Closed hegerdes closed 8 months ago
Hi, I tried reproducing the issue with the plan you have provided and I couldn't get the same results. Could you please try running with the debug flags enabled: TF_LOG=debug IONOS_LOG_LEVEL=trace
And provide us with the log of the patch request that fails, so that we can take a closer look. :exclamation::exclamation::exclamation:Note, these logs will contain your credentials in plain text, so make sure to remove them beforehand.
I run it again - It works with a Linux based image but not a Windows one. Strange since the drive shout not care what bits are one it.
What is even more strange is that if you run it a second time it says that everything is up to date. This is wrong - drive has still the old size. I thing the provider has already update the state before it actually did something (which is a bad idea) and it doesent refresh the deployed state properly either.
Second comment because two logs seem to be to much for GH to handle:
I run it again - It works with a Linux based image but not a Windows one. Strange since the drive shout not care what bits are one it.
This is good to know, it'll help us narrow down the search for the issue. Also thank you for providing us with the logs, we will start investigating this asap.
Hello,
The issue you are currently facing is actually currently caused by boot_cdrom
, which switches the boot device of the server from the inline volume to the cdrom image.
There is currently a known bug which affects updates of the inline volume when a different boot device is set, the fix is work in progress.
Is there an ETA for the fix?
the fix for this is included in a larger feature which will allow the user to select the boot device of a server resource currently we are experimenting with multiple implementations to see which fits best, so it probably won't make it in the next release
Hello, the fix will be available in the next release, we've implemented a new resource that allows the management of boot devices for the server resource.
Description
Unable to update volume inplace. Plan works, apply errors with
"Http method 'PATCH' not allowed. Expected one of [POST,GET,OPTIONS]
and when you run apply again it says that there are no changes to make because all is up to date - thats wrong to. So also a bug in state managment and actal stateExpected behavior
Volume increase works
Environment
Terraform version:
Provider version:
OS:
Configuration Files
How to Reproduce
Create infra, update volume size, apply infra - error
Apply again - TF says its all up to date and nothing to do - wrong volume still has old size
Error and Debug Output
Additional Notes
Resize over API with HTTP-Client (PATCH) works just the TF provider is broken.
References