terraform-provider-aws: terraform: Error finding route after creating it: Unable to find matching route for Route Table

This issue was originally opened by @FlorinAndrei as hashicorp/terraform#9756. It was migrated here as part of the provider split. The original body of the issue is below.

Terraform Version

Terraform v0.7.7

Terraform Configuration Files

https://dl.dropboxusercontent.com/u/29966/tf.zip

Expected Behavior

Create the infrastructure.

Actual Behavior

https://gist.github.com/FlorinAndrei/17c0a12e70c0a15671322030b419ec92

However, the routes, route tables, and NAT gateways were all created just fine. Subnet associations were set also.

Running terraform plan again produces this output:

+ module.network.private_subnet.aws_route.private_nat_gw.2
    destination_cidr_block:     "0.0.0.0/0"
    destination_prefix_list_id: "<computed>"
    gateway_id:                 "<computed>"
    instance_id:                "<computed>"
    instance_owner_id:          "<computed>"
    nat_gateway_id:             "nat-09ffe6396b7b4f821"
    network_interface_id:       "<computed>"
    origin:                     "<computed>"
    route_table_id:             "rtb-6ad9640d"
    state:                      "<computed>"


Plan: 1 to add, 0 to change, 0 to destroy.

But that route exists already. If I try to apply the plan, I get this:

module.network.private_subnet.aws_route.private_nat_gw.2: Creating...
  destination_cidr_block:     "" => "0.0.0.0/0"
  destination_prefix_list_id: "" => "<computed>"
  gateway_id:                 "" => "<computed>"
  instance_id:                "" => "<computed>"
  instance_owner_id:          "" => "<computed>"
  nat_gateway_id:             "" => "nat-09ffe6396b7b4f821"
  network_interface_id:       "" => "<computed>"
  origin:                     "" => "<computed>"
  route_table_id:             "" => "rtb-6ad9640d"
  state:                      "" => "<computed>"
Error applying plan:

1 error(s) occurred:

* aws_route.private_nat_gw.2: Error creating route: RouteAlreadyExists: The route identified by 0.0.0.0/0 already exists.
	status code: 400, request id: 31c1946f-6359-41c6-8be6-460c8d3b5d99

terraform refresh produces this output:

terraform_remote_state.aws_global: Refreshing state... (ID: 2016-10-31 20:05:38.730089879 +0000 UTC)
module.network.nat.aws_eip.nat.0: Refreshing state... (ID: eipalloc-e15ccd86)
module.network.nat.aws_eip.nat.1: Refreshing state... (ID: eipalloc-785ccd1f)
module.network.nat.aws_eip.nat.2: Refreshing state... (ID: eipalloc-795ccd1e)
module.network.vpc.aws_vpc.vpc: Refreshing state... (ID: vpc-b04abed7)
module.network.vpc.aws_vpc_peering_connection.peering: Refreshing state... (ID: pcx-327bc05b)
module.network.public_subnet.aws_internet_gateway.public: Refreshing state... (ID: igw-da109cbe)
module.network.public_subnet.aws_route_table.public: Refreshing state... (ID: rtb-6cd9640b)
module.network.public_subnet.aws_subnet.public.2: Refreshing state... (ID: subnet-a86c0ff0)
module.network.private_subnet.aws_subnet.private.0: Refreshing state... (ID: subnet-afefdad9)
module.network.private_subnet.aws_route_table.private.0: Refreshing state... (ID: rtb-6bd9640c)
module.network.private_subnet.aws_route_table.private.1: Refreshing state... (ID: rtb-6dd9640a)
module.network.public_subnet.aws_subnet.public.1: Refreshing state... (ID: subnet-933b28f7)
module.network.public_subnet.aws_subnet.public.0: Refreshing state... (ID: subnet-adefdadb)
module.network.private_subnet.aws_subnet.private.2: Refreshing state... (ID: subnet-a96c0ff1)
module.network.private_subnet.aws_route_table.private.2: Refreshing state... (ID: rtb-6ad9640d)
module.network.private_subnet.aws_subnet.private.1: Refreshing state... (ID: subnet-9d3b28f9)
module.network.public_subnet.aws_route.public_gw: Refreshing state... (ID: r-rtb-6cd9640b1080289494)
module.network.public_subnet.aws_route_table_association.public.0: Refreshing state... (ID: rtbassoc-f928c59f)
module.network.public_subnet.aws_route_table_association.public.1: Refreshing state... (ID: rtbassoc-c728c5a1)
module.network.public_subnet.aws_route_table_association.public.2: Refreshing state... (ID: rtbassoc-c628c5a0)
module.network.nat.aws_nat_gateway.nat.0: Refreshing state... (ID: nat-0b9579e6856f9f3d3)
module.network.nat.aws_nat_gateway.nat.1: Refreshing state... (ID: nat-0533b77ab8365de80)
module.network.nat.aws_nat_gateway.nat.2: Refreshing state... (ID: nat-09ffe6396b7b4f821)
module.network.private_subnet.aws_route_table_association.private.0: Refreshing state... (ID: rtbassoc-fd28c59b)
module.network.private_subnet.aws_route_table_association.private.1: Refreshing state... (ID: rtbassoc-fa28c59c)
module.network.private_subnet.aws_route_table_association.private.2: Refreshing state... (ID: rtbassoc-fb28c59d)
module.network.aws_network_acl.acl: Refreshing state... (ID: acl-9168b8f6)
module.network.private_subnet.aws_route.private_nat_gw.1: Refreshing state... (ID: r-rtb-6dd9640a1080289494)
module.network.private_subnet.aws_route.private_nat_gw.0: Refreshing state... (ID: r-rtb-6bd9640c1080289494)

Please note how gw.2 is not mentioned in the refresh.

terraform plan -destroy also does not mention gw.2

Applying that -destroy plan works well and destroys all infrastructure.

Steps to Reproduce

terraform apply

About this issue

Original URL
State: closed
Created 7 years ago
Reactions: 7
Comments: 18 (3 by maintainers)

Most upvoted comments

Still experiencing this issues, successfully resolved using:

timeouts {
    create = "5m"
  }

Terraform v0.12.8 provider “aws” (hashicorp/aws) 2.33.0

kostas-theo on Oct 30, 2019

Hey,

Just wanted to confirm that the timeout increase apparently fixed the issue:

timeouts {
    create = "5m"
  }

rafaljanicki on Apr 9, 2018

I found that my situation had to do with a malformed CIDR reference. I had 192.168.0.21/30 as my destination_cidr_block, but AWS was nicely adjusting that to the more accurate 192.168.0.20/30 when. Meanwhile, I guess terraform was looking for the one I specified to confirm completion rather than the adjusted one. Same error pattern, different cause. Hope this saves someone some trouble.

flickerfly on Sep 20, 2018

Same here, on 0.10.2.

Running apply the first time:

aws_route_table.private: Creating...
  propagating_vgws.#: "" => "<computed>"
  route.#:            "" => "<computed>"
  tags.%:             "" => "1"
  tags.Name:          "" => "stage-private"
  vpc_id:             "" => "vpc-63082c0b"
aws_route_table.private: Creation complete (ID: rtb-4a8f6021)
aws_nat_gateway.main: Creating...
  allocation_id:        "" => "eipalloc-975e5efe"
  network_interface_id: "" => "<computed>"
  private_ip:           "" => "<computed>"
  public_ip:            "" => "<computed>"
  subnet_id:            "" => "subnet-c4c42daf"
aws_nat_gateway.main: Still creating... (10s elapsed)
aws_nat_gateway.main: Still creating... (20s elapsed)
aws_nat_gateway.main: Still creating... (30s elapsed)
aws_nat_gateway.main: Still creating... (40s elapsed)
aws_nat_gateway.main: Still creating... (50s elapsed)
aws_nat_gateway.main: Still creating... (1m0s elapsed)
aws_nat_gateway.main: Still creating... (1m10s elapsed)
aws_nat_gateway.main: Still creating... (1m20s elapsed)
aws_nat_gateway.main: Still creating... (1m30s elapsed)
aws_nat_gateway.main: Still creating... (1m40s elapsed)
aws_nat_gateway.main: Creation complete (ID: nat-07f3c867c4ca19368)
aws_route.private-nat-gateway: Creating...
  destination_cidr_block:     "" => "0.0.0.0/0"
  destination_prefix_list_id: "" => "<computed>"
  egress_only_gateway_id:     "" => "<computed>"
  gateway_id:                 "" => "<computed>"
  instance_id:                "" => "<computed>"
  instance_owner_id:          "" => "<computed>"
  nat_gateway_id:             "" => "nat-07f3c867c4ca19368"
  network_interface_id:       "" => "<computed>"
  origin:                     "" => "<computed>"
  route_table_id:             "" => "rtb-4a8f6021"
  state:                      "" => "<computed>"
aws_route.private-nat-gateway: Still creating... (10s elapsed)
aws_route.private-nat-gateway: Still creating... (20s elapsed)
aws_route.private-nat-gateway: Still creating... (30s elapsed)
aws_route.private-nat-gateway: Still creating... (40s elapsed)
aws_route.private-nat-gateway: Still creating... (50s elapsed)
aws_route.private-nat-gateway: Still creating... (1m0s elapsed)
aws_route.private-nat-gateway: Still creating... (1m10s elapsed)
aws_route.private-nat-gateway: Still creating... (1m20s elapsed)
aws_route.private-nat-gateway: Still creating... (1m30s elapsed)
aws_route.private-nat-gateway: Still creating... (1m40s elapsed)
aws_route.private-nat-gateway: Still creating... (1m50s elapsed)
aws_route.private-nat-gateway: Still creating... (2m0s elapsed)
Error applying plan:

1 error(s) occurred:

* aws_route.private-nat-gateway: 1 error(s) occurred:

* aws_route.private-nat-gateway: Error finding route after creating it: Unable to find matching route for Route Table (rtb-4a8f6021) and destination CIDR block (0.0.0.0/0).

Yet, in AWS web console I can see the route was created exactly as expected.

Running apply a second time:

aws_route.private-nat-gateway: Creating...
  destination_cidr_block:     "" => "0.0.0.0/0"
  destination_prefix_list_id: "" => "<computed>"
  egress_only_gateway_id:     "" => "<computed>"
  gateway_id:                 "" => "<computed>"
  instance_id:                "" => "<computed>"
  instance_owner_id:          "" => "<computed>"
  nat_gateway_id:             "" => "nat-07f3c867c4ca19368"
  network_interface_id:       "" => "<computed>"
  origin:                     "" => "<computed>"
  route_table_id:             "" => "rtb-4a8f6021"
  state:                      "" => "<computed>"
Error applying plan:

1 error(s) occurred:

* aws_route.private-nat-gateway: 1 error(s) occurred:

* aws_route.private-nat-gateway: Error creating route: RouteAlreadyExists: The route identified by 0.0.0.0/0 already exists.
	status code: 400, request id: 35c5077a-7437-43a9-84b7-40261cb79927

After doing destroy, and then apply again, without changing anything else, the same conf works just fine.

thnee on Sep 1, 2017

@smihajlovic-frame maybe you can try to increase aws_route configurable timeout for create, it’s available since provider.aws v1.11.0. Default is 2 minutes.

eLco on Apr 4, 2018

Ran into this issue; route was created properly but terraform does not recognize it:

Terraform v0.11.5
+ provider.aws v1.11.0

AnthonyWC on Apr 3, 2018

I have the same problem in v0.10.7

gopisaba on Oct 24, 2017