fortio

README

Fortio

Fortio (Φορτίο) started as, and is, Istio's load testing tool and now graduated to be its own project.

Fortio is also used by, among others, Meshery

Fortio runs at a specified query per second (qps) and records an histogram of execution time and calculates percentiles (e.g. p99 ie the response time such as 99% of the requests take less than that number (in seconds, SI unit)). It can run for a set duration, for a fixed number of calls, or until interrupted (at a constant target QPS, or max speed/load per connection/thread).

The name fortio comes from greek φορτίο which means load/burden.

Fortio is a fast, small (3Mb docker image, minimal dependencies), reusable, embeddable go library as well as a command line tool and server process, the server includes a simple web UI and REST API to trigger run and see graphical representation of the results (both a single latency graph and a multiple results comparative min, max, avg, qps and percentiles graphs).

Fortio also includes a set of server side features (similar to httpbin) to help debugging and testing: request echo back including headers, adding latency or error codes with a probability distribution, tcp echoing, tcp proxying, http fan out/scatter and gather proxy server, GRPC echo/health in addition to http, etc...

Fortio is quite mature and very stable with no known major bugs (lots of possible improvements if you want to contribute though!), and when bugs are found they are fixed quickly, so after 1 year of development and 42 incremental releases, we reached 1.0 in June 2018.

Fortio components can be used a library even for unrelated projects, for instance the log, stats, or fhttp utilities both client and server. As well as the newly integrated Dynamic Flags support (greatly inspired/imported initially from https://github.com/mwitkow/go-flagz)

Installation

We publish a multi architecture docker image (linux/amd64, linux/arm64, linux/ppc64le, linux/s390x) fortio/fortio

For instance:

docker run -p 8080:8080 -p 8079:8079 fortio/fortio server & # For the server
docker run fortio/fortio load http://www.google.com/ # For a test run

You can install from source:

1. Install go (golang 1.18 or later)
2. go install fortio.org/fortio@latest
3. you can now run fortio (from your gopath bin/ directory, usually ~/go/bin)

The releases page has binaries for many OS/architecture combinations (see assets).

curl -L https://github.com/fortio/fortio/releases/download/v1.31.0/fortio-linux_amd64-1.31.0.tgz \
 | sudo tar -C / -xvzpf -
# or the debian package
wget https://github.com/fortio/fortio/releases/download/v1.31.0/fortio_1.31.0_amd64.deb
dpkg -i fortio_1.31.0_amd64.deb
# or the rpm
rpm -i https://github.com/fortio/fortio/releases/download/v1.31.0/fortio-1.31.0-1.x86_64.rpm
# and more, see assets in release page

On a MacOS you can also install Fortio using Homebrew:

brew install fortio

On Windows, download https://github.com/fortio/fortio/releases/download/v1.31.0/fortio_win_1.31.0.zip and extract fortio.exe to any location, then using the Windows Command Prompt:

fortio.exe server

(at the prompt, allow the windows firewall to let connections in)

Once fortio server is running, you can visit its web UI at http://localhost:8080/fortio/

You can get a preview of the reporting/graphing UI at https://demo.fortio.org/.

Command line arguments

Fortio can be an http or grpc load generator, gathering statistics using the load subcommand, or start simple http and grpc ping servers, as well as a basic web UI, result graphing, tcp/udp echo, proxies, https redirector, with the server command or issue grpc ping messages using the grpcping command. It can also fetch a single URL's for debugging when using the curl command (or the -curl flag to the load command). Likewise you can establish a single TCP (or unix domain or UDP (use udp:// prefix)) connection using the nc command (like the standalone netcat package). You can run just the redirector with redirect or just the tcp echo with tcp-echo. If you saved JSON results (using the web UI or directly from the command line), you can browse and graph those results using the report command. The version command will print version and build information, fortio version -s just the version. Lastly, you can learn which flags are available using help command.

Most important flags for http load generation:

Flag	Description, example
-qps rate	Queries Per Seconds or 0 for no wait/max qps
-nocatchup	Do not try to reach the target qps by going faster when the service falls behind and then recovers. Makes QPS an absolute ceiling even if the service has some spikes in latency, fortio will not compensate (but also won't stress the target more than the set qps). Recommended to use jointly with -uniform.
-c connections	Number of parallel simultaneous connections (and matching go routine)
-t duration	How long to run the test (for instance -t 30m for 30 minutes) or 0 to run until ^C, example (default 5s)
-n numcalls	Run for exactly this number of calls instead of duration. Default (0) is to use duration (-t).
-payload str or -payload-file fname	Switch to using POST with the given payload (see also -payload-size for random payload)
-uniform	Spread the calls in time across threads for a more uniform call distribution. Works even better in conjunction with -nocatchup.
-r resolution	Resolution of the histogram lowest buckets in seconds (default 0.001 i.e 1ms), use 1/10th of your expected typical latency
-H "header: value"	Can be specified multiple times to add headers (including Host:)
-a	Automatically save JSON result with filename based on labels and timestamp
-json filename	Filename or - for stdout to output json result (relative to -data-dir by default, should end with .json if you want fortio report to show them; using -a is typicallly a better option)
-labels "l1 l2 ..."	Additional config data/labels to add to the resulting JSON, defaults to target URL and hostname

You can switch from http GET queries to POST by setting -content-type or passing one of the -payload-* option.

Full list of command line flags (fortio help):

Φορτίο 1.31.0 usage:
where command is one of: load (load testing), server (starts ui, http-echo,
 redirect, proxies, tcp-echo and grpc ping servers), tcp-echo (only the tcp-echo
 server), report (report only UI server), redirect (only the redirect server),
 proxies (only the -M and -P configured proxies), grpcping (grpc client),
 or curl (single URL debug), or nc (single tcp or udp:// connection),
 or version (prints the version).
where target is a url (http load tests) or host:port (grpc health test).
flags are:
  -H header
        Additional header(s)
  -L    Follow redirects (implies -std-client) - do not use for load test
  -M value
        Http multi proxy to run, e.g -M "localport1 baseDestURL1 baseDestURL2"
-M ...
  -P value
        Tcp proxies to run, e.g -P "localport1 dest_host1:dest_port1" -P
"[::1]:0 www.google.com:443" ...
  -a    Automatically save JSON result with filename based on labels & timestamp
  -abort-on code
        Http code that if encountered aborts the run. e.g. 503 or -1 for socket
errors.
  -access-log-file path
        file path to log all requests to. Maybe have performance impacts
  -access-log-format format
        format for access log. Supported values: [json, influx] (default "json")
  -allow-initial-errors
        Allow and don't abort on initial warmup errors
  -base-url URL
        base URL used as prefix for data/index.tsv generation. (when empty, the
url from the first request is used)
  -c int
        Number of connections/goroutine/threads (default 4)
  -cacert Path
        Path to a custom CA certificate file to be used for the TLS client
connections, if empty, use https:// prefix for standard internet/system CAs
  -calc-qps
        Calculate the qps based on number of requests (-n) and duration (-t)
  -cert Path
        Path to the certificate file to be used for client or server TLS
  -compression
        Enable http compression
  -config path
        Config directory path to watch for changes of dynamic flags (empty for
no watch)
  -content-type string
        Sets http content type. Setting this value switches the request method
from GET to POST.
  -curl
        Just fetch the content once
  -curl-stdout-headers
        Restore pre 1.22 behavior where http headers of the fast client are
output to stdout in curl mode. now stderr by default.
  -data-dir Directory
        Directory where JSON results are stored/read (default ".")
  -dns-method method
        When a name resolves to multiple ip, which method to pick: cached-rr
for cached round robin, rnd for random, first for first answer (pre 1.30
behavior), rr for round robin. (default cached-rr)
  -echo-debug-path URI
        http echo server URI for debug, empty turns off that part (more secure)
(default "/debug")
  -echo-server-default-params value
        Default parameters/querystring to use if there isn't one provided
explicitly. E.g "status=404&delay=3s"
  -gomaxprocs int
        Setting for runtime.GOMAXPROCS, <1 doesn't change the default
  -grpc
        Use GRPC (health check by default, add -ping for ping) for load testing
  -grpc-max-streams uint
        MaxConcurrentStreams for the grpc server. Default (0) is to leave the
option unset.
  -grpc-ping-delay duration
        grpc ping delay in response
  -grpc-port port
        grpc server port. Can be in the form of host:port, ip:port or port or
/unix/domain/path or "disabled" to not start the grpc server. (default "8079")
  -h    Print usage/help on stdout
  -halfclose
        When not keepalive, whether to half close the connection (only for fast
http)
  -health
        grpc ping client mode: use health instead of ping
  -healthservice string
        which service string to pass to health check
  -http-port port
        http echo server port. Can be in the form of host:port, ip:port, port
or /unix/domain/path or "disabled". (default "8080")
  -http1.0
        Use http1.0 (instead of http 1.1)
  -httpbufferkb kbytes
        Size of the buffer (max data size) for the optimized http client in
kbytes (default 128)
  -httpccch
        Check for Connection: Close Header
  -https-insecure
        Long form of the -k flag
  -jitter
        set to true to de-synchronize parallel clients' by 10%
  -json path
        Json output to provided file path or '-' for stdout (empty = no json
output, unless -a is used)
  -k    Do not verify certs in https/tls/grpc connections
  -keepalive
        Keep connection alive (only for fast http 1.1) (default true)
  -key Path
        Path to the key file matching the -cert
  -labels string
        Additional config data/labels to add to the resulting JSON, defaults to
target URL and hostname
  -log-errors
        Log http non 2xx/418 error codes as they occur (default true)
  -logcaller
        Logs filename and line number of callers to log (default true)
  -loglevel value
        loglevel, one of [Debug Verbose Info Warning Error Critical Fatal]
(default Info)
  -logprefix string
        Prefix to log lines before logged messages (default "> ")
  -max-echo-delay value
        Maximum sleep time for delay= echo server parameter. dynamic flag.
(default 1.5s)
  -maxpayloadsizekb Kbytes
        MaxPayloadSize is the maximum size of payload to be generated by the
EchoHandler size= argument. In Kbytes. (default 256)
  -multi-mirror-origin
        Mirror the request url to the target for multi proxies (-M) (default
true)
  -multi-serial-mode
        Multi server (-M) requests one at a time instead of parallel mode
  -n int
        Run for exactly this number of calls instead of duration. Default (0)
is to use duration (-t). Default is 1 when used as grpc ping count.
  -nc-dont-stop-on-eof
        in netcat (nc) mode, don't abort as soon as remote side closes
  -nocatchup
        set to exact fixed qps and prevent fortio from trying to catchup when
the target fails to keep up temporarily
  -offset duration
        Offset of the histogram data
  -p string
        List of pXX to calculate (default "50,75,90,99,99.9")
  -payload string
        Payload string to send along
  -payload-file path
        File path to be use as payload (POST for http), replaces -payload when
set.
  -payload-size int
        Additional random payload size, replaces -payload when set > 0, must be
smaller than -maxpayloadsizekb. Setting this switches http to POST.
  -ping
        grpc load test: use ping instead of health
  -profile file
        write .cpu and .mem profiles to file
  -proxy-all-headers
        Determines if only tracing or all headers (and cookies) are copied from
request on the fetch2 ui/server endpoint (default true)
  -qps float
        Queries Per Seconds or 0 for no wait/max qps (default 8)
  -quiet
        Quiet mode: sets the loglevel to Error and reduces the output.
  -r float
        Resolution of the histogram lowest buckets in seconds (default 0.001)
  -redirect-port port
        Redirect all incoming traffic to https URL (need ingress to work
properly). Can be in the form of host:port, ip:port, port or "disabled" to
disable the feature. (default "8081")
  -resolve IP
        Resolve host name to this IP
  -resolve-ip-type type
        Resolve type: ip4 for ipv4, ip6 for ipv6 only, use ip for both (default
ip4)
  -runid int
        Optional RunID to add to json result and auto save filename, to match
server mode
  -s int
        Number of streams per grpc connection (default 1)
  -sequential-warmup
        http(s) runner warmup done in parallel instead of sequentially. When
set, restores pre 1.21 behavior
  -server-idle-timeout value
        Default IdleTimeout for servers (default 30s)
  -static-dir path
        Deprecated/unused path.
  -stdclient
        Use the slower net/http standard client (slower but supports h2)
  -sync URL
        index.tsv or s3/gcs bucket xml URL to fetch at startup for server modes.
  -sync-interval duration
        Refresh the url every given interval (default, no refresh)
  -t duration
        How long to run the test or 0 to run until ^C (default 5s)
  -tcp-port port
        tcp echo server port. Can be in the form of host:port, ip:port, port or
/unix/domain/path or "disabled". (default "8078")
  -timeout duration
        Connection and read timeout value (for http) (default 3s)
  -udp-async
        if true, udp echo server will use separate go routine to reply
  -udp-port port
        udp echo server port. Can be in the form of host:port, ip:port, port or
"disabled". (default "8078")
  -udp-timeout duration
        Udp timeout (default 750ms)
  -ui-path URI
        http server URI for UI, empty turns off that part (more secure)
(default "/fortio/")
  -uniform
        set to true to de-synchronize parallel clients' requests uniformly
  -unix-socket path
        Unix domain socket path to use for physical connection
  -user user:password
        User credentials for basic authentication (for http). Input data format
should be user:password

See also the FAQ entry about fortio flags for best results

Server URLs and features

Fortio server has the following feature for the http listening on 8080 (all paths and ports are configurable through flags above):

• A simple echo server which will echo back posted data (for any path not mentioned below).

  For instance curl -d abcdef http://localhost:8080/ returns abcdef back. It supports the following optional query argument parameters:

Parameter	Usage, example
delay	duration to delay the response by. Can be a single value or a comma separated list of probabilities, e.g delay=150us:10,2ms:5,0.5s:1 for 10% of chance of a 150 us delay, 5% of a 2ms delay and 1% of a 1/2 second delay
status	http status to return instead of 200. Can be a single value or a comma separated list of probabilities, e.g status=404:10,503:5,429:1 for 10% of chance of a 404 status, 5% of a 503 status and 1% of a 429 status
size	size of the payload to reply instead of echoing input. Also works as probabilities list. size=1024:10,512:5 10% of response will be 1k and 5% will be 512 bytes payload and the rest defaults to echoing back.
close	close the socket after answering e.g close=true to close after all requests or close=5.3 to close after approximately 5.3% of requests
header	header(s) to add to the reply e.g. &header=Foo:Bar&header=X:Y
gzip	If Accept-Encoding: gzip is passed in headers by the caller/client; and gzip=true is in the query args, all response will be gzipped; or if gzip=42.7 is passed, approximately 42.7% will

You can set a default value for all these by passing -echo-server-default-params to the server command line, for instance: fortio server -echo-server-default-params="delay=0.5s:50,1s:40&status=418" will make the server respond with http 418 and a delay of either 0.5s half of the time, 1s 40% and no delay in 10% of the calls; unless any ? query args is passed by the client. Note that the quotes (") are for the shell to escape the ampersand (&) but should not be put in a yaml nor the dynamicflag url for instance.

• /debug will echo back the request in plain text for human debugging.

• /fortio/ A UI to

  • Run/Trigger tests and graph the results.
  • A UI to browse saved results and single graph or multi graph them (comparative graph of min,avg, median, p75, p99, p99.9 and max).
  • Proxy/fetch other URLs
  • /fortio/data/index.tsv an tab separated value file conforming to Google cloud storage URL list data transfer format so you can export/backup local results to the cloud.
  • Download/sync peer to peer JSON results files from other Fortio servers (using their index.tsv URLs)
  • Download/sync from an Amazon S3 or Google Cloud compatible bucket listings XML URLs

• API to trigger and cancel runs from the running server (like the form ui but more directly and with async=on option)

  • /fortio/rest/run starts a run; the arguments are either from the command line or from POSTed JSON; jsonPath can be provided to look for in a subset of the json object, for instance jsonPath=metadata allows to use the flagger webhook meta data for fortio run parameters (see Remote Triggered load test section below).
  • /fortio/rest/stop stops all current run or by run id.

The report mode is a readonly subset of the above directly on /.

There is also the GRPC health and ping servers, as well as the http->https redirector.

Example use and output

Start the internal servers

$ fortio server &
Fortio X.Y.Z tcp-echo server listening on tcp [::]:8078
Fortio X.Y.Z udp-echo server listening on udp [::]:8078
Fortio X.Y.Z grpc 'ping' server listening on tcp [::]:8079
Fortio X.Y.Z https redirector server listening on tcp [::]:8081
Fortio X.Y.Z http-echo server listening on tcp [::]:8080
Data directory is /Users/ldemailly/dev/fortio
UI started - visit:
http://localhost:8080/fortio/
(or any host/ip reachable on this server)
14:11:05 I fortio_main.go:285> Note: not using dynamic flag watching (use -config to set watch directory)
14:11:05 I fortio_main.go:293> All fortio X.Y.Z unknown goM.m.p servers started!

Sample of the graphing UI

With the 2 histograms - total and errors overlayed

Change the port / binding address

By default, Fortio's web/echo servers listen on port 8080 on all interfaces. Use the -http-port flag to change this behavior:

$ fortio server -http-port 10.10.10.10:8088
UI starting - visit:
http://10.10.10.10:8088/fortio/
Https redirector running on :8081
Fortio X.Y.Z grpc ping server listening on port :8079
Fortio X.Y.Z echo server listening on port 10.10.10.10:8088

Unix domain sockets

You can use unix domain socket for any server/client:

$ fortio server --http-port /tmp/fortio-uds-http &
Fortio X.Y.Z grpc 'ping' server listening on [::]:8079
Fortio X.Y.Z https redirector server listening on [::]:8081
Fortio X.Y.Z echo server listening on /tmp/fortio-uds-http
UI started - visit:
fortio curl -unix-socket=/tmp/fortio-uds-http http://localhost/fortio/
14:58:45 I fortio_main.go:217> All fortio X.Y.Z unknown goM.m.p servers started!
$ fortio curl -unix-socket=/tmp/fortio-uds-http http://foo.bar/debug
15:00:48 I http_client.go:428> Using unix domain socket /tmp/fortio-uds-http instead of foo.bar http
HTTP/1.1 200 OK
Content-Type: text/plain; charset=UTF-8
Date: Wed, 08 Aug 2018 22:00:48 GMT
Content-Length: 231

Φορτίο version X.Y.Z unknown goM.m.p echo debug server up for 2m3.4s on ldemailly-macbookpro - request from

GET /debug HTTP/1.1

headers:

Host: foo.bar
User-Agent: fortio.org/fortio-X.Y.Z

body:

TCP

Start the echo-server alone and run a load (use tcp:// prefix for the load test to be for tcp echo server)

$ fortio tcp-echo &
Fortio X.Y.Z tcp-echo TCP server listening on [::]:8078
19:45:30 I fortio_main.go:238> All fortio X.Y.Z release goM.m.p servers started!
$ fortio load -qps -1 -n 100000 tcp://localhost:8078
Fortio X.Y.Z running at -1 queries per second, 16->16 procs, for 100000 calls: tcp://localhost:8078
20:01:31 I tcprunner.go:218> Starting tcp test for tcp://localhost:8078 with 4 threads at -1.0 qps
Starting at max qps with 4 thread(s) [gomax 16] for exactly 100000 calls (25000 per thread + 0)
20:01:32 I periodic.go:558> T003 ended after 1.240585427s : 25000 calls. qps=20151.77629520873
20:01:32 I periodic.go:558> T002 ended after 1.241141084s : 25000 calls. qps=20142.75437521493
20:01:32 I periodic.go:558> T001 ended after 1.242066385s : 25000 calls. qps=20127.7486468648
20:01:32 I periodic.go:558> T000 ended after 1.24227731s : 25000 calls. qps=20124.331176909283
Ended after 1.242312567s : 100000 calls. qps=80495
Aggregated Function Time : count 100000 avg 4.9404876e-05 +/- 1.145e-05 min 2.7697e-05 max 0.000887051 sum 4.94048763
# range, mid point, percentile, count
>= 2.7697e-05 <= 0.000887051 , 0.000457374 , 100.00, 100000
# target 50% 0.00045737
# target 75% 0.00067221
# target 90% 0.000801115
# target 99% 0.000878457
# target 99.9% 0.000886192
Sockets used: 4 (for perfect no error run, would be 4)
Total Bytes sent: 2400000, received: 2400000
tcp OK : 100000 (100.0 %)
All done 100000 calls (plus 0 warmup) 0.049 ms avg, 80495.0 qps

UDP

Start the udp-echo server alone and run a load (use tcp:// prefix for the load test to be for tcp echo server)

$ fortio udp-echo &
Fortio X.Y.Z udp-echo UDP server listening on [::]:8078
21:54:52 I fortio_main.go:273> Note: not using dynamic flag watching (use -config to set watch directory)
21:54:52 I fortio_main.go:281> All fortio X.Y.Z release goM.m.p servers started!
$ fortio load -qps -1 -n 100000 udp://localhost:8078/
Fortio X.Y.Z running at -1 queries per second, 16->16 procs, for 100000 calls: udp://localhost:8078/
21:56:48 I udprunner.go:222> Starting udp test for udp://localhost:8078/ with 4 threads at -1.0 qps
Starting at max qps with 4 thread(s) [gomax 16] for exactly 100000 calls (25000 per thread + 0)
21:56:49 I periodic.go:558> T003 ended after 969.635695ms : 25000 calls. qps=25782.879208051432
21:56:49 I periodic.go:558> T000 ended after 969.906228ms : 25000 calls. qps=25775.687667818544
21:56:49 I periodic.go:558> T002 ended after 970.543935ms : 25000 calls. qps=25758.751457243405
21:56:49 I periodic.go:558> T001 ended after 970.737665ms : 25000 calls. qps=25753.610786287973
Ended after 970.755702ms : 100000 calls. qps=1.0301e+05
Aggregated Function Time : count 100000 avg 3.8532238e-05 +/- 1.7e-05 min 2.0053e-05 max 0.000881827 sum 3.85322376
# range, mid point, percentile, count
>= 2.0053e-05 <= 0.000881827 , 0.00045094 , 100.00, 100000
# target 50% 0.000450936
# target 75% 0.000666381
# target 90% 0.000795649
# target 99% 0.000873209
# target 99.9% 0.000880965
Sockets used: 4 (for perfect no error run, would be 4)
Total Bytes sent: 2400000, received: 2400000
udp OK : 100000 (100.0 %)
All done 100000 calls (plus 0 warmup) 0.039 ms avg, 103012.5 qps

GRPC

Simple grpc ping

$ fortio grpcping -n 5 localhost
22:36:55 I pingsrv.go:150> Ping RTT 212000 (avg of 259000, 217000, 160000 ns) clock skew -10500
22:36:55 I pingsrv.go:150> Ping RTT 134333 (avg of 170000, 124000, 109000 ns) clock skew 5000
22:36:55 I pingsrv.go:150> Ping RTT 112000 (avg of 111000, 122000, 103000 ns) clock skew 5000
22:36:55 I pingsrv.go:150> Ping RTT 157000 (avg of 136000, 158000, 177000 ns) clock skew 6000
22:36:55 I pingsrv.go:150> Ping RTT 108333 (avg of 118000, 106000, 101000 ns) clock skew 1000
Clock skew histogram usec : count 5 avg 1.3 +/- 6.145 min -10.5 max 6 sum 6.5
# range, mid point, percentile, count
>= -10.5 <= -10 , -10.25 , 20.00, 1
> 0 <= 2 , 1 , 40.00, 1
> 4 <= 6 , 5 , 100.00, 3
# target 50% 4.33333
RTT histogram usec : count 15 avg 144.73333 +/- 44.48 min 101 max 259 sum 2171
# range, mid point, percentile, count
>= 101 <= 110 , 105.5 , 26.67, 4
> 110 <= 120 , 115 , 40.00, 2
> 120 <= 140 , 130 , 60.00, 3
> 140 <= 160 , 150 , 73.33, 2
> 160 <= 180 , 170 , 86.67, 2
> 200 <= 250 , 225 , 93.33, 1
> 250 <= 259 , 254.5 , 100.00, 1
# target 50% 130

Change the target port for grpc

The value of -grpc-port (default 8079) is used when specifying a hostname or an IP address in grpcping. Add :port to the grpcping destination to change this behavior:

$ fortio grpcping 10.10.10.100:8078 # Connects to gRPC server 10.10.10.100 listening on port 8078
02:29:27 I pingsrv.go:116> Ping RTT 305334 (avg of 342970, 293515, 279517 ns) clock skew -2137
Clock skew histogram usec : count 1 avg -2.137 +/- 0 min -2.137 max -2.137 sum -2.137
# range, mid point, percentile, count
>= -4 < -2 , -3 , 100.00, 1
# target 50% -2.137
RTT histogram usec : count 3 avg 305.334 +/- 27.22 min 279.517 max 342.97 sum 916.002
# range, mid point, percentile, count
>= 250 < 300 , 275 , 66.67, 2
>= 300 < 350 , 325 , 100.00, 1
# target 50% 294.879

grpcping using TLS

• First, start Fortio server with the -cert and -key flags:

/path/to/fortio/server.crt and /path/to/fortio/server.key are paths to the TLS certificate and key that you must provide.

$ fortio server -cert /path/to/fortio/server.crt -key /path/to/fortio/server.key
UI starting - visit:
http://localhost:8080/fortio/
Https redirector running on :8081
Fortio X.Y.Z grpc ping server listening on port :8079
Fortio X.Y.Z echo server listening on port localhost:8080
Using server certificate /path/to/fortio/server.crt to construct TLS credentials
Using server key /path/to/fortio/server.key to construct TLS credentials

• Next, use grpcping with the -cacert flag:

/path/to/fortio/ca.crt is the path to the CA certificate that issued the server certificate for localhost. In our example, the server certificate is /path/to/fortio/server.crt:

$ fortio grpcping -cacert /path/to/fortio/ca.crt localhost
Using server certificate /path/to/fortio/ca.crt to construct TLS credentials
16:00:10 I pingsrv.go:129> Ping RTT 501452 (avg of 595441, 537088, 371828 ns) clock skew 31094
Clock skew histogram usec : count 1 avg 31.094 +/- 0 min 31.094 max 31.094 sum 31.094
# range, mid point, percentile, count
>= 31.094 <= 31.094 , 31.094 , 100.00, 1
# target 50% 31.094
RTT histogram usec : count 3 avg 501.45233 +/- 94.7 min 371.828 max 595.441 sum 1504.357
# range, mid point, percentile, count
>= 371.828 <= 400 , 385.914 , 33.33, 1
> 500 <= 595.441 , 547.721 , 100.00, 2
# target 50% 523.86

GRPC to standard https service

grpcping can connect to a non-Fortio TLS server by prefacing the destination with https://:

$ fortio grpcping https://grpc.fortio.org
13:48:20 I grpcrunner.go:276> stripping https scheme. grpc destination: grpc.fortio.org. grpc port: 443
13:48:26 I pingsrv.go:152> Ping RTT 63101562 (avg of 63577000, 63192688, 62535000 ns) clock skew 32021375
Clock skew histogram usec : count 1 avg 32021.375 +/- 0 min 32021.375 max 32021.375 sum 32021.375
# range, mid point, percentile, count
>= 32021.4 <= 32021.4 , 32021.4 , 100.00, 1
# target 50% 32021.4
RTT histogram usec : count 3 avg 63101.563 +/- 430.2 min 62535 max 63577 sum 189304.688
# range, mid point, percentile, count
>= 62535 <= 63577 , 63056 , 100.00, 3
# target 50% 62795.5

Simple load test

Load (low default qps/threading) test:

$ fortio load http://www.google.com
Fortio X.Y.Z running at 8 queries per second, 8->8 procs, for 5s: http://www.google.com
19:10:33 I httprunner.go:84> Starting http test for http://www.google.com with 4 threads at 8.0 qps
Starting at 8 qps with 4 thread(s) [gomax 8] for 5s : 10 calls each (total 40)
19:10:39 I periodic.go:314> T002 ended after 5.056753279s : 10 calls. qps=1.9775534712220633
19:10:39 I periodic.go:314> T001 ended after 5.058085991s : 10 calls. qps=1.9770324224999916
19:10:39 I periodic.go:314> T000 ended after 5.058796046s : 10 calls. qps=1.9767549252963101
19:10:39 I periodic.go:314> T003 ended after 5.059557593s : 10 calls. qps=1.9764573910247019
Ended after 5.059691387s : 40 calls. qps=7.9056
Sleep times : count 36 avg 0.49175757 +/- 0.007217 min 0.463508712 max 0.502087879 sum 17.7032725
Aggregated Function Time : count 40 avg 0.060587641 +/- 0.006564 min 0.052549016 max 0.089893269 sum 2.42350566
# range, mid point, percentile, count
>= 0.052549 < 0.06 , 0.0562745 , 47.50, 19
>= 0.06 < 0.07 , 0.065 , 92.50, 18
>= 0.07 < 0.08 , 0.075 , 97.50, 2
>= 0.08 <= 0.0898933 , 0.0849466 , 100.00, 1
# target 50% 0.0605556
# target 75% 0.0661111
# target 99% 0.085936
# target 99.9% 0.0894975
Code 200 : 40
Response Header Sizes : count 40 avg 690.475 +/- 15.77 min 592 max 693 sum 27619
Response Body/Total Sizes : count 40 avg 12565.2 +/- 301.9 min 12319 max 13665 sum 502608
All done 40 calls (plus 4 warmup) 60.588 ms avg, 7.9 qps

Remote triggered load test (server mode rest API)

New since 1.18 the server has a fortio/rest/run endpoint similar to what the form UI submit in fortio/ to start a run.

• plus async query arg or json value "on" will make the run asynchronous (returns just the runid of the run instead of waiting for the result)
• plus read all the run configuration from either query args or jsonPath POSTed info
• compatible with flagger and other webhooks
• New in 1.22: use headers json array to send headers (or multiple &H= query args)

Examples:

$ curl -v -d '{"metadata": {"url":"localhost:8080", "c":"1", "n":"1", "async":"on", "save":"on"}}' \
     "localhost:8080/fortio/rest/run?jsonPath=.metadata"
{"started": 3}

makes a 1 connection 1 query run for localhost:8080 url asynchronously and saves results

or minimally:

curl -s -d '{"url":"localhost:8080"}' "localhost:8080/fortio/rest/run" | jq

More complete example:

With sample.json (all values must be strings, even the numbers):

{
    "metadata": {
        "url": "localhost:8080",
        "payload": "foo",
        "qps": "40",
        "c": "2",
        "t": "0.1s",
        "headers": [
            "Foo:Bar",
            "X-Blah: Something else"
        ],
        "save": "on"
    }
}

You can run:

$ fortio curl -stdclient -payload-file sample.json "http://localhost:8080/fortio/rest/run?jsonPath=.metadata" > result.json

which makes requests like this:

POST / HTTP/1.1
Host: localhost:8080
Content-Length: 3
Content-Type: application/octet-stream
Foo: Bar
X-Blah: Something else
X-On-Behalf-Of: [::1]:62629

foo

and you get in result.json

{
  "RunType": "HTTP",
  "Labels": "",
  "StartTime": "2022-03-19T15:34:23.279389-07:00",
  "RequestedQPS": "40",
  "RequestedDuration": "100ms",
  "ActualQPS": 38.44836361217263,
  "ActualDuration": 104035637,
  "NumThreads": 2,
  "Version": "X.Y.Z",
  "DurationHistogram": {
    "Count": 4,
    "Min": 0.00027292,
    "Max": 0.000930407,
    "Sum": 0.002332047,
    "Avg": 0.00058301175,
    "StdDev": 0.00028491034912527755,
    "Data": [
      {
        "Start": 0.00027292,
        "End": 0.000930407,
        "Percent": 100,
        "Count": 4
      }
    ],
    "Percentiles": [
      {
        "Percentile": 50,
        "Value": 0.0004920823333333334
      },
      {
        "Percentile": 75,
        "Value": 0.0007112446666666667
      },
      {
        "Percentile": 90,
        "Value": 0.0008427420666666666
      },
      {
        "Percentile": 99,
        "Value": 0.0009216405066666668
      },
      {
        "Percentile": 99.9,
        "Value": 0.0009295303506666667
      }
    ]
  },
  "Exactly": 0,
  "Jitter": false,
  "Uniform": false,
  "RunID": 7,
  "AccessLoggerInfo": "",
  "RetCodes": {
    "200": 4
  },
  "URL": "http://localhost:8080",
  "NumConnections": 1,
  "Compression": false,
  "DisableFastClient": false,
  "HTTP10": false,
  "DisableKeepAlive": false,
  "AllowHalfClose": false,
  "Insecure": false,
  "FollowRedirects": false,
  "CACert": "",
  "Cert": "",
  "Key": "",
  "Resolve": "",
  "HTTPReqTimeOut": 3000000000,
  "UserCredentials": "",
  "ContentType": "",
  "Payload": "Zm9v",
  "UnixDomainSocket": "",
  "LogErrors": false,
  "ID": 0,
  "SequentialWarmup": false,
  "Sizes": {
    "Count": 4,
    "Min": 118,
    "Max": 118,
    "Sum": 472,
    "Avg": 118,
    "StdDev": 0,
    "Data": [
      {
        "Start": 118,
        "End": 118,
        "Percent": 100,
        "Count": 4
      }
    ],
    "Percentiles": null
  },
  "HeaderSizes": {
    "Count": 4,
    "Min": 115,
    "Max": 115,
    "Sum": 460,
    "Avg": 115,
    "StdDev": 0,
    "Data": [
      {
        "Start": 115,
        "End": 115,
        "Percent": 100,
        "Count": 4
      }
    ],
    "Percentiles": null
  },
  "SocketCount": 2,
  "AbortOn": 0
}

• There is also the fortio/rest/stop endpoint to stop a run by its id or all runs if not specified

GRPC load test

Uses -s to use multiple (h2/grpc) streams per connection (-c), request to hit the fortio ping grpc endpoint with a delay in replies of 0.25s and an extra payload for 10 bytes and auto save the json result:

$ fortio load -a -grpc -ping -grpc-ping-delay 0.25s -payload "01234567890" -c 2 -s 4 https://fortio-stage.istio.io
Fortio X.Y.Z running at 8 queries per second, 8->8 procs, for 5s: https://fortio-stage.istio.io
16:32:56 I grpcrunner.go:139> Starting GRPC Ping Delay=250ms PayloadLength=11 test for https://fortio-stage.istio.io with 4*2 threads at 8.0 qps
16:32:56 I grpcrunner.go:261> stripping https scheme. grpc destination: fortio-stage.istio.io. grpc port: 443
16:32:57 I grpcrunner.go:261> stripping https scheme. grpc destination: fortio-stage.istio.io. grpc port: 443
Starting at 8 qps with 8 thread(s) [gomax 8] for 5s : 5 calls each (total 40)
16:33:04 I periodic.go:533> T005 ended after 5.283227589s : 5 calls. qps=0.9463911814835126
[...]
Ended after 5.28514474s : 40 calls. qps=7.5684
Sleep times : count 32 avg 0.97034752 +/- 0.002338 min 0.967323561 max 0.974838789 sum 31.0511206
Aggregated Function Time : count 40 avg 0.27731944 +/- 0.001606 min 0.2741372 max 0.280604967 sum 11.0927778
# range, mid point, percentile, count
>= 0.274137 <= 0.280605 , 0.277371 , 100.00, 40
# target 50% 0.277288
# target 75% 0.278947
# target 90% 0.279942
# target 99% 0.280539
# target 99.9% 0.280598
Ping SERVING : 40
All done 40 calls (plus 2 warmup) 277.319 ms avg, 7.6 qps
Successfully wrote 1210 bytes of Json data to 2018-04-03-163258_fortio_stage_istio_io_ldemailly_macbookpro.json

And the JSON saved is

{
  "RunType": "GRPC Ping Delay=250ms PayloadLength=11",
  "Labels": "fortio-stage.istio.io , ldemailly-macbookpro",
  "StartTime": "2018-04-03T16:32:58.895472681-07:00",
  "RequestedQPS": "8",
  "RequestedDuration": "5s",
  "ActualQPS": 7.568383075162479,
  "ActualDuration": 5285144740,
  "NumThreads": 8,
  "Version": "0.9.0",
  "DurationHistogram": {
    "Count": 40,
    "Min": 0.2741372,
    "Max": 0.280604967,
    "Sum": 11.092777797,
    "Avg": 0.277319444925,
    "StdDev": 0.0016060870789948905,
    "Data": [
      {
        "Start": 0.2741372,
        "End": 0.280604967,
        "Percent": 100,
        "Count": 40
      }
    ],
    "Percentiles": [
      {
        "Percentile": 50,
        "Value": 0.2772881634102564
      },
      {
        "Percentile": 75,
        "Value": 0.27894656520512817
      },
      {
        "Percentile": 90,
        "Value": 0.2799416062820513
      },
      {
        "Percentile": 99,
        "Value": 0.28053863092820513
      },
      {
        "Percentile": 99.9,
        "Value": 0.2805983333928205
      }
    ]
  },
  "Exactly": 0,
  "RetCodes": {
    "1": 40
  },
  "Destination": "https://fortio-stage.istio.io",
  "Streams": 4,
  "Ping": true
}

• Load test using gRPC and TLS security. First, start Fortio server with the -cert and -key flags:

fortio server -cert /etc/ssl/certs/server.crt -key /etc/ssl/certs/server.key

Next, run the load command with the -cacert flag:

fortio load -cacert /etc/ssl/certs/ca.crt -grpc localhost:8079

Curl like (single request) mode

$ fortio load -curl -H Foo:Bar http://localhost:8080/debug
14:26:26 I http.go:133> Setting regular extra header Foo: Bar
HTTP/1.1 200 OK
Content-Type: text/plain; charset=UTF-8
Date: Mon, 08 Jan 2018 22:26:26 GMT
Content-Length: 230

Φορτίο version X.Y.Z echo debug server up for 39s on ldemailly-macbookpro - request from [::1]:65055

GET /debug HTTP/1.1

headers:

Host: localhost:8080
User-Agent: fortio.org/fortio-X.Y.Z
Foo: Bar

body:

Report only UI

If you have json files saved from running the full UI or downloaded, using the -sync option, from an amazon or google cloud storage bucket or from a peer fortio server (to synchronize from a peer fortio, use http://peer:8080/data/index.tsv as the sync URL). You can then serve just the reports:

$ fortio report -sync-interval 15m -sync "https://storage.googleapis.com/fortio-data?prefix=fortio.istio.io/"
Browse only UI starting - visit:
http://localhost:8080/
Https redirector running on :8081

Using the HTTP fan out / multi proxy feature

Example listen on 1 extra port and every request sent to that 1 port is forward to 2:

# in one window or &
$ fortio server -M "5554 http://localhost:8080 http://localhost:8080"
[...]
Fortio X.Y.Z Multi on 5554 server listening on [::]:5554
10:09:56 I http_forwarder.go:152> Multi-server on [::]:5554 running with &{Targets:[{Destination:http://localhost:8080 MirrorOrigin:true} {Destination:http://localhost:8080 MirrorOrigin:true}] Name:Multi on [::]:5554 client:0xc0001ccc00}

Call the debug endpoint on both

# in new window
$ fortio curl -payload "a test" http://localhost:5554/debug
HTTP/1.1 200 OK
Date: Wed, 07 Oct 2020 17:11:06 GMT
Content-Length: 684
Content-Type: text/plain; charset=utf-8

Φορτίο version X.Y.Z unknown goM.m.p echo debug server up for 1m9.3s on C02C77BHMD6R - request from [::1]:51020

POST /debug HTTP/1.1

headers:

Host: localhost:8080
Accept-Encoding: gzip
Content-Type: application/octet-stream
User-Agent: fortio.org/fortio-X.Y.Z
X-Fortio-Multi-Id: 1
X-On-Behalf-Of: [::1]:51019

body:

a test
Φορτίο version X.Y.Z unknown goM.m.p echo debug server up for 1m9.3s on C02C77BHMD6R - request from [::1]:51020

POST /debug HTTP/1.1

headers:

Host: localhost:8080
Accept-Encoding: gzip
Content-Type: application/octet-stream
User-Agent: fortio.org/fortio-X.Y.Z
X-Fortio-Multi-Id: 2
X-On-Behalf-Of: [::1]:51019

body:

a test

There are 2 flags to further control the behaviour of the multi server proxies:

• pass -mirrorOriginFlag=false to not mirror all headers and request type to targets.
• pass -multi-serial-mode to stream request response serially instead of fetching in parallel and writing combined data after completion

Also remember you can pass multiple -M.

Using the TCP proxy server(s) feature

Example: open 2 additional listening ports and forward all requests received on 8888 and 8889 (ipv6) to 8080 (regular http server)

$ fortio server -P "8888 [::1]:8080" -P "[::1]:8889 [::1]:8080"
Fortio X.Y.Z grpc 'ping' server listening on [::]:8079
Fortio X.Y.Z https redirector server listening on [::]:8081
Fortio X.Y.Z echo server listening on [::]:8080
Data directory is /home/dl
UI started - visit:
http://localhost:8080/fortio/
(or any host/ip reachable on this server)
Fortio X.Y.Z proxy for [::1]:8080 server listening on [::]:8888
Fortio X.Y.Z proxy for [::1]:8080 server listening on [::1]:8889

Implementation details

Fortio is written in the Go language and includes a scalable semi log histogram in stats.go and a periodic runner engine in periodic.go with specializations for http and grpc. The http/ package includes a very high performance specialized http 1.1 client. You may find fortio's logger useful as well.

You can run the histogram code standalone as a command line in histogram/, a basic echo http server in echosrv/, or both the http echo and GRPC ping server through fortio server, the fortio command line interface lives in this top level directory fortio_main.go

There is also fcurl/ which is the fortio curl part of the code (if you need a light http client without grpc or server side). A matching tiny (2Mb compressed) docker image is fortio/fortio.fcurl

More examples

You can get the data on the console, for instance, with 5k qps: (includes envoy and mixer in the calls)

$ time fortio load -qps 5000 -t 60s -c 8 -r 0.0001 -H "Host: perf-cluster" http://benchmark-2:9090/echo
2017/07/09 02:31:05 Will be setting special Host header to perf-cluster
Fortio running at 5000 queries per second for 1m0s: http://benchmark-2:9090/echo
Starting at 5000 qps with 8 thread(s) [gomax 4] for 1m0s : 37500 calls each (total 300000)
2017/07/09 02:32:05 T004 ended after 1m0.000907812s : 37500 calls. qps=624.9905437680746
2017/07/09 02:32:05 T000 ended after 1m0.000922222s : 37500 calls. qps=624.9903936684861
2017/07/09 02:32:05 T005 ended after 1m0.00094454s : 37500 calls. qps=624.9901611965524
2017/07/09 02:32:05 T006 ended after 1m0.000944816s : 37500 calls. qps=624.9901583216429
2017/07/09 02:32:05 T001 ended after 1m0.00102094s : 37500 calls. qps=624.9893653892883
2017/07/09 02:32:05 T007 ended after 1m0.001096292s : 37500 calls. qps=624.9885805003184
2017/07/09 02:32:05 T003 ended after 1m0.001045342s : 37500 calls. qps=624.9891112105419
2017/07/09 02:32:05 T002 ended after 1m0.001044416s : 37500 calls. qps=624.9891208560392
Ended after 1m0.00112695s : 300000 calls. qps=4999.9
Aggregated Sleep Time : count 299992 avg 8.8889218e-05 +/- 0.002326 min -0.03490402 max 0.001006041 sum 26.6660543
# range, mid point, percentile, count
< 0 , 0 , 8.58, 25726
>= 0 < 0.001 , 0.0005 , 100.00, 274265
>= 0.001 < 0.002 , 0.0015 , 100.00, 1
# target 50% 0.000453102
WARNING 8.58% of sleep were falling behind
Aggregated Function Time : count 300000 avg 0.00094608764 +/- 0.0007901 min 0.000510522 max 0.029267604 sum 283.826292
# range, mid point, percentile, count
>= 0.0005 < 0.0006 , 0.00055 , 0.15, 456
>= 0.0006 < 0.0007 , 0.00065 , 3.25, 9295
>= 0.0007 < 0.0008 , 0.00075 , 24.23, 62926
>= 0.0008 < 0.0009 , 0.00085 , 62.73, 115519
>= 0.0009 < 0.001 , 0.00095 , 85.68, 68854
>= 0.001 < 0.0011 , 0.00105 , 93.11, 22293
>= 0.0011 < 0.0012 , 0.00115 , 95.38, 6792
>= 0.0012 < 0.0014 , 0.0013 , 97.18, 5404
>= 0.0014 < 0.0016 , 0.0015 , 97.94, 2275
>= 0.0016 < 0.0018 , 0.0017 , 98.34, 1198
>= 0.0018 < 0.002 , 0.0019 , 98.60, 775
>= 0.002 < 0.0025 , 0.00225 , 98.98, 1161
>= 0.0025 < 0.003 , 0.00275 , 99.21, 671
>= 0.003 < 0.0035 , 0.00325 , 99.36, 449
>= 0.0035 < 0.004 , 0.00375 , 99.47, 351
>= 0.004 < 0.0045 , 0.00425 , 99.57, 290
>= 0.0045 < 0.005 , 0.00475 , 99.66, 280
>= 0.005 < 0.006 , 0.0055 , 99.79, 380
>= 0.006 < 0.007 , 0.0065 , 99.82, 92
>= 0.007 < 0.008 , 0.0075 , 99.83, 15
>= 0.008 < 0.009 , 0.0085 , 99.83, 5
>= 0.009 < 0.01 , 0.0095 , 99.83, 1
>= 0.01 < 0.012 , 0.011 , 99.83, 8
>= 0.012 < 0.014 , 0.013 , 99.84, 35
>= 0.014 < 0.016 , 0.015 , 99.92, 231
>= 0.016 < 0.018 , 0.017 , 99.94, 65
>= 0.018 < 0.02 , 0.019 , 99.95, 26
>= 0.02 < 0.025 , 0.0225 , 100.00, 139
>= 0.025 < 0.03 , 0.0275 , 100.00, 14
# target 50% 0.000866935
# target 75% 0.000953452
# target 99% 0.00253875
# target 99.9% 0.0155152
Code 200 : 300000
Response Body Sizes : count 300000 avg 0 +/- 0 min 0 max 0 sum 0

Or you can get the data in JSON format (using -json result.json)

Web/Graphical UI

Or graphically (through the http://localhost:8080/fortio/ web UI):

Simple form/UI:

Sample requests with responses delayed by 250us and 0.5% of 503 and 1.5% of 429 simulated http errors.

Run result:

Code 200 : 2929 (97.6 %)
Code 429 : 56 (1.9 %)
Code 503 : 15 (0.5 %)

There are live examples on demo.fortio.org

Contributing

Contributions whether through issues, documentation, bug fixes, or new features are most welcome !

Please also see Contributing to Istio and Getting started contributing to Fortio in the FAQ.

If you are not using the binary releases, please do make pull to pull/update to the latest of the current branch.

And make sure to go strict format (go get mvdan.cc/gofumpt and gofumpt -s -w *.go) and run those commands successfully before sending your PRs:

make test
make lint
make release-test

When modifying Javascript, check with standard:

standard --fix ui/static/js/fortio_chart.js

New features and bug fixes should include a test.

See also

Our wiki and the Fortio FAQ (including for instance differences between fortio and wrk or httpbin)

Disclaimer

This is not an officially supported Google product.