alvarocurt
commented
1 year ago I solved the problem with a little roundabout as shown before
`resource "null_resource" "espera_datastore" { provisioner "local-exec" { command = "sleep 10" } depends_on = [opennebula_datastore.ceph_image] }
resource "opennebula_image" "imagen" { name = var.image.name path = var.image.path datastore_id = opennebula_datastore.ceph_image.id dev_prefix = "vd" depends_on = [null_resource.espera_datastore] }`
The solution is not perfect and 10s is an arbitrary time. It takes longer than necessary and the optimal solution would be for terraform to know when the oned knows the actual size of the datastore, but for now this will work for me.
Description
I am currently trying to make a definition with all cluster resources in order to raise them all with only one terraform apply. The problem is that when I create simultaneously a resource for the datastore (ceph in this case) and another one for the image to download from the marketplace, OpenNebula does not have time to measure the available storage in the Ceph and I get this error:
│ Error: Failed to create the image │ │ with opennebula_image.imagen, │ on main.tf line 92, in resource "opennebula_image" "imagen": │ 92: resource "opennebula_image" "imagen" { │ │ OpenNebula error [ACTION]: [one.image.allocate] Not enough space in datastoreOf course this error disappears at a second terraform apply, but that's not what I'm looking for What can be done for now to solve this?
Terraform and Provider version
Terraform v1.4.6 on darwin_arm64
Affected resources and data sources
resource "opennebula_datastore" "ceph_image" { name = var.ceph_image_datastore type = "image" cluster_ids = [tonumber(opennebula_cluster.cluster_CPD.id)] bridge_list = var.ceph_pool.libvirt_hosts tags = { "BRIDGE_LIST" = join(" ", var.ceph_pool.libvirt_hosts) "CEPH_HOST" = var.ceph_pool.ceph_mons "CEPH_SECRET" = var.ceph_pool.ceph_secret "POOL_NAME" = var.ceph_pool.ceph_pool "CEPH_USER" = var.ceph_pool.ceph_client "DS_MAD" = "ceph" "TM_MAD" = "ceph" "DISK_TYPE" = "RBD" }
resource "opennebula_image" "imagen" { name = var.image.name path = var.image.path datastore_id = opennebula_datastore.ceph_image.id dev_prefix = "vd" }
Terraform configuration
No response
Expected behavior
Datastore and image created succesfully
Actual behavior
Datastore created but error raised when provisioning the image
Steps to Reproduce
Create an image datastore and create an image in the same file. In the code provided I also previously created a cluster for the datastore, but I think that's not necesary
Debug output
No response
Panic output
No response
Important factoids
No response
References
No response