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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452
|
/*
Copyright 2022 The Kubernetes 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 factory
import (
"context"
"errors"
"fmt"
"strings"
"sync"
"sync/atomic"
"testing"
"time"
clientv3 "go.etcd.io/etcd/client/v3"
"go.etcd.io/etcd/client/v3/kubernetes"
"k8s.io/apiserver/pkg/storage/etcd3/testserver"
"k8s.io/apiserver/pkg/storage/storagebackend"
)
type mockKV struct {
get func(ctx context.Context) (*clientv3.GetResponse, error)
}
func (mkv mockKV) Put(ctx context.Context, key, val string, opts ...clientv3.OpOption) (*clientv3.PutResponse, error) {
return nil, nil
}
func (mkv mockKV) Get(ctx context.Context, key string, opts ...clientv3.OpOption) (*clientv3.GetResponse, error) {
return mkv.get(ctx)
}
func (mockKV) Delete(ctx context.Context, key string, opts ...clientv3.OpOption) (*clientv3.DeleteResponse, error) {
return nil, nil
}
func (mockKV) Compact(ctx context.Context, rev int64, opts ...clientv3.CompactOption) (*clientv3.CompactResponse, error) {
return nil, nil
}
func (mockKV) Do(ctx context.Context, op clientv3.Op) (clientv3.OpResponse, error) {
return clientv3.OpResponse{}, nil
}
func (mockKV) Txn(ctx context.Context) clientv3.Txn {
return nil
}
func TestCreateHealthcheck(t *testing.T) {
etcdConfig := testserver.NewTestConfig(t)
client := testserver.RunEtcd(t, etcdConfig)
newETCD3ClientFn := newETCD3Client
defer func() {
newETCD3Client = newETCD3ClientFn
}()
tests := []struct {
name string
cfg storagebackend.Config
want error
responseTime time.Duration
}{
{
name: "ok if response time lower than default timeout",
cfg: storagebackend.Config{
Type: storagebackend.StorageTypeETCD3,
Transport: storagebackend.TransportConfig{},
},
responseTime: 1 * time.Second,
want: nil,
},
{
name: "ok if response time lower than custom timeout",
cfg: storagebackend.Config{
Type: storagebackend.StorageTypeETCD3,
Transport: storagebackend.TransportConfig{},
HealthcheckTimeout: 5 * time.Second,
},
responseTime: 3 * time.Second,
want: nil,
},
{
name: "timeouts if response time higher than default timeout",
cfg: storagebackend.Config{
Type: storagebackend.StorageTypeETCD3,
Transport: storagebackend.TransportConfig{},
},
responseTime: 3 * time.Second,
want: context.DeadlineExceeded,
},
{
name: "timeouts if response time higher than custom timeout",
cfg: storagebackend.Config{
Type: storagebackend.StorageTypeETCD3,
Transport: storagebackend.TransportConfig{},
HealthcheckTimeout: 3 * time.Second,
},
responseTime: 5 * time.Second,
want: context.DeadlineExceeded,
},
}
for _, tc := range tests {
t.Run(tc.name, func(t *testing.T) {
ready := make(chan struct{})
tc.cfg.Transport.ServerList = client.Endpoints()
newETCD3Client = func(c storagebackend.TransportConfig) (*kubernetes.Client, error) {
defer close(ready)
dummyKV := mockKV{
get: func(ctx context.Context) (*clientv3.GetResponse, error) {
select {
case <-ctx.Done():
return nil, ctx.Err()
case <-time.After(tc.responseTime):
return nil, nil
}
},
}
client.KV = dummyKV
return client, nil
}
stop := make(chan struct{})
defer close(stop)
healthcheck, err := CreateHealthCheck(tc.cfg, stop)
if err != nil {
t.Fatal(err)
}
// Wait for healthcheck to establish connection
<-ready
got := healthcheck()
if !errors.Is(got, tc.want) {
t.Errorf("healthcheck() missmatch want %v got %v", tc.want, got)
}
})
}
}
func TestCreateReadycheck(t *testing.T) {
etcdConfig := testserver.NewTestConfig(t)
client := testserver.RunEtcd(t, etcdConfig)
newETCD3ClientFn := newETCD3Client
defer func() {
newETCD3Client = newETCD3ClientFn
}()
tests := []struct {
name string
cfg storagebackend.Config
want error
responseTime time.Duration
}{
{
name: "ok if response time lower than default timeout",
cfg: storagebackend.Config{
Type: storagebackend.StorageTypeETCD3,
Transport: storagebackend.TransportConfig{},
},
responseTime: 1 * time.Second,
want: nil,
},
{
name: "ok if response time lower than custom timeout",
cfg: storagebackend.Config{
Type: storagebackend.StorageTypeETCD3,
Transport: storagebackend.TransportConfig{},
ReadycheckTimeout: 5 * time.Second,
},
responseTime: 3 * time.Second,
want: nil,
},
{
name: "timeouts if response time higher than default timeout",
cfg: storagebackend.Config{
Type: storagebackend.StorageTypeETCD3,
Transport: storagebackend.TransportConfig{},
},
responseTime: 3 * time.Second,
want: context.DeadlineExceeded,
},
{
name: "timeouts if response time higher than custom timeout",
cfg: storagebackend.Config{
Type: storagebackend.StorageTypeETCD3,
Transport: storagebackend.TransportConfig{},
ReadycheckTimeout: 3 * time.Second,
},
responseTime: 5 * time.Second,
want: context.DeadlineExceeded,
},
{
name: "timeouts if response time higher than default timeout with custom healthcheck timeout",
cfg: storagebackend.Config{
Type: storagebackend.StorageTypeETCD3,
Transport: storagebackend.TransportConfig{},
HealthcheckTimeout: 10 * time.Second,
},
responseTime: 3 * time.Second,
want: context.DeadlineExceeded,
},
}
for _, tc := range tests {
t.Run(tc.name, func(t *testing.T) {
ready := make(chan struct{})
tc.cfg.Transport.ServerList = client.Endpoints()
newETCD3Client = func(c storagebackend.TransportConfig) (*kubernetes.Client, error) {
defer close(ready)
dummyKV := mockKV{
get: func(ctx context.Context) (*clientv3.GetResponse, error) {
select {
case <-ctx.Done():
return nil, ctx.Err()
case <-time.After(tc.responseTime):
return nil, nil
}
},
}
client.KV = dummyKV
return client, nil
}
stop := make(chan struct{})
defer close(stop)
healthcheck, err := CreateReadyCheck(tc.cfg, stop)
if err != nil {
t.Fatal(err)
}
// Wait for healthcheck to establish connection
<-ready
got := healthcheck()
if !errors.Is(got, tc.want) {
t.Errorf("healthcheck() missmatch want %v got %v", tc.want, got)
}
})
}
}
func TestRateLimitHealthcheck(t *testing.T) {
etcdConfig := testserver.NewTestConfig(t)
client := testserver.RunEtcd(t, etcdConfig)
newETCD3ClientFn := newETCD3Client
defer func() {
newETCD3Client = newETCD3ClientFn
}()
cfg := storagebackend.Config{
Type: storagebackend.StorageTypeETCD3,
Transport: storagebackend.TransportConfig{},
HealthcheckTimeout: 5 * time.Second,
}
cfg.Transport.ServerList = client.Endpoints()
tests := []struct {
name string
want error
}{
{
name: "etcd ok",
},
{
name: "etcd down",
want: errors.New("etcd down"),
},
}
for _, tc := range tests {
t.Run(tc.name, func(t *testing.T) {
ready := make(chan struct{})
var counter uint64
newETCD3Client = func(c storagebackend.TransportConfig) (*kubernetes.Client, error) {
defer close(ready)
dummyKV := mockKV{
get: func(ctx context.Context) (*clientv3.GetResponse, error) {
atomic.AddUint64(&counter, 1)
select {
case <-ctx.Done():
return nil, ctx.Err()
default:
return nil, tc.want
}
},
}
client.KV = dummyKV
return client, nil
}
stop := make(chan struct{})
defer close(stop)
healthcheck, err := CreateHealthCheck(cfg, stop)
if err != nil {
t.Fatal(err)
}
// Wait for healthcheck to establish connection
<-ready
// run a first request to obtain the state
err = healthcheck()
if !errors.Is(err, tc.want) {
t.Errorf("healthcheck() mismatch want %v got %v", tc.want, err)
}
// run multiple request in parallel, they should have the same state that the first one
var wg sync.WaitGroup
for i := 0; i < 100; i++ {
wg.Add(1)
go func() {
defer wg.Done()
err := healthcheck()
if !errors.Is(err, tc.want) {
t.Errorf("healthcheck() mismatch want %v got %v", tc.want, err)
}
}()
}
// check the counter once the requests have finished
wg.Wait()
if counter != 1 {
t.Errorf("healthcheck() called etcd %d times, expected only one call", counter)
}
// wait until the rate limit allows new connections
time.Sleep(cfg.HealthcheckTimeout / 2)
// a new run on request should increment the counter only once
// run multiple request in parallel, they should have the same state that the first one
for i := 0; i < 100; i++ {
wg.Add(1)
go func() {
defer wg.Done()
err := healthcheck()
if !errors.Is(err, tc.want) {
t.Errorf("healthcheck() mismatch want %v got %v", tc.want, err)
}
}()
}
wg.Wait()
if counter != 2 {
t.Errorf("healthcheck() called etcd %d times, expected only two calls", counter)
}
})
}
}
func TestTimeTravelHealthcheck(t *testing.T) {
etcdConfig := testserver.NewTestConfig(t)
client := testserver.RunEtcd(t, etcdConfig)
newETCD3ClientFn := newETCD3Client
defer func() {
newETCD3Client = newETCD3ClientFn
}()
cfg := storagebackend.Config{
Type: storagebackend.StorageTypeETCD3,
Transport: storagebackend.TransportConfig{},
HealthcheckTimeout: 5 * time.Second,
}
cfg.Transport.ServerList = client.Endpoints()
ready := make(chan struct{})
signal := make(chan struct{})
var counter uint64
newETCD3Client = func(c storagebackend.TransportConfig) (*kubernetes.Client, error) {
defer close(ready)
dummyKV := mockKV{
get: func(ctx context.Context) (*clientv3.GetResponse, error) {
atomic.AddUint64(&counter, 1)
val := atomic.LoadUint64(&counter)
// the first request wait for a custom timeout to trigger an error.
// We don't use the context timeout because we want to check that
// the cached answer is not overridden, and since the rate limit is
// based on cfg.HealthcheckTimeout / 2, the timeout will race with
// the race limiter to server the new request from the cache or allow
// it to go through
if val == 1 {
select {
case <-ctx.Done():
return nil, ctx.Err()
case <-time.After((2 * cfg.HealthcheckTimeout) / 3):
return nil, fmt.Errorf("etcd down")
}
}
// subsequent requests will always work
return nil, nil
},
}
client.KV = dummyKV
return client, nil
}
stop := make(chan struct{})
defer close(stop)
healthcheck, err := CreateHealthCheck(cfg, stop)
if err != nil {
t.Fatal(err)
}
// Wait for healthcheck to establish connection
<-ready
// run a first request that fails after 2 seconds
go func() {
err := healthcheck()
if !strings.Contains(err.Error(), "etcd down") {
t.Errorf("healthcheck() mismatch want %v got %v", fmt.Errorf("etcd down"), err)
}
close(signal)
}()
// wait until the rate limit allows new connections
time.Sleep(cfg.HealthcheckTimeout / 2)
select {
case <-signal:
t.Errorf("first request should not return yet")
default:
}
// a new run on request should succeed and increment the counter
err = healthcheck()
if err != nil {
t.Errorf("unexpected error: %v", err)
}
c := atomic.LoadUint64(&counter)
if c != 2 {
t.Errorf("healthcheck() called etcd %d times, expected only two calls", c)
}
// cached request should be success and not be overridden by the late error
<-signal
err = healthcheck()
if err != nil {
t.Errorf("unexpected error: %v", err)
}
c = atomic.LoadUint64(&counter)
if c != 2 {
t.Errorf("healthcheck() called etcd %d times, expected only two calls", c)
}
}
|