Reduce timeout for scrapping IMDS and give instruction when fail to scrape IMDS inside container

.github/workflows/build-test-linux.yml

@@ -49,7 +49,7 @@ jobs:
         path: |
               ~/.cache/go-build
               ~/go/pkg/mod
-        key: ${{ runner.os }}-go-${{ hashFiles('**/go.sum') }}
+        key: v1-go-pkg-mod-${{ runner.os }}-${{ hashFiles('**/go.sum') }}
 
     - name: Test
       run: make test

.github/workflows/build-test-macos.yml

@@ -41,7 +41,7 @@ jobs:
         path: |
               ~/Library/Caches/go-build
               ~/go/pkg/mod
-        key: ${{ runner.os }}-go-${{ hashFiles('**/go.sum') }}
+        key: v1-go-pkg-mod-${{ runner.os }}-${{ hashFiles('**/go.sum') }}
 
     - name: Test
       run: make test

.github/workflows/build-test-windows.yml

@@ -39,7 +39,7 @@ jobs:
           path: |
             %LocalAppData%\go-build
             ~/go/pkg/mod
-          key: ${{ runner.os }}-go-${{ hashFiles('**/go.sum') }}
+          key: v1-go-pkg-mod-${{ runner.os }}-${{ hashFiles('**/go.sum') }}
 
       - name: Install make
         run: choco install make

.github/workflows/integrationTest.yml

@@ -116,7 +116,7 @@ jobs:
           path: |
             ~/go/pkg/mod
             ~/.cache/go-build
-          key: v1-go-pkg-mod-${{ hashFiles('**/go.sum') }}
+          key: v1-go-pkg-mod-${{ runner.os }}-${{ hashFiles('**/go.sum') }}
 
       - name: Cache binaries
         id: cached_binaries

amazon-cloudwatch-container-insights/cloudwatch-agent-dockerfile/source/Dockerfile

@@ -16,6 +16,7 @@ ENV GO111MODULE=${GO111MODULE}
 COPY go.mod /go/src/github.com/aws/amazon-cloudwatch-agent/
 COPY go.sum /go/src/github.com/aws/amazon-cloudwatch-agent/
 RUN go mod download -x
+
 COPY . /go/src/github.com/aws/amazon-cloudwatch-agent/
 RUN make build-for-docker-${TARGETARCH}
 

amazon-cloudwatch-container-insights/k8s-yaml-templates/cwagent-kubernetes-monitoring/cwagent-daemonset.yaml

@@ -89,4 +89,4 @@ spec:
           hostPath:
             path: /dev/disk/
       terminationGracePeriodSeconds: 60
-      serviceAccountName: cloudwatch-agent
+      serviceAccountName: cloudwatch-agent

plugins/processors/ec2tagger/constants.go

@@ -0,0 +1,79 @@
+// Copyright Amazon.com, Inc. or its affiliates. All Rights Reserved.
+// SPDX-License-Identifier: MIT
+
+package ec2tagger
+
+import (
+	"time"
+)
+
+// Reminder, keep this in sync with the plugin's README.md
+const sampleConfig = `
+  ##
+  ## ec2tagger calls AWS api to fetch EC2 Metadata and Instance Tags and EBS Volumes associated with the
+  ## current EC2 Instance and attached those values as tags to the metric.
+  ##
+  ## Frequency for the plugin to refresh the EC2 Instance Tags and ebs Volumes associated with this Instance.
+  ## Defaults to 0 (no refresh).
+  ## When it is zero, ec2tagger doesn't do refresh to keep the ec2 tags and ebs volumes updated. However, as the
+  ## AWS api request made by ec2tagger might not return the complete values (e.g. initial api call might return a
+  ## subset of ec2 tags), ec2tagger will retry every 3 minutes until all the tags/volumes (as specified by
+  ## "ec2_instance_tag_keys"/"ebs_device_keys") are retrieved successfully. (Note when the specified list is ["*"],
+  ## there is no way to check if all tags/volumes are retrieved, so there is no retry in that case)
+  # refresh_interval_seconds = 60
+  ##
+  ## Add tags for EC2 Metadata fields.
+  ## Supported fields are: "InstanceId", "ImageId" (aka AMI), "InstanceType"
+  ## If the configuration is not provided or it has an empty list, no EC2 Metadata tags are applied.
+  # ec2_metadata_tags = ["InstanceId", "ImageId", "InstanceType"]
+  ##
+  ## Add tags retrieved from the EC2 Instance Tags associated with this instance.
+  ## If this configuration is not provided, or has an empty list, no EC2 Instance Tags are applied.
+  ## If this configuration contains one entry and its value is "*", then ALL EC2 Instance Tags for the instance are applied.
+  ## Note: This plugin renames the "aws:autoscaling:groupName" EC2 Instance Tag key to be spelled "AutoScalingGroupName".
+  ## This aligns it with the AutoScaling dimension-name seen in AWS CloudWatch.
+  # ec2_instance_tag_keys = ["aws:autoscaling:groupName", "Name"]
+  ##
+  ## Retrieve ebs_volume_id for the specified devices, add ebs_volume_id as tag. The specified devices are
+  ## the values corresponding to the tag key "disk_device_tag_key" in the input metric.
+  ## If this configuration is not provided, or has an empty list, no ebs volume is applied.
+  ## If this configuration contains one entry and its value is "*", then all ebs volume for the instance are applied.
+  # ebs_device_keys = ["/dev/xvda", "/dev/nvme0n1"]
+  ##
+  ## Specify which tag to use to get the specified disk device name from input Metric
+  # disk_device_tag_key = "device"
+  ##
+  ## Amazon Credentials
+  ## Credentials are loaded in the following order
+  ## 1) Assumed credentials via STS if role_arn is specified
+  ## 2) explicit credentials from 'access_key' and 'secret_key'
+  ## 3) shared profile from 'profile'
+  ## 4) environment variables
+  ## 5) shared credentials file
+  ## 6) EC2 Instance Profile
+  # access_key = ""
+  # secret_key = ""
+  # token = ""
+  # role_arn = ""
+  # profile = ""
+  # shared_credential_file = ""
+`
+
+const (
+	ec2InstanceTagKeyASG = "aws:autoscaling:groupName"
+	cwDimensionASG       = "AutoScalingGroupName"
+	mdKeyInstanceId      = "InstanceId"
+	mdKeyImageId         = "ImageId"
+	mdKeyInstanceType    = "InstanceType"
+	ebsVolumeId          = "EBSVolumeId"
+)
+
+const (
+	allowedIMDSRetries = 2
+)
+
+var (
+	defaultIMDSTimeout     = 1 * time.Second
+	defaultRefreshInterval = 180 * time.Second
+	backoffSleepArray      = []time.Duration{0, 1 * time.Minute, 1 * time.Minute, 3 * time.Minute, 3 * time.Minute, 3 * time.Minute, 10 * time.Minute} // backoff retry for ec2 describe instances API call. Assuming the throttle limit is 20 per second. 10 mins allow 12000 API calls.
+)