equinix / terraform-equinix-metal-anthos-on-vsphere

[Deprecated] Automated Anthos Installation via Terraform for Equinix Metal with vSphere
https://registry.terraform.io/modules/equinix/anthos-on-vsphere/metal/latest
Apache License 2.0
62 stars 41 forks source link

Fix bug with null #90

Closed c0dyhi11 closed 4 years ago

c0dyhi11 commented 4 years ago

And change facility to SJC1

displague commented 4 years ago

Noting that the quoted nulls were inappropriately introduced via https://github.com/packet-labs/google-anthos/pull/76/files#diff-c9ac8098c5ea9d3e6a9a596ff0c512a4L66-R76

The problem that this created, and is being addressed here, is the effect of "null" on one of the python scripts, esx_host_networking.py.

This error is not breaking the overall terraform apply, but does prevent the provisioned system from working:

null_resource.apply_esx_network_config[0] (remote-exec): Trying to connect to ESX Host . . .
null_resource.apply_esx_network_config[0] (remote-exec): Connected to ESX Host !
null_resource.apply_esx_network_config[0]: Still creating... [40s elapsed]
null_resource.apply_esx_network_config[0] (remote-exec): Successfully created vSwitch  vSwitch1
null_resource.apply_esx_network_config[0] (remote-exec): Successfully created PortGroup  Management
null_resource.apply_esx_network_config[0] (remote-exec): Successfully created PortGroup  Management Net
null_resource.apply_esx_network_config[0] (remote-exec): Successfully created PortGroup  vMotion
null_resource.apply_esx_network_config[0] (remote-exec): Successfully created PortGroup  vSAN
null_resource.apply_esx_network_config[0] (remote-exec): Successfully created PortGroup  VM Private Net
null_resource.apply_esx_network_config[0] (remote-exec): Traceback (most recent call last):
null_resource.apply_esx_network_config[0] (remote-exec):   File "/root/esx_host_networking.py", line 323, in <module>
null_resource.apply_esx_network_config[0] (remote-exec):     main()
null_resource.apply_esx_network_config[0] (remote-exec):   File "/root/esx_host_networking.py", line 210, in main
null_resource.apply_esx_network_config[0] (remote-exec):     enable_service_on_virtual_nic(host, virtual_nic, subnet['vsphere_service_type'])
null_resource.apply_esx_network_config[0] (remote-exec):   File "/root/esx_host_networking.py", line 119, in enable_service_on_virtual_nic
null_resource.apply_esx_network_config[0] (remote-exec):     host.configManager.virtualNicManager.SelectVnicForNicType(service_type, virtual_nic)
null_resource.apply_esx_network_config[0] (remote-exec):   File "/usr/local/lib/python3.6/dist-packages/pyVmomi/VmomiSupport.py", line 706, in <lambda>
null_resource.apply_esx_network_config[0] (remote-exec):     self.f(*(self.args + (obj,) + args), **kwargs)
null_resource.apply_esx_network_config[0] (remote-exec):   File "/usr/local/lib/python3.6/dist-packages/pyVmomi/VmomiSupport.py", line 512, in _InvokeMethod
null_resource.apply_esx_network_config[0] (remote-exec):     return self._stub.InvokeMethod(self, info, args)
null_resource.apply_esx_network_config[0] (remote-exec):   File "/usr/local/lib/python3.6/dist-packages/pyVmomi/SoapAdapter.py", line 1397, in InvokeMethod
null_resource.apply_esx_network_config[0] (remote-exec):     raise obj # pylint: disable-msg=E0702
null_resource.apply_esx_network_config[0] (remote-exec): pyVmomi.VmomiSupport.InvalidArgument: (vmodl.fault.InvalidArgument) {
null_resource.apply_esx_network_config[0] (remote-exec):    dynamicType = <unset>,
null_resource.apply_esx_network_config[0] (remote-exec):    dynamicProperty = (vmodl.DynamicProperty) [],
null_resource.apply_esx_network_config[0] (remote-exec):    msg = 'A specified parameter was not correct: ',
null_resource.apply_esx_network_config[0] (remote-exec):    faultCause = <unset>,
null_resource.apply_esx_network_config[0] (remote-exec):    faultMessage = (vmodl.LocalizableMessage) [],
null_resource.apply_esx_network_config[0] (remote-exec):    invalidProperty = <unset>
null_resource.apply_esx_network_config[0] (remote-exec): }

Problems like these will need to be avoided in future tests. terraform apply success is not currently sufficient to indicate a working build.

I've created https://github.com/packet-labs/google-anthos/issues/91 to represent the need for CI within this project.