ServiceMonitor 'APPCMD failed with error code 183' @ docker start for aspnet:4.7.2-windowsservercore-ltsc2016

[liberdis]

FROM microsoft/aspnet:4.7.2-windowsservercore-ltsc2016 create a container with some docker run -e environment variables. docker stop container docker start container expected result: container enters and maintains running state actual result: container crashes, producing error at tail of log when servicemonitor attempts to start up: 'APPCMD failed with error code 183'

known workarounds: The July-2017 1.1.0.1 version also caused the same symptom and I have had to roll back to 1.0.0.1 in order to restart my containers because apparently that version does not attempt to update app.config

I have found numerous "duplicate issues" that have been closed yet the issue remains without a clear answer other than to roll back. Most seem to refer back to iis-docker#42 which is closed, citing a July-2017 update however it seems some problems remain

If anyone has any ideas as to what I might try to allow 2.0.1.6 to work in ltsc2016 please let me know and thanks in advance for the help

[bariscaglar]

Can you post your exact dockerfile and the docker run command you use? I recently closed a similar issue in this repo. Perhaps it could provide an answer?

[pf56]

For what it's worth, I am running into the same error message if I try to restart a container after editing the web.config file.

$ docker run -d -p 8000:80 --name aspnet_sample mcr.microsoft.com/dotnet/framework/samples:aspnetapp
$ docker exec -it aspnet_sample powershell
<edit web.config through powershell>
<refreshing the page in the browser confirms that it is still functional>
$ docker stop aspnet_sample
$ docker start aspnet_sample
$ docker logs aspnet_sample

 Service 'w3svc' has been stopped

APPCMD failed with error code 183

Failed to update IIS configuration

The container is now broken and refuses to start.

[dbaltor]

Run into the same error with dotnet/framework/aspnet:3.5 image when trying to restart IIS. As already said, the workaround is to downgrade to an old ServiceMonitor version found here.

[triynko]

This is preventing tasks from launching correctly in AWS ECS, using the Microsoft base image. I don't understand how it's possible for something so simple to be this broken. Importing environment variables from a command line. Seriously? It's broke in the latest version, but some old/obsolete version 1.0.0.1 works? I'm pretty sure that 1.1.0.1 works and 1.0.0.1 does not. Neither does 2.0.1.4. 1.1.0.1 also outputs a ton of logs to the command lines that the others do not. There's no consistency with any of this. What a mess.