[BUG] citrixadc_lbvserver_servicegroup_binding throws an error if SG has something bound

[rhelfand]

Contact us

For any immediate issues or help , reach out to us at NetScaler-AutomationToolkit@cloud.com !

Bug Report

Describe the bug We create some lb_vservers using "count", then bind each to one of a few different serviceGroups depending on which datacenter we're in (we use a map variable for this).

We get an error when attempting to run the apply: │ Error: Provider produced inconsistent result after apply │ │ When applying changes to citrixadc_lbvserver_servicegroup_binding.dcb_sg_remote_bindings["2"], provider "provider[\"registry.terraform.io/citrix/citrixadc\"]" produced an unexpected new value: Root resource was present, but now absent. │ │ This is a bug in the provider, which should be reported in the provider's own issue tracker.

To Reproduce Steps to reproduce the behaviour:

1. My terraform files are

   
   # main
   resource "citrixadc_lbvserver" "create_my_rstacks" {
   count           = 4
   name            = "my_rstack_${count.index + 1}_lb_vs"
   ipv46           = "0.0.0.0"
   servicetype     = "SSL"
   persistencetype = "NONE"
   clttimeout      = 300
   somethod        = "DYNAMICCONNECTION"
   sobackupaction  = "DROP"
   downstateflush  = "DISABLED"
   appflowlog      = "DISABLED"
   }

resource "citrixadc_servicegroup" "dcb_sg" { count = (var.datacenter != "DC_B") ? 1 : 0 servicegroupname = "my_dcb_sg" servicetype = "SSL" maxclient = 0 maxreq = 0 cip = "ENABLED" cipheader = "X-Forwarded-For" usip = "NO" useproxyport = "YES" clttimeout = 300 svrtimeout = 360 cka = "YES" tcpb = "YES" cmp = "NO" downstateflush = "DISABLED" appflowlog = "DISABLED" }

resource "citrixadc_lbvserver_servicegroup_binding" "dcb_sg_remote_bindings" { for_each = { for sgkey, sgvalue in var.primary_stack_locations : sgkey => sgvalue if(sgvalue == "DC_B" && var.datacenter != "DC_B") }

servicegroupname = citrixadc_servicegroup.dcb_sg[0].servicegroupname name = "myrstack${each.key}_lb_vs"

depends_on = [ citrixadc_lbvserver.create_my_rstacks ] }

variables

variable "datacenter" { type = string }

variable "ns_hostname" { type = string }

variable "primary_stack_locations" { description = "The datacenter where the primary database resides for a given stack" type = map(any)

default = { 1 = "DC_A" 2 = "DC_B" 3 = "DC_A" 4 = "DC_C" } }

variables

tfns_pw = "*****" datacenter = "DC_A" ns_hostname = "coresredev1000"

2. which terraform command I am getting the error
```bash
## This errors out if we have a server bound to a serviceGroup, and it works if the serviceGroup does not have anything bound
terraform apply -var-file ./myvars

3. Terraform output logs to the log file (./tf.log). Append TF_LOG=TRACE NS_LOG=TRACE TF_LOG_PATH=./tf.log to your terraform command

   Attached.

   bound_sg.tf.log unbound_sg.tf.log

4. Error I am facing on the console

   │ Error: Provider produced inconsistent result after apply
   │
   │ When applying changes to citrixadc_lbvserver_servicegroup_binding.dcb_sg_remote_bindings["2"], provider "provider[\"registry.terraform.io/citrix/citrixadc\"]" produced an unexpected new value: Root resource was present, but now absent.
   │
   │ This is a bug in the provider, which should be reported in the provider's own issue tracker.

Expected behaviour My serviceGroup should be bound to my lb_vservers.

Screenshots If applicable, add screenshots to help explain your problem.

Environment (please fill the following information):

• OS: Mac
• terraform version output

  
  $ terraform -version
  Terraform v1.5.7
  on darwin_arm64
  + provider registry.terraform.io/citrix/citrixadc v1.37.0

• terraform-provider-citrixadc version v1.37.0 (also tried on v1.33.0)
• go version output go version go1.22.0 darwin/arm64

Additional context Add any other context about the problem here.

[rhelfand]

At first I thought perhaps this was related to: https://github.com/citrix/terraform-provider-citrixadc/issues/275

However I'm not sure of that now.

[rhelfand]

Hi @sumanth-lingappa, I have been working with my colleague to try to figure this out and we believe this is a Citrix bug. We are going to submit a ticket to Citrix. I will try to summarize.

If you try to bind a servicegroup to an lb_vserver, and the servicegroup has members in it: bind lb vserver my_lb_vs my_svcgrp

This command will work, HOWEVER, if you examine the lb_vserver either on the CLI or using the NITRO API, it does not show you the serviceGroup information. The output looks like this:

{ "errorcode": 0, "message": "Done", "severity": "NONE" }

And if you run show lb vserver my_lb_vs, there is no Bound Service Groups section at the bottom of the output.

If you bind a servicegroup to an lb_vserver, and the servicegroup is empty, and you examine the lb_vserver on the CLI or using the NITRO API, you get more output:

{ "errorcode": 0, "message": "Done", "severity": "NONE", "lbvserver_servicegroup_binding": [ { "name": "my_lb_vs", "servicegroupname": "my_svcgrp", "stateflag": "536936464", "servicename": "my_svcgrp" } ] }

Running show lb vserver my_lb_vs on the command line shows:

Bound Service Groups:
1)  Group Name: my_svcgrp
 Done

at the bottom of the output.