Open jdaugherty opened 1 year ago
Hi @jdaugherty , thanks for reporting.
Accroding to the description, I may suggest to use ignore_change
to ignore the staging_storage_account_id
prop chaning like the following.
resource "azurerm_site_recovery_replicated_vm" "example" {
# ...
lifecycle {
ignore_changes = [
managed_disk
]
}
}
The reason is that changing this prop (staging_storage_account_id
) will force a new resource created, and the provider can't tell wether the diff is caused by user changing the config or by the Azure.
But, I have tried to crate an azurerm_site_recovery_replicated_vm
resource and triggered "failover" manually via Portal, and the staging_storage_account_id
did not change. Could you please provide more details about how the situation formed so that I can dig further?
have an issue with staging_storage_account_id where in its /storageaccounts/ and then /storageAccounts/ this causes a recreation of DR
Does this not need support for Target storage account , ignoring the managed disk is not a good idea at all any disks changes to the virtual machine will not update the replication.
Is there an existing issue for this?
Community Note
Description
When managing site recovery using the azurerm_site_recovery_replicated_vm resource, the staging_storage_account_id is set on the initial create. After failing over the VM, the staging_storage_account_id is changed by Azure, which causes terraform to recreate the resource. The site recovery is still configured and works. Is it possible to add support for not refreshing the resource if the staging_storage_account_id changes?
New or Affected Resource(s)/Data Source(s)
azurerm_site_recovery_replicated_vm
Potential Terraform Configuration
No response
References
No response