Commit Graph

24 Commits

Author SHA1 Message Date
6b4c313413 Merge pull request #15886 from xiekeyang/master
Fix: Docker Daemon didn't send actual actions client ask for to Token Server
Upstream-commit: d25dada639fe475ff6496b1d357d2250a0e014e3
Component: engine
2015-09-14 09:47:19 -07:00
189e37d591 Remove io.Pipe used for download progress
The original purpose of this was to cancel downloads if pullV2Tag
returns an error, preventing an associated crash (see #15353). The
broadcaster now accomplishes the same thing that the pipe does, making
the pipe redundant. When pullV2Tag returns, all broadcasters are closed,
which means all further writes to those broadcasters will return errors.

Signed-off-by: Aaron Lehmann <aaron.lehmann@docker.com>
Upstream-commit: 50c355777e31c08d3a97ec70b67bbe5a3bc883af
Component: engine
2015-09-11 10:54:40 -07:00
0534c3bd2c Fix races accessing d.poolKey and d.tmpFile when pullV2Tag returns
Signed-off-by: Aaron Lehmann <aaron.lehmann@docker.com>
Upstream-commit: 64faec8269f33be3ac3f3a80322a619668bcb9d5
Component: engine
2015-09-11 10:54:35 -07:00
e7b8d5068c Fix race condition when waiting for a concurrent layer pull
Before, this only waited for the download to complete. There was no
guarantee that the layer had been registered in the graph and was ready
use. This is especially problematic with v2 pulls, which wait for all
downloads before extracting layers.

Change Broadcaster to allow an error value to be propagated from Close
to the waiters.

Make the wait stop when the extraction is finished, rather than just the
download.

This also fixes v2 layer downloads to prefix the pool key with "layer:"
instead of "img:". "img:" is the wrong prefix, because this is what v1
uses for entire images. A v1 pull waiting for one of these operations to
finish would only wait for that particular layer, not all its
dependencies.

Signed-off-by: Aaron Lehmann <aaron.lehmann@docker.com>
Upstream-commit: 23e68679f080fee7ceb25cf791832f523a3a024a
Component: engine
2015-09-11 10:52:34 -07:00
075e783df0 Fix goroutine leak on pull
Close the pipeWriter even if there was no error.

Signed-off-by: Tonis Tiigi <tonistiigi@gmail.com>
Upstream-commit: 3037e4f7c4eec06cf3a5afca6f58ec58b4bdd63e
Component: engine
2015-08-31 15:56:19 -07:00
e9d8afb4a3 Fix: Docker Daemon didn't send actual actions client ask for to Token Server
The Docker Daemon should send actual actions client ask for to issue tokens,

not all the permissions that client is guaranteed.

Signed-off-by: xiekeyang <xiekeyang@huawei.com>
Upstream-commit: e451ddbf4657cbe72f539e1f9513a5bc3a31f3d6
Component: engine
2015-08-30 16:40:26 +08:00
49179849ea Remove "...already being pulled" messages
These don't get seen, so they are unnecessary.

Signed-off-by: Aaron Lehmann <aaron.lehmann@docker.com>
Upstream-commit: 9b9d70ad271b46a67bde57b8a807121f6e85b75f
Component: engine
2015-08-27 13:22:08 -07:00
d4bbf5d640 Clean up ProgressStatus
- Rename to Broadcaster

- Document exported types

- Change Wait function to just wait. Writing a message to the writer and
  adding the writer to the observers list are now handled by separate
  function calls.

- Avoid importing logrus (the condition where it was used should never
  happen, anyway).

- Make writes non-blocking

Signed-off-by: Aaron Lehmann <aaron.lehmann@docker.com>
Upstream-commit: 26c9b585042ac7dce8db83478a69fd01a4b003d7
Component: engine
2015-08-27 13:22:08 -07:00
4366d3278e Change poolAdd to return a boolean instead of an error
Previously, its other return value was used even when it returned an
error. This is awkward and goes against the convention. It also could
have resulted in a nil pointer dereference when an error was returned
because of an unknown pool type. This changes the unknown pool type
error to a panic (since the pool types are hardcoded at call sites and
must always be "push" or "pull"), and returns a "found" boolean instead
of an error.

Signed-off-by: Aaron Lehmann <aaron.lehmann@docker.com>
Upstream-commit: 80513d85cfc0e46f8202fc3030f11052bbfeea7a
Component: engine
2015-08-27 11:18:42 -07:00
2b816e64e2 Show pull progress in terminal for inflight pull requests
Based on #12874 from Sam Abed <sam.abed@gmail.com>. His original commit
was brought up to date by manually porting the changes in pull.go into
the new code in pull_v1.go and pull_v2.go.

Fixes #8385

Signed-off-by: Aaron Lehmann <aaron.lehmann@docker.com>
Upstream-commit: 572f008e892b06262963ccb75c631f2f5d6c6fcd
Component: engine
2015-08-27 11:18:40 -07:00
bab25931f6 Merge pull request #15646 from tonistiigi/15589-pull-cancellation
Fix pull on client disconnect
Upstream-commit: 0634c653d5214c7d8e5f1a154ce35d40ac8d25d7
Component: engine
2015-08-27 17:54:08 +02:00
9c87b64804 Remove nil contexts
Causes daemon panic because loggers can’t be found.

Fixes #15724

Signed-off-by: Tonis Tiigi <tonistiigi@gmail.com>
Upstream-commit: 0c08913d523da2a935e65ade93884194c1906991
Component: engine
2015-08-21 10:31:27 -07:00
2a748ae2e6 Fix pull on client disconnect
Fixes #15589

Signed-off-by: Tonis Tiigi <tonistiigi@gmail.com>
Upstream-commit: 8eeafa058c762b303436422a4682c364fe469985
Component: engine
2015-08-18 09:41:37 -07:00
8771bcf9d5 Refactoring of pullV2Tag
* use downloadInfo pointers everywhere
* use downloads slice only for things that we really download
* cleanup tmp files in all cases

Signed-off-by: Alexander Morozov <lk4d4@docker.com>
Upstream-commit: 9b2426c606b92fa37b7484d9d884edd43352b69c
Component: engine
2015-08-13 11:50:19 -07:00
fc48931914 [graph] Use a pipe for downloads to write progress
The process of pulling an image spawns a new goroutine for each layer in the
image manifest. If any of these downloads fail we would stop everything and
return the error, even though other goroutines would still be running and
writing output through a progress reader which is attached to an http response
writer. Since the request handler had already returned from the first error,
the http server panics when one of these download goroutines makes a write to
the response writer buffer.

This patch prevents this crash in the daemon http server by waiting for all of
the download goroutines to complete, even if one of them fails. Only then does
it return, terminating the request handler.

Docker-DCO-1.1-Signed-off-by: Josh Hawn <josh.hawn@docker.com> (github: jlhawn)
Upstream-commit: d80c4244d322fe0a9cbdd996d23e37fb5b089544
Component: engine
2015-08-05 18:13:39 -07:00
aa73b79fe2 Merge pull request #14921 from aaronlehmann/int64
Fix uses of "int" where "int64" should be used instead
Upstream-commit: 7374852be9def787921aea2ca831771982badecf
Component: engine
2015-08-04 19:16:13 -07:00
d08d95f41d [graph] Enforce manifest/layer digest verification
We noticed a regression since the 1.7.1 patch after some refactoring. This
patch corrects the behavior and adds integration tests for modified manifest
and rootfs layer blobs.

Docker-DCO-1.1-Signed-off-by: Josh Hawn <josh.hawn@docker.com> (github: jlhawn)
Upstream-commit: de52a3bcaa1742054be184cc3465f0933f6d383a
Component: engine
2015-08-03 11:41:23 -07:00
761b152e62 Fix uses of "int" where "int64" should be used instead
Some structures use int for sizes and UNIX timestamps. On some
platforms, int is 32 bits, so this can lead to the year 2038 issues and
overflows when dealing with large containers or layers.

Consistently use int64 to store sizes and UNIX timestamps in
api/types/types.go. Update related to code accordingly (i.e.
strconv.FormatInt instead of strconv.Itoa).

Use int64 in progressreader package to avoid integer overflow when
dealing with large quantities. Update related code accordingly.

Signed-off-by: Aaron Lehmann <aaron.lehmann@docker.com>
Upstream-commit: 1f61084d83aea37b212468aaa975020094b7f7c9
Component: engine
2015-07-31 16:31:40 -07:00
e1b2148585 /graph fix lin errors/warnings
Addresses #14756
Signed-off-by: Srini Brahmaroutu <srbrahma@us.ibm.com>
Upstream-commit: 1d6e44311916b26e1cd128e96ea9f0e5ede5f9df
Component: engine
2015-07-29 20:59:36 +00:00
7e6fe45a4f Use notary library for trusted image fetch and signing
Add a trusted flag to force the cli to resolve a tag into a digest via the notary trust library and pull by digest.
On push the flag the trust flag will indicate the digest and size of a manifest should be signed and push to a notary server.
If a tag is given, the cli will resolve the tag into a digest and pull by digest.
After pulling, if a tag is given the cli makes a request to tag the image.

Use certificate directory for notary requests

Read certificates using same logic used by daemon for registry requests.

Catch JSON syntax errors from Notary client

When an uncaught error occurs in Notary it may show up in Docker as a JSON syntax error, causing a confusing error message to the user.
Provide a generic error when a JSON syntax error occurs.

Catch expiration errors and wrap in additional context.

Signed-off-by: Derek McGowan <derek@mcgstyle.net> (github: dmcgowan)
Upstream-commit: ed13c3abfb242905ec012e8255dc6f26dcf122f6
Component: engine
2015-07-24 14:08:20 -07:00
f7a16948b8 Do Docker edits so we can use the new distribution code
Signed-off-by: Doug Davis <dug@us.ibm.com>
Upstream-commit: b874ef8f43754e5c0fe40b90c3533db93491753e
Component: engine
2015-07-23 10:24:46 -07:00
b70343dc2b Fixing Image struct to no longer use Graph.
Signed-off-by:  John Howard <jhoward@microsoft.com>
Upstream-commit: 9001ea26e7821239207586ad23c78c67d6926bfc
Component: engine
2015-07-20 13:59:53 -07:00
371476f565 fix 8926: rmi dangling is unsafe when pulling
Signed-off-by: Ma Shimiao <mashimiao.fnst@cn.fujitsu.com>
Signed-off-by: Tibor Vass <tibor@docker.com>
Upstream-commit: 1b67c38f6f4a709e1d159445b56f0beee8cfad3a
Component: engine
2015-07-17 11:39:57 -04:00
1492d5ce64 Update graph to use vendored distribution client for the v2 codepath
Signed-off-by: Derek McGowan <derek@mcgstyle.net> (github: dmcgowan)
Signed-off-by: Tibor Vass <tibor@docker.com>
Upstream-commit: 19515a7ad859b28c474d81e756ac245afcd968e3
Component: engine
2015-07-16 13:13:47 -04:00