log_config can't be disabled

[awasilyev]

Affected Resource(s)

• google_compute_backend_service

(based on https://github.com/terraform-google-modules/terraform-google-lb-http/issues/92)

I'm tryng to create a load balancer and set:

log_config { enable = false sample_rate = null } The configuration doesn't work correctly as it is always set to enable->true sample_rate->1

I can see this in the logs

2020-03-18T12:45:50.787+0100 [DEBUG] plugin.terraform-provider-google-beta_v3.13.0_x5: }, 2020-03-18T12:45:50.787+0100 [DEBUG] plugin.terraform-provider-google-beta_v3.13.0_x5: "logConfig": { 2020-03-18T12:45:50.787+0100 [DEBUG] plugin.terraform-provider-google-beta_v3.13.0_x5: "enable": true, 2020-03-18T12:45:50.787+0100 [DEBUG] plugin.terraform-provider-google-beta_v3.13.0_x5: "sampleRate": 1 2020-03-18T12:45:50.787+0100 [DEBUG] plugin.terraform-provider-google-beta_v3.13.0_x5: }, 2020/03/18 12:45:50 [WARN] Provider "registry.terraform.io/-/google-beta" produced an unexpected new value for module.app_lb.google_compute_backend_service.default["default"], but we are tolerating it because it is using the legacy plugin SDK. The following problems may be the cause of any confusing errors from downstream operations:

• .log_config[0].enable: was cty.False, but now cty.True
• .log_config[0].sample_rate: was null, but now cty.NumberIntVal(1)

same results with 3.18.0 and 4.xx

[venkykuberan]

Can your share your config or full debug log to see what's going on ?

[joscabmar]

I am having the same issue. I always get suggested changes like:

      + log_config {
          + enable      = false
          + sample_rate = 0
        }

or:

      + log_config {}

In my TF code I used:

      log_config = {
        enable      = false
        sample_rate = 0
      }

[stolendog]

the same here, I cannot disable logging on existing google_compute_backend_service

[artazar]

Don't want to file a new issue, as the cause seems to be the same. Provider version 3.38.0. When I have log_config section as follows:

log_config
  enable      = false
  sample_rate = null
}

for backend_service resource and

logging             = false

for health checks, terraform always suggests adding these to resources, while it finds it empty querying API:

  # module.global-loadbalancer.google_compute_health_check.default["sb2integration"] will be updated in-place
  ~ resource "google_compute_health_check" "default" {
        check_interval_sec  = 5
        ...
        http_health_check {
            port         = 8080
            proxy_header = "NONE"
            request_path = "/hc"
        }

      + log_config {
          + enable = false
        }
    }

and inside terraform provider log:

2020-09-15T10:37:53.323+0700 [DEBUG] plugin.terraform-provider-google-beta_v3.38.0_x5: {
2020-09-15T10:37:53.323+0700 [DEBUG] plugin.terraform-provider-google-beta_v3.38.0_x5:   "id": "6970670192020790769",
2020-09-15T10:37:53.323+0700 [DEBUG] plugin.terraform-provider-google-beta_v3.38.0_x5:   "creationTimestamp": "2020-09-14T19:12:46.246-07:00",
2020-09-15T10:37:53.323+0700 [DEBUG] plugin.terraform-provider-google-beta_v3.38.0_x5:   "name": "global-loadbalancer-hc-sb2integration",
2020-09-15T10:37:53.324+0700 [DEBUG] plugin.terraform-provider-google-beta_v3.38.0_x5:   "checkIntervalSec": 5,
2020-09-15T10:37:53.324+0700 [DEBUG] plugin.terraform-provider-google-beta_v3.38.0_x5:   "timeoutSec": 5,
2020-09-15T10:37:53.324+0700 [DEBUG] plugin.terraform-provider-google-beta_v3.38.0_x5:   "unhealthyThreshold": 2,
2020-09-15T10:37:53.324+0700 [DEBUG] plugin.terraform-provider-google-beta_v3.38.0_x5:   "healthyThreshold": 2,
2020-09-15T10:37:53.324+0700 [DEBUG] plugin.terraform-provider-google-beta_v3.38.0_x5:   "type": "HTTP",
2020-09-15T10:37:53.324+0700 [DEBUG] plugin.terraform-provider-google-beta_v3.38.0_x5:   "httpHealthCheck": {
2020-09-15T10:37:53.324+0700 [DEBUG] plugin.terraform-provider-google-beta_v3.38.0_x5:     "port": 8080,
2020-09-15T10:37:53.324+0700 [DEBUG] plugin.terraform-provider-google-beta_v3.38.0_x5:     "requestPath": "/hc",
2020-09-15T10:37:53.324+0700 [DEBUG] plugin.terraform-provider-google-beta_v3.38.0_x5:     "proxyHeader": "NONE"
2020-09-15T10:37:53.324+0700 [DEBUG] plugin.terraform-provider-google-beta_v3.38.0_x5:   },
...
2020-09-15T10:37:53.324+0700 [DEBUG] plugin.terraform-provider-google-beta_v3.38.0_x5:   "logConfig": {},
2020-09-15T10:37:53.324+0700 [DEBUG] plugin.terraform-provider-google-beta_v3.38.0_x5:   "kind": "compute#healthCheck"
2020-09-15T10:37:53.324+0700 [DEBUG] plugin.terraform-provider-google-beta_v3.38.0_x5: }

we can see

"logConfig": {},

which is not treated well.

[rantoniuk]

Same problem with 3.43.0. I would even ask, why do I have to define logging in the first place instead of using the default values.

[yegle]

Still have the same problem with 3.51.0

[diasjorge]

Dear maintainers, any possibility to get this fixed, or at least some guidance where to try to fix it please?

[Nicarim]

Problem still persists. (lb-http/serverless_neg 6.1.0 here)

[gabriuspa]

Hey guys, I faced the same problem with the GCP module about firewall rules (source="terraform-google-modules/network/google//modules/firewall-rules") with the "log_config {}" property.

I SOLVED it simply putting it to null and, in my case, the firewall's rule logging was disabled (On GCP Dashboard, Logging is set to Off).

Put here the code for clarity:

module "firewall_rules" { source = "terraform-google-modules/network/google//modules/firewall-rules" project_id = var.gcp_project network_name = module.vpc.network_name

rules = [{ name = "terraform-allow-internal" description = null direction = "INGRESS" priority = 1000 ranges = ["10.0.0.0/24"]

source_tags             = null
source_service_accounts = null
target_tags             = null
target_service_accounts = null
allow = [{
  protocol = "all"
  ports    = []
}]
deny = []
log_config = null

#ATTENTION: to disable logging from here, "log_config" have to be null

}] }

Hoping this can help you

[Nicarim]

Module that I use (from terraform-google-modules) doesn't support setting it to null. (lb-http/serverless_neg 6.1.0 here)

[franck-ornikar]

Still have the same problem with 4.54.0

[NickElliot]

@franck-ornikar can you provide the details of your config? While there was a change since your comment it shouldn't have affected the core issue of being unable to disable log_config, which I have been unable to reproduce.