services/crunch-run/crunchrun.go

   1 package main
   2
   3 import (
   4         "encoding/json"
   5         "errors"
   6         "flag"
   7         "fmt"
   8         "io"
   9         "io/ioutil"
  10         "log"
  11         "os"
  12         "os/exec"
  13         "os/signal"
  14         "path"
  15         "path/filepath"
  16         "sort"
  17         "strings"
  18         "sync"
  19         "syscall"
  20         "time"
  21
  22         "git.curoverse.com/arvados.git/lib/crunchstat"
  23         "git.curoverse.com/arvados.git/sdk/go/arvados"
  24         "git.curoverse.com/arvados.git/sdk/go/arvadosclient"
  25         "git.curoverse.com/arvados.git/sdk/go/keepclient"
  26         "git.curoverse.com/arvados.git/sdk/go/manifest"
  27         "github.com/curoverse/dockerclient"
  28 )
  29
  30 // IArvadosClient is the minimal Arvados API methods used by crunch-run.
  31 type IArvadosClient interface {
  32         Create(resourceType string, parameters arvadosclient.Dict, output interface{}) error
  33         Get(resourceType string, uuid string, parameters arvadosclient.Dict, output interface{}) error
  34         Update(resourceType string, uuid string, parameters arvadosclient.Dict, output interface{}) error
  35         Call(method, resourceType, uuid, action string, parameters arvadosclient.Dict, output interface{}) error
  36         CallRaw(method string, resourceType string, uuid string, action string, parameters arvadosclient.Dict) (reader io.ReadCloser, err error)
  37         Discovery(key string) (interface{}, error)
  38 }
  39
  40 // ErrCancelled is the error returned when the container is cancelled.
  41 var ErrCancelled = errors.New("Cancelled")
  42
  43 // IKeepClient is the minimal Keep API methods used by crunch-run.
  44 type IKeepClient interface {
  45         PutHB(hash string, buf []byte) (string, int, error)
  46         ManifestFileReader(m manifest.Manifest, filename string) (keepclient.Reader, error)
  47 }
  48
  49 // NewLogWriter is a factory function to create a new log writer.
  50 type NewLogWriter func(name string) io.WriteCloser
  51
  52 type RunArvMount func(args []string, tok string) (*exec.Cmd, error)
  53
  54 type MkTempDir func(string, string) (string, error)
  55
  56 // ThinDockerClient is the minimal Docker client interface used by crunch-run.
  57 type ThinDockerClient interface {
  58         StopContainer(id string, timeout int) error
  59         InspectImage(id string) (*dockerclient.ImageInfo, error)
  60         LoadImage(reader io.Reader) error
  61         CreateContainer(config *dockerclient.ContainerConfig, name string, authConfig *dockerclient.AuthConfig) (string, error)
  62         StartContainer(id string, config *dockerclient.HostConfig) error
  63         AttachContainer(id string, options *dockerclient.AttachOptions) (io.ReadCloser, error)
  64         Wait(id string) <-chan dockerclient.WaitResult
  65         RemoveImage(name string, force bool) ([]*dockerclient.ImageDelete, error)
  66 }
  67
  68 // ContainerRunner is the main stateful struct used for a single execution of a
  69 // container.
  70 type ContainerRunner struct {
  71         Docker    ThinDockerClient
  72         ArvClient IArvadosClient
  73         Kc        IKeepClient
  74         arvados.Container
  75         dockerclient.ContainerConfig
  76         dockerclient.HostConfig
  77         token       string
  78         ContainerID string
  79         ExitCode    *int
  80         NewLogWriter
  81         loggingDone   chan bool
  82         CrunchLog     *ThrottledLogger
  83         Stdout        io.WriteCloser
  84         Stderr        *ThrottledLogger
  85         LogCollection *CollectionWriter
  86         LogsPDH       *string
  87         RunArvMount
  88         MkTempDir
  89         ArvMount       *exec.Cmd
  90         ArvMountPoint  string
  91         HostOutputDir  string
  92         CleanupTempDir []string
  93         Binds          []string
  94         OutputPDH      *string
  95         SigChan        chan os.Signal
  96         ArvMountExit   chan error
  97         finalState     string
  98
  99         statLogger   io.WriteCloser
 100         statReporter *crunchstat.Reporter
 101         statInterval time.Duration
 102         cgroupRoot   string
 103         // What we expect the container's cgroup parent to be.
 104         expectCgroupParent string
 105         // What we tell docker to use as the container's cgroup
 106         // parent. Note: Ideally we would use the same field for both
 107         // expectCgroupParent and setCgroupParent, and just make it
 108         // default to "docker". However, when using docker < 1.10 with
 109         // systemd, specifying a non-empty cgroup parent (even the
 110         // default value "docker") hits a docker bug
 111         // (https://github.com/docker/docker/issues/17126). Using two
 112         // separate fields makes it possible to use the "expect cgroup
 113         // parent to be X" feature even on sites where the "specify
 114         // cgroup parent" feature breaks.
 115         setCgroupParent string
 116
 117         cStateLock sync.Mutex
 118         cStarted   bool // StartContainer() succeeded
 119         cCancelled bool // StopContainer() invoked
 120 }
 121
 122 // SetupSignals sets up signal handling to gracefully terminate the underlying
 123 // Docker container and update state when receiving a TERM, INT or QUIT signal.
 124 func (runner *ContainerRunner) SetupSignals() {
 125         runner.SigChan = make(chan os.Signal, 1)
 126         signal.Notify(runner.SigChan, syscall.SIGTERM)
 127         signal.Notify(runner.SigChan, syscall.SIGINT)
 128         signal.Notify(runner.SigChan, syscall.SIGQUIT)
 129
 130         go func(sig chan os.Signal) {
 131                 <-sig
 132                 runner.stop()
 133                 signal.Stop(sig)
 134         }(runner.SigChan)
 135 }
 136
 137 // stop the underlying Docker container.
 138 func (runner *ContainerRunner) stop() {
 139         runner.cStateLock.Lock()
 140         defer runner.cStateLock.Unlock()
 141         if runner.cCancelled {
 142                 return
 143         }
 144         runner.cCancelled = true
 145         if runner.cStarted {
 146                 err := runner.Docker.StopContainer(runner.ContainerID, 10)
 147                 if err != nil {
 148                         log.Printf("StopContainer failed: %s", err)
 149                 }
 150         }
 151 }
 152
 153 // LoadImage determines the docker image id from the container record and
 154 // checks if it is available in the local Docker image store.  If not, it loads
 155 // the image from Keep.
 156 func (runner *ContainerRunner) LoadImage() (err error) {
 157
 158         runner.CrunchLog.Printf("Fetching Docker image from collection '%s'", runner.Container.ContainerImage)
 159
 160         var collection arvados.Collection
 161         err = runner.ArvClient.Get("collections", runner.Container.ContainerImage, nil, &collection)
 162         if err != nil {
 163                 return fmt.Errorf("While getting container image collection: %v", err)
 164         }
 165         manifest := manifest.Manifest{Text: collection.ManifestText}
 166         var img, imageID string
 167         for ms := range manifest.StreamIter() {
 168                 img = ms.FileStreamSegments[0].Name
 169                 if !strings.HasSuffix(img, ".tar") {
 170                         return fmt.Errorf("First file in the container image collection does not end in .tar")
 171                 }
 172                 imageID = img[:len(img)-4]
 173         }
 174
 175         runner.CrunchLog.Printf("Using Docker image id '%s'", imageID)
 176
 177         _, err = runner.Docker.InspectImage(imageID)
 178         if err != nil {
 179                 runner.CrunchLog.Print("Loading Docker image from keep")
 180
 181                 var readCloser io.ReadCloser
 182                 readCloser, err = runner.Kc.ManifestFileReader(manifest, img)
 183                 if err != nil {
 184                         return fmt.Errorf("While creating ManifestFileReader for container image: %v", err)
 185                 }
 186
 187                 err = runner.Docker.LoadImage(readCloser)
 188                 if err != nil {
 189                         return fmt.Errorf("While loading container image into Docker: %v", err)
 190                 }
 191         } else {
 192                 runner.CrunchLog.Print("Docker image is available")
 193         }
 194
 195         runner.ContainerConfig.Image = imageID
 196
 197         return nil
 198 }
 199
 200 func (runner *ContainerRunner) ArvMountCmd(arvMountCmd []string, token string) (c *exec.Cmd, err error) {
 201         c = exec.Command("arv-mount", arvMountCmd...)
 202
 203         // Copy our environment, but override ARVADOS_API_TOKEN with
 204         // the container auth token.
 205         c.Env = nil
 206         for _, s := range os.Environ() {
 207                 if !strings.HasPrefix(s, "ARVADOS_API_TOKEN=") {
 208                         c.Env = append(c.Env, s)
 209                 }
 210         }
 211         c.Env = append(c.Env, "ARVADOS_API_TOKEN="+token)
 212
 213         nt := NewThrottledLogger(runner.NewLogWriter("arv-mount"))
 214         c.Stdout = nt
 215         c.Stderr = nt
 216
 217         err = c.Start()
 218         if err != nil {
 219                 return nil, err
 220         }
 221
 222         statReadme := make(chan bool)
 223         runner.ArvMountExit = make(chan error)
 224
 225         keepStatting := true
 226         go func() {
 227                 for keepStatting {
 228                         time.Sleep(100 * time.Millisecond)
 229                         _, err = os.Stat(fmt.Sprintf("%s/by_id/README", runner.ArvMountPoint))
 230                         if err == nil {
 231                                 keepStatting = false
 232                                 statReadme <- true
 233                         }
 234                 }
 235                 close(statReadme)
 236         }()
 237
 238         go func() {
 239                 runner.ArvMountExit <- c.Wait()
 240                 close(runner.ArvMountExit)
 241         }()
 242
 243         select {
 244         case <-statReadme:
 245                 break
 246         case err := <-runner.ArvMountExit:
 247                 runner.ArvMount = nil
 248                 keepStatting = false
 249                 return nil, err
 250         }
 251
 252         return c, nil
 253 }
 254
 255 func (runner *ContainerRunner) SetupArvMountPoint(prefix string) (err error) {
 256         if runner.ArvMountPoint == "" {
 257                 runner.ArvMountPoint, err = runner.MkTempDir("", prefix)
 258         }
 259         return
 260 }
 261
 262 func (runner *ContainerRunner) SetupMounts() (err error) {
 263         err = runner.SetupArvMountPoint("keep")
 264         if err != nil {
 265                 return fmt.Errorf("While creating keep mount temp dir: %v", err)
 266         }
 267
 268         runner.CleanupTempDir = append(runner.CleanupTempDir, runner.ArvMountPoint)
 269
 270         pdhOnly := true
 271         tmpcount := 0
 272         arvMountCmd := []string{"--foreground", "--allow-other", "--read-write"}
 273
 274         if runner.Container.RuntimeConstraints.KeepCacheRAM > 0 {
 275                 arvMountCmd = append(arvMountCmd, "--file-cache", fmt.Sprintf("%d", runner.Container.RuntimeConstraints.KeepCacheRAM))
 276         }
 277
 278         collectionPaths := []string{}
 279         runner.Binds = nil
 280         needCertMount := true
 281
 282         var binds []string
 283         for bind, _ := range runner.Container.Mounts {
 284                 binds = append(binds, bind)
 285         }
 286         sort.Strings(binds)
 287
 288         for _, bind := range binds {
 289                 mnt := runner.Container.Mounts[bind]
 290                 if bind == "stdout" {
 291                         // Is it a "file" mount kind?
 292                         if mnt.Kind != "file" {
 293                                 return fmt.Errorf("Unsupported mount kind '%s' for stdout. Only 'file' is supported.", mnt.Kind)
 294                         }
 295
 296                         // Does path start with OutputPath?
 297                         prefix := runner.Container.OutputPath
 298                         if !strings.HasSuffix(prefix, "/") {
 299                                 prefix += "/"
 300                         }
 301                         if !strings.HasPrefix(mnt.Path, prefix) {
 302                                 return fmt.Errorf("Stdout path does not start with OutputPath: %s, %s", mnt.Path, prefix)
 303                         }
 304                 }
 305
 306                 if bind == "/etc/arvados/ca-certificates.crt" {
 307                         needCertMount = false
 308                 }
 309
 310                 if strings.HasPrefix(bind, runner.Container.OutputPath+"/") && bind != runner.Container.OutputPath+"/" {
 311                         if mnt.Kind != "collection" {
 312                                 return fmt.Errorf("Only mount points of kind 'collection' are supported underneath the output_path: %v", bind)
 313                         }
 314                 }
 315
 316                 switch {
 317                 case mnt.Kind == "collection":
 318                         var src string
 319                         if mnt.UUID != "" && mnt.PortableDataHash != "" {
 320                                 return fmt.Errorf("Cannot specify both 'uuid' and 'portable_data_hash' for a collection mount")
 321                         }
 322                         if mnt.UUID != "" {
 323                                 if mnt.Writable {
 324                                         return fmt.Errorf("Writing to existing collections currently not permitted.")
 325                                 }
 326                                 pdhOnly = false
 327                                 src = fmt.Sprintf("%s/by_id/%s", runner.ArvMountPoint, mnt.UUID)
 328                         } else if mnt.PortableDataHash != "" {
 329                                 if mnt.Writable {
 330                                         return fmt.Errorf("Can never write to a collection specified by portable data hash")
 331                                 }
 332                                 idx := strings.Index(mnt.PortableDataHash, "/")
 333                                 if idx > 0 {
 334                                         mnt.Path = path.Clean(mnt.PortableDataHash[idx:])
 335                                         mnt.PortableDataHash = mnt.PortableDataHash[0:idx]
 336                                         runner.Container.Mounts[bind] = mnt
 337                                 }
 338                                 src = fmt.Sprintf("%s/by_id/%s", runner.ArvMountPoint, mnt.PortableDataHash)
 339                                 if mnt.Path != "" && mnt.Path != "." {
 340                                         if strings.HasPrefix(mnt.Path, "./") {
 341                                                 mnt.Path = mnt.Path[2:]
 342                                         } else if strings.HasPrefix(mnt.Path, "/") {
 343                                                 mnt.Path = mnt.Path[1:]
 344                                         }
 345                                         src += "/" + mnt.Path
 346                                 }
 347                         } else {
 348                                 src = fmt.Sprintf("%s/tmp%d", runner.ArvMountPoint, tmpcount)
 349                                 arvMountCmd = append(arvMountCmd, "--mount-tmp")
 350                                 arvMountCmd = append(arvMountCmd, fmt.Sprintf("tmp%d", tmpcount))
 351                                 tmpcount += 1
 352                         }
 353                         if mnt.Writable {
 354                                 if bind == runner.Container.OutputPath {
 355                                         runner.HostOutputDir = src
 356                                 } else if strings.HasPrefix(bind, runner.Container.OutputPath+"/") {
 357                                         return fmt.Errorf("Writable mount points are not permitted underneath the output_path: %v", bind)
 358                                 }
 359                                 runner.Binds = append(runner.Binds, fmt.Sprintf("%s:%s", src, bind))
 360                         } else {
 361                                 runner.Binds = append(runner.Binds, fmt.Sprintf("%s:%s:ro", src, bind))
 362                         }
 363                         collectionPaths = append(collectionPaths, src)
 364
 365                 case mnt.Kind == "tmp" && bind == runner.Container.OutputPath:
 366                         runner.HostOutputDir, err = runner.MkTempDir("", "")
 367                         if err != nil {
 368                                 return fmt.Errorf("While creating mount temp dir: %v", err)
 369                         }
 370                         st, staterr := os.Stat(runner.HostOutputDir)
 371                         if staterr != nil {
 372                                 return fmt.Errorf("While Stat on temp dir: %v", staterr)
 373                         }
 374                         err = os.Chmod(runner.HostOutputDir, st.Mode()|os.ModeSetgid|0777)
 375                         if staterr != nil {
 376                                 return fmt.Errorf("While Chmod temp dir: %v", err)
 377                         }
 378                         runner.CleanupTempDir = append(runner.CleanupTempDir, runner.HostOutputDir)
 379                         runner.Binds = append(runner.Binds, fmt.Sprintf("%s:%s", runner.HostOutputDir, bind))
 380
 381                 case mnt.Kind == "tmp":
 382                         runner.Binds = append(runner.Binds, bind)
 383
 384                 case mnt.Kind == "json":
 385                         jsondata, err := json.Marshal(mnt.Content)
 386                         if err != nil {
 387                                 return fmt.Errorf("encoding json data: %v", err)
 388                         }
 389                         // Create a tempdir with a single file
 390                         // (instead of just a tempfile): this way we
 391                         // can ensure the file is world-readable
 392                         // inside the container, without having to
 393                         // make it world-readable on the docker host.
 394                         tmpdir, err := runner.MkTempDir("", "")
 395                         if err != nil {
 396                                 return fmt.Errorf("creating temp dir: %v", err)
 397                         }
 398                         runner.CleanupTempDir = append(runner.CleanupTempDir, tmpdir)
 399                         tmpfn := filepath.Join(tmpdir, "mountdata.json")
 400                         err = ioutil.WriteFile(tmpfn, jsondata, 0644)
 401                         if err != nil {
 402                                 return fmt.Errorf("writing temp file: %v", err)
 403                         }
 404                         runner.Binds = append(runner.Binds, fmt.Sprintf("%s:%s:ro", tmpfn, bind))
 405                 }
 406         }
 407
 408         if runner.HostOutputDir == "" {
 409                 return fmt.Errorf("Output path does not correspond to a writable mount point")
 410         }
 411
 412         if wantAPI := runner.Container.RuntimeConstraints.API; needCertMount && wantAPI != nil && *wantAPI {
 413                 for _, certfile := range arvadosclient.CertFiles {
 414                         _, err := os.Stat(certfile)
 415                         if err == nil {
 416                                 runner.Binds = append(runner.Binds, fmt.Sprintf("%s:/etc/arvados/ca-certificates.crt:ro", certfile))
 417                                 break
 418                         }
 419                 }
 420         }
 421
 422         if pdhOnly {
 423                 arvMountCmd = append(arvMountCmd, "--mount-by-pdh", "by_id")
 424         } else {
 425                 arvMountCmd = append(arvMountCmd, "--mount-by-id", "by_id")
 426         }
 427         arvMountCmd = append(arvMountCmd, runner.ArvMountPoint)
 428
 429         token, err := runner.ContainerToken()
 430         if err != nil {
 431                 return fmt.Errorf("could not get container token: %s", err)
 432         }
 433
 434         runner.ArvMount, err = runner.RunArvMount(arvMountCmd, token)
 435         if err != nil {
 436                 return fmt.Errorf("While trying to start arv-mount: %v", err)
 437         }
 438
 439         for _, p := range collectionPaths {
 440                 _, err = os.Stat(p)
 441                 if err != nil {
 442                         return fmt.Errorf("While checking that input files exist: %v", err)
 443                 }
 444         }
 445
 446         return nil
 447 }
 448
 449 func (runner *ContainerRunner) ProcessDockerAttach(containerReader io.Reader) {
 450         // Handle docker log protocol
 451         // https://docs.docker.com/engine/reference/api/docker_remote_api_v1.15/#attach-to-a-container
 452
 453         header := make([]byte, 8)
 454         for {
 455                 _, readerr := io.ReadAtLeast(containerReader, header, 8)
 456
 457                 if readerr == nil {
 458                         readsize := int64(header[7]) | (int64(header[6]) << 8) | (int64(header[5]) << 16) | (int64(header[4]) << 24)
 459                         if header[0] == 1 {
 460                                 // stdout
 461                                 _, readerr = io.CopyN(runner.Stdout, containerReader, readsize)
 462                         } else {
 463                                 // stderr
 464                                 _, readerr = io.CopyN(runner.Stderr, containerReader, readsize)
 465                         }
 466                 }
 467
 468                 if readerr != nil {
 469                         if readerr != io.EOF {
 470                                 runner.CrunchLog.Printf("While reading docker logs: %v", readerr)
 471                         }
 472
 473                         closeerr := runner.Stdout.Close()
 474                         if closeerr != nil {
 475                                 runner.CrunchLog.Printf("While closing stdout logs: %v", closeerr)
 476                         }
 477
 478                         closeerr = runner.Stderr.Close()
 479                         if closeerr != nil {
 480                                 runner.CrunchLog.Printf("While closing stderr logs: %v", closeerr)
 481                         }
 482
 483                         if runner.statReporter != nil {
 484                                 runner.statReporter.Stop()
 485                                 closeerr = runner.statLogger.Close()
 486                                 if closeerr != nil {
 487                                         runner.CrunchLog.Printf("While closing crunchstat logs: %v", closeerr)
 488                                 }
 489                         }
 490
 491                         runner.loggingDone <- true
 492                         close(runner.loggingDone)
 493                         return
 494                 }
 495         }
 496 }
 497
 498 func (runner *ContainerRunner) StartCrunchstat() {
 499         runner.statLogger = NewThrottledLogger(runner.NewLogWriter("crunchstat"))
 500         runner.statReporter = &crunchstat.Reporter{
 501                 CID:          runner.ContainerID,
 502                 Logger:       log.New(runner.statLogger, "", 0),
 503                 CgroupParent: runner.expectCgroupParent,
 504                 CgroupRoot:   runner.cgroupRoot,
 505                 PollPeriod:   runner.statInterval,
 506         }
 507         runner.statReporter.Start()
 508 }
 509
 510 type infoCommand struct {
 511         label string
 512         cmd   []string
 513 }
 514
 515 // Gather node information and store it on the log for debugging
 516 // purposes.
 517 func (runner *ContainerRunner) LogNodeInfo() (err error) {
 518         w := runner.NewLogWriter("node-info")
 519         logger := log.New(w, "node-info", 0)
 520
 521         commands := []infoCommand{
 522                 infoCommand{
 523                         label: "Host Information",
 524                         cmd:   []string{"uname", "-a"},
 525                 },
 526                 infoCommand{
 527                         label: "CPU Information",
 528                         cmd:   []string{"cat", "/proc/cpuinfo"},
 529                 },
 530                 infoCommand{
 531                         label: "Memory Information",
 532                         cmd:   []string{"cat", "/proc/meminfo"},
 533                 },
 534                 infoCommand{
 535                         label: "Disk Space",
 536                         cmd:   []string{"df", "-m", "/"},
 537                 },
 538                 infoCommand{
 539                         label: "Disk Space",
 540                         cmd:   []string{"df", "-m", os.TempDir()},
 541                 },
 542                 infoCommand{
 543                         label: "Disk INodes",
 544                         cmd:   []string{"df", "-i", "/"},
 545                 },
 546                 infoCommand{
 547                         label: "Disk INodes",
 548                         cmd:   []string{"df", "-i", os.TempDir()},
 549                 },
 550         }
 551
 552         // Run commands with informational output to be logged.
 553         var out []byte
 554         for _, command := range commands {
 555                 out, err = exec.Command(command.cmd[0], command.cmd[1:]...).CombinedOutput()
 556                 if err != nil {
 557                         return fmt.Errorf("While running command %q: %v",
 558                                 command.cmd, err)
 559                 }
 560                 logger.Println(command.label)
 561                 for _, line := range strings.Split(string(out), "\n") {
 562                         logger.Println(" ", line)
 563                 }
 564         }
 565
 566         err = w.Close()
 567         if err != nil {
 568                 return fmt.Errorf("While closing node-info logs: %v", err)
 569         }
 570         return nil
 571 }
 572
 573 // Get and save the raw JSON container record from the API server
 574 func (runner *ContainerRunner) LogContainerRecord() (err error) {
 575         w := &ArvLogWriter{
 576                 runner.ArvClient,
 577                 runner.Container.UUID,
 578                 "container",
 579                 runner.LogCollection.Open("container.json"),
 580         }
 581         // Get Container record JSON from the API Server
 582         reader, err := runner.ArvClient.CallRaw("GET", "containers", runner.Container.UUID, "", nil)
 583         if err != nil {
 584                 return fmt.Errorf("While retrieving container record from the API server: %v", err)
 585         }
 586         // Read the API server response as []byte
 587         json_bytes, err := ioutil.ReadAll(reader)
 588         if err != nil {
 589                 return fmt.Errorf("While reading container record API server response: %v", err)
 590         }
 591         // Decode the JSON []byte
 592         var cr map[string]interface{}
 593         if err = json.Unmarshal(json_bytes, &cr); err != nil {
 594                 return fmt.Errorf("While decoding the container record JSON response: %v", err)
 595         }
 596         // Re-encode it using indentation to improve readability
 597         enc := json.NewEncoder(w)
 598         enc.SetIndent("", "    ")
 599         if err = enc.Encode(cr); err != nil {
 600                 return fmt.Errorf("While logging the JSON container record: %v", err)
 601         }
 602         err = w.Close()
 603         if err != nil {
 604                 return fmt.Errorf("While closing container.json log: %v", err)
 605         }
 606         return nil
 607 }
 608
 609 // AttachLogs connects the docker container stdout and stderr logs to the
 610 // Arvados logger which logs to Keep and the API server logs table.
 611 func (runner *ContainerRunner) AttachStreams() (err error) {
 612
 613         runner.CrunchLog.Print("Attaching container streams")
 614
 615         var containerReader io.Reader
 616         containerReader, err = runner.Docker.AttachContainer(runner.ContainerID,
 617                 &dockerclient.AttachOptions{Stream: true, Stdout: true, Stderr: true})
 618         if err != nil {
 619                 return fmt.Errorf("While attaching container stdout/stderr streams: %v", err)
 620         }
 621
 622         runner.loggingDone = make(chan bool)
 623
 624         if stdoutMnt, ok := runner.Container.Mounts["stdout"]; ok {
 625                 stdoutPath := stdoutMnt.Path[len(runner.Container.OutputPath):]
 626                 index := strings.LastIndex(stdoutPath, "/")
 627                 if index > 0 {
 628                         subdirs := stdoutPath[:index]
 629                         if subdirs != "" {
 630                                 st, err := os.Stat(runner.HostOutputDir)
 631                                 if err != nil {
 632                                         return fmt.Errorf("While Stat on temp dir: %v", err)
 633                                 }
 634                                 stdoutPath := path.Join(runner.HostOutputDir, subdirs)
 635                                 err = os.MkdirAll(stdoutPath, st.Mode()|os.ModeSetgid|0777)
 636                                 if err != nil {
 637                                         return fmt.Errorf("While MkdirAll %q: %v", stdoutPath, err)
 638                                 }
 639                         }
 640                 }
 641                 stdoutFile, err := os.Create(path.Join(runner.HostOutputDir, stdoutPath))
 642                 if err != nil {
 643                         return fmt.Errorf("While creating stdout file: %v", err)
 644                 }
 645                 runner.Stdout = stdoutFile
 646         } else {
 647                 runner.Stdout = NewThrottledLogger(runner.NewLogWriter("stdout"))
 648         }
 649         runner.Stderr = NewThrottledLogger(runner.NewLogWriter("stderr"))
 650
 651         go runner.ProcessDockerAttach(containerReader)
 652
 653         return nil
 654 }
 655
 656 // CreateContainer creates the docker container.
 657 func (runner *ContainerRunner) CreateContainer() error {
 658         runner.CrunchLog.Print("Creating Docker container")
 659
 660         runner.ContainerConfig.Cmd = runner.Container.Command
 661         if runner.Container.Cwd != "." {
 662                 runner.ContainerConfig.WorkingDir = runner.Container.Cwd
 663         }
 664
 665         for k, v := range runner.Container.Environment {
 666                 runner.ContainerConfig.Env = append(runner.ContainerConfig.Env, k+"="+v)
 667         }
 668         if wantAPI := runner.Container.RuntimeConstraints.API; wantAPI != nil && *wantAPI {
 669                 tok, err := runner.ContainerToken()
 670                 if err != nil {
 671                         return err
 672                 }
 673                 runner.ContainerConfig.Env = append(runner.ContainerConfig.Env,
 674                         "ARVADOS_API_TOKEN="+tok,
 675                         "ARVADOS_API_HOST="+os.Getenv("ARVADOS_API_HOST"),
 676                         "ARVADOS_API_HOST_INSECURE="+os.Getenv("ARVADOS_API_HOST_INSECURE"),
 677                 )
 678                 runner.ContainerConfig.NetworkDisabled = false
 679         } else {
 680                 runner.ContainerConfig.NetworkDisabled = true
 681         }
 682
 683         var err error
 684         runner.ContainerID, err = runner.Docker.CreateContainer(&runner.ContainerConfig, "", nil)
 685         if err != nil {
 686                 return fmt.Errorf("While creating container: %v", err)
 687         }
 688
 689         runner.HostConfig = dockerclient.HostConfig{
 690                 Binds:        runner.Binds,
 691                 CgroupParent: runner.setCgroupParent,
 692                 LogConfig: dockerclient.LogConfig{
 693                         Type: "none",
 694                 },
 695         }
 696
 697         return runner.AttachStreams()
 698 }
 699
 700 // StartContainer starts the docker container created by CreateContainer.
 701 func (runner *ContainerRunner) StartContainer() error {
 702         runner.CrunchLog.Printf("Starting Docker container id '%s'", runner.ContainerID)
 703         runner.cStateLock.Lock()
 704         defer runner.cStateLock.Unlock()
 705         if runner.cCancelled {
 706                 return ErrCancelled
 707         }
 708         err := runner.Docker.StartContainer(runner.ContainerID, &runner.HostConfig)
 709         if err != nil {
 710                 return fmt.Errorf("could not start container: %v", err)
 711         }
 712         runner.cStarted = true
 713         return nil
 714 }
 715
 716 // WaitFinish waits for the container to terminate, capture the exit code, and
 717 // close the stdout/stderr logging.
 718 func (runner *ContainerRunner) WaitFinish() error {
 719         runner.CrunchLog.Print("Waiting for container to finish")
 720
 721         waitDocker := runner.Docker.Wait(runner.ContainerID)
 722         waitMount := runner.ArvMountExit
 723         for waitDocker != nil {
 724                 select {
 725                 case err := <-waitMount:
 726                         runner.CrunchLog.Printf("arv-mount exited before container finished: %v", err)
 727                         waitMount = nil
 728                         runner.stop()
 729                 case wr := <-waitDocker:
 730                         if wr.Error != nil {
 731                                 return fmt.Errorf("While waiting for container to finish: %v", wr.Error)
 732                         }
 733                         runner.ExitCode = &wr.ExitCode
 734                         waitDocker = nil
 735                 }
 736         }
 737
 738         // wait for stdout/stderr to complete
 739         <-runner.loggingDone
 740
 741         return nil
 742 }
 743
 744 // HandleOutput sets the output, unmounts the FUSE mount, and deletes temporary directories
 745 func (runner *ContainerRunner) CaptureOutput() error {
 746         if runner.finalState != "Complete" {
 747                 return nil
 748         }
 749
 750         if wantAPI := runner.Container.RuntimeConstraints.API; wantAPI != nil && *wantAPI {
 751                 // Output may have been set directly by the container, so
 752                 // refresh the container record to check.
 753                 err := runner.ArvClient.Get("containers", runner.Container.UUID,
 754                         nil, &runner.Container)
 755                 if err != nil {
 756                         return err
 757                 }
 758                 if runner.Container.Output != "" {
 759                         // Container output is already set.
 760                         runner.OutputPDH = &runner.Container.Output
 761                         return nil
 762                 }
 763         }
 764
 765         if runner.HostOutputDir == "" {
 766                 return nil
 767         }
 768
 769         _, err := os.Stat(runner.HostOutputDir)
 770         if err != nil {
 771                 return fmt.Errorf("While checking host output path: %v", err)
 772         }
 773
 774         var manifestText string
 775
 776         collectionMetafile := fmt.Sprintf("%s/.arvados#collection", runner.HostOutputDir)
 777         _, err = os.Stat(collectionMetafile)
 778         if err != nil {
 779                 // Regular directory
 780                 cw := CollectionWriter{0, runner.Kc, nil, nil, sync.Mutex{}}
 781                 manifestText, err = cw.WriteTree(runner.HostOutputDir, runner.CrunchLog.Logger)
 782                 if err != nil {
 783                         return fmt.Errorf("While uploading output files: %v", err)
 784                 }
 785         } else {
 786                 // FUSE mount directory
 787                 file, openerr := os.Open(collectionMetafile)
 788                 if openerr != nil {
 789                         return fmt.Errorf("While opening FUSE metafile: %v", err)
 790                 }
 791                 defer file.Close()
 792
 793                 var rec arvados.Collection
 794                 err = json.NewDecoder(file).Decode(&rec)
 795                 if err != nil {
 796                         return fmt.Errorf("While reading FUSE metafile: %v", err)
 797                 }
 798                 manifestText = rec.ManifestText
 799         }
 800
 801         // Pre-populate output from the configured mount points
 802         var binds []string
 803         for bind, _ := range runner.Container.Mounts {
 804                 binds = append(binds, bind)
 805         }
 806         sort.Strings(binds)
 807
 808         for _, bind := range binds {
 809                 mnt := runner.Container.Mounts[bind]
 810
 811                 bindSuffix := strings.TrimPrefix(bind, runner.Container.OutputPath)
 812
 813                 if bindSuffix == bind || len(bindSuffix) <= 0 {
 814                         // either does not start with OutputPath or is OutputPath itself
 815                         continue
 816                 }
 817
 818                 if mnt.ExcludeFromOutput == true {
 819                         continue
 820                 }
 821
 822                 // append to manifest_text
 823                 m, err := runner.getCollectionManifestForPath(mnt, bindSuffix)
 824                 if err != nil {
 825                         return err
 826                 }
 827
 828                 manifestText = manifestText + m
 829         }
 830
 831         // Save output
 832         var response arvados.Collection
 833         manifest := manifest.Manifest{Text: manifestText}
 834         manifestText = manifest.Extract(".", ".").Text
 835         err = runner.ArvClient.Create("collections",
 836                 arvadosclient.Dict{
 837                         "ensure_unique_name": true,
 838                         "collection": arvadosclient.Dict{
 839                                 "is_trashed":    true,
 840                                 "name":          "output for " + runner.Container.UUID,
 841                                 "manifest_text": manifestText}},
 842                 &response)
 843         if err != nil {
 844                 return fmt.Errorf("While creating output collection: %v", err)
 845         }
 846         runner.OutputPDH = &response.PortableDataHash
 847         return nil
 848 }
 849
 850 var outputCollections = make(map[string]arvados.Collection)
 851
 852 // Fetch the collection for the mnt.PortableDataHash
 853 // Return the manifest_text fragment corresponding to the specified mnt.Path
 854 //  after making any required updates.
 855 //  Ex:
 856 //    If mnt.Path is not specified,
 857 //      return the entire manifest_text after replacing any "." with bindSuffix
 858 //    If mnt.Path corresponds to one stream,
 859 //      return the manifest_text for that stream after replacing that stream name with bindSuffix
 860 //    Otherwise, check if a filename in any one stream is being sought. Return the manifest_text
 861 //      for that stream after replacing stream name with bindSuffix minus the last word
 862 //      and the file name with last word of the bindSuffix
 863 //  Allowed path examples:
 864 //    "path":"/"
 865 //    "path":"/subdir1"
 866 //    "path":"/subdir1/subdir2"
 867 //    "path":"/subdir/filename" etc
 868 func (runner *ContainerRunner) getCollectionManifestForPath(mnt arvados.Mount, bindSuffix string) (string, error) {
 869         collection := outputCollections[mnt.PortableDataHash]
 870         if collection.PortableDataHash == "" {
 871                 err := runner.ArvClient.Get("collections", mnt.PortableDataHash, nil, &collection)
 872                 if err != nil {
 873                         return "", fmt.Errorf("While getting collection for %v: %v", mnt.PortableDataHash, err)
 874                 }
 875                 outputCollections[mnt.PortableDataHash] = collection
 876         }
 877
 878         if collection.ManifestText == "" {
 879                 runner.CrunchLog.Printf("No manifest text for collection %v", collection.PortableDataHash)
 880                 return "", nil
 881         }
 882
 883         mft := manifest.Manifest{Text: collection.ManifestText}
 884         extracted := mft.Extract(mnt.Path, bindSuffix)
 885         if extracted.Err != nil {
 886                 return "", fmt.Errorf("Error parsing manifest for %v: %v", mnt.PortableDataHash, extracted.Err.Error())
 887         }
 888         return extracted.Text, nil
 889 }
 890
 891 func (runner *ContainerRunner) CleanupDirs() {
 892         if runner.ArvMount != nil {
 893                 umount := exec.Command("fusermount", "-z", "-u", runner.ArvMountPoint)
 894                 umnterr := umount.Run()
 895                 if umnterr != nil {
 896                         runner.CrunchLog.Printf("While running fusermount: %v", umnterr)
 897                 }
 898
 899                 mnterr := <-runner.ArvMountExit
 900                 if mnterr != nil {
 901                         runner.CrunchLog.Printf("Arv-mount exit error: %v", mnterr)
 902                 }
 903         }
 904
 905         for _, tmpdir := range runner.CleanupTempDir {
 906                 rmerr := os.RemoveAll(tmpdir)
 907                 if rmerr != nil {
 908                         runner.CrunchLog.Printf("While cleaning up temporary directory %s: %v", tmpdir, rmerr)
 909                 }
 910         }
 911 }
 912
 913 // CommitLogs posts the collection containing the final container logs.
 914 func (runner *ContainerRunner) CommitLogs() error {
 915         runner.CrunchLog.Print(runner.finalState)
 916         runner.CrunchLog.Close()
 917
 918         // Closing CrunchLog above allows it to be committed to Keep at this
 919         // point, but re-open crunch log with ArvClient in case there are any
 920         // other further (such as failing to write the log to Keep!) while
 921         // shutting down
 922         runner.CrunchLog = NewThrottledLogger(&ArvLogWriter{runner.ArvClient, runner.Container.UUID,
 923                 "crunch-run", nil})
 924
 925         if runner.LogsPDH != nil {
 926                 // If we have already assigned something to LogsPDH,
 927                 // we must be closing the re-opened log, which won't
 928                 // end up getting attached to the container record and
 929                 // therefore doesn't need to be saved as a collection
 930                 // -- it exists only to send logs to other channels.
 931                 return nil
 932         }
 933
 934         mt, err := runner.LogCollection.ManifestText()
 935         if err != nil {
 936                 return fmt.Errorf("While creating log manifest: %v", err)
 937         }
 938
 939         var response arvados.Collection
 940         err = runner.ArvClient.Create("collections",
 941                 arvadosclient.Dict{
 942                         "ensure_unique_name": true,
 943                         "collection": arvadosclient.Dict{
 944                                 "is_trashed":    true,
 945                                 "name":          "logs for " + runner.Container.UUID,
 946                                 "manifest_text": mt}},
 947                 &response)
 948         if err != nil {
 949                 return fmt.Errorf("While creating log collection: %v", err)
 950         }
 951         runner.LogsPDH = &response.PortableDataHash
 952         return nil
 953 }
 954
 955 // UpdateContainerRunning updates the container state to "Running"
 956 func (runner *ContainerRunner) UpdateContainerRunning() error {
 957         runner.cStateLock.Lock()
 958         defer runner.cStateLock.Unlock()
 959         if runner.cCancelled {
 960                 return ErrCancelled
 961         }
 962         return runner.ArvClient.Update("containers", runner.Container.UUID,
 963                 arvadosclient.Dict{"container": arvadosclient.Dict{"state": "Running"}}, nil)
 964 }
 965
 966 // ContainerToken returns the api_token the container (and any
 967 // arv-mount processes) are allowed to use.
 968 func (runner *ContainerRunner) ContainerToken() (string, error) {
 969         if runner.token != "" {
 970                 return runner.token, nil
 971         }
 972
 973         var auth arvados.APIClientAuthorization
 974         err := runner.ArvClient.Call("GET", "containers", runner.Container.UUID, "auth", nil, &auth)
 975         if err != nil {
 976                 return "", err
 977         }
 978         runner.token = auth.APIToken
 979         return runner.token, nil
 980 }
 981
 982 // UpdateContainerComplete updates the container record state on API
 983 // server to "Complete" or "Cancelled"
 984 func (runner *ContainerRunner) UpdateContainerFinal() error {
 985         update := arvadosclient.Dict{}
 986         update["state"] = runner.finalState
 987         if runner.LogsPDH != nil {
 988                 update["log"] = *runner.LogsPDH
 989         }
 990         if runner.finalState == "Complete" {
 991                 if runner.ExitCode != nil {
 992                         update["exit_code"] = *runner.ExitCode
 993                 }
 994                 if runner.OutputPDH != nil {
 995                         update["output"] = *runner.OutputPDH
 996                 }
 997         }
 998         return runner.ArvClient.Update("containers", runner.Container.UUID, arvadosclient.Dict{"container": update}, nil)
 999 }
1000
1001 // IsCancelled returns the value of Cancelled, with goroutine safety.
1002 func (runner *ContainerRunner) IsCancelled() bool {
1003         runner.cStateLock.Lock()
1004         defer runner.cStateLock.Unlock()
1005         return runner.cCancelled
1006 }
1007
1008 // NewArvLogWriter creates an ArvLogWriter
1009 func (runner *ContainerRunner) NewArvLogWriter(name string) io.WriteCloser {
1010         return &ArvLogWriter{runner.ArvClient, runner.Container.UUID, name, runner.LogCollection.Open(name + ".txt")}
1011 }
1012
1013 // Run the full container lifecycle.
1014 func (runner *ContainerRunner) Run() (err error) {
1015         runner.CrunchLog.Printf("Executing container '%s'", runner.Container.UUID)
1016
1017         hostname, hosterr := os.Hostname()
1018         if hosterr != nil {
1019                 runner.CrunchLog.Printf("Error getting hostname '%v'", hosterr)
1020         } else {
1021                 runner.CrunchLog.Printf("Executing on host '%s'", hostname)
1022         }
1023
1024         // Clean up temporary directories _after_ finalizing
1025         // everything (if we've made any by then)
1026         defer runner.CleanupDirs()
1027
1028         runner.finalState = "Queued"
1029
1030         defer func() {
1031                 // checkErr prints e (unless it's nil) and sets err to
1032                 // e (unless err is already non-nil). Thus, if err
1033                 // hasn't already been assigned when Run() returns,
1034                 // this cleanup func will cause Run() to return the
1035                 // first non-nil error that is passed to checkErr().
1036                 checkErr := func(e error) {
1037                         if e == nil {
1038                                 return
1039                         }
1040                         runner.CrunchLog.Print(e)
1041                         if err == nil {
1042                                 err = e
1043                         }
1044                 }
1045
1046                 // Log the error encountered in Run(), if any
1047                 checkErr(err)
1048
1049                 if runner.finalState == "Queued" {
1050                         runner.CrunchLog.Close()
1051                         runner.UpdateContainerFinal()
1052                         return
1053                 }
1054
1055                 if runner.IsCancelled() {
1056                         runner.finalState = "Cancelled"
1057                         // but don't return yet -- we still want to
1058                         // capture partial output and write logs
1059                 }
1060
1061                 checkErr(runner.CaptureOutput())
1062                 checkErr(runner.CommitLogs())
1063                 checkErr(runner.UpdateContainerFinal())
1064
1065                 // The real log is already closed, but then we opened
1066                 // a new one in case we needed to log anything while
1067                 // finalizing.
1068                 runner.CrunchLog.Close()
1069         }()
1070
1071         err = runner.ArvClient.Get("containers", runner.Container.UUID, nil, &runner.Container)
1072         if err != nil {
1073                 err = fmt.Errorf("While getting container record: %v", err)
1074                 return
1075         }
1076
1077         // setup signal handling
1078         runner.SetupSignals()
1079
1080         // check for and/or load image
1081         err = runner.LoadImage()
1082         if err != nil {
1083                 runner.finalState = "Cancelled"
1084                 err = fmt.Errorf("While loading container image: %v", err)
1085                 return
1086         }
1087
1088         // set up FUSE mount and binds
1089         err = runner.SetupMounts()
1090         if err != nil {
1091                 runner.finalState = "Cancelled"
1092                 err = fmt.Errorf("While setting up mounts: %v", err)
1093                 return
1094         }
1095
1096         err = runner.CreateContainer()
1097         if err != nil {
1098                 return
1099         }
1100
1101         // Gather and record node information
1102         err = runner.LogNodeInfo()
1103         if err != nil {
1104                 return
1105         }
1106         // Save container.json record on log collection
1107         err = runner.LogContainerRecord()
1108         if err != nil {
1109                 return
1110         }
1111
1112         runner.StartCrunchstat()
1113
1114         if runner.IsCancelled() {
1115                 return
1116         }
1117
1118         err = runner.UpdateContainerRunning()
1119         if err != nil {
1120                 return
1121         }
1122         runner.finalState = "Cancelled"
1123
1124         err = runner.StartContainer()
1125         if err != nil {
1126                 return
1127         }
1128
1129         err = runner.WaitFinish()
1130         if err == nil {
1131                 runner.finalState = "Complete"
1132         }
1133         return
1134 }
1135
1136 // NewContainerRunner creates a new container runner.
1137 func NewContainerRunner(api IArvadosClient,
1138         kc IKeepClient,
1139         docker ThinDockerClient,
1140         containerUUID string) *ContainerRunner {
1141
1142         cr := &ContainerRunner{ArvClient: api, Kc: kc, Docker: docker}
1143         cr.NewLogWriter = cr.NewArvLogWriter
1144         cr.RunArvMount = cr.ArvMountCmd
1145         cr.MkTempDir = ioutil.TempDir
1146         cr.LogCollection = &CollectionWriter{0, kc, nil, nil, sync.Mutex{}}
1147         cr.Container.UUID = containerUUID
1148         cr.CrunchLog = NewThrottledLogger(cr.NewLogWriter("crunch-run"))
1149         cr.CrunchLog.Immediate = log.New(os.Stderr, containerUUID+" ", 0)
1150         return cr
1151 }
1152
1153 func main() {
1154         statInterval := flag.Duration("crunchstat-interval", 10*time.Second, "sampling period for periodic resource usage reporting")
1155         cgroupRoot := flag.String("cgroup-root", "/sys/fs/cgroup", "path to sysfs cgroup tree")
1156         cgroupParent := flag.String("cgroup-parent", "docker", "name of container's parent cgroup (ignored if -cgroup-parent-subsystem is used)")
1157         cgroupParentSubsystem := flag.String("cgroup-parent-subsystem", "", "use current cgroup for given subsystem as parent cgroup for container")
1158         caCertsPath := flag.String("ca-certs", "", "Path to TLS root certificates")
1159         flag.Parse()
1160
1161         containerId := flag.Arg(0)
1162
1163         if *caCertsPath != "" {
1164                 arvadosclient.CertFiles = []string{*caCertsPath}
1165         }
1166
1167         api, err := arvadosclient.MakeArvadosClient()
1168         if err != nil {
1169                 log.Fatalf("%s: %v", containerId, err)
1170         }
1171         api.Retries = 8
1172
1173         var kc *keepclient.KeepClient
1174         kc, err = keepclient.MakeKeepClient(api)
1175         if err != nil {
1176                 log.Fatalf("%s: %v", containerId, err)
1177         }
1178         kc.Retries = 4
1179
1180         var docker *dockerclient.DockerClient
1181         docker, err = dockerclient.NewDockerClient("unix:///var/run/docker.sock", nil)
1182         if err != nil {
1183                 log.Fatalf("%s: %v", containerId, err)
1184         }
1185
1186         cr := NewContainerRunner(api, kc, docker, containerId)
1187         cr.statInterval = *statInterval
1188         cr.cgroupRoot = *cgroupRoot
1189         cr.expectCgroupParent = *cgroupParent
1190         if *cgroupParentSubsystem != "" {
1191                 p := findCgroup(*cgroupParentSubsystem)
1192                 cr.setCgroupParent = p
1193                 cr.expectCgroupParent = p
1194         }
1195
1196         err = cr.Run()
1197         if err != nil {
1198                 log.Fatalf("%s: %v", containerId, err)
1199         }
1200
1201 }