* Correct SetMatrix documentation
The SetMatrix is a generic data structure, so the description
should not be tight to any specific use
Signed-off-by: Flavio Crisciani <flavio.crisciani@docker.com>
* Service Discovery reuse name and serviceBindings deletion
- Added logic to handle name reuse from different services
- Moved the deletion from the serviceBindings map at the end
of the rmServiceBindings body to avoid race with new services
Signed-off-by: Flavio Crisciani <flavio.crisciani@docker.com>
* Avoid race on network cleanup
Use the locker to avoid the race between the network
deletion and new endpoints being created
Signed-off-by: Flavio Crisciani <flavio.crisciani@docker.com>
* CleanupServiceBindings to clean the SD records
Allow the cleanupServicebindings to take care of the service discovery
cleanup. Also avoid to trigger the cleanup for each endpoint from an SD
point of view
LB and SD will be separated in the future
Signed-off-by: Flavio Crisciani <flavio.crisciani@docker.com>
* Addressed comments
Signed-off-by: Flavio Crisciani <flavio.crisciani@docker.com>
* NetworkDB deleteEntry has to happen
If there is an error locally guarantee that the delete entry
on network DB is still honored
Signed-off-by: Flavio Crisciani <flavio.crisciani@docker.com>
changed the ipMap to SetMatrix to allow transient states
Compacted the addSvc and deleteSvc into a one single method
Updated the datastructure for backends to allow storing all the information needed
to cleanup properly during the cleanupServiceBindings
Removed the enable/disable Service logic that was racing with sbLeave/sbJoin logic
Add some debug logs to track further race conditions
Signed-off-by: Flavio Crisciani <flavio.crisciani@docker.com>
- They are configuration-only networks which
can be used to supply the configuration
when creating regular networks.
- They do not get allocated and do net get plumbed.
Drivers do not get to know about them.
- They can be removed, once no other network is
using them.
- When user creates a network specifying a
configuration network for the config, no
other network specific configuration field
is are accepted. User can only specify
network operator fields (attachable, internal,...)
- They do not need to have a driver field, that
field gets actually reset upon creation.
Signed-off-by: Alessandro Boch <aboch@docker.com>
With the introduction of a driver generic gossip in libnetwork it is not
necessary for drivers to run their own gossip protocol (like what
overlay driver is doing currently) but instead rely on the gossip
instance run centrally in libnetwork. In order to achieve this, certain
enhancements to driver api are needed. This api aims to provide these
enhancements.
The new api provides a way for drivers to register interest on table
names of their choice by returning a list of table names of interest as
a response to CreateNetwork. By doing that they will get notified if a
CRUD operation happened on the tables of their interest, via the newly
added EventNotify call.
Drivers themselves can add entries to any table during a Join call by
invoking AddTableEntry method any number of times during the Join
call. These entries lifetime is the same as the endpoint itself. As soon
as the container leaves the endpoint, those entries added by driver
during that endpoint's Join call will be automatically removed by
libnetwork. This action may trigger notification of such deletion to all
driver instances in the cluster who have registered interest in that
table's notification.
Signed-off-by: Jana Radhakrishnan <mrjana@docker.com>
Currently the libnetwork function `NewNetwork` does not allow
caller to pass a network ID and it is always generated internally.
This is sufficient for engine use. But it doesn't satisfy the needs
of libnetwork being used as an independent library in programs other
than the engine. This enhancement is one of the many needed to
facilitate a generic libnetwork.
Signed-off-by: Jana Radhakrishnan <mrjana@docker.com>
Because overlay is a builtin driver and global allocation of overlay
resources is probably going to happen in a different node (a single
node) and the actual plumbing of the network is probably going to happen
in all nodes, it makes sense to split the functionality of allocation
into two different packages. The central component(this package) only
implements the NetworkAllocate/Free apis while the distributed
component(the existing overlay driver) implements the rest of the driver
api. This way we can reduce the memory footprint overall.
Signed-off-by: Jana Radhakrishnan <mrjana@docker.com>
Added NetworkAllocate and NetworkFree apis to the list of
driver apis. The intention of the api is to provide a
centralized way of allocating and freeing network resources
for a network which is cross-host.
Signed-off-by: Jana Radhakrishnan <mrjana@docker.com>
Currently driver management logic is tightly coupled with
libnetwork package and that makes it very difficult to
modularize it and use it separately. This PR modularizes
the driver management logic by creating a driver registry
package.
Signed-off-by: Jana Radhakrishnan <mrjana@docker.com>
- Move DiscoverNew() and DiscoverDelete() methods into the new interface
- Add DatastoreUpdate notification
- Now this interface can be implemented by any drivers, not only network drivers
Signed-off-by: Alessandro Boch <aboch@docker.com>
- So that a DHCP based plugin can express it needs
the endpoint MAC address when requested for an IP address.
- In such case libnetwork will allocate one if not already
provided by user
Signed-off-by: Alessandro Boch <aboch@docker.com>
- Allow to create an endpoint as anonymous.
An anonymous endpoint does not get added
to the network service records.
Signed-off-by: Alessandro Boch <aboch@docker.com>
Currently endpoint count is maintained as part of
network object and the endpoint count gets updated
frequently while the rest of network is quite stable.
Because of the frequent updates to endpoint count the
network object is getting marshalled and unmarshalled
ferquently. This is causing a lot of churn and transient
memory usage. Fix this by creating a deparate object of
endpoint count so that only that gets updated.
Signed-off-by: Jana Radhakrishnan <mrjana@docker.com>
- Both an empty and nil list of IpamConf object
will trigger auto-allocation for ipv4.
Auto-allocation for ipv6 will still be excluded
in the two cases above.
Signed-off-by: Alessandro Boch <aboch@docker.com>
- libnetwork should reserve only the auxiliary
addresses which belong to the container
addresable pool. And should fail the network
creation if the aux addr does not belong to
the master pool.
Signed-off-by: Alessandro Boch <aboch@docker.com>
Always on watching of networks and endpoints can
affect scalability of the cluster beyond a few nodes.
Remove pro active watching and watch only the objects
you are interested in.
Signed-off-by: Jana Radhakrishnan <mrjana@docker.com>
The configuration format for docker runtime is based on daemon flags and
hence adjusting the libnetwork configuration to accomodate it by moving
the TOML based configuration to the dnet tool.
Also changed the controller configuration via options
Signed-off-by: Madhu Venugopal <madhu@docker.com>
Currently store makes use of a static isReservedNetwork check to decide
if a network needs to be stored in the distributed store or not. But it
is better if the check is not static, but be determined based on the
capability of the driver that backs the network.
Hence introducing a new capability mechanism to the driver which it can
express its capability during registration. Making use of first such
capability : Scope. This can be expanded in the future for more such cases.
Signed-off-by: Madhu Venugopal <madhu@docker.com>
In the present code, each driver package provides a `New()` method
which constructs a driver of its type, which is then registered with
the controller.
However, this is not suitable for the `drivers/remote` package, since
it does not provide a (singleton) driver, but a mechanism for drivers
to be added dynamically. As a result, the implementation is oddly
dual-purpose, and a spurious `"remote"` driver is added to the
controller's list of available drivers.
Instead, it is better to provide the registration callback to each
package and let it register its own driver or drivers. That way, the
singleton driver packages can construct one and register it, and the
remote package can hook the callback up with whatever the dynamic
driver mechanism turns out to be.
NB there are some method signature changes; in particular to
controller.New, which can return an error if the built-in driver
packages fail to initialise.
Signed-off-by: Michael Bridgen <mikeb@squaremobius.net>
This commits brings in a functionality for remote drivers to register
with LibNetwork. The Built-In remote driver is responsible for the
actual "remote" plugin to be made available.
Having such a mechanism makes libnetwork core not dependent on any
external plugin mechanism and also the Libnetwork NB apis are free of
Driver interface.
Signed-off-by: Madhu Venugopal <madhu@docker.com>
Flavio Crisciani
Alessandro Boch
Santhosh Manohar
Alessandro Boch
Madhu Venugopal
Akihiro Suda
msabansal
Jana Radhakrishnan
Stefan Weil
Vincent Demeester
Ryan Belgrave
Tim Düsterhus
Chun Chen
Michael Bridgen