_last_modified (optional)
String UNIX time since epoch in microseconds. Units(MICROSECONDS).
accelerated_networking (optional)
Boolean Enable accelerated networking option for Azure SE. Accelerated networking enables single root I/O virtualization (SR-IOV) to a SE VM. This improves networking performance. Field introduced in 17.2.14,18.1.5,18.2.1.
active_standby (optional)
Boolean Service Engines in active/standby mode for HA failover.
additional_config_memory (optional)
Integer Indicates the percent of config memory used for config updates. Allowed values are 0-90. Field deprecated in 18.1.2. Field introduced in 18.1.1. Unit is PERCENT. format: int32
advertise_backend_networks (optional)
Boolean Advertise reach-ability of backend server networks via ADC through BGP for default gateway feature. Field deprecated in 18.2.5.
aggressive_failure_detection (optional)
Boolean Enable aggressive failover configuration for ha. Allowed in Basic(Allowed values- false) edition, Essentials(Allowed values- false) edition, Enterprise edition.
algo (optional)
String In compact placement, Virtual Services are placed on existing SEs until max_vs_per_se limit is reached. Enum options - PLACEMENT_ALGO_PACKED, PLACEMENT_ALGO_DISTRIBUTED.
allow_burst (optional)
Boolean Allow SEs to be created using burst license. Field introduced in 17.2.5.
app_cache_percent (optional)
Integer A percent value of total SE memory reserved for applicationcaching. This is an SE bootup property and requires SE restart.Requires SE Reboot. Allowed values are 0 - 100. Special values are 0- 'disable'. Field introduced in 18.2.3. Unit is PERCENT. Allowed in Basic(Allowed values- 0) edition, Essentials(Allowed values- 0) edition, Enterprise edition. Special default for Basic edition is 0, Essentials edition is 0, Enterprise is 10. format: int32
app_cache_threshold (optional)
Integer The max memory that can be allocated for the app cache. This value will act as an upper bound on the cache size specified in app_cache_percent. Special values are 0- 'disable'. Field introduced in 20.1.1. Unit is GB. format: int32
app_learning_memory_percent (optional)
Integer A percent value of total SE memory reserved for Application learning. This is an SE bootup property and requires SE restart. Allowed values are 0 - 10. Field introduced in 18.2.3. Unit is PERCENT. format: int32
archive_shm_limit (optional)
Integer Amount of SE memory in GB until which shared memory is collected in core archive. Field introduced in 17.1.3. Unit is GB. format: int32
async_ssl (optional)
Boolean SSL handshakes will be handled by dedicated SSL Threads.Requires SE Reboot. Allowed in Basic(Allowed values- false) edition, Essentials(Allowed values- false) edition, Enterprise edition.
async_ssl_threads (optional)
Integer Number of Async SSL threads per se_dp.Requires SE Reboot. Allowed values are 1-16. format: int32
auto_rebalance (optional)
Boolean If set, Virtual Services will be automatically migrated when load on an SE is less than minimum or more than maximum thresholds. Only Alerts are generated when the auto_rebalance is not set. Allowed in Basic(Allowed values- false) edition, Essentials(Allowed values- false) edition, Enterprise edition.
auto_rebalance_capacity_per_se (optional)
array[Integer] Capacities of SE for auto rebalance for each criteria. Field introduced in 17.2.4.
auto_rebalance_criteria (optional)
array[String] Set of criteria for SE Auto Rebalance. Enum options - SE_AUTO_REBALANCE_CPU, SE_AUTO_REBALANCE_PPS, SE_AUTO_REBALANCE_MBPS, SE_AUTO_REBALANCE_OPEN_CONNS, SE_AUTO_REBALANCE_CPS. Field introduced in 17.2.3.
auto_rebalance_interval (optional)
Integer Frequency of rebalance, if 'Auto rebalance' is enabled. Unit is SEC. format: int32
auto_redistribute_active_standby_load (optional)
Boolean Redistribution of virtual services from the takeover SE to the replacement SE can cause momentary traffic loss. If the auto-redistribute load option is left in its default off state, any desired rebalancing requires calls to REST API. Allowed in Basic(Allowed values- false) edition, Essentials(Allowed values- false) edition, Enterprise edition.
availability_zone_refs (optional)
array[String] Availability zones for Virtual Service High Availability. It is a reference to an object of type AvailabilityZone. Field introduced in 20.1.1.
bgp_state_update_interval (optional)
Integer BGP peer state update interval. Allowed values are 5-100. Field introduced in 17.2.14,18.1.5,18.2.1. Unit is SEC. format: int32
buffer_se (optional)
Integer Excess Service Engine capacity provisioned for HA failover. format: int32
cloud_ref (optional)
String It is a reference to an object of type Cloud.
compress_ip_rules_for_each_ns_subnet (optional)
Boolean Compress IP rules into a single subnet based IP rule for each north-south IPAM subnet configured in PCAP mode in OpenShift/Kubernetes node. Field introduced in 18.2.9, 20.1.1.
config_debugs_on_all_cores (optional)
Boolean Enable config debugs on all cores of SE. Field introduced in 17.2.13,18.1.5,18.2.1.
connection_memory_percentage (optional)
Integer Percentage of memory for connection state. This will come at the expense of memory used for HTTP in-memory cache. Allowed values are 10-90. Unit is PERCENT. format: int32
core_shm_app_cache (optional)
Boolean Include shared memory for app cache in core file.Requires SE Reboot. Field introduced in 18.2.8, 20.1.1.
core_shm_app_learning (optional)
Boolean Include shared memory for app learning in core file.Requires SE Reboot. Field introduced in 18.2.8, 20.1.1.
cpu_reserve (optional)
Boolean Placeholder for description of property cpu_reserve of obj type ServiceEngineGroup field type str type boolean
cpu_socket_affinity (optional)
Boolean Allocate all the CPU cores for the Service Engine Virtual Machines on the same CPU socket. Applicable only for vCenter Cloud.
custom_securitygroups_data (optional)
array[String] Custom Security Groups to be associated with data vNics for SE instances in OpenStack and AWS Clouds. Field introduced in 17.1.3.
custom_securitygroups_mgmt (optional)
array[String] Custom Security Groups to be associated with management vNic for SE instances in OpenStack and AWS Clouds. Field introduced in 17.1.3.
custom_tag (optional)
array[CustomTag] Custom tag will be used to create the tags for SE instance in AWS. Note this is not the same as the prefix for SE name.
data_network_id (optional)
String Subnet used to spin up the data nic for Service Engines, used only for Azure cloud. Overrides the cloud level setting for Service Engine subnet. Field introduced in 18.2.3.
datascript_timeout (optional)
Long Number of instructions before datascript times out. Allowed values are 0-100000000. Field introduced in 18.2.3. format: int64
dedicated_dispatcher_core (optional)
Boolean Dedicate the core that handles packet receive/transmit from the network to just the dispatching function. Don't use it for TCP/IP and SSL functions.
description (optional)
String User defined description for the object.
disable_avi_securitygroups (optional)
Boolean By default, Avi creates and manages security groups along with custom sg provided by user. Set this to True to disallow Avi to create and manage new security groups. Avi will only make use of custom security groups provided by user. This option is supported for AWS and OpenStack cloud types. Field introduced in 17.2.13,18.1.4,18.2.1.
disable_csum_offloads (optional)
Boolean Stop using TCP/UDP and IP checksum offload features of NICs. Field introduced in 17.1.14, 17.2.5, 18.1.1.
disable_flow_probes (optional)
Boolean Disable Flow Probes for Scaled out VS'es. Field introduced in 20.1.3.
disable_gro (optional)
Boolean Disable Generic Receive Offload (GRO) in DPDK poll-mode driver packet receive path. GRO is on by default on NICs that do not support LRO (Large Receive Offload) or do not gain performance boost from LRO. Field introduced in 17.2.5, 18.1.1.
disable_se_memory_check (optional)
Boolean If set, disable the config memory check done in service engine. Field introduced in 18.1.2.
disable_tso (optional)
Boolean Disable TCP Segmentation Offload (TSO) in DPDK poll-mode driver packet transmit path. TSO is on by default on NICs that support it. Field introduced in 17.2.5, 18.1.1.
disk_per_se (optional)
Integer Amount of disk space for each of the Service Engine virtual machines. Unit is GB. format: int32
distribute_load_active_standby (optional)
Boolean Use both the active and standby Service Engines for Virtual Service placement in the legacy active standby HA mode. Allowed in Basic(Allowed values- false) edition, Essentials(Allowed values- false) edition, Enterprise edition.
distribute_queues (optional)
Boolean Distributes queue ownership among cores so multiple cores handle dispatcher duties. Requires SE Reboot. Deprecated from 18.2.8, instead use max_queues_per_vnic. Field introduced in 17.2.8. Allowed in Basic(Allowed values- false) edition, Essentials(Allowed values- false) edition, Enterprise edition.
distribute_vnics (optional)
Boolean Distributes vnic ownership among cores so multiple cores handle dispatcher duties.Requires SE Reboot. Field introduced in 18.2.5. Allowed in Basic(Allowed values- false) edition, Essentials(Allowed values- false) edition, Enterprise edition.
dp_aggressive_hb_frequency (optional)
Integer Frequency of SE - SE HB messages when aggressive failure mode detection is enabled. Field introduced in 20.1.3. Unit is MILLISECONDS. format: int32
dp_aggressive_hb_timeout_count (optional)
Integer Consecutive HB failures after which failure is reported to controller,when aggressive failure mode detection is enabled. Field introduced in 20.1.3. format: int32
dp_hb_frequency (optional)
Integer Frequency of SE - SE HB messages when aggressive failure mode detection is not enabled. Field introduced in 20.1.3. Unit is MILLISECONDS. format: int32
dp_hb_timeout_count (optional)
Integer Consecutive HB failures after which failure is reported to controller, when aggressive failure mode detection is not enabled. Field introduced in 20.1.3. format: int32
enable_gratarp_permanent (optional)
Boolean Enable GratArp for VIP_IP. Field introduced in 18.2.3.
enable_hsm_priming (optional)
Boolean (This is a beta feature). Enable HSM key priming. If enabled, key handles on the hsm will be synced to SE before processing client connections. Field introduced in 17.2.7, 18.1.1.
enable_multi_lb (optional)
Boolean Applicable only for Azure cloud with Basic SKU LB. If set, additional Azure LBs will be automatically created if resources in existing LB are exhausted. Field introduced in 17.2.10, 18.1.2.
enable_pcap_tx_ring (optional)
Boolean Enable TX ring support in pcap mode of operation. TSO feature is not supported with TX Ring enabled. Deprecated from 18.2.8, instead use pcap_tx_mode. Requires SE Reboot. Field introduced in 18.2.5.
enable_routing (optional)
Boolean Enable routing for this ServiceEngineGroup . Field deprecated in 18.2.5. Allowed in Basic(Allowed values- false) edition, Essentials(Allowed values- false) edition, Enterprise edition.
enable_vip_on_all_interfaces (optional)
Boolean Enable VIP on all interfaces of SE. Field deprecated in 18.2.5. Field introduced in 17.1.1.
enable_vmac (optional)
Boolean Use Virtual MAC address for interfaces on which floating interface IPs are placed. Field deprecated in 18.2.5.
ephemeral_portrange_end (optional)
Integer End local ephemeral port number for outbound connections. Field introduced in 17.2.13, 18.1.5, 18.2.1. format: int32
ephemeral_portrange_start (optional)
Integer Start local ephemeral port number for outbound connections. Field introduced in 17.2.13, 18.1.5, 18.2.1. format: int32
extra_config_multiplier (optional)
Double Multiplier for extra config to support large VS/Pool config. format: double
extra_shared_config_memory (optional)
Integer Extra config memory to support large Geo DB configuration. Field introduced in 17.1.1. Unit is MB. format: int32
floating_intf_ip (optional)
array[IpAddr] If ServiceEngineGroup is configured for Legacy 1+1 Active Standby HA Mode, Floating IP's will be advertised only by the Active SE in the Pair. Virtual Services in this group must be disabled/enabled for any changes to the Floating IP's to take effect. Only active SE hosting VS tagged with Active Standby SE 1 Tag will advertise this floating IP when manual load distribution is enabled. Field deprecated in 18.2.5. Maximum of 32 items allowed.
floating_intf_ip_se_2 (optional)
array[IpAddr] If ServiceEngineGroup is configured for Legacy 1+1 Active Standby HA Mode, Floating IP's will be advertised only by the Active SE in the Pair. Virtual Services in this group must be disabled/enabled for any changes to the Floating IP's to take effect. Only active SE hosting VS tagged with Active Standby SE 2 Tag will advertise this floating IP when manual load distribution is enabled. Field deprecated in 18.2.5. Maximum of 32 items allowed.
flow_table_new_syn_max_entries (optional)
Integer Maximum number of flow table entries that have not completed TCP three-way handshake yet. Field introduced in 17.2.5. format: int32
free_list_size (optional)
Integer Number of entries in the free list. Field introduced in 17.2.10, 18.1.2. format: int32
gcp_config (optional)
GCPSeGroupConfig Google Cloud Platform, Service Engine Group Configuration. Field introduced in 20.1.3.
gratarp_permanent_periodicity (optional)
Integer GratArp periodicity for VIP-IP. Allowed values are 5-30. Field introduced in 18.2.3. Unit is MIN. format: int32
ha_mode (optional)
String High Availability mode for all the Virtual Services using this Service Engine group. Enum options - HA_MODE_SHARED_PAIR, HA_MODE_SHARED, HA_MODE_LEGACY_ACTIVE_STANDBY. Allowed in Basic(Allowed values- HA_MODE_LEGACY_ACTIVE_STANDBY) edition, Essentials(Allowed values- HA_MODE_LEGACY_ACTIVE_STANDBY) edition, Enterprise edition. Special default for Basic edition is HA_MODE_LEGACY_ACTIVE_STANDBY, Essentials edition is HA_MODE_LEGACY_ACTIVE_STANDBY, Enterprise is HA_MODE_SHARED.
handle_per_pkt_attack (optional)
Boolean Configuration to handle per packet attack handling.For example, DNS Reflection Attack is a type of attack where a response packet is sent to the DNS VS.This configuration tells if such packets should be dropped without further processing. Field introduced in 20.1.3.
hardwaresecuritymodulegroup_ref (optional)
String It is a reference to an object of type HardwareSecurityModuleGroup.
heap_minimum_config_memory (optional)
Integer Minimum required heap memory to apply any configuration. Allowed values are 0-100. Field introduced in 18.1.2. Unit is MB. format: int32
hm_on_standby (optional)
Boolean Enable active health monitoring from the standby SE for all placed virtual services. Allowed in Basic(Allowed values- false) edition, Essentials(Allowed values- false) edition, Enterprise edition. Special default for Basic edition is false, Essentials edition is false, Enterprise is True.
host_attribute_key (optional)
String Key of a (Key, Value) pair identifying a label for a set of Nodes usually in Container Clouds. Needs to be specified together with host_attribute_value. SEs can be configured differently including HA modes across different SE Groups. May also be used for isolation between different classes of VirtualServices. VirtualServices' SE Group may be specified via annotations/labels. A OpenShift/Kubernetes namespace maybe annotated with a matching SE Group label as openshift.io/node-selector apptype=prod. When multiple SE Groups are used in a Cloud with host attributes specified,just a single SE Group can exist as a match-all SE Group without a host_attribute_key.
host_attribute_value (optional)
String Value of a (Key, Value) pair identifying a label for a set of Nodes usually in Container Clouds. Needs to be specified together with host_attribute_key.
host_gateway_monitor (optional)
Boolean Enable the host gateway monitor when service engine is deployed as docker container. Disabled by default. Field introduced in 17.2.4.
hypervisor (optional)
String Override default hypervisor. Enum options - DEFAULT, VMWARE_ESX, KVM, VMWARE_VSAN, XEN.
ignore_rtt_threshold (optional)
Integer Ignore RTT samples if it is above threshold. Field introduced in 17.1.6,17.2.2. Unit is MILLISECONDS. format: int32
ingress_access_data (optional)
String Program SE security group ingress rules to allow VIP data access from remote CIDR type. Enum options - SG_INGRESS_ACCESS_NONE, SG_INGRESS_ACCESS_ALL, SG_INGRESS_ACCESS_VPC. Field introduced in 17.1.5.
ingress_access_mgmt (optional)
String Program SE security group ingress rules to allow SSH/ICMP management access from remote CIDR type. Enum options - SG_INGRESS_ACCESS_NONE, SG_INGRESS_ACCESS_ALL, SG_INGRESS_ACCESS_VPC. Field introduced in 17.1.5.
instance_flavor (optional)
String Instance/Flavor name for SE instance.
instance_flavor_info (optional)
CloudFlavor Additional information associated with instance_flavor. Field introduced in 20.1.1.
iptables (optional)
labels (optional)
array[KeyValue] Labels associated with this SE group. Field introduced in 20.1.1. Maximum of 1 items allowed.
least_load_core_selection (optional)
Boolean Select core with least load for new flow.
license_tier (optional)
String Specifies the license tier which would be used. This field by default inherits the value from cloud. Enum options - ENTERPRISE_16, ENTERPRISE, ENTERPRISE_18, BASIC, ESSENTIALS. Field introduced in 17.2.5.
license_type (optional)
String If no license type is specified then default license enforcement for the cloud type is chosen. Enum options - LIC_BACKEND_SERVERS, LIC_SOCKETS, LIC_CORES, LIC_HOSTS, LIC_SE_BANDWIDTH, LIC_METERED_SE_BANDWIDTH. Field introduced in 17.2.5.
log_disksz (optional)
Integer Maximum disk capacity (in MB) to be allocated to an SE. This is exclusively used for debug and log data. Unit is MB. format: int32
log_malloc_failure (optional)
Boolean SE will log memory allocation related failure to the se_trace file, wherever available. Field introduced in 20.1.2. Allowed in Basic(Allowed values- true) edition, Essentials(Allowed values- true) edition, Enterprise edition.
markers (optional)
array[RoleFilterMatchLabel] List of labels to be used for granular RBAC. Field introduced in 20.1.7. Allowed in Basic edition, Essentials edition, Enterprise edition.
max_concurrent_external_hm (optional)
Integer Maximum number of external health monitors that can run concurrently in a service engine. This helps control the CPU and memory use by external health monitors. Special values are 0- 'Value will be internally calculated based on cpu and memory'. Field introduced in 18.2.7. format: int32
max_cpu_usage (optional)
Integer When CPU usage on an SE exceeds this threshold, Virtual Services hosted on this SE may be rebalanced to other SEs to reduce load. A new SE may be created as part of this process. Allowed values are 40-90. Unit is PERCENT. format: int32
max_memory_per_mempool (optional)
Integer Max bytes that can be allocated in a single mempool. Field introduced in 18.1.5. Unit is MB. format: int32
max_num_se_dps (optional)
Integer Configures the maximum number of se_dp processes created on the SE, requires SE reboot. If not configured, defaults to the number of CPUs on the SE. This should only be used if user wants to limit the number of se_dps to less than the available CPUs on the SE. Allowed values are 1-128. Field introduced in 20.1.1. Allowed in Basic(Allowed values- 0) edition, Essentials(Allowed values- 0) edition, Enterprise edition. format: int32
max_public_ips_per_lb (optional)
Integer Applicable to Azure platform only. Maximum number of public IPs per Azure LB. . Field introduced in 17.2.12, 18.1.2. format: int32
max_queues_per_vnic (optional)
Integer Maximum number of queues per vnic Setting to '0' utilises all queues that are distributed across dispatcher cores. Allowed values are 0,1,2,4,8,16. Field introduced in 18.2.7, 20.1.1. Allowed in Basic(Allowed values- 1) edition, Essentials(Allowed values- 1) edition, Enterprise edition. format: int32
max_rules_per_lb (optional)
Integer Applicable to Azure platform only. Maximum number of rules per Azure LB. . Field introduced in 17.2.12, 18.1.2. format: int32
max_scaleout_per_vs (optional)
Integer Maximum number of active Service Engines for the Virtual Service. Allowed values are 1-64. format: int32
max_se (optional)
Integer Maximum number of Services Engines in this group. Allowed values are 0-1000. format: int32
max_vs_per_se (optional)
Integer Maximum number of Virtual Services that can be placed on a single Service Engine. Allowed values are 1-1000. format: int32
mem_reserve (optional)
Boolean Placeholder for description of property mem_reserve of obj type ServiceEngineGroup field type str type boolean
memory_for_config_update (optional)
Integer Indicates the percent of memory reserved for config updates. Allowed values are 0-100. Field introduced in 18.1.2. Unit is PERCENT. format: int32
memory_per_se (optional)
Integer Amount of memory for each of the Service Engine virtual machines. Changes to this setting do not affect existing SEs. format: int32
mgmt_network_ref (optional)
String Management network to use for Avi Service Engines. It is a reference to an object of type Network.
mgmt_subnet (optional)
IpAddrPrefix Management subnet to use for Avi Service Engines.
min_cpu_usage (optional)
Integer When CPU usage on an SE falls below the minimum threshold, Virtual Services hosted on the SE may be consolidated onto other underutilized SEs. After consolidation, unused Service Engines may then be eligible for deletion. . Allowed values are 20-60. Unit is PERCENT. format: int32
min_scaleout_per_vs (optional)
Integer Minimum number of active Service Engines for the Virtual Service. Allowed values are 1-64. format: int32
min_se (optional)
Integer Minimum number of Services Engines in this group (relevant for SE AutoRebalance only). Allowed values are 0-1000. Field introduced in 17.2.13,18.1.3,18.2.1. format: int32
minimum_connection_memory (optional)
Integer Indicates the percent of memory reserved for connections. Allowed values are 0-100. Field introduced in 18.1.2. Unit is PERCENT. format: int32
minimum_required_config_memory (optional)
Integer Required available config memory to apply any configuration. Allowed values are 0-90. Field deprecated in 18.1.2. Field introduced in 18.1.1. Unit is PERCENT. format: int32
n_log_streaming_threads (optional)
Integer Number of threads to use for log streaming. Allowed values are 1-100. Field introduced in 17.2.12, 18.1.2. format: int32
name
nat_flow_tcp_closed_timeout (optional)
Integer Idle timeout in seconds for nat tcp flows in closed state. Allowed values are 1-3600. Field deprecated in 18.2.5. Field introduced in 18.2.5. Unit is SECONDS. format: int32
nat_flow_tcp_established_timeout (optional)
Integer Idle timeout in seconds for nat tcp flows in established state. Allowed values are 1-3600. Field deprecated in 18.2.5. Field introduced in 18.2.5. Unit is SECONDS. format: int32
nat_flow_tcp_half_closed_timeout (optional)
Integer Idle timeout in seconds for nat tcp flows in half closed state. Allowed values are 1-3600. Field deprecated in 18.2.5. Field introduced in 18.2.5. Unit is SECONDS. format: int32
nat_flow_tcp_handshake_timeout (optional)
Integer Idle timeout in seconds for nat tcp flows in handshake state. Allowed values are 1-3600. Field deprecated in 18.2.5. Field introduced in 18.2.5. Unit is SECONDS. format: int32
nat_flow_udp_noresponse_timeout (optional)
Integer Idle timeout in seconds for nat udp flows in noresponse state. Allowed values are 1-3600. Field deprecated in 18.2.5. Field introduced in 18.2.5. Unit is SECONDS. format: int32
nat_flow_udp_response_timeout (optional)
Integer Idle timeout in seconds for nat udp flows in response state. Allowed values are 1-3600. Field deprecated in 18.2.5. Field introduced in 18.2.5. Unit is SECONDS. format: int32
non_significant_log_throttle (optional)
Integer This setting limits the number of non-significant logs generated per second per core on this SE. Default is 100 logs per second. Set it to zero (0) to deactivate throttling. Field introduced in 17.1.3. Unit is PER_SECOND. format: int32
num_dispatcher_cores (optional)
Integer Number of dispatcher cores (0,1,2,4,8 or 16). If set to 0, then number of dispatcher cores is deduced automatically.Requires SE Reboot. Allowed values are 0,1,2,4,8,16. Field introduced in 17.2.12, 18.1.3, 18.2.1. Allowed in Basic(Allowed values- 0) edition, Essentials(Allowed values- 0) edition, Enterprise edition. format: int32
num_flow_cores_sum_changes_to_ignore (optional)
Integer Number of changes in num flow cores sum to ignore. format: int32
objsync_config (optional)
ObjSyncConfig Configuration knobs for InterSE Object Distribution. Field introduced in 20.1.3.
objsync_port (optional)
Integer TCP port on SE management interface for InterSE Object Distribution. Supported only for externally managed security groups. Not supported on full access deployments. Requires SE reboot. Field introduced in 20.1.3. format: int32
openstack_availability_zone (optional)
String Field deprecated in 17.1.1.
openstack_availability_zones (optional)
array[String] Field introduced in 17.1.1. Maximum of 5 items allowed.
openstack_mgmt_network_name (optional)
String Avi Management network name.
openstack_mgmt_network_uuid (optional)
String Management network UUID.
os_reserved_memory (optional)
Integer Amount of extra memory to be reserved for use by the Operating System on a Service Engine. Unit is MB. format: int32
pcap_tx_mode (optional)
String Determines the PCAP transmit mode of operation. Requires SE Reboot. Enum options - PCAP_TX_AUTO, PCAP_TX_SOCKET, PCAP_TX_RING. Field introduced in 18.2.8, 20.1.1.
pcap_tx_ring_rd_balancing_factor (optional)
Integer In PCAP mode, reserve a configured portion of TX ring resources for itself and the remaining portion for the RX ring to achieve better balance in terms of queue depth. Requires SE Reboot. Allowed values are 10-100. Field introduced in 20.1.3. Unit is PERCENT. format: int32
per_app (optional)
Boolean Per-app SE mode is designed for deploying dedicated load balancers per app (VS). In this mode, each SE is limited to a max of 2 VSs. vCPUs in per-app SEs count towards licensing usage at 25% rate. Allowed in Basic(Allowed values- false) edition, Essentials(Allowed values- false) edition, Enterprise edition.
per_vs_admission_control (optional)
Boolean Enable/Disable per VS level admission control.Enabling this feature will cause the connection and packet throttling on a particular VS that has high packet buffer consumption. Field introduced in 20.1.3.
placement_mode (optional)
String If placement mode is 'Auto', Virtual Services are automatically placed on Service Engines. Enum options - PLACEMENT_MODE_AUTO.
realtime_se_metrics (optional)
reboot_on_panic (optional)
Boolean Reboot the VM or host on kernel panic. Field introduced in 18.2.5.
reboot_on_stop (optional)
Boolean Reboot the system if the SE is stopped. Field deprecated in 18.2.5.
resync_time_interval (optional)
Integer Time interval to re-sync SE's time with wall clock time. Allowed values are 8-600000. Field introduced in 20.1.1. Unit is MILLISECONDS. format: int32
se_bandwidth_type (optional)
String Select the SE bandwidth for the bandwidth license. Enum options - SE_BANDWIDTH_UNLIMITED, SE_BANDWIDTH_25M, SE_BANDWIDTH_200M, SE_BANDWIDTH_1000M, SE_BANDWIDTH_10000M. Field introduced in 17.2.5. Allowed in Basic(Allowed values- SE_BANDWIDTH_UNLIMITED) edition, Essentials(Allowed values- SE_BANDWIDTH_UNLIMITED) edition, Enterprise edition.
se_delayed_flow_delete (optional)
Boolean Delay the cleanup of flowtable entry. To be used under surveillance of Avi Support. Field introduced in 20.1.2. Allowed in Basic(Allowed values- true) edition, Essentials(Allowed values- true) edition, Enterprise edition.
se_deprovision_delay (optional)
Integer Duration to preserve unused Service Engine virtual machines before deleting them. If traffic to a Virtual Service were to spike up abruptly, this SE would still be available to be utilized again rather than creating a new SE. If this value is set to 0, Controller will never delete any SEs and administrator has to manually cleanup unused SEs. Allowed values are 0-525600. Unit is MIN. format: int32
se_dos_profile (optional)
DosThresholdProfile Placeholder for description of property se_dos_profile of obj type ServiceEngineGroup field type str type object
se_dp_hm_drops (optional)
Integer Internal only. Used to simulate SE - SE HB failure. Field introduced in 20.1.3. format: int32
se_dp_isolation (optional)
Boolean Toggle support to run SE datapath instances in isolation on exclusive CPUs. This improves latency and performance. However, this could reduce the total number of se_dp instances created on that SE instance. Supported for >= 8 CPUs. Requires SE reboot. Field introduced in 20.1.4.
se_dp_isolation_num_non_dp_cpus (optional)
Integer Number of CPUs for non se-dp tasks in SE datapath isolation mode. Translates Total cpus minus 'num_non_dp_cpus' for datapath use. It is recommended to reserve an even number of CPUs for hyper-threaded processors. Requires SE reboot. Allowed values are 1-8. Special values are 0- 'auto'. Field introduced in 20.1.4. format: int32
se_dp_max_hb_version (optional)
Integer The highest supported SE-SE Heartbeat protocol version. This version is reported by Secondary SE to Primary SE in Heartbeat response messages. Allowed values are 1-2. Field introduced in 20.1.1. format: int32
se_dp_vnic_queue_stall_event_sleep (optional)
Integer Time (in seconds) service engine waits for after generating a Vnic transmit queue stall event before resetting theNIC. Field introduced in 18.2.5. format: int32
se_dp_vnic_queue_stall_threshold (optional)
Integer Number of consecutive transmit failures to look for before generating a Vnic transmit queue stall event. Field introduced in 18.2.5. format: int32
se_dp_vnic_queue_stall_timeout (optional)
Integer Time (in milliseconds) to wait for network/NIC recovery on detecting a transmit queue stall after which service engine resets the NIC. Field introduced in 18.2.5. format: int32
se_dp_vnic_restart_on_queue_stall_count (optional)
Integer Number of consecutive transmit queue stall events in se_dp_vnic_stall_se_restart_window to look for before restarting SE. Field introduced in 18.2.5. format: int32
se_dp_vnic_stall_se_restart_window (optional)
Integer Window of time (in seconds) during which se_dp_vnic_restart_on_queue_stall_count number of consecutive stalls results in a SE restart. Field introduced in 18.2.5. format: int32
se_dpdk_pmd (optional)
Integer Determines if DPDK pool mode driver should be used or not 0 Automatically determine based on hypervisor/NIC type 1 Unconditionally use DPDK poll mode driver 2 Don't use DPDK poll mode driver.Requires SE Reboot. Allowed values are 0-2. Field introduced in 18.1.3. format: int32
se_flow_probe_retries (optional)
Integer Flow probe retry count if no replies are received.Requires SE Reboot. Allowed values are 0-5. Field introduced in 18.1.4, 18.2.1. format: int32
se_flow_probe_retry_timer (optional)
Integer Timeout in milliseconds for flow probe retries.Requires SE Reboot. Allowed values are 20-50. Field introduced in 18.2.5. Unit is MILLISECONDS. format: int32
se_flow_probe_timer (optional)
Integer Timeout in milliseconds for flow probe entries. Allowed values are 10-200. Field deprecated in 18.2.5. Field introduced in 18.1.4, 18.2.1. Unit is MILLISECONDS. format: int32
se_group_analytics_policy (optional)
se_hyperthreaded_mode (optional)
String Controls the distribution of SE data path processes on CPUs which support hyper-threading. Requires hyper-threading to be enabled at host level. Requires SE Reboot. For more details please refer to SE placement KB. Enum options - SE_CPU_HT_AUTO, SE_CPU_HT_SPARSE_DISPATCHER_PRIORITY, SE_CPU_HT_SPARSE_PROXY_PRIORITY, SE_CPU_HT_PACKED_CORES. Field introduced in 20.1.1.
se_ip_encap_ipc (optional)
Integer Determines if SE-SE IPC messages are encapsulated in an IP header 0 Automatically determine based on hypervisor type 1 Use IP encap unconditionally ~[0,1] Don't use IP encapRequires SE Reboot. Field introduced in 20.1.3. format: int32
se_ipc_udp_port (optional)
Integer UDP Port for SE_DP IPC in Docker bridge mode. Field deprecated in 20.1.1. Field introduced in 17.1.2. format: int32
se_kni_burst_factor (optional)
Integer This knob controls the resource availability and burst size used between SE datapath and KNI. This helps in minimising packet drops when there is higher KNI traffic (non-VIP traffic from and to Linux). The factor takes the following values 0-default. 1-doubles the burst size and KNI resources. 2-quadruples the burst size and KNI resources. Allowed values are 0-2. Field introduced in 18.2.6. format: int32
se_l3_encap_ipc (optional)
Integer Determines if SE-SE IPC messages use SE interface IP instead of VIP 0 Automatically determine based on hypervisor type 1 Use SE interface IP unconditionally ~[0,1] Don't use SE interface IPRequires SE Reboot. Field introduced in 20.1.3. format: int32
se_lro (optional)
Boolean Enable or disable Large Receive Optimization for vnics. Requires SE Reboot. Field introduced in 18.2.5.
se_mp_ring_retry_count (optional)
Integer The retry count for the multi-producer enqueue before yielding the CPU. To be used under surveillance of Avi Support. Field introduced in 20.1.3. Allowed in Basic(Allowed values- 500) edition, Essentials(Allowed values- 500) edition, Enterprise edition. format: int32
se_mtu (optional)
Integer MTU for the VNICs of SEs in the SE group. Allowed values are 512-9000. Field introduced in 18.2.8, 20.1.1. format: int32
se_name_prefix (optional)
String Prefix to use for virtual machine name of Service Engines.
se_pcap_lookahead (optional)
Boolean Enables lookahead mode of packet receive in PCAP mode. Introduced to overcome an issue with hv_netvsc driver. Lookahead mode attempts to ensure that application and kernel's view of the receive rings are consistent. Field introduced in 18.2.3.
se_pcap_pkt_count (optional)
Integer Max number of packets the pcap interface can hold and if the value is 0 the optimum value will be chosen. The optimum value will be chosen based on SE-memory, Cloud Type and Number of Interfaces.Requires SE Reboot. Field introduced in 18.2.5. format: int32
se_pcap_pkt_sz (optional)
Integer Max size of each packet in the pcap interface. Requires SE Reboot. Field introduced in 18.2.5. format: int32
se_pcap_qdisc_bypass (optional)
Boolean Bypass the kernel's traffic control layer, to deliver packets directly to the driver. Enabling this feature results in egress packets not being captured in host tcpdump. Note brief packet reordering or loss may occur upon toggle. Field introduced in 18.2.6.
se_pcap_reinit_frequency (optional)
Integer Frequency in seconds at which periodically a PCAP reinit check is triggered. May be used in conjunction with the configuration pcap_reinit_threshold. (Valid range 15 mins - 12 hours, 0 - disables). Allowed values are 900-43200. Special values are 0- 'disable'. Field introduced in 17.2.13, 18.1.3, 18.2.1. Unit is SEC. format: int32
se_pcap_reinit_threshold (optional)
Integer Threshold for input packet receive errors in PCAP mode exceeding which a PCAP reinit is triggered. If not set, an unconditional reinit is performed. This value is checked every pcap_reinit_frequency interval. Field introduced in 17.2.13, 18.1.3, 18.2.1. Unit is METRIC_COUNT. format: int32
se_probe_port (optional)
Integer TCP port on SE where echo service will be run. Field introduced in 17.2.2. format: int32
se_remote_punt_udp_port (optional)
Integer UDP Port for punted packets in Docker bridge mode. Field deprecated in 20.1.1. Field introduced in 17.1.2. format: int32
se_rl_prop (optional)
se_routing (optional)
Boolean Enable routing via Service Engine Datapath. When disabled, routing is done by the Linux kernel. IP Routing needs to be enabled in Service Engine Group for SE Routing to be effective. Field deprecated in 18.2.5. Field introduced in 18.2.3.
se_rum_sampling_nav_interval (optional)
Integer Minimum time to wait on server between taking sampleswhen sampling the navigation timing data from the end user client. Field introduced in 18.2.6. Unit is SEC. format: int32
se_rum_sampling_nav_percent (optional)
Integer Percentage of navigation timing data from the end user client, used for sampling to get client insights. Field introduced in 18.2.6. format: int32
se_rum_sampling_res_interval (optional)
Integer Minimum time to wait on server between taking sampleswhen sampling the resource timing data from the end user client. Field introduced in 18.2.6. Unit is SEC. format: int32
se_rum_sampling_res_percent (optional)
Integer Percentage of resource timing data from the end user client used for sampling to get client insight. Field introduced in 18.2.6. format: int32
se_sb_dedicated_core (optional)
Boolean Sideband traffic will be handled by a dedicated core.Requires SE Reboot. Field introduced in 16.5.2, 17.1.9, 17.2.3.
se_sb_threads (optional)
Integer Number of Sideband threads per SE.Requires SE Reboot. Allowed values are 1-128. Field introduced in 16.5.2, 17.1.9, 17.2.3. format: int32
se_thread_multiplier (optional)
Integer Multiplier for SE threads based on vCPU. Allowed values are 1-10. Allowed in Basic(Allowed values- 1) edition, Essentials(Allowed values- 1) edition, Enterprise edition. format: int32
se_tracert_port_range (optional)
PortRange Traceroute port range. Field introduced in 17.2.8.
se_tunnel_mode (optional)
Integer Determines if Direct Secondary Return (DSR) from secondary SE is active or not 0 Automatically determine based on hypervisor type. 1 Enable tunnel mode - DSR is unconditionally disabled. 2 Disable tunnel mode - DSR is unconditionally enabled. Tunnel mode can be enabled or disabled at run-time. Allowed values are 0-2. Field introduced in 17.1.1. Allowed in Basic(Allowed values- 0) edition, Essentials(Allowed values- 0) edition, Enterprise edition. format: int32
se_tunnel_udp_port (optional)
Integer UDP Port for tunneled packets from secondary to primary SE in Docker bridge mode.Requires SE Reboot. Field introduced in 17.1.3. format: int32
se_tx_batch_size (optional)
Integer Number of packets to batch for transmit to the nic. Requires SE Reboot. Field introduced in 18.2.5. format: int32
se_txq_threshold (optional)
Integer Once the TX queue of the dispatcher reaches this threshold, hardware queues are not polled for further packets. To be used under surveillance of Avi Support. Allowed values are 512-32768. Field introduced in 20.1.2. Allowed in Basic(Allowed values- 2048) edition, Essentials(Allowed values- 2048) edition, Enterprise edition. format: int32
se_udp_encap_ipc (optional)
Integer Determines if SE-SE IPC messages are encapsulated in a UDP header 0 Automatically determine based on hypervisor type. 1 Use UDP encap unconditionally.Requires SE Reboot. Allowed values are 0-1. Field introduced in 17.1.2. format: int32
se_use_dpdk (optional)
Integer Determines if DPDK library should be used or not 0 Automatically determine based on hypervisor type 1 Use DPDK if PCAP is not enabled 2 Don't use DPDK. Allowed values are 0-2. Field introduced in 18.1.3. format: int32
se_vnic_tx_sw_queue_flush_frequency (optional)
Integer Configure the frequency in milliseconds of software transmit spillover queue flush when enabled. This is necessary to flush any packets in the spillover queue in the absence of a packet transmit in the normal course of operation. Allowed values are 50-500. Special values are 0- 'disable'. Field introduced in 20.1.1. Unit is MILLISECONDS. format: int32
se_vnic_tx_sw_queue_size (optional)
Integer Configure the size of software transmit spillover queue when enabled. Requires SE Reboot. Allowed values are 128-2048. Field introduced in 20.1.1. format: int32
se_vs_hb_max_pkts_in_batch (optional)
Integer Maximum number of aggregated vs heartbeat packets to send in a batch. Allowed values are 1-256. Field introduced in 17.1.1. format: int32
se_vs_hb_max_vs_in_pkt (optional)
Integer Maximum number of virtualservices for which heartbeat messages are aggregated in one packet. Allowed values are 1-1024. Field introduced in 17.1.1. format: int32
self_se_election (optional)
Boolean Enable SEs to elect a primary amongst themselves in the absence of a connectivity to controller. Field introduced in 18.1.2. Allowed in Basic(Allowed values- false) edition, Essentials(Allowed values- false) edition, Enterprise edition.
service_ip6_subnets (optional)
array[IpAddrPrefix] IPv6 Subnets assigned to the SE group. Required for VS group placement. Field introduced in 18.1.1. Maximum of 128 items allowed.
service_ip_subnets (optional)
array[IpAddrPrefix] Subnets assigned to the SE group. Required for VS group placement. Field introduced in 17.1.1. Maximum of 128 items allowed.
shm_minimum_config_memory (optional)
Integer Minimum required shared memory to apply any configuration. Allowed values are 0-100. Field introduced in 18.1.2. Unit is MB. format: int32
significant_log_throttle (optional)
Integer This setting limits the number of significant logs generated per second per core on this SE. Default is 100 logs per second. Set it to zero (0) to deactivate throttling. Field introduced in 17.1.3. Unit is PER_SECOND. format: int32
ssl_preprocess_sni_hostname (optional)
Boolean (Beta) Preprocess SSL Client Hello for SNI hostname extension.If set to True, this will apply SNI child's SSL protocol(s), if they are different from SNI Parent's allowed SSL protocol(s). Field introduced in 17.2.12, 18.1.3.
tenant_ref (optional)
String It is a reference to an object of type Tenant.
transient_shared_memory_max (optional)
Integer The threshold for the transient shared config memory in the SE. Allowed values are 0-100. Field introduced in 20.1.1. Unit is PERCENT. format: int32
udf_log_throttle (optional)
Integer This setting limits the number of UDF logs generated per second per core on this SE. UDF logs are generated due to the configured client log filters or the rules with logging enabled. Default is 100 logs per second. Set it to zero (0) to deactivate throttling. Field introduced in 17.1.3. Unit is PER_SECOND. format: int32
url (optional)
use_hyperthreaded_cores (optional)
Boolean Enables the use of hyper-threaded cores on SE. Requires SE Reboot. Field introduced in 20.1.1.
use_objsync (optional)
Boolean Enable InterSE Objsyc distribution framework. Field introduced in 20.1.3. Allowed in Basic edition, Essentials edition, Enterprise edition.
use_standard_alb (optional)
Boolean Use Standard SKU Azure Load Balancer. By default cloud level flag is set. If not set, it inherits/uses the use_standard_alb flag from the cloud. Field introduced in 18.2.3.
uuid (optional)
String Unique object identifier of the object.
vcenter_clusters (optional)
VcenterClusters Placeholder for description of property vcenter_clusters of obj type ServiceEngineGroup field type str type object
vcenter_datastore_mode (optional)
String Enum options - VCENTER_DATASTORE_ANY, VCENTER_DATASTORE_LOCAL, VCENTER_DATASTORE_SHARED.
vcenter_datastores (optional)
array[VcenterDatastore] Placeholder for description of property vcenter_datastores of obj type ServiceEngineGroup field type str type object
vcenter_datastores_include (optional)
Boolean Placeholder for description of property vcenter_datastores_include of obj type ServiceEngineGroup field type str type boolean
vcenter_folder (optional)
String Folder to place all the Service Engine virtual machines in vCenter.
vcenter_hosts (optional)
VcenterHosts Placeholder for description of property vcenter_hosts of obj type ServiceEngineGroup field type str type object
vcenters (optional)
vcpus_per_se (optional)
Integer Number of vcpus for each of the Service Engine virtual machines. Changes to this setting do not affect existing SEs. format: int32
vip_asg (optional)
VipAutoscaleGroup When vip_asg is set, Vip configuration will be managed by Avi.User will be able to configure vip_asg or Vips individually at the time of create. Field introduced in 17.2.12, 18.1.2.
vs_host_redundancy (optional)
Boolean Ensure primary and secondary Service Engines are deployed on different physical hosts. Allowed in Basic(Allowed values- true) edition, Essentials(Allowed values- true) edition, Enterprise edition. Special default for Basic edition is true, Essentials edition is true, Enterprise is True.
vs_scalein_timeout (optional)
Integer Time to wait for the scaled in SE to drain existing flows before marking the scalein done. Unit is SEC. format: int32
vs_scalein_timeout_for_upgrade (optional)
Integer During SE upgrade, Time to wait for the scaled-in SE to drain existing flows before marking the scalein done. Unit is SEC. format: int32
vs_scaleout_timeout (optional)
Integer Time to wait for the scaled out SE to become ready before marking the scaleout done. Unit is SEC. format: int32
vs_se_scaleout_additional_wait_time (optional)
Integer Wait time for sending scaleout ready notification after Virtual Service is marked UP. In certain deployments, there may be an additional delay to accept traffic. For example, for BGP, some time is needed for route advertisement. Allowed values are 0-20. Field introduced in 18.1.5,18.2.1. Unit is SEC. format: int32
vs_se_scaleout_ready_timeout (optional)
Integer Timeout in seconds for Service Engine to sendScaleout Ready notification of a Virtual Service. Allowed values are 0-90. Field introduced in 18.1.5,18.2.1. Unit is SEC. format: int32
vs_switchover_timeout (optional)
Integer During SE upgrade in a legacy active/standby segroup, Time to wait for the new primary SE to accept flows before marking the switchover done. Field introduced in 17.2.13,18.1.4,18.2.1. Unit is SEC. format: int32
vss_placement (optional)
VssPlacement Parameters to place Virtual Services on only a subset of the cores of an SE. Field introduced in 17.2.5.
vss_placement_enabled (optional)
Boolean If set, Virtual Services will be placed on only a subset of the cores of an SE. Field introduced in 18.1.1.
waf_learning_interval (optional)
Integer Frequency with which SE publishes WAF learning. Allowed values are 1-43200. Field deprecated in 18.2.3. Field introduced in 18.1.2. Unit is MIN. format: int32
waf_learning_memory (optional)
Integer Amount of memory reserved on SE for WAF learning. This can be atmost 5% of SE memory. Field deprecated in 18.2.3. Field introduced in 18.1.2. Unit is MB. format: int32
waf_mempool (optional)
Boolean Enable memory pool for WAF.Requires SE Reboot. Field introduced in 17.2.3.
waf_mempool_size (optional)
Integer Memory pool size used for WAF.Requires SE Reboot. Field introduced in 17.2.3. Unit is KB. format: int32