ApisixUpstream Reference

Field	Type	Description
scheme	string	The protocol used to talk to the Service, can be http, https, grpc, grpcs default is http.
loadbalancer	object	The load balancing algorithm of this upstream service
loadbalancer.type	string	The load balancing type, can be roundrobin, ewma, least_conn, chash, default is roundrobin.
loadbalancer.hashOn	string	The hash value source scope, only take effects if the chash algorithm is in use. Values can vars, header, vars_combinations, cookie and consumers, default is vars.
loadbalancer.key	string	The hash key, only in valid if the chash algorithm is used.
retries	int	The retry count.
timeout	object	The timeout settings.
timeout.connect	time duration in the form "72h3m0.5s"	The connect timeout.
timeout.read	time duration in the form "72h3m0.5s"	The read timeout.
timeout.send	time duration in the form "72h3m0.5s"	The send timeout.
healthCheck	object	The health check parameters, see Health Check for more details.
healthCheck.active	object	active health check configuration, which is a mandatory field.
healthCheck.active.type	string	health check type, can be http, https and tcp, default is http.
healthCheck.active.timeout	time duration in the form "72h3m0.5s"	the timeout settings for the probe, default is 1s.
healthCheck.active.concurrency	int	how many probes can be sent simultaneously, default is 10.
healthCheck.active.host	string	host header in http probe request, only in valid if the active health check type is http or https.
healthCheck.active.port	int	target port to receive probes, it's necessary to specify this field if the health check service exposes by different port, note the port value here is the container port, not the service port.
healthCheck.active.httpPath	string	the HTTP URI path in http probe, only in valid if the active health check type is http or https.
healthCheck.active.strictTLS	boolean	whether to use the strict mode when use TLS, only in valid if the active health check type is https, default is true.
healthCheck.active.requestHeaders	array of string	Extra HTTP requests carried in the http probe, only in valid if the active health check type is http or https.
healthCheck.active.healthy	object	The conditions to judge an endpoint is healthy.
healthCheck.active.healthy.successes	int	The number of consecutive requests needed to set an endpoint as healthy, default is 2.
healthCheck.active.healthy.httpCodes	array of integer	Good status codes list to check whether a probe is successful, only in valid if the active health check type is http or https, default is [200, 302].
healthCheck.active.healthy.interval	time duration in the form "72h3m0.5s"	The probes sent interval (for healthy endpoints).
healthCheck.active.unhealthy	object	The conditions to judge an endpoint is unhealthy.
healthCheck.active.unhealthy.httpFailures	int	The number of consecutive http requests needed to set an endpoint as unhealthy, only in valid if the active health check type is http or https, default is 5.
healthCheck.active.unhealthy.tcpFailures	int	The number of consecutive tcp connections needed to set an endpoint as unhealthy, only in valid if the active health check type is tcp, default is 2.
healthCheck.active.unhealthy.httpCodes	array of integer	Bad status codes list to check whether a probe is failed, only in valid if the active health check type is http or https, default is [429, 404, 500, 501, 502, 503, 504, 505].
healthCheck.active.unhealthy.interval	time duration in the form "72h3m0.5s"	The probes sent interval (for unhealthy endpoints).
healthCheck.passive	object	passive health check configuration, which is an optional field.
healthCheck.passive.type	string	health check type, can be http, https and tcp, default is http.
healthCheck.passive.healthy	object	The conditions to judge an endpoint is healthy.
healthCheck.passive.healthy.successes	int	The number of consecutive requests needed to set an endpoint as healthy, default is 5.
healthCheck.passive.healthy.httpCodes	array of integer	Good status codes list to check whether a probe is successful, only in valid if the active health check type is http or https, default is [200, 201, 202, 203, 204, 205, 206, 207, 208, 226, 300, 301, 302, 303, 304, 305, 306, 307, 308].
healthCheck.passive.unhealthy	object	The conditions to judge an endpoint is unhealthy.
healthCheck.passive.unhealthy.httpFailures	int	The number of consecutive http requests needed to set an endpoint as unhealthy, only in valid if the active health check type is http or https, default is 5.
healthCheck.passive.unhealthy.tcpFailures	int	The number of consecutive tcp connections needed to set an endpoint as unhealthy, only in valid if the active health check type is tcp, default is 2.
healthCheck.passive.unhealthy.httpCodes	array of integer	Bad status codes list to check whether a probe is failed, only in valid if the active health check type is http or https, default is [429, 404, 500, 501, 502, 503, 504, 505].
portLevelSettings	array	Settings for each individual port.
portLevelSettings.port	int	The port number defined in the Kubernetes Service, must be a valid port.
portLevelSettings.scheme	string	same as scheme but takes higher precedence.
portLevelSettings.loadbalancer	object	same as loadbalancer but takes higher precedence.
portLevelSettings.healthCheck	object	same as healthCheck but takes higher precedence.
subsets	array	service subset list, use pod labels to organize service endpoints to different groups.
subsets[].name	string	the subset name.
subsets[].labels	object	the subset label map.