`--cluster-store` is of form KV-PROVIDER://KV-URL, this commit makes
sure that KV-URL contains no "://"
Signed-off-by: Zhang Wei <zhangwei555@huawei.com>
(cherry picked from commit 0ebdac8bee0c587a8a10e3bcdec46e9d8360e2fb)
There is an extreme corner case where when the daemon
panics at the same time as a container is stopping
and cleaning up the sandbox and the sandbox may have been
left with an inconsistent state. This libnetwork vendoring
fixes that case.
Vendoring in libnetwork @ 5305ea570b85d61dd0fd261cd7e1680da1884678
Signed-off-by: Jana Radhakrishnan <mrjana@docker.com>
(cherry picked from commit 79cffa5ce4a71b29eed66e26c1fb7f153be5c045)
the example command uses `/world` but the description refers to `/src`
Signed-off-by: Harold Cooper <hrldcpr@gmail.com>
(cherry picked from commit f67da614e9963e626d4ec12fd624249947b97d6a)
- During concurrent operations in multihost environment,
it is possible that the implementer of `EndpointInfo`
is nil. It simply means the endpoint is no longer
available in the datastore.
Signed-off-by: Alessandro Boch <aboch@docker.com>
(cherry picked from commit 54d22cbd9a04a965c935a693bf403d2c87109b5a)
Currently, we get the network stats each time per subscriber, causing a
high load of cpu when there are several subscribers per container.
This change makes the daemon to collect once and publish N times, where N is the
number of subscribers per container.
Signed-off-by: David Calavera <david.calavera@gmail.com>
(cherry picked from commit 453473d6a964e58f4a61599bd90d62d0088f0cc2)
By adding a (*WriteFlusher).Close, we limit the Write calls to possibly
deallocated http response buffers to the lifetime of an http request.
Typically, this is seen as a very confusing panic, the cause is usually a
situation where an http.ResponseWriter is held after request completion. We
avoid the panic by disallowing further writes to the response writer after the
request is completed.
Signed-off-by: Stephen J Day <stephen.day@docker.com>
(cherry picked from commit ec2289b2d9ac79fd5e0f69f56f023dfe8ee78bf8)
- On `docker run --net <network id> ...`
the bug would cause the container to attempt
to connect to the network two times
- Also made sure endpoint creation rollback will
be executed on failures in `func (container *Container) connectToNetwork()`
Signed-off-by: Alessandro Boch <aboch@docker.com>
(cherry picked from commit 45e71a79840887b7974387bedba0544015ab4e3b)
And do not try to unmount empty paths.
Because nobody should be woken up in the middle of the night for them.
Signed-off-by: David Calavera <david.calavera@gmail.com>
(cherry picked from commit a54d5932e3a644317c77d59bc5aee562841d5c20)
Instead of using `MNT_DETACH` to unmount the container's mqueue/shm
mounts, force it... but only on daemon init and shutdown.
This makes sure that these IPC mounts are cleaned up even when the
daemon is killed.
Signed-off-by: Brian Goff <cpuguy83@gmail.com>
(cherry picked from commit 78bd17e805b7514505455b10f2fd90962505a3ff)
- the /etc/hosts read caveat due to dynamic update
- information about docker_gwbridge
- Carries and closes#17654
- Updating with last change by Madhu
- Updating with the IPAM api 1.22
Signed-off-by: Mary Anthony <mary@docker.com>
(cherry picked from commit 39dfc536d444df8745124ffa38504241a0a985f7)
Conflicts:
docs/reference/api/docker_remote_api_v1.22.md
updating after assignment for Nigel
Adding in some notes from Nigel work
Updating with the storage driver content Nigel added
Updating with Nigel's polishing tech
Adding in Nigel graphics
First pass of aufs material
Capturing Nigel's latest
Comments back to Nigel on devicemapper
Incorporating Nigel's comments v3
Converting images for dm
Entering comments into aufs page
Adding the btfs storage driver
Moving to userguide
Adding in two new driver articles from Nigel
Optimized images
Updating with comments
Signed-off-by: Mary Anthony <mary@docker.com>
(cherry picked from commit 950fbf99b1214e77c0dcc27e8f9a16e01b2c630a)
Removing old networking.md
Updating dockernetworks.md with images
Adding information on network plugins
Adding blurb about links to docker networking
Updating the working documentation
Adding Overlay Getting Started
Downplaying links by removing refs/examples, adding refs/examples for network.
Updating getting started to reflect networks not links
Pulling out old network material
Updating per discussion with Madhu to add Default docs section
Updating with bridge default
Fix bad merge
Updating with new cluster-advertise behavior
Update working and NetworkSettings examples
Correcting example for default bridge discovery behavior
Entering comments
Fixing broken Markdown Syntax
Updating with comments
Updating all the links
Signed-off-by: Mary Anthony <mary@docker.com>
(cherry picked from commit 9ef855f9e5fa8077468bda5ce43155318c58e60e)
Fixing the weight
Signed-off-by: Mary Anthony <mary@docker.com>
(cherry picked from commit 5ce093e94565a5f32df9adf78dcdb6deeda75f79)
Conflicts:
docs/reference/api/docker_remote_api.md
docs/reference/api/docker_remote_api_v1.21.md
docs/reference/api/docker_remote_api_v1.22.md
For graceful restart case it was done when the container was brought
down. But for ungraceful cases, the persistence is missing for nw
connect
Signed-off-by: Madhu Venugopal <madhu@docker.com>
(cherry picked from commit 401632c7568408ee9689adc1da855cfb1409e906)
This fixes a bug introduced in #15786:
* if a pre-v1.20 client requested docker stats, the daemon
would return both an API-compatible JSON blob *and* an API-incompatible JSON
blob: see https://gist.github.com/donhcd/338a5b3681cd6a071629
Signed-off-by: Donald Huang <don.hcd@gmail.com>
(cherry picked from commit d2c04f844b8258d712da4b8feac25df7590b037c)
The commit title wrongfully mentioned API v1.22, when it meant to mention v1.21.
- So that it complies with docker convention for inspect
Signed-off-by: Alessandro Boch <aboch@docker.com>
(cherry picked from commit d795bc7d53db4b9e9194d0570d65e22d39f58384)
- Keep old fields in NetworkSetting to respect the deprecation policy.
Signed-off-by: David Calavera <david.calavera@gmail.com>
(cherry picked from commit f301c5765a0d7f4b6866cedfdface6f87874ff53)
- So that they comply with docker inspect convention
Which is allowing camel case for json field names
Signed-off-by: Alessandro Boch <aboch@docker.com>
(cherry picked from commit b2d0b75018c5be5c671a15950546a79cda307cf2)
You cannot do this for individual cgroups for all the containers. Only
set the reservation if the user requested it. The error you will
receive is an EINTVAL when you try to set a large limit like we were in
the memory limit.
Signed-off-by: Michael Crosby <crosbymichael@gmail.com>
(cherry picked from commit ecb87ed0a5128a3d9ab3b0a620463b7b56e82bab)
When a container restarts after a ungraceful daemon restart, first
cleanup any unclean sandbox before trying to allocate network resources.
Signed-off-by: Madhu Venugopal <madhu@docker.com>
(cherry picked from commit 0c07096b7da6dc28df10df21283fc21b3b052e1a)
That way swarm can understand the user's intention.
Signed-off-by: David Calavera <david.calavera@gmail.com>
(cherry picked from commit 34668ad68bb42d7488649e41b32818f3f1a7cff2)
- As the retrieved info may not be available at
Endpoint creation time for certain network drivers
- Also retrieve the MAC address from Endpoint.Info().Iface()
Signed-off-by: Alessandro Boch <aboch@docker.com>
(cherry picked from commit e03daebb48221bae84bfc48b0f9b652ced6d93f3)
When an image has multiple tags and rmi is called with force on a tag, only the single tag should be removed.
The current behavior is broken and removes all tags and the image.
Signed-off-by: Derek McGowan <derek@mcgstyle.net> (github: dmcgowan)
(cherry picked from commit 48e7f7963e142a0a45b583c5e65015d896c59433)
Added Note to show users that signals will not propagate to the container if the preferred exec form isn't used.
Signed-off-by: Eric Rosenberg <ehaydenr@gmail.com>
(cherry picked from commit c1a5ee53c19d62918f5ef245b5d2268fb9c89ca8)
Nathan LeClaire