Windows pods are in crashloopback

[knabben]

Investigate why Windows pods are on crashloopback, logs attached:

vagrant@controlplane:~$ kubectl logs windows-server-iis-7985c648cc-gmtq9 

Success Restart Needed Exit Code      Feature Result                           
------- -------------- ---------      --------------                           
True    No             Success        {Common HTTP Features, Default Documen...
Invoke-WebRequest : The remote name could not be resolved: 
'dotnetbinaries.blob.core.windows.net'
At line:1 char:32
+ ... Web-Server; Invoke-WebRequest -UseBasicParsing -Uri 'https://dotnetbi ...
+                 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
    + CategoryInfo          : InvalidOperation: (System.Net.HttpWebRequest:Htt 
   pWebRequest) [Invoke-WebRequest], WebException
    + FullyQualifiedErrorId : WebCmdletWebResponseException,Microsoft.PowerShe 
   ll.Commands.InvokeWebRequestCommand
 
C:\ServiceMonitor.exe : The term 'C:\ServiceMonitor.exe' is not recognized as 
the name of a cmdlet, function, script file, or operable program. Check the 
spelling of the name, or if a path was included, verify that the path is 
correct and try again.
At line:1 char:311
+ ... ml>' > C:\inetpub\wwwroot\default.html; C:\ServiceMonitor.exe 'w3svc' ...
+                                             ~~~~~~~~~~~~~~~~~~~~~
    + CategoryInfo          : ObjectNotFound: (C:\ServiceMonitor.exe:String) [ 
   ], CommandNotFoundException
    + FullyQualifiedErrorId : CommandNotFoundException

from Jay:

I guess its failing bc it does some weird DNS request. probably bc that private-network doesnt have public DNS. Thats fine, maybe we can just remove thet Invoke-WebRequest call from the IIS pod

[jayunit100]

this is probably my fault, i copy/pasted the wrong windows YAML from the internet :)

[jayunit100]

@knabben .... lets get rid of this ServiceMonitor nonsense ?

[jayunit100]

Lets use this pattern sais @jsturtevant ,... but DONT USE THIS IMAGE , use IIS ?
https://github.com/jsturtevant/windows-k8s-playground/blob/master/deployments/iis/iis-prewarm.yaml

maybe ?

[knabben]

/assign

[johnSchnake]

@knabben is there something I can do to help on this? I was just about to report this same issue as I was working on the sonobuoy integration.

[knabben]

@johnSchnake I'm trying a few images without servicemonitor.exe if you have one more stable feel free to replace the one we have in the yaml

[johnSchnake]

So I'm trying to just "fix" the existing code and found 2 things of note:

• 1. The version of servicemonitor that was specified did not exist on the downloads page for some reason (https://github.com/microsoft/IIS.ServiceMonitor/releases)
• 2. After updating it to an existing release I still got the same error. However, from the NODE it worked so I think something is wrong with the networking on the pod.

 Invoke-WebRequest -UseBasicParsing -Uri 'https://dotnetbinaries.blob.core.windows.net/servicemonitor/2.0.1.9/ServiceMonitor.exe'
...
StatusCode        : 200
StatusDescription : OK

Maybe ya'll already knew that was the case, but it seemed like progress to me. It seems to me that if we cut out the servicemonitor then maybe the iis server will start up but perhaps all the windows nodes will still be effectively broken.

[johnSchnake]

https://stackoverflow.com/questions/60885492/cant-access-internet-from-within-windows-gke-pod says that MAC spoofing needs to be enabled but I'm out of my norm here and unsure if this is something to be done at the virtualbox/vagrant/windows level.

Am I off base here or does this seem like a trail to follow?

[knabben]

We agreed in the last sig-windows pair to change the image here to the official IIS, instead of changing the command of the current YAML.

Probably a better try IMO. We could go with: https://hub.docker.com/_/microsoft-windows-servercore-iis

[johnSchnake]

I think this issue is effectively resolved and we've moved onto having other issues related to the windows networking; I think this can be closed. Right?

[knabben]

yes, we changed the pod spec.

/close

[k8s-ci-robot]

@knabben: Closing this issue.

In response to this:

yes, we changed the pod spec.

/close

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository.