$ trillium help
The trillium.rs cli
Usage: trillium <COMMAND>
Commands:
serve Static file server and reverse proxy
client Make http requests using the trillium client
bench Generate http load and report latency/throughput statistics
proxy Run a http proxy
help Print this message or the help of the given subcommand(s)
Options:
-h, --help Print help
-V, --version Print version
HTTP Client
$ trillium help client
Make http requests using the trillium client
Usage: trillium client [OPTIONS] <METHOD> <URL>
Arguments:
<METHOD>
<URL>
Options:
-f, --file <FILE>
provide a file system path to a file to use as the request body
alternatively, you can use an operating system pipe to pass a file in
three equivalent examples:
trillium client post http://httpbin.org/anything -f ./body.json
trillium client post http://httpbin.org/anything < ./body.json
cat ./body.json | trillium client post http://httpbin.org/anything
-o, --output-file [<OUTPUT_FILE>]
write the body to a file
-b, --body <BODY>
provide a request body on the command line
example:
trillium client post http://httpbin.org/post -b '{"hello": "world"}'
-H, --headers <HEADERS>
provide headers in the form -h KEY1=VALUE1 KEY2=VALUE2
example:
trillium client get http://httpbin.org/headers -H Accept=application/json Authorization="Basic u:p"
-t, --tls <TLS>
tls implementation
requests to https:// urls with `none` will fail
[default: rustls]
[possible values: none, rustls]
--http-version <HTTP_VERSION>
http version
Possible values:
- 0.9: HTTP/0.9
- 1.0: HTTP/1.0
- 1.1: HTTP/1.1
- 2: HTTP/2
- 3: HTTP/3
[default: 1.1]
-v, --verbose...
Increase logging verbosity
-q, --quiet...
Decrease logging verbosity
-h, --help
Print help (see a summary with '-h')
Proxy (reverse and forward)
$ trillium help proxy
Run a http proxy
Usage: trillium proxy [OPTIONS] [UPSTREAM]...
Arguments:
[UPSTREAM]...
[env: UPSTREAM=]
Options:
-s, --strategy <STRATEGY>
[env: STRATEGY=]
[default: round-robin]
[possible values: round-robin, connection-counting, random, forward]
-o, --host <HOST>
Local host or ip to listen on
[env: HOST=]
[default: localhost]
-p, --port <PORT>
Local port to listen on
[env: PORT=]
[default: 8080]
--cert <CERT>
Path to a tls certificate file
This will fail unless key is also provided. Providing both cert and key enables tls.
Example: `--cert ./cert.pem --key ./key.pem` For development, try using mkcert or rcgen
[env: CERT=]
--key <KEY>
The path to a tls key file
This will fail unless cert is also provided. Providing both cert and key enables tls.
Example: `--cert ./cert.pem --key ./key.pem` For development, try using mkcert or rcgen
[env: KEY=]
--tls <TLS>
[env: TLS=]
[default: rustls]
[possible values: none, rustls]
-c, --client-tls <CLIENT_TLS>
tls implementation
required if the upstream url is https.
[default: rustls]
[possible values: none, rustls]
-v, --verbose...
Increase logging verbosity
-q, --quiet...
Decrease logging verbosity
-h, --help
Print help (see a summary with '-h')
Static file server
$ trillium help serve
Static file server and reverse proxy
Usage: trillium serve [OPTIONS] [ROOT]
Arguments:
[ROOT]
Filesystem path to serve
Defaults to the current working directory
[default: {your working directory}]
Options:
-o, --host <HOST>
Local host or ip to listen on
[env: HOST=]
[default: localhost]
-p, --port <PORT>
Local port to listen on
[env: PORT=]
[default: 8080]
--cert <CERT>
Path to a tls certificate file
This will fail unless key is also provided. Providing both cert and key enables tls.
Example: `--cert ./cert.pem --key ./key.pem` For development, try using mkcert or rcgen
[env: CERT=]
--key <KEY>
The path to a tls key file
This will fail unless cert is also provided. Providing both cert and key enables tls.
Example: `--cert ./cert.pem --key ./key.pem` For development, try using mkcert or rcgen
[env: KEY=]
--tls <TLS>
[env: TLS=]
[default: rustls]
[possible values: none, rustls]
-f, --forward <FORWARD>
Host to forward (reverse proxy) not-found requests to
This forwards any request that would otherwise be a 404 Not Found to the specified listener spec.
Examples: `--forward localhost:8081` `--forward http://localhost:8081` `--forward https://localhost:8081`
Note: http+unix:// schemes are not yet supported
[env: FORWARD=]
-i, --index <INDEX>
[env: INDEX=]
-v, --verbose...
Increase logging verbosity
-q, --quiet...
Decrease logging verbosity
-h, --help
Print help (see a summary with '-h')
Load generation / testing
$ trillium help bench
Generate http load and report latency/throughput statistics
Usage: trillium bench [OPTIONS] <URL>
Arguments:
<URL>
target URL to benchmark
Options:
-m, --method <METHOD>
HTTP method
[default: GET]
-c, --connections <CONNECTIONS>
number of concurrent connections (closed-loop) or initial pool size (open-loop)
[default: 50]
-d, --duration <DURATION>
total test duration (e.g. 10s, 1m, 30s500ms)
mutually exclusive with --requests; default is 10s when neither is specified.
-n, --requests <REQUESTS>
total number of requests to send (closed-loop only)
-r, --rate <RATE>
target rate in requests per second; switches to open-loop scheduling
--pacing <PACING>
open-loop pacing strategy
Possible values:
- uniform: fixed inter-arrival interval = 1 / rate
- poisson: exponentially-distributed inter-arrival times with mean 1 / rate
[default: uniform]
--max-concurrency <MAX_CONCURRENCY>
in open-loop mode, hard cap on simultaneous in-flight requests
scheduled tickets that would exceed this cap are dropped and counted as saturation.
-w, --warmup <WARMUP>
discard statistics collected during this initial period
--timeout <TIMEOUT>
per-request timeout
-H, --headers <HEADERS>
request headers in KEY=VALUE form, repeatable
-f, --file <FILE>
path to a file to use as the request body
-b, --body <BODY>
inline request body string
--body-size <BODY_SIZE>
synthesize a zero-filled request body of the given size (e.g. 4kb, 1mb)
--http-version <HTTP_VERSION>
http version
Possible values:
- 0.9: HTTP/0.9
- 1.0: HTTP/1.0
- 1.1: HTTP/1.1
- 2: HTTP/2
- 3: HTTP/3
[default: 1.1]
-t, --tls <TLS>
tls implementation
[default: rustls]
[possible values: none, rustls]
--no-keepalive
disable http/1.1 connection reuse
--response-buffer-len <RESPONSE_BUFFER_LEN>
HttpConfig: initial response buffer length (bytes)
--response-buffer-max-len <RESPONSE_BUFFER_MAX_LEN>
HttpConfig: maximum response buffer length under backpressure (bytes)
--head-max-len <HEAD_MAX_LEN>
HttpConfig: max length of the http head (request line + headers)
--copy-loops-per-yield <COPY_LOOPS_PER_YIELD>
HttpConfig: cooperative yield interval for the copy loop
--received-body-max-len <RECEIVED_BODY_MAX_LEN>
HttpConfig: maximum allowed received body length (bytes)
--json
emit the final report as JSON to stdout
--csv <CSV>
write per-request timing data as CSV to this path
--no-progress
suppress the live progress display even when stdout is a tty
-v, --verbose...
Increase logging verbosity
-q, --quiet...
Decrease logging verbosity
-h, --help
Print help (see a summary with '-h')
