NTP servers respond?
if responses.tooFewServerResponses {
pkgapi.MetricNTPErrorCount.With(map[string]string{
"reason": "err_too_few",
}).Inc()
}
if responses.tooManyInvalidResponses {
pkgapi.MetricNTPErrorCount.With(map[string]string{
"reason": "err_inv_time",
}).Inc()
}
// Local time is in sync. Wait for next poll.
time.Sleep(time.Duration(n.cfg.Period) * time.Second)
}
log.Logger.Info("ntp monitoring stopped")
}
// Stop the monitoring.
func (n *NTPMonitor) Stop() {
log.Logger.Info("stopping ntp monitoring")
n.run.Store(false)
}
// queryNTPServer queries a provided ntp server, trying up to a configured
// amount of times. There is one second sleep between each attempt.
func (n *NTPMonitor) queryNTPServer(srv string) (*ntp.Response, error) {
var i = 1
for {
log.Logger.Debugf("querying ntp server %s", srv)
start := time.Now()
opts := ntp.QueryOptions{
Timeout: time.Duration(n.cfg.RequestTimeout) * time.Second,
}
resp, err := n.ntpClient.QueryWithOptions(srv, opts)
pkgapi.MetricNTPLatency.With(map[string]string{
"host": srv,
}).Observe(float64(time.Since(start)))
if err == nil {
pkgapi.MetricNTPSyncCount.With(map[string]string{
"failed": "false",
"host": srv,
}).Inc()
return resp, nil
}
pkgapi.MetricNTPSyncCount.With(map[string]string{
"failed": "true",
"host": srv,
}).Inc()
log.Logger.Infof("ntp timeout from %s, attempt %d/%d",
srv, i, n.cfg.RequestAttempts)
if i == n.cfg.RequestAttempts {
break
}
i++
time.Sleep(time.Second)
}
return nil, fmt.Errorf("ntp timeout: %s", srv)
}
|