You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Please vote on this issue by adding a 👍 reaction to the original issue to help the community and maintainers prioritize this request
Please do not leave "+1" or "me too" comments, they generate extra noise for issue followers and do not help prioritize the request
If you are interested in working on this issue or have submitted a pull request, please leave a comment
Description
I am running terrafom code that creats VM with 3 NIC and 3 subnet resources in Azure. We challenge a strange behaviour where ~3th run ends up in a loop on "azurerm_subnet.acme_subnet_public: Still creating.". Workaround and possible issues describe below. Actuall final logs ends with:
...
module.re_master_primary.azurerm_network_interface.compute_nic_private: Still creating... (9m49s elapsed)
module.re_network.azurerm_subnet.acme_subnet_public: Still creating... (9m49s elapsed)
module.re_master_primary.azurerm_network_interface.compute_nic_private: Still creating... (9m59s elapsed)
module.re_network.azurerm_subnet.acme_subnet_public: Still creating... (9m59s elapsed)
module.re_master_primary.azurerm_network_interface.compute_nic_private: Still creating... (10m9s elapsed)
module.re_network.azurerm_subnet.acme_subnet_public: Still creating... (10m9s elapsed)
module.re_master_primary.azurerm_network_interface.compute_nic_private: Still creating... (10m19s elapsed)
module.re_network.azurerm_subnet.acme_subnet_public: Still creating... (10m19s elapsed)
module.re_master_primary.azurerm_network_interface.compute_nic_private: Still creating... (10m29s elapsed)
module.re_network.azurerm_subnet.acme_subnet_public: Still creating... (10m29s elapsed)
...
Workaround/Finding
Our investigation showed that it sucesfully passes when compute_nic_private is created early before acme_subnet_public (second subnet) gets in creation state. The reason might be updating network resource. Therefore we introduced delay (a sleep for 30 seconds) to be run on each resource as well as proper dependencies between terraform resources.
Related question:
Can private interface and network modification can be done in parallel?
Why terraforme azure provider does not enforece correct order?
Terraform (and AzureRM Provider) Version
terraform 0.11.11
issue identified with latest stable azurerm v1.21.0 as well as with master (rev: f398858) and a version 4months old
azure-cli==2.0.55
terraform -v
2019/01/24 18:56:11 [INFO] Terraform version: 0.11.11 ac4fff416318bf0915a0ab80e062a99ef3724334
2019/01/24 18:56:11 [INFO] Go runtime version: go1.11.1
2019/01/24 18:56:11 [INFO] CLI args: []string{"/go/bin/terraform", "-v"}
2019/01/24 18:56:11 [DEBUG] Attempting to open CLI config file: /root/.terraformrc
2019/01/24 18:56:11 [DEBUG] File doesn't exist, but doesn't need to. Ignoring.
2019/01/24 18:56:11 [INFO] CLI command args: []string{"version", "-v"}
Terraform v0.11.11
2019/01/24 18:56:11 [DEBUG] checking for provider in "."
2019/01/24 18:56:11 [DEBUG] checking for provider in "/go/bin"
2019/01/24 18:56:11 [WARN] found legacy provider "terraform-provider-azurerm"
2019/01/24 18:56:11 [WARN] found legacy provider "terraform-provider-local"
2019/01/24 18:56:11 [WARN] found legacy provider "terraform-provider-null"
2019/01/24 18:56:11 [WARN] found legacy provider "terraform-provider-random"
2019/01/24 18:56:11 [WARN] found legacy provider "terraform-provider-template"
2019/01/24 18:56:11 [WARN] found legacy provider "terraform-provider-tls"
2019/01/24 18:56:11 [DEBUG] checking for provider in ".terraform/plugins/linux_amd64"
2019/01/24 18:56:11 [DEBUG] found valid plugin: "local", "0.0.0", "/go/bin/terraform-provider-local"
2019/01/24 18:56:11 [DEBUG] found valid plugin: "null", "0.0.0", "/go/bin/terraform-provider-null"
2019/01/24 18:56:11 [DEBUG] found valid plugin: "random", "0.0.0", "/go/bin/terraform-provider-random"
2019/01/24 18:56:11 [DEBUG] found valid plugin: "template", "0.0.0", "/go/bin/terraform-provider-template"
2019/01/24 18:56:11 [DEBUG] found valid plugin: "tls", "0.0.0", "/go/bin/terraform-provider-tls"
2019/01/24 18:56:11 [DEBUG] found valid plugin: "azurerm", "0.0.0", "/go/bin/terraform-provider-azurerm"
2019/01/24 18:56:12 [DEBUG] plugin: waiting for all plugin processes to complete...
+ provider.azurerm (unversioned)
+ provider.local (unversioned)
+ provider.null (unversioned)
+ provider.random (unversioned)
+ provider.template (unversioned)
+ provider.tls (unversioned)
Affected Resource(s)
azurerm_network_interface
azurerm_subnet
Terraform Configuration Files
Corresponding parts of the TF template:
Note: sleep 0.1s leads to an issue (in my case on every ~3rd deployment), with sleep set to 30 (second) the deployment succeeds.
Thanks for opening this issue - apologies for the delayed response here!
Taking a look into this I believe this issue is a duplicate of #2489 which was fixed via #3673 and shipped in v1.33.1 of the Azure Provider - would you be able to upgrade to that version of the Provider and see if this works for you?
Since I believe upgrading to the latest version of the Azure Provider should fix this issue I'm going to close this issue for the moment - however please let us know if that doesn't work for you and we'll take another look.
I'm going to lock this issue because it has been closed for 30 days ⏳. This helps our maintainers find and focus on the active issues.
If you feel this issue should be reopened, we encourage creating a new issue linking back to this one for added context. If you feel I made an error 🤖 🙉 , please reach out to my human friends 👉 [email protected]. Thanks!
ghost
locked and limited conversation to collaborators
Oct 2, 2019
Sign up for freeto subscribe to this conversation on GitHub.
Already have an account?
Sign in.
Community Note
Description
I am running terrafom code that creats VM with 3 NIC and 3 subnet resources in Azure. We challenge a strange behaviour where ~3th run ends up in a loop on "azurerm_subnet.acme_subnet_public: Still creating.". Workaround and possible issues describe below. Actuall final logs ends with:
Workaround/Finding
Our investigation showed that it sucesfully passes when compute_nic_private is created early before acme_subnet_public (second subnet) gets in creation state. The reason might be updating network resource. Therefore we introduced delay (a sleep for 30 seconds) to be run on each resource as well as proper dependencies between terraform resources.
Related question:
Terraform (and AzureRM Provider) Version
Affected Resource(s)
Terraform Configuration Files
Corresponding parts of the TF template:
Note: sleep 0.1s leads to an issue (in my case on every ~3rd deployment), with sleep set to 30 (second) the deployment succeeds.
Debug Output
For testing consistency I am using this container:
Usage (depends on actuall container used), but for example:
Expected Behavior
Actual Behavior
Forever hangs on:
Steps to Reproduce
Above
References
The text was updated successfully, but these errors were encountered: