distribution/registry/handlers/blobupload.go

356 lines
11 KiB
Go
Raw Normal View History

Refactor Blob Service API This PR refactors the blob service API to be oriented around blob descriptors. Identified by digests, blobs become an abstract entity that can be read and written using a descriptor as a handle. This allows blobs to take many forms, such as a ReadSeekCloser or a simple byte buffer, allowing blob oriented operations to better integrate with blob agnostic APIs (such as the `io` package). The error definitions are now better organized to reflect conditions that can only be seen when interacting with the blob API. The main benefit of this is to separate the much smaller metadata from large file storage. Many benefits also follow from this. Reading and writing has been separated into discrete services. Backend implementation is also simplified, by reducing the amount of metadata that needs to be picked up to simply serve a read. This also improves cacheability. "Opening" a blob simply consists of an access check (Stat) and a path calculation. Caching is greatly simplified and we've made the mapping of provisional to canonical hashes a first-class concept. BlobDescriptorService and BlobProvider can be combined in different ways to achieve varying effects. Recommend Review Approach ------------------------- This is a very large patch. While apologies are in order, we are getting a considerable amount of refactoring. Most changes follow from the changes to the root package (distribution), so start there. From there, the main changes are in storage. Looking at (*repository).Blobs will help to understand the how the linkedBlobStore is wired. One can explore the internals within and also branch out into understanding the changes to the caching layer. Following the descriptions below will also help to guide you. To reduce the chances for regressions, it was critical that major changes to unit tests were avoided. Where possible, they are left untouched and where not, the spirit is hopefully captured. Pay particular attention to where behavior may have changed. Storage ------- The primary changes to the `storage` package, other than the interface updates, were to merge the layerstore and blobstore. Blob access is now layered even further. The first layer, blobStore, exposes a global `BlobStatter` and `BlobProvider`. Operations here provide a fast path for most read operations that don't take access control into account. The `linkedBlobStore` layers on top of the `blobStore`, providing repository- scoped blob link management in the backend. The `linkedBlobStore` implements the full `BlobStore` suite, providing access-controlled, repository-local blob writers. The abstraction between the two is slightly broken in that `linkedBlobStore` is the only channel under which one can write into the global blob store. The `linkedBlobStore` also provides flexibility in that it can act over different link sets depending on configuration. This allows us to use the same code for signature links, manifest links and blob links. Eventually, we will fully consolidate this storage. The improved cache flow comes from the `linkedBlobStatter` component of `linkedBlobStore`. Using a `cachedBlobStatter`, these combine together to provide a simple cache hierarchy that should streamline access checks on read and write operations, or at least provide a single path to optimize. The metrics have been changed in a slightly incompatible way since the former operations, Fetch and Exists, are no longer relevant. The fileWriter and fileReader have been slightly modified to support the rest of the changes. The most interesting is the removal of the `Stat` call from `newFileReader`. This was the source of unnecessary round trips that were only present to look up the size of the resulting reader. Now, one must simply pass in the size, requiring the caller to decide whether or not the `Stat` call is appropriate. In several cases, it turned out the caller already had the size already. The `WriterAt` implementation has been removed from `fileWriter`, since it is no longer required for `BlobWriter`, reducing the number of paths which writes may take. Cache ----- Unfortunately, the `cache` package required a near full rewrite. It was pretty mechanical in that the cache is oriented around the `BlobDescriptorService` slightly modified to include the ability to set the values for individual digests. While the implementation is oriented towards caching, it can act as a primary store. Provisions are in place to have repository local metadata, in addition to global metadata. Fallback is implemented as a part of the storage package to maintain this flexibility. One unfortunate side-effect is that caching is now repository-scoped, rather than global. This should have little effect on performance but may increase memory usage. Handlers -------- The `handlers` package has been updated to leverage the new API. For the most part, the changes are superficial or mechanical based on the API changes. This did expose a bug in the handling of provisional vs canonical digests that was fixed in the unit tests. Configuration ------------- One user-facing change has been made to the configuration and is updated in the associated documentation. The `layerinfo` cache parameter has been deprecated by the `blobdescriptor` cache parameter. Both are equivalent and configuration files should be backward compatible. Notifications ------------- Changes the `notification` package are simply to support the interface changes. Context ------- A small change has been made to the tracing log-level. Traces have been moved from "info" to "debug" level to reduce output when not needed. Signed-off-by: Stephen J Day <stephen.day@docker.com>
2015-05-12 09:10:29 +02:00
package handlers
import (
"fmt"
"io"
"net/http"
"net/url"
"os"
"github.com/docker/distribution"
ctxu "github.com/docker/distribution/context"
"github.com/docker/distribution/digest"
"github.com/docker/distribution/registry/api/v2"
"github.com/gorilla/handlers"
)
// blobUploadDispatcher constructs and returns the blob upload handler for the
// given request context.
func blobUploadDispatcher(ctx *Context, r *http.Request) http.Handler {
buh := &blobUploadHandler{
Context: ctx,
UUID: getUploadUUID(ctx),
}
handler := http.Handler(handlers.MethodHandler{
"POST": http.HandlerFunc(buh.StartBlobUpload),
"GET": http.HandlerFunc(buh.GetUploadStatus),
"HEAD": http.HandlerFunc(buh.GetUploadStatus),
"PATCH": http.HandlerFunc(buh.PatchBlobData),
"PUT": http.HandlerFunc(buh.PutBlobUploadComplete),
"DELETE": http.HandlerFunc(buh.CancelBlobUpload),
})
if buh.UUID != "" {
state, err := hmacKey(ctx.Config.HTTP.Secret).unpackUploadState(r.FormValue("_state"))
if err != nil {
return http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
ctxu.GetLogger(ctx).Infof("error resolving upload: %v", err)
w.WriteHeader(http.StatusBadRequest)
buh.Errors.Push(v2.ErrorCodeBlobUploadInvalid, err)
})
}
buh.State = state
if state.Name != ctx.Repository.Name() {
return http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
ctxu.GetLogger(ctx).Infof("mismatched repository name in upload state: %q != %q", state.Name, buh.Repository.Name())
w.WriteHeader(http.StatusBadRequest)
buh.Errors.Push(v2.ErrorCodeBlobUploadInvalid, err)
})
}
if state.UUID != buh.UUID {
return http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
ctxu.GetLogger(ctx).Infof("mismatched uuid in upload state: %q != %q", state.UUID, buh.UUID)
w.WriteHeader(http.StatusBadRequest)
buh.Errors.Push(v2.ErrorCodeBlobUploadInvalid, err)
})
}
blobs := ctx.Repository.Blobs(buh)
upload, err := blobs.Resume(buh, buh.UUID)
if err != nil {
ctxu.GetLogger(ctx).Errorf("error resolving upload: %v", err)
if err == distribution.ErrBlobUploadUnknown {
return http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
w.WriteHeader(http.StatusNotFound)
buh.Errors.Push(v2.ErrorCodeBlobUploadUnknown, err)
})
}
return http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
w.WriteHeader(http.StatusInternalServerError)
buh.Errors.Push(v2.ErrorCodeUnknown, err)
})
}
buh.Upload = upload
if state.Offset > 0 {
// Seek the blob upload to the correct spot if it's non-zero.
// These error conditions should be rare and demonstrate really
// problems. We basically cancel the upload and tell the client to
// start over.
if nn, err := upload.Seek(buh.State.Offset, os.SEEK_SET); err != nil {
defer upload.Close()
ctxu.GetLogger(ctx).Infof("error seeking blob upload: %v", err)
return http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
w.WriteHeader(http.StatusBadRequest)
buh.Errors.Push(v2.ErrorCodeBlobUploadInvalid, err)
upload.Cancel(buh)
})
} else if nn != buh.State.Offset {
defer upload.Close()
ctxu.GetLogger(ctx).Infof("seek to wrong offest: %d != %d", nn, buh.State.Offset)
return http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
w.WriteHeader(http.StatusBadRequest)
buh.Errors.Push(v2.ErrorCodeBlobUploadInvalid, err)
upload.Cancel(buh)
})
}
}
handler = closeResources(handler, buh.Upload)
}
return handler
}
// blobUploadHandler handles the http blob upload process.
type blobUploadHandler struct {
*Context
// UUID identifies the upload instance for the current request. Using UUID
// to key blob writers since this implementation uses UUIDs.
UUID string
Upload distribution.BlobWriter
State blobUploadState
}
// StartBlobUpload begins the blob upload process and allocates a server-side
// blob writer session.
func (buh *blobUploadHandler) StartBlobUpload(w http.ResponseWriter, r *http.Request) {
blobs := buh.Repository.Blobs(buh)
upload, err := blobs.Create(buh)
if err != nil {
w.WriteHeader(http.StatusInternalServerError) // Error conditions here?
buh.Errors.Push(v2.ErrorCodeUnknown, err)
return
}
buh.Upload = upload
defer buh.Upload.Close()
if err := buh.blobUploadResponse(w, r, true); err != nil {
w.WriteHeader(http.StatusInternalServerError) // Error conditions here?
buh.Errors.Push(v2.ErrorCodeUnknown, err)
return
}
w.Header().Set("Docker-Upload-UUID", buh.Upload.ID())
w.WriteHeader(http.StatusAccepted)
}
// GetUploadStatus returns the status of a given upload, identified by id.
func (buh *blobUploadHandler) GetUploadStatus(w http.ResponseWriter, r *http.Request) {
if buh.Upload == nil {
w.WriteHeader(http.StatusNotFound)
buh.Errors.Push(v2.ErrorCodeBlobUploadUnknown)
return
}
// TODO(dmcgowan): Set last argument to false in blobUploadResponse when
// resumable upload is supported. This will enable returning a non-zero
// range for clients to begin uploading at an offset.
if err := buh.blobUploadResponse(w, r, true); err != nil {
w.WriteHeader(http.StatusInternalServerError) // Error conditions here?
buh.Errors.Push(v2.ErrorCodeUnknown, err)
return
}
w.Header().Set("Docker-Upload-UUID", buh.UUID)
w.WriteHeader(http.StatusNoContent)
}
// PatchBlobData writes data to an upload.
func (buh *blobUploadHandler) PatchBlobData(w http.ResponseWriter, r *http.Request) {
if buh.Upload == nil {
w.WriteHeader(http.StatusNotFound)
buh.Errors.Push(v2.ErrorCodeBlobUploadUnknown)
return
}
ct := r.Header.Get("Content-Type")
if ct != "" && ct != "application/octet-stream" {
w.WriteHeader(http.StatusBadRequest)
// TODO(dmcgowan): encode error
return
}
// TODO(dmcgowan): support Content-Range header to seek and write range
// Copy the data
if _, err := io.Copy(buh.Upload, r.Body); err != nil {
ctxu.GetLogger(buh).Errorf("unknown error copying into upload: %v", err)
w.WriteHeader(http.StatusInternalServerError)
buh.Errors.Push(v2.ErrorCodeUnknown, err)
return
}
if err := buh.blobUploadResponse(w, r, false); err != nil {
w.WriteHeader(http.StatusInternalServerError) // Error conditions here?
buh.Errors.Push(v2.ErrorCodeUnknown, err)
return
}
w.WriteHeader(http.StatusAccepted)
}
// PutBlobUploadComplete takes the final request of a blob upload. The
// request may include all the blob data or no blob data. Any data
// provided is received and verified. If successful, the blob is linked
// into the blob store and 201 Created is returned with the canonical
// url of the blob.
func (buh *blobUploadHandler) PutBlobUploadComplete(w http.ResponseWriter, r *http.Request) {
if buh.Upload == nil {
w.WriteHeader(http.StatusNotFound)
buh.Errors.Push(v2.ErrorCodeBlobUploadUnknown)
return
}
dgstStr := r.FormValue("digest") // TODO(stevvooe): Support multiple digest parameters!
if dgstStr == "" {
// no digest? return error, but allow retry.
w.WriteHeader(http.StatusBadRequest)
buh.Errors.Push(v2.ErrorCodeDigestInvalid, "digest missing")
return
}
dgst, err := digest.ParseDigest(dgstStr)
if err != nil {
// no digest? return error, but allow retry.
w.WriteHeader(http.StatusNotFound)
buh.Errors.Push(v2.ErrorCodeDigestInvalid, "digest parsing failed")
return
}
// Read in the data, if any.
if _, err := io.Copy(buh.Upload, r.Body); err != nil {
ctxu.GetLogger(buh).Errorf("unknown error copying into upload: %v", err)
w.WriteHeader(http.StatusInternalServerError)
buh.Errors.Push(v2.ErrorCodeUnknown, err)
return
}
desc, err := buh.Upload.Commit(buh, distribution.Descriptor{
Digest: dgst,
// TODO(stevvooe): This isn't wildly important yet, but we should
// really set the length and mediatype. For now, we can let the
// backend take care of this.
})
if err != nil {
switch err := err.(type) {
case distribution.ErrBlobInvalidDigest:
w.WriteHeader(http.StatusBadRequest)
buh.Errors.Push(v2.ErrorCodeDigestInvalid, err)
default:
switch err {
case distribution.ErrBlobInvalidLength, distribution.ErrBlobDigestUnsupported:
w.WriteHeader(http.StatusBadRequest)
buh.Errors.Push(v2.ErrorCodeBlobUploadInvalid, err)
default:
ctxu.GetLogger(buh).Errorf("unknown error completing upload: %#v", err)
w.WriteHeader(http.StatusInternalServerError)
buh.Errors.Push(v2.ErrorCodeUnknown, err)
}
}
// Clean up the backend blob data if there was an error.
if err := buh.Upload.Cancel(buh); err != nil {
// If the cleanup fails, all we can do is observe and report.
ctxu.GetLogger(buh).Errorf("error canceling upload after error: %v", err)
}
return
}
// Build our canonical blob url
blobURL, err := buh.urlBuilder.BuildBlobURL(buh.Repository.Name(), desc.Digest)
if err != nil {
buh.Errors.Push(v2.ErrorCodeUnknown, err)
w.WriteHeader(http.StatusInternalServerError)
return
}
w.Header().Set("Location", blobURL)
w.Header().Set("Content-Length", "0")
w.Header().Set("Docker-Content-Digest", desc.Digest.String())
w.WriteHeader(http.StatusCreated)
}
// CancelBlobUpload cancels an in-progress upload of a blob.
func (buh *blobUploadHandler) CancelBlobUpload(w http.ResponseWriter, r *http.Request) {
if buh.Upload == nil {
w.WriteHeader(http.StatusNotFound)
buh.Errors.Push(v2.ErrorCodeBlobUploadUnknown)
return
}
w.Header().Set("Docker-Upload-UUID", buh.UUID)
if err := buh.Upload.Cancel(buh); err != nil {
ctxu.GetLogger(buh).Errorf("error encountered canceling upload: %v", err)
w.WriteHeader(http.StatusInternalServerError)
buh.Errors.PushErr(err)
}
w.WriteHeader(http.StatusNoContent)
}
// blobUploadResponse provides a standard request for uploading blobs and
// chunk responses. This sets the correct headers but the response status is
// left to the caller. The fresh argument is used to ensure that new blob
// uploads always start at a 0 offset. This allows disabling resumable push by
// always returning a 0 offset on check status.
func (buh *blobUploadHandler) blobUploadResponse(w http.ResponseWriter, r *http.Request, fresh bool) error {
var offset int64
if !fresh {
var err error
offset, err = buh.Upload.Seek(0, os.SEEK_CUR)
if err != nil {
ctxu.GetLogger(buh).Errorf("unable get current offset of blob upload: %v", err)
return err
}
}
// TODO(stevvooe): Need a better way to manage the upload state automatically.
buh.State.Name = buh.Repository.Name()
buh.State.UUID = buh.Upload.ID()
buh.State.Offset = offset
buh.State.StartedAt = buh.Upload.StartedAt()
token, err := hmacKey(buh.Config.HTTP.Secret).packUploadState(buh.State)
if err != nil {
ctxu.GetLogger(buh).Infof("error building upload state token: %s", err)
return err
}
uploadURL, err := buh.urlBuilder.BuildBlobUploadChunkURL(
buh.Repository.Name(), buh.Upload.ID(),
url.Values{
"_state": []string{token},
})
if err != nil {
ctxu.GetLogger(buh).Infof("error building upload url: %s", err)
return err
}
endRange := offset
if endRange > 0 {
endRange = endRange - 1
}
w.Header().Set("Docker-Upload-UUID", buh.UUID)
w.Header().Set("Location", uploadURL)
w.Header().Set("Content-Length", "0")
w.Header().Set("Range", fmt.Sprintf("0-%d", endRange))
return nil
}