Modify VLAN attachments

This document describes how to modify Dedicated Interconnect VLAN attachments and their Border Gateway Protocol (BGP) sessions.

The following table lists VLAN attachment and BGP session properties. It also provides information about how to modify them.

Property How to modify

Description

Capacity

MTU

For details, see the next section Modify description, capacity, and MTU.

Modifications are immediate and in rare cases, can cause traffic disruption. If traffic is disrupted, the disruption lasts less than a minute.

Stack type

You can change the stack type of an existing VLAN attachment to IPv4 only (single stack) or IPv4 and IPv6 (dual stack).

If you change the stack type of your VLAN attachment, your existing traffic might be rerouted or dropped. For more details, see Modify stack type.

Whether a VLAN attachment is enabled or disabled To disable or re-enable a VLAN attachment, see Disable VLAN attachments.

VLAN ID

BGP IP addresses (Cloud Router IP address, on-premises router IP address)

You can't modify these properties for existing attachments. Instead, you must recreate the VLAN attachment.

To create VLAN attachments for Dedicated Interconnect, see Create VLAN attachments.

Custom learned routes Follow the steps in Update an existing session to use custom learned routes.
MD5 authentication Follow the steps in Add authentication to an existing session.

The peer ASN is not a property of the VLAN attachment—it is part of the BGP configuration of the Cloud Router. To modify the peer ASN, update the Cloud Router's BGP session for the VLAN attachment.

For more information, see the following resources:

Modify description, capacity, and MTU

The section provides instructions for updating the description, capacity, and maximum transmission unit (MTU) of a VLAN attachment. Changing the attributes of a VLAN attachment, in rare cases, can cause traffic disruption. If traffic is disrupted, the disruption lasts less than a minute.

To make use of a 1460-, 1500-, or 8896-byte MTU, the VPC network that uses the attachment must have the same MTU value. In addition, the on-premises virtual machine (VM) instances and routers must also have the same MTU value. If your network has the default MTU of 1460, then select an MTU of 1460 for your VLAN attachment.

You cannot modify the MTU (1440) of an encrypted VLAN attachment that is used in an HA VPN over Cloud Interconnect deployment.

Console

  1. In the Google Cloud console, go to the Cloud Interconnect VLAN attachments tab.

    Go to VLAN attachments

  2. Select the VLAN attachment to modify.

  3. On the attachment's detail page, click Edit.

  4. Modify your VLAN attachment:

    • Description: Information about the VLAN attachment.
    • Capacity: The maximum bandwidth of the attachment. If you modify the capacity of an encrypted VLAN attachment, you might need to add HA VPN tunnels to accommodate increased traffic. For more information, see Configure HA VPN over Cloud Interconnect.
    • Maximum transmission unit (MTU): The MTU of the attachment.
  5. Click Save.

gcloud

Modify an interconnectAttachment by specifying the name of the existing VLAN attachment and the attributes to modify:

gcloud compute interconnects attachments dedicated update NAME \
    --region=REGION \
    --description=DESCRIPTION \
    --bandwidth=BANDWIDTH \
    --mtu=MTU

Replace the following:

  • NAME: the name of the existing VLAN attachment
  • REGION: the region of the attachment
  • DESCRIPTION: a description of the attachment
  • BANDWIDTH: the provisioned capacity of the attachment. If you modify the capacity of an encrypted VLAN attachment, you might need to add HA VPN tunnels to accommodate the increased traffic. For more information, see Configure HA VPN over Cloud Interconnect.
  • MTU: the MTU of the attachment

For more information, see the gcloud compute interconnects attachments dedicated update reference.

Modify stack type

You can change the stack type of an existing VLAN attachment to single stack (IPv4 only) or dual stack (IPv4 and IPv6).

To support IPv6 traffic over a Dedicated Interconnect connection, you must use a dual-stack VLAN attachment. With a dual-stack VLAN attachment, you can configure an IPv6 BGP session (Preview), or you can enable IPv6 route exchange in an IPv4 BGP session.

With a dual-stack VLAN attachment, you can also configure an IPv4 BGP session and an IPv6 BGP session that run in parallel. For more information, see Parallel IPv4 and IPv6 BGP sessions.

Disable IPv6 traffic

You might want to disable IPv6 traffic temporarily in your Dedicated Interconnect VLAN attachment for troubleshooting purposes.

Disabling IPv6 route exchange in the IPv4 BGP session or disabling your IPv6 BGP session is less disruptive to your Cloud Interconnect configuration than changing the stack type of your VLAN attachment.

To disable IPv6 traffic temporarily in an IPv4 BGP session, you can disable IPv6 route exchange in the IPv4 BGP session. For information about how to disable IPv6 route exchange in an IPv4 BGP session, see Configure multiprotocol BGP in IPv4 or IPv6 BGP sessions.

If you are using an IPv6 BGP session, you can disable your entire BGP session instead of switching the stack type of the VLAN attachment. For information about how to disable a BGP session, see Disable or remove BGP sessions

Modification impact on IPv4 and IPv6 traffic

The following table describes how the stack type configuration of the VLAN attachment combined with the BGP session configuration affects IPv4 and IPv6 traffic.

Original stack type of VLAN attachment Modified stack type of VLAN attachment BGP session MP-BGP configuration Impact on traffic
Dual stack (IPv4 and IPv6) IPv4 only IPv4 BGP IPv6 disabled No IPv6 traffic supported; only IPv4 traffic. This is the default state for Dedicated Interconnect and Cloud Router.
Dual stack (IPv4 and IPv6) IPv4 only IPv4 BGP IPv6 enabled

No IPv6 traffic supported. Because the VLAN attachment is configured as IPv4 only, Cloud Router automatically disables IPv6 route exchange for the BGP session.

However, if the VLAN attachment previously had IPv6 route exchange enabled for the IPv4 BGP session, then Google Cloud might retain IPv6 BGP addresses for the VLAN attachment. Google Cloud can reuse these addresses if IPv6 route exchange is later re-enabled for the VLAN attachment.

IPv4 only Dual stack (IPv4 and IPv6) IPv4 BGP IPv6 enabled

Both IPv4 and IPv6 traffic are supported.

This is the supported configuration for an IPv4 BGP session that uses MP-BGP and exchanges IPv6 routes in Dedicated Interconnect.

Dual stack (IPv4 and IPv6) IPv4 only IPv6 BGP IPv4 enabled

No IPv6 traffic supported; only IPv4 traffic supported.

Dual stack (IPv4 and IPv6) IPv4 only IPv6 BGP IPv4 disabled

No IPv6 traffic and no IPv4 traffic supported. The IPv6 BGP session goes down.

However, if the VLAN attachment previously had IPv4 route exchange enabled for the IPv6 BGP session, then Google Cloud might retain its IPv4 BGP addresses. Google Cloud can reuse the addresses if an IPv4 route exchange is later re-enabled for the VLAN attachment.

Dual stack (IPv4 and IPv6) IPv4 only Both IPv4 BGP and IPv6 BGP Not applicable

No IPv6 traffic supported; only IPv4 traffic supported. The IPv6 BGP session goes down, but the IPv4 BGP session stays up.

However, the VLAN attachment might retain its IPv6 BGP addresses. Google Cloud can reuse the addresses if an IPv6 BGP session is later reconfigured for the VLAN attachment.

Console

  1. In the Google Cloud console, go to the Cloud Interconnect VLAN attachments tab.

    Go to VLAN attachments

  2. Select the VLAN attachment to modify.

  3. On the attachment's detail page, click Edit.

  4. Modify the stack type of the VLAN attachment by selecting IPv4 (single-stack) or IPv4 and IPv6 (dual-stack).

  5. Click Save.

If you are modifying a VLAN attachment to exchange IPv6 traffic, complete the configuration by enabling IPv6 in the BGP session.

  1. On the attachment's detail page, click Edit BGP session.
  2. Under Multiprotocol BGP, enable or disable IPv6 traffic for the VLAN attachment configured with the IPv4 and IPv6 stack type.
    • To allow the exchange of IPv6 address prefixes over the BGP session, select Enable IPv6 traffic. Google Cloud automatically assigns BGP IPv6 next hop addresses for Cloud Router and your on-premises router. Google Cloud assigns an unused address.
    • To disable IPv6 route exchange, deselect Enable IPv6 traffic. You can enable IPv6 traffic in the BGP session at a later time. For more information, see Configure multiprotocol BGP in IPv4 or IPv6 BGP sessions.
  3. Click Save and Continue.

gcloud

Modify a VLAN attachment by specifying the name of the existing attachment and the attributes to modify:

gcloud compute interconnects attachments dedicated update NAME \
    --stack-type=IP_STACK

Replace the following:

  • IP_STACK: the IP stack to use on this attachment. Specify either IPV4_ONLY or IPV4_IPV6.

    If you are modifying an IPv4-only VLAN attachment to use the dual-stack (IPV4_IPV6) stack type, you can complete the configuration by choosing one of the following configurations.

What's next