Closed andreaswolf closed 2 months ago
Hi @andreaswolf 👋🏼
Thanks for the report, I was able to reproduce it on local-lvm
as well, so not ZFS-only.
When adding mp
at create time, the argument in POST looks like
mount_point {
volume = "local-lvm"
size = "4G"
path = "mnt/local1"
}
mp0: mp=mnt/local1,volume=local-lvm:4
But when in update it is:
mount_point {
volume = "local-lvm"
size = "4G"
path = "mnt/local2"
}
mp1: acl=0,backup=0,mp=mnt/local2,quota=0,ro=0,replicate=1,shared=0,volume=local-lvm
The size suffix (i.e. :4
) is missing
@all-contributors please add @andreaswolf for bug
@bpg
I've put up a pull request to add @andreaswolf! :tada:
Describe the bug Adding a new volume to an existing LXC container fails. (Presumably this is only an issue on ZFS?).
The error returned is "Error: error updating container: received an HTTP 500 response - Reason: unable to parse volume ID 'local-zfs'"
To Reproduce Steps to reproduce the behavior:
terraform apply
terraform apply
Please also provide a minimal Terraform configuration that reproduces the issue.
The mount point I added in step 3:
Expected behavior The subvolume is created on "local-zfs" and attached to the container.
Additional context Adding the volume works if it is present on container creation already. Removing the volume later on then fails (it is not removed although Terraform reports success in removing it).
TF_LOG=DEBUG terraform apply
):