pingap.toml

[basic]
# Specify the application instance name, which is used to easily identify the application when sending notifications. Default `pingap`
# name = "pingap"

# Error response templates, which can be configured as HTML or JSON. The following characters in the template will be replaced.
# {{version}} - the version of the application
# {{error_type}} - the type of the error
# {{content}} - the content of the error
# Default `the content of error.html`
# error_template = ""

# Process id file. Default `/run/pingap.pid`
# pid_file = "/run/pingap.pid"

# Unix socket path used for application upgrades. Default `/tmp/pingap_upgrade.sock`
# upgrade_sock = "/tmp/pingap_upgrade.sock"

# After daemonization, this process will switch to the given user before starting to serve traffic. Default `none`
# user = ""

# Similar to `user`, the group this process should switch to. Default `none`
# group = ""

# Set the number of threads per server. If set to 0, it will match the CPU core count. Default `1`
# threads = 1

# Allow work stealing between threads of the same service. Default `true`.
# work_stealing = true

# Grace period before starting the final step of the graceful shutdown after signaling shutdown. Default `5m`
# grace_period = "5m"

# Timeout in seconds of the final step for the graceful shutdown. Default `5s`
# graceful_shutdown_timeout = "5s"

# Keepalive pool size for client connections to upstream. Default `128`
# upstream_keepalive_pool_size = 128

# HTTP webhook URL for notifications, default `none`
# webhook = "https://example.com/webhook"

# Webhook type, supports 'wecom', 'dingtalk' and 'normal' types. Default `none`
# webhook_type = "wecom"

# Available events: "backend_status" (upstream backend status changes), "lets_encrypt" (Let's Encrypt certificate operations),
# "diff_config" (configuration changes), "restart" (application restarts), "restart_fail" (application restart fails),
# "reload_config" (configuration reloads), "reload_config_fail" (configuration reload fails), "tls_validity" (TLS certificate validity changes),
# "service_discover_fail" (service discovery failures). Default `none`
# webhook_notifications = ["backend_status"]

# Set log level for application. 
# Available levels: "debug", "info", "warn", "error"
# Default `None`
# log_level = "info"

# Specifies how much log data to buffer in memory before writing to disk.
# Larger buffer size can improve performance but uses more memory and may lose some logs if the application crashes.
# Default `None`
# log_buffered_size = "10MB"

# Enable json log format. Default `false`
# log_format_json = false

# Sentry connection URI for error tracking and monitoring.
# Enable this to track errors and exceptions in your application.
# It's available in the full feature release. Default `none`
# sentry = ""

# Pyroscope is a continuous profiling platform that helps you understand performance
# bottlenecks in your application. When configured, it will collect and analyze CPU profiles.
# Example value: "http://pyroscope:4040"
# It's available in the full feature release. Default `none`
# pyroscope = ""

# This setting controls how often Pingap checks for configuration file changes.
# When changes are detected, the process will automatically restart to apply the new configuration.
# Default `90s`
# auto_restart_check_interval = "90s"

# Set the file cache directory. Format: "/path/to/cache?reading_max=1000&writing_max=200&cache_max=100&cache_file_max_size=40960"
# - reading_max: maximum number of concurrent read operations
# - writing_max: maximum number of concurrent write operations
# - cache_max: maximum number of tinyufo items for file cache
# - cache_file_max_size: maximum size of file for tinyufo, if file size is greater than this value, it will be cached in file not tinyufo.
# cache_directory = ""

# Maximum size of the in-memory cache for file caching. This limit applies when no cache_directory is set.
# Increasing this value will allow more files to be cached but consume more memory. It is not limited for file cache.
# Default `100MB`
# cache_max_size = "100MB"

[upstreams.charts]
# Upstream address list
# Format: "address:port [weight]"
# weight is optional, default is 1
# Example: "127.0.0.1:5000" has weight 1
#          "127.0.0.1:5001 10" has weight 10
addrs = ["127.0.0.1:5000", "127.0.0.1:5001 10"]

# Service discovery, support "dns", "docker", "common", "transparent".
# Default `none`
# discovery = ""

# How often to refresh the list of upstream servers when using service discovery.
# Format: duration string (e.g. "1m", "30s", "1h").
# It should be set when discovery is `dns` or `docker`. Default `none`
# update_frequency = "1m"


# Loadbalancer selection algorithm
# Supported algorithms:
# - "hash:url" - Consistent hashing based on request URL
# - "hash:ip" - Consistent hashing based on request IP
# - "hash:header:header_name" - Consistent hashing based on specified header
# - "hash:cookie:cookie_name" - Consistent hashing based on specified cookie
# - "hash:query:query_name" - Consistent hashing based on specified query
# - "hash" - Consistent hashing based on request path
# - "round_robin" - Simple round robin distribution
# Default `round_robin`
# algo = "hash:cookie:sid"

# Server Name Indication (SNI) for HTTPS upstream connections. 
# Specify the hostname to be used in the TLS handshake when connecting to upstream HTTPS servers.
# Example: "example.com"
# It should be set if the upstream server is https. Default `none`
# sni = ""

# Whether to check if upstream' server cert is valid and validated
# Default `true`
# verify_cert = true

# Upstream http health check, if not set, tcp health check will be used.
# - http: `http://upstreamname/path?connection_timeout=3s&read_timeout=3s&check_frequency=10s&success=1&failure=2&reuse=true`
# - tcp: `tcp://upstreamname?connection_timeout=3s&read_timeout=3s&check_frequency=10s&success=1&failure=2&reuse=true`
# - grpc: `grpc://upstreamname/path?connection_timeout=3s&read_timeout=3s&check_frequency=10s&success=1&failure=2&reuse=true&tls=true&service=pingap`
# The default parameters are:
# - connection_timeout: 3s
# - read_timeout: 3s
# - check_frequency: 10s
# - success: 1
# - failure: 2
# - reuse: false
# - tls: false
# health_check = "http://charts/ping?connection_timeout=3s&read_timeout=3s"

# When set to true, forces upstream connections to only use IPv4 addresses,
# ignoring any IPv6 addresses even if available. Useful for environments
# where IPv6 connectivity is problematic or not supported. 
# Default `false`
# ipv4_only = false

# Enable distributed tracing for upstream connections. When enabled, the connected connection will be traced.
# Default `false`
# enable_tracer = false

# Application-Layer Protocol Negotiation (ALPN) protocol for upstream connections.
# Supported values: "h1" (HTTP/1.1), "h2" (HTTP/2), "h2h1" (HTTP/2 over TLS with HTTP/1.1 fallback).
# This setting determines which protocol version to use when communicating with upstream servers.
# Default `h1`
# alpn = "h1"

# How long to wait before giving up establishing a TCP connection
# Default `none`
connection_timeout = "10s"

# How long to wait before giving up establishing a connection including TLS handshake time
# Default `none`
total_connection_timeout = "30s"

# How long to wait before each individual `read()` from upstream. The timer is reset after each `read()`
# Default `none`
read_timeout = "10s"

# How long to wait before closing a idle connection waiting for connection reuse
# Default `none`
idle_timeout = "2m"

# How long to wait before a `write()` to upstream finishes
# Default `none`
write_timeout = "10s"

# TCP keepalive idle time:
# - Duration a connection must be idle before TCP starts sending keepalive probes
# - Helps detect dead peer connections while minimizing unnecessary network traffic
# - Default: none (system default)
tcp_idle = "2m"

# TCP keepalive probe interval:
# - Time between consecutive keepalive probe attempts
# - Lower values detect connection failures faster but increase network overhead
# - Default: none (system default)
tcp_interval = "1m"

# TCP keepalive probe count:
# - Maximum number of failed probes before declaring connection dead
# - Higher values increase tolerance to temporary network issues
# - Lower values detect failures faster but may cause premature disconnections
# - Default: none (system default)
tcp_probe_count = 9

# TCP receive buffer size:
# - Size of the kernel buffer for receiving TCP packets
# - Larger values can improve throughput for high-bandwidth connections
# - Smaller values conserve memory but may impact performance
# - Default: none (system default)
tcp_recv_buf = "4kb"

# TCP Fast Open:
# - Enables TCP Fast Open (TFO) protocol for faster connection establishment
# - Reduces latency by sending data in the initial SYN packet
# - Requires kernel support and may not work with all clients/networks
# - Default: none (system default)
tcp_fast_open = true


[upstreams.diving]
addrs = ["127.0.0.1:5001"]


[locations.lo]
# upstream of location (default none)
upstream = "charts"

# location match path (default none)
path = "/"

# location match host, multiple domain names are separated by commas (default none)
host = ""

# set headers to request (default none)
includes = ["proxySetHeader"]

# add headers to request (default none)
proxy_add_headers = ["name:value"]

# rewrite the request path, e.g. `^/api/ /` replace prefix /api/ to / (default none)
rewrite = ""

# the weigh of location (default none)
weight = 1024

# client max body size limit (default none)
client_max_body_size = "1mb"

# plugin list for location
plugins = ["pingap:requestId", "stats"]

[servers.test]
# server linsten address, multiple addresses are separated by commas (default none)
addr = "0.0.0.0:6188"

# access log format (default none)
access_log = "tiny"

# the locations for server
locations = ["lo"]

# the threads count for server (default 1)
threads = 1

# tls certificate(pem) (default none)
# tls_cert = ""
# tle private key(pem) (default none)
# tls_key = ""

# the list of supported ciphers for protocols before TLSv1.3 (default none)
tls_cipher_list = ""

# the list of supported ciphers for protocols TLSv1.3 (default none)
tls_ciphersuites = ""

# tls min version (default none)
tls_min_version = ""

# tls max version (default none)
tls_max_version = ""

# get domain certificates from let's encrypt (default none)
# lets_encrypt = ""

# the file for save certificate (default none)
certificate_file = ""

# using global certificates (default false)
global_certificates = false

# enable http2 (default false)
enabled_h2 = false

# the time a connection needs to be idle before TCP
# begins sending out keep-alive probes (default none)
tcp_idle = "2m"

# duration between TCP keep-alive probes (default none)
tcp_interval = "1m"

# the maximum number of TCP keep-alive probes to send
# before giving up and killing the connection
tcp_probe_count = 9

# enable TCP fast open and set the backlog size of it (default none)
tcp_fast_open = 10

# enable prometheus metrics, it can be a push gateway url or pull metrics path (default none)
prometheus_metrics = ""


[plugins.stats]
value = "/stats"
category = "stats"

[storages.authToken]
category = "secret"
secret = "123123"
value = "PLpKJqvfkjTcYTDpauJf+2JnEayP+bm+0Oe60Jk="

[storages.proxySetHeader]
category = "config"
value = 'proxy_set_headers = ["name:value"]'