PKS (now VMware Tanzu Kubernetes Grid Integrated Edition or TKGI) with NSX-T configured does not update the node DNS on existing clusters when updated on the tile.
What actions can the user take to resolve or work around the described problem? This article will describe a workaround for the issue.
export BOSH_CLIENT=pivotal-container-service-cedfe8e.....cadcd BOSH_CLIENT_SECRET=135a9......83458bbf BOSH_CA_CERT=/var/tempest/workspaces/default/root_ca_certificate BOSH_ENVIRONMENT=<DIRECTOR_IP>
bosh configs | grep <ID>Get the ID for the config (first column) <configID>:
bosh config <configID>Confirm the correct <ID> is specified in the output.
bosh config <configID> > config.yml
Origin: ID 76 Type cloud Name service-instance_3f05330a-43a5-41ae-ad79-a6c16a931ccb Created At 2020-07-10 09:36:07 UTC Content networks: - name: pks-3f05330a-43a5-41ae-ad79-a6c16a931ccb subnets: - azs: - PKS-MGMT-1 cloud_properties: name: pks-3f05330a-43a5-41ae-ad79-a6c16a931ccb dns: - 10.192.2.10 - 10.192.2.11 gateway: 172.34.16.1 range: 172.34.16.0/24 type: manual vm_extensions: - cloud_properties: {} name: service-instance-3f05330a-43a5-41ae-ad79-a6c16a931ccb-master-nsgroup
Content networks: - name: pks-3f05330a-43a5-41ae-ad79-a6c16a931ccb subnets: - azs: - PKS-MGMT-1 cloud_properties: name: pks-3f05330a-43a5-41ae-ad79-a6c16a931ccb dns: - 10.192.2.10 - 10.192.2.11 - 8.8.8.8 gateway: 172.34.16.1 range: 172.34.16.0/24 type: manual vm_extensions: - cloud_properties: {} name: service-instance-3f05330a-43a5-41ae-ad79-a6c16a931ccb-master-nsgroup
bosh update-config --type=cloud --name=service-instance_<ID> config.yml
bosh configs | grep <ID>Get the ID for the config (first column) <configID>.
bosh config <configID>Confirm the correct <ID> is specified in the output.
pks upgrade-cluster <clustername>
cat /etc/resolv.conf # This file was automatically updated by bosh-dns nameserver 169.254.0.2 nameserver 10.192.2.10 nameserver 10.192.2.11 nameserver 8.8.8.8