[ec2_vpc_nat_gateway] CI pipeline fails with RequestLimitExceeded errors

[alinabuzachis]

SUMMARY

CI pipeline not successful for the ec2_vpc_nat_gateway module due due to RequestLimitExceedederrors.

fatal: [testhost]: FAILED! => {"changed": false, "msg": "An error occurred (RequestLimitExceeded) when calling the DescribeNatGateways operation (reached max retries: 4): Request limit exceeded.", "resource_actions": ["ec2:DescribeNatGateways"]}

For example, creating a NAT Gateway can throw this error (see Shippable result page for further details):

- name: Create new NAT gateway with eip allocation-id
    ec2_vpc_nat_gateway:
      subnet_id: "{{ subnet_id }}"
      allocation_id: "{{ allocation_id }}"
      wait: yes
    register: create_ngw

I quick workaround is to include an until loop as follow:

- name: Create new NAT gateway with eip allocation-id
    ec2_vpc_nat_gateway:
      subnet_id: "{{ subnet_id }}"
      allocation_id: "{{ allocation_id }}"
      wait: yes
    register: create_ngw
    retries: 10
    until: create_ngw is not failed

ISSUE TYPE

• Bug Report

COMPONENT NAME

ec2_vpc_nat_gateway

ANSIBLE VERSION

2.10

[ansibullbot]

Files identified in the description:

• plugins/modules/ec2_vpc_nat_gateway.py

If these files are inaccurate, please update the component name section of the description or use the !component bot command.

click here for bot help

[ansibullbot]

cc @Etherdaemon @jillr @jonhadfield @linuxdynasty @s-hertel @tremble @wimnat
click here for bot help

[jillr]

@alinabuzachis have we seen this since increasing the quotas? The zuul migration should also help since we'll be controlling test concurrency better.

[alinabuzachis]

@jillr I don't think we've seen the error lately.