[
https://issues.apache.org/jira/browse/MESOS-3013?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Connor Doyle updated MESOS-3013:
--------------------------------
Comment: was deleted
(was: Jie, was basing this on the code in MesosContainerizerProcess::launch in
src/slave/containerizer/mesos/containerizer.cpp. See
https://github.com/apache/mesos/blob/0.24.0/src/slave/containerizer/mesos/containerizer.cpp#L575.
Did we misunderstand this?
)
> Extend ContainerInfo to include "NetworkInfo" message
> -----------------------------------------------------
>
> Key: MESOS-3013
> URL: https://issues.apache.org/jira/browse/MESOS-3013
> Project: Mesos
> Issue Type: Task
> Reporter: Kapil Arya
> Assignee: Kapil Arya
> Labels: mesosphere
>
> As per the [design
> doc|https://docs.google.com/document/d/17mXtAmdAXcNBwp_JfrxmZcQrs7EO6ancSbejrqjLQ0g],
> we need to enable frameworks to specify network requirements. The proposed
> message could be along the lines of:
> {code}
> message NetworkInfo {
> enum Protocol {
> IPv4,
> IPv6
> }
> required Protocol protocol;
> // A netgroup is the name given to a set of logically-related IPs that are
> // allowed to communicate within themselves. For example, one might want
> // to create separate netgroups for dev, testing, qa and prod deployment
> // environments.
> repeated string netgroups;
> // Sticky IPs allow a framwork to re-launch a task with the same IP on a
> // different Slave/Node.
> optional bool sticky [default = false];
> // A unique id that the framework uses to "tag" the assigned IP. This "tag"
> // can be later used to "reclaim" IP while relaunching the task.
> optional string id;
> };
> {code}
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
