OSDI'20 [0] [0] https://sites.google.com/site/tangchq/papers/Twine-OSDI20.pdf twagent Container (aka “task”): ● namespaces: cgroup, mount, pid and optionally: ipc, net, user, uts ● cgroup v2 ● ... other ): cgroup-bpf 3 Task IP assignment (aka IP-per-task) ● Facebook DC network is IPv6 only ● Every server has /64 IPv6 prefix ● Convenient to have a unique IPv6 per twagent task (e.g. for QoS tagging) Solution: ● Make task use specified IP by a set of BPF_PROG_TYPE_CGROUP_SOCK_ADDR and BPF_CGROUP_SOCK_OPS programs Move TCP/UDP servers to task IP: ● bind(2): ctx.user_ip6 = task_ip Make TCP/UDP clients

overall vision of eBPF and as of 2023, there is work in progress with the IETF (Internet Engineering Task Force) for standardizing eBPF. Under this new banner, bees now help innovate Teams on various vessels

marathon... ... ... ... ... Done Step 6: Simulate a Web-Server and Clients Use curl to submit a task to Marathon for scheduling, with data to run the simulated web-server provided by the web-server.json 5113 mesos:id=badclie Marathon runs the tasks as batch jobs with stdout logged to task-specific files located in /var/lib/mesos . To simplify the retrieval of the stdout log, use the tail_client pod is started. Note In order for the the cni-install.sh script to work properly, the kubelet task must either be running on the host filesystem of the worker node, or the /etc/cni/net.d and /opt/cni/bin

marathon... ... ... ... ... Done Step 6: Simulate a Web-Server and Clients Use curl to submit a task to Marathon for scheduling, with data to run the simulated web-server provided by the web-server.json f00d::a0f:0:0:c76d 10.15.34.97 ready Marathon runs the tasks as batch jobs with stdout logged to task-specific files located in /var/lib/mesos. To simplify the retrieval of the stdout log, use the tail_client cilium pod is started. Note In order for the the cni-install.sh script to work properly, the kubelet task must either be running on the host filesystem of the worker node, or the /etc/cni/net.d and /opt/cni/bin

marathon... ... ... ... ... Done Step 6: Simulate a Web-Server and Clients Use curl to submit a task to Marathon for scheduling, with data to run the simulated web-server provided by the web-server.json f00d::a0f:0:0:c76d 10.15.34.97 ready Marathon runs the tasks as batch jobs with stdout logged to task-specific files located in /var/lib/mesos. To simplify the retrieval of the stdout log, use the tail_client cilium pod is started. Note In order for the the cni-install.sh script to work properly, the kubelet task must either be running on the host filesystem of the worker node, or the /etc/cni/net.d and /opt/cni/bin

cilium pod is started. Note In order for the the cni-install.sh script to work properly, the kubelet task must either be running on the host filesystem of the worker node, or the /etc/cni/net.d and /opt/cni/bin can backport the PR themselves. Follow the Backporting process guide to know how to perform this task. Coordination In general, coordinating in the #launchpad Slack channel with the other hat owner use the new release Please reach out on the #development channel on Slack for assistance with this task. 16. Update the stable tags for cilium, cilium-operator, cilium-operator-aws, cilium-operator-azure

cilium pod is started. Note In order for the cni-install.sh script to work properly, the kubelet task must either be running on the host filesystem of the worker node, or the /etc/cni/net.d and /opt/cni/bin can backport the PR themselves. Follow the Backporting process guide to know how to perform this task. Coordination In general, coordinating in the #launchpad Slack channel with the other hat owner use the new release Please reach out on the #development channel on Slack for assistance with this task. 19. Update the stable tags for cilium, cilium-operator, cilium-operator-aws, cilium-operator-azure

cilium pod is started. Note In order for the cni-install.sh script to work properly, the kubelet task must either be running on the host filesystem of the worker node, or the /etc/cni/net.d and /opt/cni/bin can backport the PR themselves. Follow the Backporting process guide to know how to perform this task. Coordination In general, coordinating in the #launchpad Slack channel with the other hat owner use the new release Please reach out on the #development channel on Slack for assistance with this task. 16. Update the stable tags for cilium, cilium-operator, cilium-operator-aws, cilium-operator-azure

cilium pod is started. Note In order for the cni-install.sh script to work properly, the kubelet task must either be running on the host filesystem of the worker node, or the /etc/cni/net.d and /opt/cni/bin can backport the PR themselves. Follow the Backporting process guide to know how to perform this task. Coordination In general, coordinating in the #launchpad Slack channel with the other hat owner use the new release Please reach out on the #development channel on Slack for assistance with this task. 20. Update the stable tags for cilium, cilium-operator, cilium-operator-aws, cilium-operator-azure