- Split these options to a separate struct, so that we can handle them in isolation.
- Change some tests to use subtests, and improve coverage
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
This makes it more transparent that it's unused for Linux,
and we don't pass "root", which has no relation with the
path on Linux.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
The containerd client is very chatty at the best of times. Because the
libcontained API is stateless and references containers and processes by
string ID for every method call, the implementation is essentially
forced to use the containerd client in a way which amplifies the number
of redundant RPCs invoked to perform any operation. The libcontainerd
remote implementation has to reload the containerd container, task
and/or process metadata for nearly every operation. This in turn
amplifies the number of context switches between dockerd and containerd
to perform any container operation or handle a containerd event,
increasing the load on the system which could otherwise be allocated to
workloads.
Overhaul the libcontainerd interface to reduce the impedance mismatch
with the containerd client so that the containerd client can be used
more efficiently. Split the API out into container, task and process
interfaces which the consumer is expected to retain so that
libcontainerd can retain state---especially the analogous containerd
client objects---without having to manage any state-store inside the
libcontainerd client.
Signed-off-by: Cory Snider <csnider@mirantis.com>
Commit 483aa6294b introduced a regression, causing
spurious warnings to be shown when starting a daemon for the first time after
a fresh install:
docker info
...
WARNING: IPv4 forwarding is disabled
WARNING: bridge-nf-call-iptables is disabled
WARNING: bridge-nf-call-ip6tables is disabled
The information shown is incorrect, as checking the corresponding options on
the system, shows that these options are available:
cat /proc/sys/net/ipv4/ip_forward
1
cat /proc/sys/net/bridge/bridge-nf-call-iptables
1
cat /proc/sys/net/bridge/bridge-nf-call-ip6tables
1
The reason this is failing is because the daemon itself reconfigures those
options during networking initialization in `configureIPForwarding()`;
cf4595265e/libnetwork/drivers/bridge/setup_ip_forwarding.go (L14-L25)
Network initialization happens in the `daemon.restore()` function within `daemon.NewDaemon()`:
cf4595265e/daemon/daemon.go (L475-L478)
However, 483aa6294b moved detection of features
earlier in the `daemon.NewDaemon()` function, and collects the system information
(`d.RawSysInfo()`) before we enter `daemon.restore()`;
cf4595265e/daemon/daemon.go (L1008-L1011)
For optimization (collecting the system information comes at a cost), those
results are cached on the daemon, and will only be performed once (using a
`sync.Once`).
This patch:
- introduces a `getSysInfo()` utility, which collects system information without
caching the results
- uses `getSysInfo()` to collect the preliminary information needed at that
point in the daemon's lifecycle.
- moves printing warnings to the end of `daemon.NewDaemon()`, after all information
can be read correctly.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
This method returned the network controller, only to set it on the daemon.
While making this change, also;
- update some error messages to be in the correct format
- use errors.Wrap() where possible
- extract configuring networks into a separate function to make the flow
slightly easier to follow.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
This is a method on the daemon, which itself holds the Config, so
there's no need to pass the same configuration as an argument.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
This function was only used in a single place, and pkg/parsers/operatingsystem
already copied the `verNTWorkstation` const, so we might as well move this function
there as well to "unclutter" pkg/system.
The function had no external users, so not adding an alias / stub.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
Finish the refactor which was partially completed with commit
34536c498d, passing around IdentityMapping structs instead of pairs of
[]IDMap slices.
Existing code which uses []IDMap relies on zero-valued fields to be
valid, empty mappings. So in order to successfully finish the
refactoring without introducing bugs, their replacement therefore also
needs to have a useful zero value which represents an empty mapping.
Change IdentityMapping to be a pass-by-value type so that there are no
nil pointers to worry about.
The functionality provided by the deprecated NewIDMappingsFromMaps
function is required by unit tests to to construct arbitrary
IdentityMapping values. And the daemon will always need to access the
mappings to pass them to the Linux kernel. Accommodate these use cases
by exporting the struct fields instead. BuildKit currently depends on
the UIDs and GIDs methods so we cannot get rid of them yet.
Signed-off-by: Cory Snider <csnider@mirantis.com>
Windows Server 2016 (RS1) reached end of support, and Docker Desktop requires
Windows 10 V19H2 (version 1909, build 18363) as a minimum.
This patch makes Windows Server RS5 / ltsc2019 (build 17763) the minimum version
to run the daemon, and removes some hacks for older versions of Windows.
There is one check remaining that checks for Windows RS3 for a workaround
on older versions, but recent changes in Windows seemed to have regressed
on the same issue, so I kept that code for now to check if we may need that
workaround (again);
085c6a98d5/daemon/graphdriver/windows/windows.go (L319-L341)
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
All regular, non-EOL Linux distros now come with more recent kernels
out of the box. There may still be users trying to run on kernel 3.10
or older (some embedded systems, e.g.), but those should be a rare
exception, which we don't have to take into account.
This patch removes the kernel version check on Linux, and the corresponding
DOCKER_NOWARN_KERNEL_VERSION environment that was there to skip this
check.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
The `daemon.RawSysInfo()` function can be a heavy operation, as it collects
information about all cgroups on the host, networking, AppArmor, Seccomp, etc.
While looking at our code, I noticed that various parts in the code call this
function, potentially even _multiple times_ per container, for example, it is
called from:
- `verifyPlatformContainerSettings()`
- `oci.WithCgroups()` if the daemon has `cpu-rt-period` or `cpu-rt-runtime` configured
- in `ContainerDecoder.DecodeConfig()`, which is called on boith `container create` and `container commit`
Given that this information is not expected to change during the daemon's
lifecycle, and various information coming from this (such as seccomp and
apparmor status) was already cached, we may as well load it once, and cache
the results in the daemon instance.
This patch updates `daemon.RawSysInfo()` to use a `sync.Once()` so that
it's only executed once for the daemon's lifecycle.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
This adds support for 2 runtimes on Windows, one that uses the built-in
HCSv1 integration and another which uses containerd with the runhcs
shim.
Signed-off-by: Brian Goff <cpuguy83@gmail.com>
The "quiet" argument was only used in a single place (at daemon startup), and
every other use had to pass "false" to prevent this function from logging
warnings.
Now that SysInfo contains the warnings that occurred when collecting the
system information, we can make leave it up to the caller to use those
warnings (and log them if wanted).
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
The LCOW implementation in dockerd has been deprecated in favor of re-implementation
in containerd (in progress). Microsoft started removing the LCOW V1 code from the
build dependencies we use in Microsoft/opengcs (soon to be part of Microsoft/hcshhim),
which means that we need to start removing this code.
This first step removes the lcow graphdriver, the LCOW initialization code, and
some LCOW-related utilities.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
After moving libnetwork to this repo, we need to update all the import
paths for libnetwork to point to docker/docker/libnetwork instead of
docker/libnetwork.
This change implements that.
Signed-off-by: Brian Goff <cpuguy83@gmail.com>
This function was removed in the Linux code as part of
f63f73a4a8, but was not removed in
the Windows code.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
Looks like a wrong copy-paste using `RuntimeKernel100ns` twice instead of `RuntimeUser100ns`
Signed-off-by: Vincent Boulineau <vincent.boulineau@datadoghq.com>
* Requires containerd binaries from containerd/containerd#3799 . Metrics are unimplemented yet.
* Works with crun v0.10.4, but `--security-opt seccomp=unconfined` is needed unless using master version of libseccomp
( containers/crun#156, seccomp/libseccomp#177 )
* Doesn't work with master runc yet
* Resource limitations are unimplemented
Signed-off-by: Akihiro Suda <akihiro.suda.cz@hco.ntt.co.jp>
This fix was added in 8e71b1e210 to work around
a go issue (https://github.com/golang/go/issues/20506).
That issue was fixed in
66c03d39f3,
which is part of Go 1.10 and up. This reverts the changes that were made in
8e71b1e210, and are no longer needed.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
Signed-off-by: John Howard <jhoward@microsoft.com>
This is the first step in refactoring moby (dockerd) to use containerd on Windows.
Similar to the current model in Linux, this adds the option to enable it for runtime.
It does not switch the graphdriver to containerd snapshotters.
- Refactors libcontainerd to a series of subpackages so that either a
"local" containerd (1) or a "remote" (2) containerd can be loaded as opposed
to conditional compile as "local" for Windows and "remote" for Linux.
- Updates libcontainerd such that Windows has an option to allow the use of a
"remote" containerd. Here, it communicates over a named pipe using GRPC.
This is currently guarded behind the experimental flag, an environment variable,
and the providing of a pipename to connect to containerd.
- Infrastructure pieces such as under pkg/system to have helper functions for
determining whether containerd is being used.
(1) "local" containerd is what the daemon on Windows has used since inception.
It's not really containerd at all - it's simply local invocation of HCS APIs
directly in-process from the daemon through the Microsoft/hcsshim library.
(2) "remote" containerd is what docker on Linux uses for it's runtime. It means
that there is a separate containerd service running, and docker communicates over
GRPC to it.
To try this out, you will need to start with something like the following:
Window 1:
containerd --log-level debug
Window 2:
$env:DOCKER_WINDOWS_CONTAINERD=1
dockerd --experimental -D --containerd \\.\pipe\containerd-containerd
You will need the following binary from github.com/containerd/containerd in your path:
- containerd.exe
You will need the following binaries from github.com/Microsoft/hcsshim in your path:
- runhcs.exe
- containerd-shim-runhcs-v1.exe
For LCOW, it will require and initrd.img and kernel in `C:\Program Files\Linux Containers`.
This is no different to the current requirements. However, you may need updated binaries,
particularly initrd.img built from Microsoft/opengcs as (at the time of writing), Linuxkit
binaries are somewhat out of date.
Note that containerd and hcsshim for HCS v2 APIs do not yet support all the required
functionality needed for docker. This will come in time - this is a baby (although large)
step to migrating Docker on Windows to containerd.
Note that the HCS v2 APIs are only called on RS5+ builds. RS1..RS4 will still use
HCS v1 APIs as the v2 APIs were not fully developed enough on these builds to be usable.
This abstraction is done in HCSShim. (Referring specifically to runtime)
Note the LCOW graphdriver still uses HCS v1 APIs regardless.
Note also that this does not migrate docker to use containerd snapshotters
rather than graphdrivers. This needs to be done in conjunction with Linux also
doing the same switch.
for windows all networks are re-populated in the store during network controller initialization. In current version it also regenerate network Ids which may be referenced by other components and it may cause broken references to a networks. This commit avoids regeneration of network ids.
Signed-off-by: Andrey Kolomentsev <andrey.kolomentsev@docker.com>
Many startup tasks have to run for each container, and thus using a
WaitGroup (which doesn't have a limit to the number of parallel tasks)
can result in Docker exceeding the NOFILE limit quite trivially. A more
optimal solution is to have a parallelism limit by using a semaphore.
In addition, several startup tasks were not parallelised previously
which resulted in very long startup times. According to my testing, 20K
dead containers resulted in ~6 minute startup times (during which time
Docker is completely unusable).
This patch fixes both issues, and the parallelStartupTimes factor chosen
(128 * NumCPU) is based on my own significant testing of the 20K
container case. This patch (on my machines) reduces the startup time
from 6 minutes to less than a minute (ideally this could be further
reduced by removing the need to scan all dead containers on startup --
but that's beyond the scope of this patchset).
In order to avoid the NOFILE limit problem, we also detect this
on-startup and if NOFILE < 2*128*NumCPU we will reduce the parallelism
factor to avoid hitting NOFILE limits (but also emit a warning since
this is almost certainly a mis-configuration).
Signed-off-by: Aleksa Sarai <asarai@suse.de>
