search

aws / containers-roadmap

This is the public roadmap for AWS container services (ECS, ECR, Fargate, and EKS).
https://aws.amazon.com/about-aws/whats-new/containers/
Other
5.22k stars 321 forks source link

[EKS] [Windows] Invalid Hostname override passed to kube-proxy #273

Open mfkuntz opened 5 years ago

mfkuntz commented 5 years ago

There seems to be an issue with kube-proxy on windows containers receiving the wrong Hostname Override parameter from the bootstrap script. Looking at kube-proxy logs, it fails to lookup the ip (from the kubernetes api) for the node and falls back to 127.0.0.1, causing issues.

From there there are errors creating the proxy endpoints. The nodes do appear to work, until you create a deployment with a service and more then 1 replica. With a single replica, everything works, but adding a second (or more) all pods fail.

I can see what causes the error, and have a patch for my specific cluster configuration, but I am not sure why it is necessary for this cluster vs others not having the issues. One guess - this is an older cluster that has been upgraded from 1.10 all the way to current 1.11

Start-EKSBootstrap.ps1 has a $HostName variable from calling the metadata endpoint. The value of that, for an example node, is ip-10-20-66-170. That value is then passed as an arg to kubelet and kube-proxy. The issue comes from the nodes actual name ip-10-20-66-170.ec2.internal.

That value is then used by kube-proxy to lookup the IP here. Using ip-10-20-66-170 returns nil, because that lookup is by name. Using ip-10-20-66-170.ec2.internal works as expected, returning the correct id.

As a quick patch, I added this script block to patch the Bootstrap script to use the "correct" hostname for kube-proxy

function patchBootstrap() {
  $target = "[string]`$EniMACAddress = Get-EC2MetaData 'latest/meta-data/mac'"
  $newStanza = "`$HostName = `$HostName + '.ec2.internal'"
  $replacement = "$target `n $newStanza"

  $script = [System.IO.File]::ReadAllText("C:\Program Files\Amazon\EKS\Start-EKSBootstrap.ps1").Replace($target, $replacement)
  $script | Set-Content -Path "C:\Program Files\Amazon\EKS\Start-EKSBootstrap.ps1" -Force
}

I didn't see the repo for the AMI, so I wasn't sure if there was more documentation for the Bootstrap that would help explain some of this.

AMI - ami-032bdf5292844295a EKS - 1.11 - eks.2 - Updated from 1.10 to 1.11 Kubelet - v1.11.5 Kube-Proxy - v1.11.5 aws-node - amazon-k8s-cni:v1.3.3

Example windows node: 

Name:               ip-10-20-66-170.ec2.internal
Roles:              <none>
Labels:             beta.kubernetes.io/arch=amd64
                    beta.kubernetes.io/instance-type=m3.large
                    beta.kubernetes.io/os=windows
                    failure-domain.beta.kubernetes.io/region=us-east-1
                    failure-domain.beta.kubernetes.io/zone=us-east-1d
                    kubernetes.io/hostname=ip-10-20-66-170
                    mhelpdesk.com/ocean-node=true
Annotations:        node.alpha.kubernetes.io/ttl=0
                    volumes.kubernetes.io/controller-managed-attach-detach=true
CreationTimestamp:  Wed, 01 May 2019 11:33:15 -0400
Taints:             os=Win1809:NoSchedule
Unschedulable:      false
Conditions:
  Type             Status  LastHeartbeatTime                 LastTransitionTime                Reason                       Message
  ----             ------  -----------------                 ------------------                ------                       -------
  OutOfDisk        False   Wed, 01 May 2019 13:37:26 -0400   Wed, 01 May 2019 11:33:14 -0400   KubeletHasSufficientDisk     kubelet has sufficient disk space available
  MemoryPressure   False   Wed, 01 May 2019 13:37:26 -0400   Wed, 01 May 2019 11:33:14 -0400   KubeletHasSufficientMemory   kubelet has sufficient memory available
  DiskPressure     False   Wed, 01 May 2019 13:37:26 -0400   Wed, 01 May 2019 11:33:14 -0400   KubeletHasNoDiskPressure     kubelet has no disk pressure
  PIDPressure      False   Wed, 01 May 2019 13:37:26 -0400   Wed, 01 May 2019 11:33:14 -0400   KubeletHasSufficientPID      kubelet has sufficient PID available
  Ready            True    Wed, 01 May 2019 13:37:26 -0400   Wed, 01 May 2019 11:33:14 -0400   KubeletReady                 kubelet is posting ready status
Addresses:
  InternalIP:   10.20.66.170
  InternalDNS:  ip-10-20-66-170
  Hostname:     ip-10-20-66-170
Capacity:
 cpu:                                   2
 ephemeral-storage:                     0
 memory:                                7863920Ki
 pods:                                  110
 vpc.amazonaws.com/ENI:                 1
 vpc.amazonaws.com/PrivateIPv4Address:  9
Allocatable:
 cpu:                                   2
 ephemeral-storage:                     0
 memory:                                7863920Ki
 pods:                                  110
 vpc.amazonaws.com/ENI:                 1
 vpc.amazonaws.com/PrivateIPv4Address:  9
System Info:
 Machine ID:      QA-EKSW-07e4f48
 System UUID:     
 Boot ID:         
 Kernel Version:  10.0.17763.316

 OS Image:                   Windows Server 2019 Datacenter
 Operating System:           windows
 Architecture:               amd64
 Container Runtime Version:  docker://18.9.2
 Kubelet Version:            v1.11.5
 Kube-Proxy Version:         v1.11.5
ExternalID:                  ip-10-20-66-170.ec2.internal
ProviderID:                  aws:///us-east-1d/i-07e4f48c4572a74dd
Non-terminated Pods:         (2 in total)
  Namespace                  Name                     CPU Requests  CPU Limits  Memory Requests  Memory Limits
  ---------                  ----                     ------------  ----------  ---------------  -------------
  qa                         mapi-55f5bf5886-zl7qz    0 (0%)        0 (0%)      0 (0%)           0 (0%)
  qa                         mhdv2-8645bd495-ngch4    0 (0%)        0 (0%)      0 (0%)           0 (0%)
Allocated resources:
  (Total limits may be over 100 percent, i.e., overcommitted.)
  CPU Requests  CPU Limits  Memory Requests  Memory Limits
  ------------  ----------  ---------------  -------------
  0 (0%)        0 (0%)      0 (0%)           0 (0%)
Events:         <none>
somujay commented 5 years ago

Hostname in Start-EKSBootstrap.ps1 script should represent internal private DNS name of your EC2 instance. You can verify that by running Invoke-RestMethod 'http://169.254.169.254/latest/meta-data/local-hostname' . This should return private DNS name (This is what Get-EC2Metadata call returns in Register-KubernetesServices). You shouldn't hardcode the Hostname as above. Let us know if you need additional information.

mfkuntz commented 5 years ago

I agree, I am not sure where the difference in the host name comes from.

The host name returned from the metadata endpoint does not contain the ec2.internal suffix, so I guess I will scan through the kubelet args and code to see where that comes from.

There is an interesting thing though: In the AWS console UI, the private dns does have the correct suffix, but the meta data endpoint does not. I'll look into why that is as well.

image

image

The VPC does have the dns resolution and hostnames flags set.

edit; yeah, it looks like it uses the cloud controller manager to lookup the nodename from the API directly, not from the instance endpoint.

kubelet cloud provider

Could be related to https://github.com/kubernetes/kubernetes/issues/11543, need to dig more

somujay commented 5 years ago

I agree, I am not sure where the difference in the host name comes from.

The host name returned from the metadata endpoint does not contain the ec2.internal suffix, so I guess I will scan through the kubelet args and code to see where that comes from.

There is an interesting thing though: In the AWS console UI, the private dns does have the correct suffix, but the meta data endpoint does not. I'll look into why that is as well.

image

image

The VPC does have the dns resolution and hostnames flags set.

edit; yeah, it looks like it uses the cloud controller manager to lookup the nodename from the API directly, not from the instance endpoint.

kubelet cloud provider

Could be related to kubernetes/kubernetes#11543, need to dig more

Interesting. can you run ipconfig /all on this instance and send it to me?

vorantz commented 5 years ago

Couold this be related? https://github.com/aws/containers-roadmap/issues/236 I had DNS resolution issues because of this

somujay commented 5 years ago

Couold this be related? #236 I had DNS resolution issues because of this The above mentioned DNS issue can occur when accessing kubernetes services inside pod. In your case, metadata service doesn't return the fully qualified name. These are two different ones.

mfkuntz commented 5 years ago

Sorry for the big delay in responding, but I am back on the project now.

Seems that having kube-proxy lookup the IP with a different method is a design choice: https://github.com/kubernetes/kubernetes/issues/71851

ipconfig /all

Windows IP Configuration

   Host Name . . . . . . . . . . . . : QA-EKSW-0d955d1
   Primary Dns Suffix  . . . . . . . :
   Node Type . . . . . . . . . . . . : Hybrid
   IP Routing Enabled. . . . . . . . : No
   WINS Proxy Enabled. . . . . . . . : No
   DNS Suffix Search List. . . . . . : us-east-1.ec2-utilities.amazonaws.com
                                       us-west-2.ec2-utilities.amazonaws.com
                                       us-west-2.compute.internal

Ethernet adapter vEthernet (Ethernet 3):

   Connection-specific DNS Suffix  . :
   Description . . . . . . . . . . . : Hyper-V Virtual Ethernet Adapter #2
   Physical Address. . . . . . . . . : 12-12-16-14-BD-56
   DHCP Enabled. . . . . . . . . . . : Yes
   Autoconfiguration Enabled . . . . : Yes
   Link-local IPv6 Address . . . . . : fe80::3049:d3a:fbad:3f5c%16(Preferred)
   IPv4 Address. . . . . . . . . . . : 10.20.202.92(Preferred)
   Subnet Mask . . . . . . . . . . . : 255.255.192.0
   Lease Obtained. . . . . . . . . . : Wednesday, July 17, 2019 5:25:08 PM
   Lease Expires . . . . . . . . . . : Wednesday, July 17, 2019 6:55:10 PM
   Default Gateway . . . . . . . . . : 10.20.192.1
   DHCP Server . . . . . . . . . . . : 10.20.192.1
   DHCPv6 IAID . . . . . . . . . . . : 269619734
   DHCPv6 Client DUID. . . . . . . . : 00-01-00-01-24-C1-11-63-0E-64-B2-6D-B6-CC
   DNS Servers . . . . . . . . . . . : 10.20.0.2
   NetBIOS over Tcpip. . . . . . . . : Enabled

Ethernet adapter vEthernet (nat):

   Connection-specific DNS Suffix  . :
   Description . . . . . . . . . . . : Hyper-V Virtual Ethernet Adapter
   Physical Address. . . . . . . . . : 00-15-5D-C6-8D-C6
   DHCP Enabled. . . . . . . . . . . : No
   Autoconfiguration Enabled . . . . : Yes
   Link-local IPv6 Address . . . . . : fe80::18f:f87c:c681:6158%11(Preferred)
   IPv4 Address. . . . . . . . . . . : 172.31.208.1(Preferred)
   Subnet Mask . . . . . . . . . . . : 255.255.240.0
   Default Gateway . . . . . . . . . :
   DHCPv6 IAID . . . . . . . . . . . : 184554845
   DHCPv6 Client DUID. . . . . . . . : 00-01-00-01-24-C1-11-63-0E-64-B2-6D-B6-CC
   DNS Servers . . . . . . . . . . . : fec0:0:0:ffff::1%1
                                       fec0:0:0:ffff::2%1
                                       fec0:0:0:ffff::3%1
   NetBIOS over Tcpip. . . . . . . . : Enabled