IP pool

An IP pool resource (IPPool) represents a collection of IP addresses from which Calico expects endpoint IPs to be assigned.

Sample YAML

apiVersion: projectcalico.org/v3
kind: IPPool
metadata:
  name: my.ippool-1
spec:
  cidr: 10.1.0.0/16
  ipipMode: CrossSubnet
  natOutgoing: true
  disabled: false
  nodeSelector: all()
  allowedUses:
    - Workload
    - Tunnel

IP pool definition

Metadata

Field	Description	Accepted Values	Schema
name	The name of this IPPool resource. Required.	Alphanumeric string with optional ., _, or -.	string

Spec

Field	Description	Accepted Values	Schema	Default
cidr	IP range to use for this pool.	A valid IPv4 or IPv6 CIDR. Subnet length must be at least big enough to fit a single block (by default /26 for IPv4 or /122 for IPv6). Must not overlap with the Link Local range 169.254.0.0/16 or fe80::/10.	string
blockSize	The CIDR size of allocation blocks used by this pool. Blocks are allocated on demand to hosts and are used to aggregate routes. The value can only be set when the pool is created.	20 to 32 (inclusive) for IPv4 and 116 to 128 (inclusive) for IPv6	int	26 for IPv4 pools and 122 for IPv6 pools.
ipipMode	The mode defining when IPIP will be used. Cannot be set at the same time as vxlanMode.	Always, CrossSubnet, Never	string	Never
vxlanMode	The mode defining when VXLAN will be used. Cannot be set at the same time as ipipMode.	Always, CrossSubnet, Never	string	Never
natOutgoing	When enabled, packets sent from Calico networked containers in this pool to destinations outside of any Calico IP pools will be masqueraded.	true, false	boolean	false
disabled	When set to true, Calico IPAM will not assign addresses from this pool.	true, false	boolean	false
disableBGPExport (since v3.21.0)	Disable exporting routes from this IP Pool’s CIDR over BGP.	true, false	boolean	false
nodeSelector	Selects the nodes that Calico IPAM should assign addresses from this pool to.		selector	all()
allowedUses (since v3.21.0)	Controls whether the pool will be used for automatic assignments of certain types. See below.	Workload, Tunnel	list of strings	["Workload", "Tunnel"]

note

Do not use a custom blockSize until all Calico components have been updated to a version that supports it (at least v3.3.0). Older versions of components do not understand the field so they may corrupt the IP pool by creating blocks of incorrect size.

Allowed uses

When automatically assigning IP addresses to workloads, only pools with "Workload" in their allowedUses field are consulted. Similarly, when assigning IPs for tunnel devices, only "Tunnel" pools are eligible.

If the allowedUses field is not specified, it defaults to ["Workload", "Tunnel"] for compatibility with older versions of Calico. It is not possible to specify a pool with no allowed uses.

The allowedUses field is only consulted for new allocations, changing the field has no effect on previously allocated addresses.

Calico supports Kubernetes annotations that force the use of specific IP addresses. These annotations take precedence over the allowedUses field.

IPIP

Routing of packets using IP-in-IP will be used when the destination IP address is in an IP Pool that has IPIP enabled. In addition, if the ipipMode is set to CrossSubnet, Calico will only route using IP-in-IP if the IP address of the destination node is in a different subnet. The subnet of each node is configured on the node resource (which may be automatically determined when running the calico/node service).

For details on configuring IP-in-IP on your deployment, please refer to Configuring IP-in-IP.

note

Setting natOutgoing is recommended on any IP Pool with ipip enabled. When ipip is enabled without natOutgoing routing between Workloads and Hosts running Calico is asymmetric and may cause traffic to be filtered due to RPF checks failing.

VXLAN

Routing of packets using VXLAN will be used when the destination IP address is in an IP Pool that has VXLAN enabled.. In addition, if the vxlanMode is set to CrossSubnet, Calico will only route using VXLAN if the IP address of the destination node is in a different subnet. The subnet of each node is configured on the node resource (which may be automatically determined when running the calico/node service).

note

Setting natOutgoing is recommended on any IP Pool with vxlan enabled. When vxlan is enabled without natOutgoing routing between Workloads and Hosts running Calico is asymmetric and may cause traffic to be filtered due to RPF checks failing.

Block sizes

The default block sizes of 26 for IPv4 and 122 for IPv6 provide blocks of 64 addresses. This allows addresses to be allocated in groups to workloads running on the same host. By grouping addresses, fewer routes need to be exchanged between hosts and to other BGP peers. If a host allocates all of the addresses in a block then it will be allocated an additional block. If there are no more blocks available then the host can take addresses from blocks allocated to other hosts. Specific routes are added for the borrowed addresses which has an impact on route table size.

Increasing the block size from the default (e.g., using 24 for IPv4 to give 256 addresses per block) means fewer blocks per host, and potentially fewer routes. But try to ensure that there are at least as many blocks in the pool as there are hosts.

Reducing the block size from the default (e.g., using 28 for IPv4 to give 16 addresses per block) means more blocks per host and therefore potentially more routes. This can be beneficial if it allows the blocks to be more fairly distributed amongst the hosts.

Node Selector

For details on configuring IP pool node selectors, please read the Assign IP addresses based on topology guide..

tip

To prevent an IP pool from being used automatically by Calico IPAM, while still allowing it to be used manually for static assignments, set the IPPool's nodeSelector to !all(). Since the selector matches no nodes, the IPPool will not be used automatically and, unlike setting disabled: true, it can still be used for manual assignments.

Selector reference

A label selector is an expression which either matches or does not match a resource based on its labels.

Calico label selectors support a number of operators, which can be combined into larger expressions using the boolean operators and parentheses.

Expression	Meaning
Logical operators
( <expression> )	Matches if and only if <expression> matches. (Parentheses are used for grouping expressions.)
! <expression>	Matches if and only if <expression> does not match. Tip: ! is a special character at the start of a YAML string, if you need to use ! at the start of a YAML string, enclose the string in quotes.
<expression 1> && <expression 2>	"And": matches if and only if both <expression 1>, and, <expression 2> matches
<expression 1> || <expression 2>	"Or": matches if and only if either <expression 1>, or, <expression 2> matches.
Match operators
all()	Match all in-scope resources. To match no resources, combine this operator with ! to form !all().
global()	Match all non-namespaced resources. Useful in a namespaceSelector to select global resources such as global network sets.
k == 'v'	Matches resources with the label 'k' and value 'v'.
k != 'v'	Matches resources without label 'k' or with label 'k' and value not equal to v
has(k)	Matches resources with label 'k', independent of value. To match pods that do not have label k, combine this operator with ! to form !has(k)
k in { 'v1', 'v2' }	Matches resources with label 'k' and value in the given set
k not in { 'v1', 'v2' }	Matches resources without label 'k' or with label 'k' and value not in the given set
k contains 's'	Matches resources with label 'k' and value containing the substring 's'
k starts with 's'	Matches resources with label 'k' and value starting with the substring 's'
k ends with 's'	Matches resources with label 'k' and value ending with the substring 's'

Operators have the following precedence:

• Highest: all the match operators
• Parentheses ( ... )
• Negation with !
• Conjunction with &&
• Lowest: Disjunction with ||

For example, the expression

! has(my-label) || my-label starts with 'prod' && role in {'frontend','business'}

Would be "bracketed" like this:

((!(has(my-label)) || ((my-label starts with 'prod') && (role in {'frontend','business'}))

It would match:

• Any resource that did not have label "my-label".
• Any resource that both:
  • Has a value for my-label that starts with "prod", and,
  • Has a role label with value either "frontend", or "business".

Supported operations

Datastore type	Create/Delete	Update	Get/List	Notes
etcdv3	Yes	Yes	Yes
Kubernetes API server	Yes	Yes	Yes

See also

The IPReservation resource allows for small parts of an IP pool to be reserved so that they will not be used for automatic IPAM assignments.