v0.1.0 - first cut

What's Changed

• Update OWNERS - Add dims by @dims in #1
• Initial drop of remote test-e2e-node equivalent for AWS by @dims in #2
• Ubuntu 2204 Instance config by @upodroid in #5
• switch default user id for ubuntu image by @dims in #6
• Update dependencies to latest k8s master branch by @dims in #7
• Force the old style cgroup non systemd by @dims in #9
• Add ec2-user to root group to allow access to containerd.sock by @dims in #8
• Fix k/k path by @upodroid in #10
• Allow access to SSH by @upodroid in #11
• Revert "Fix k/k path" by @upodroid in #13
• Fix bugs and abstract some parameters by @upodroid in #12
• Add a new image for e2e-node testing with AL2023 (kernel 6.1) by @dims in #14
• Enable systemd cgroup driver by @dims in #15
• Drop Temporary SG hack for aws ec2 authorize-security-group-ingress by @dims in #16
• Adding configuration for arm64 based CI jobs by @dims in #19
• Prep for arm64 based ci jobs by @dims in #18
• set USE_DOCKERIZED_BUILD to false explicitly by @dims in #21
• Pick suitable instance type for arm64 by @dims in #22
• create the CNI directory and template by @dims in #23
• Drop dependency on "jq" by @dims in #24
• Add better defaults for arm64/e2e-node ci jobs by @dims in #25
• cleanup cni.template by @dims in #26
• forgot to delete one line by @dims in #27
• Updates to ebs disk size and dealing with instances by @dims in #28
• Tweak inotify settings by @dims in #29
• Use specific version of containerd for arm64 - v1.6.21 by @dims in #30
• if there are serial tests use one ginkgo node by @dims in #31
• whoops we lost "focus" param by @dims in #32
• Switch to newer Ubuntu - focal to jammy by @dims in #33
• Bump instance type size and switch back to newest containerd by @dims in #34
• Bump disk still running out of space by @dims in #35
• Lookup root device name for AMI by @dims in #36
• Default to 4 hours of test timeout by @dims in #37
• whoops! set the actual timeout by @dims in #38
• Drop back to lower size of instance type by @dims in #40
• Ensure instance-id resolves to the host ip address on ubuntu by @dims in #41
• Graceful node shutdown conflicts with unattended-upgrades package by @dims in #42
• Add kubetest2 support by @upodroid in #39
• Move EKS images to another yaml file by @dims in #43
• install nfs-common package as a e2e node test is mounting an nfs server by @dims in #44
• Capture exit code from piped process by @dims in #45
• Update vendor/ to latest k/k (June 29th) by @dims in #46
• Install ecr-credential-provider during node e2e tests by @dims in #47
• Drop down to 50 gb disk by @dims in #48
• Add a GitHub action to update kubernetes periodically by @dims in #49
• forgot to checkout kubernetes by @dims in #50
• drop path and fix typo by @dims in #51
• hmm we will need path to ensure it goes to k8s.io/kubernetes by @dims in #52
• use relative root instead of GOPATH by @dims in #53
• better detection of kubernetes directory (take 3) by @dims in #54
• try checking out both github repos correctly again (take 4) by @dims in #55
• change working-directory to path by @dims in #57
• path should be indented right by @dims in #58
• provider-aws-test-infra uses main not master by @dims in #59
• Update dependencies by @github-actions in #60
• Update dependencies by @github-actions in #62
• Scripts to build amazon-eks-ami based AMI from scratch for testing by @dims in #61
• option to BUILD_EKS_AMI by @dims in #63
• use sudo to install 10-testcni.conflist by @dims in #66
• Build EKS AMI and then run node e2e tests using that new AMI by @dims in #67
• fix check for existing image by @dims in #68
• fix setting TEST_INFRA_ROOT by @dims in #69
• Update dependencies by @github-actions in #65
• Add ec2-instance-connect package for al2023 by @dims in #73
• Support for testing al2023 based eks images by @dims in #75
• add support to build arm64 based EKS AMI(s) by @dims in #76
• fix eks instance type for arm64 by @dims in #77
• fixup references to x86_64 irrespective of BUILD_EKS_AMI_OS setting by @dims in #78
• separate the instance type determination from the _OS check by @dims in #79
• Ensure we use systemd cgroup with al2023 by @dims in #81
• Default arch is amd64 when building eks image by @dims in #82
• Use the right containerd config.toml by @dims in #83
• set packer variables when BUILD_EKS_AMI_OS is NOT arm64 by @dims in #84
• fix typo in file name by @dims in #85
• remove the systemd hack on al2023 by @tzneal in #86
• Trying to fix update dependencies action by @dims in #87
• Add a new dependency from staging - k8s.io/apiextensions-apiserver by @dims in #88
• Update dependencies by @github-actions in #89
• Run e2e tests without specifying an image config file by @upodroid in #80
• Drop sha256 sum check for CI by @dims in #90
• Update dependencies by @github-actions in #91
• remove the runtime.slice configuration by @tzneal in #92
• ensure userdata works when image-config file isn't being used by @upodroid in #93
• Adding a kubetest2 deployer for EC2 by @dims in #95
• move kubetest2-ec2 into a standalone go module by @dims in #97
• Drop the dirty suffix in version number by @dims in #98
• implement DumpClusterLogs - Download kubelet/cloud-init/cluster-info logs by @dims in #99
• Support new parameter "num-nodes" for worker nodes by @dims in #101
• wait for and print info on nodes when Up() wraps up by @dims in #102
• capture containerd related logs by @dims in #103
• Add roles and instance profiles when creating ec2 instances by @dims in #105
• Install ecr credential provider by @dims in #106
• Install aws vpc cni by @dims in #108
• use the usual directory for CNI as AWS VPC CNI does not like alternate directories by @dims in #109
• Update dependencies by @github-actions in #110
• Maintain scripts outside yaml file by @dims in #111
• Disable default cni and stop apparmor by @dims in #112
• Switch to cilium from AWS VPC CNI by @dims in #113
• use singular --image on the command line, also make it optional to specify by @dims in #114
• Do not bail out after creating logs directory by @dims in #115
• Better directory to pick up binaries from by @dims in #116
• Embed the configure.sh and ubuntu2204.yaml into the binary by @dims in #117
• Ability to specify a version that already exists in the s3 bucket by @dims in #118
• Switch back to aws vpc cni - Revert systemd-udev change (MACAddressPolicy) that disables networking from pods by @dims in #119
• Dump logs when Deployer Up() fails for some reason by @dims in #120
• Lookup version when one is NOT specified by @dims in #121
• Avoid adding instance to array multiple times by @dims in #122
• Better defaults for arm64 by @dims in #123
• Grab vpc vni related logs by @dims in #124
• Update dependencies by @github-actions in #125
• Tweak network settings to avoid problems during huge pod churn by @dims in #126
• Dump vpc cni logs first by @dims in #127
• Instance types are way too small by @dims in #128
• Add a README.md for kubetest2-ec2 by @dims in #129
• Update dependencies by @github-actions in #131
• Validate s3 bucket and version directory by @dims in #132
• Allow standing up clusters from release artifacts by @dims in #133
• Switch to using kubeadm config by @dims in #139
• fix staging bucket variations by @dims in #140
• Drop external cloud provider for now by @dims in #141

New Contributors

• @dims made their first contribution in #1
• @upodroid made their first contribution in #5
• @github-actions made their first contribution in #60
• @tzneal made their first contribution in #86

Full Changelog: https://github.com/kubernetes-sigs/provider-aws-test-infra/commits/v0.1.0