Files
postgrest/nix/tools/loadtest.nix
T
Wolfgang Walther 59dffc258a nix(loadtest): report latency instead of rate
We previously used "rate", i.e. number of requests per second, as the
primary metric to judge loadtest results. However, this has always been
varying from run to run quite a bit, especially in CI where other jobs
possibly run on the same VM host.

The run-to-run variance has massively increased after splitting the
results up per request. Example run in CI with rate on the PR
introducing this change (on which we would expect no change at all):

| rate [1/s]                         |   main |   head |    Δ |
|:-----------------------------------|-------:|-------:|-----:|
| /                                  |    9.4 |    9.5 |   1% |
| /actors                            |  870.4 | 1023.0 |  18% |
| /actors?actor=eq.1                 |  188.5 |  198.6 |   5% |
| /actors?actor=eq.1&columns=name    |  197.3 |  167.1 | -15% |
| /actors?select=*,roles(*,films(*)) |  153.9 |  144.9 |  -6% |
| /films?columns=id,title            |  157.9 |  182.6 |  16% |
| /films?columns=id,title,year,...   |   87.0 |   87.1 |   0% |
| /roles                             |  204.5 |  267.3 |  31% |
| /rpc/call_me                       |  231.3 |  208.8 | -10% |
| /rpc/call_me?name=John             |  212.2 |  201.7 |  -5% |

From the data we can easily tell that the very reason that rate as a
paramter has only worked, so far, because the data was *heavily*
dominated by the requests on the root endpoint for OpenAPI. The longer
duration makes the request much less vulnerable for concurrent activity.
For all other requests its essentially not possible to judge the effect
of a PR this way.

One way to counter this would be to massively increase the time the
loadtest runs. More samples will result in a smoother average. However,
that's not practical for usability of CI. In the original PR #1812 I
already evaluated using the *minimum latency* as the most reliable
criteriumi, but this has never really caught on. The theory behind this
is: The variation in timings between requests is happening because of
concurrent activity, priority chosen by the scheduler, availability of
resources and such - all factors *outside* our control, and *irrelevant*
to the Haskell code we're writing.

Using the minimum latency is an estimation of how fast the code can run
*in the best case*. This might not be a number relevant for production,
but it's much more directly related to the code we write.

Here's to show how variation becomes *much* smaller with minimum latency
as the parameter:

| min latency [μs]                   |     main |   head |    Δ |
|:-----------------------------------|---------:|-------:|-----:|
| /                                  |   1275.3 | 1263.6 |  -1% |
| /actors                            |     10.0 |    9.9 |  -1% |
| /actors?actor=eq.1                 |     50.7 |   48.3 |  -5% |
| /actors?actor=eq.1&columns=name    |     54.1 |   54.0 |   0% |
| /actors?select=*,roles(*,films(*)) |     63.2 |   61.9 |  -2% |
| /films?columns=id,title            |     51.1 |   50.7 |  -1% |
| /films?columns=id,title,year,...   |    121.9 |  121.8 |   0% |
| /roles                             |     42.9 |   42.6 |  -1% |
| /rpc/call_me                       |     45.6 |   45.4 |   0% |
| /rpc/call_me?name=John             |     44.4 |   44.2 |   0% |

Since we're separating results per request now, we can only sensibly
focus on *one* parameter - otherwise this would get really clunky
UI-wise. Especially for automated CI failures, minimum latency is the
logical choice.

This commit starts using minimum latency, i.e. P0, but any percentile
should be an improvement over the status quo. A later commit will change
to a different P-value.
2026-06-01 20:49:35 +00:00

347 lines
13 KiB
Nix

{ buildToolbox
, checkedShellScript
, jq
, libfaketime
, python3Packages
, vegeta
, withTools
, writers
}:
let
runner =
checkedShellScript
{
name = "postgrest-loadtest-runner";
docs = "Run vegeta. Assume PostgREST to be running.";
args = [
"ARG_LEFTOVERS([additional vegeta arguments])"
"ARG_USE_ENV([PGRST_SERVER_UNIX_SOCKET], [], [Unix socket to connect to running PostgREST instance])"
];
}
''
echo "Starting vegeta loadtest..."
# ARG_USE_ENV only adds defaults or docs for environment variables
# We manually implement a required check here
# See also: https://github.com/matejak/argbash/issues/80
: "''${PGRST_SERVER_UNIX_SOCKET:?PGRST_SERVER_UNIX_SOCKET is required}"
${vegeta}/bin/vegeta -cpus 1 attack \
-dns-ttl -1 \
-unix-socket "$PGRST_SERVER_UNIX_SOCKET" \
-max-workers 1 \
-workers 1 \
-rate 0 \
-duration 60s \
"''${_arg_leftovers[@]}"
'';
loadtest =
checkedShellScript
{
name = "postgrest-loadtest";
docs = "Run the vegeta loadtests with PostgREST.";
args = [
"ARG_OPTIONAL_SINGLE([output], [o], [Filename to dump json output to], [./loadtest/result.bin])"
"ARG_OPTIONAL_SINGLE([testdir], [t], [Directory to load tests and fixtures from], [./test/load])"
"ARG_OPTIONAL_SINGLE([kind], [k], [Kind of loadtest], [mixed])"
"ARG_OPTIONAL_SINGLE([method],, [HTTP method used for the jwt loadtests], [OPTIONS])"
"ARG_TYPE_GROUP_SET([KIND], [KIND], [kind], [mixed,errors,jwt-hs,jwt-hs-cache,jwt-hs-cache-worst,jwt-rsa,jwt-rsa-cache,jwt-rsa-cache-worst])"
"ARG_TYPE_GROUP_SET([METHOD], [METHOD], [method], [OPTIONS,GET])"
"ARG_OPTIONAL_SINGLE([monitor], [m], [Monitoring file], [./loadtest/result.csv])"
"ARG_LEFTOVERS([additional vegeta arguments])"
];
workingDir = "/";
}
''
export PGRST_DB_ANON_ROLE="postgrest_test_anonymous"
export PGRST_DB_CONFIG="false"
export PGRST_DB_POOL="1"
export PGRST_DB_SCHEMAS="test"
export PGRST_DB_TX_END="rollback-allow-override"
export PGRST_LOG_LEVEL="crit"
export PGRST_JWT_SECRET="reallyreallyreallyreallyverysafe"
mkdir -p "$(dirname "$_arg_output")"
abs_output="$(realpath "$_arg_output")"
case "$_arg_kind" in
jwt-hs)
export PGRST_JWT_CACHE_MAX_ENTRIES="0"
${genTargets} --method "$_arg_method" "$_arg_testdir"/gen_targets.http
# shellcheck disable=SC2145
${withTools.withPg} -f "$_arg_testdir"/fixtures.sql \
${withTools.withPgrst} -m "$_arg_monitor" \
sh -c "cd \"$_arg_testdir\" && \
${runner} -lazy -targets gen_targets.http -output \"$abs_output\" \"''${_arg_leftovers[@]}\""
;;
jwt-hs-cache)
${genTargets} --method "$_arg_method" "$_arg_testdir"/gen_targets.http
# shellcheck disable=SC2145
${withTools.withPg} -f "$_arg_testdir"/fixtures.sql \
${withTools.withPgrst} -m "$_arg_monitor" \
sh -c "cd \"$_arg_testdir\" && \
${runner} -lazy -targets gen_targets.http -output \"$abs_output\" \"''${_arg_leftovers[@]}\""
;;
jwt-hs-cache-worst)
${libfaketime}/bin/faketime '2000-01-01 00:00:00' ${genTargets} --method "$_arg_method" --worst "$_arg_testdir"/gen_targets.http
# shellcheck disable=SC2145
${withTools.withPg} -f "$_arg_testdir"/fixtures.sql \
${withTools.withPgrst} --faketime '2000-01-01 00:00:00' -m "$_arg_monitor" \
sh -c "cd \"$_arg_testdir\" && \
${runner} -lazy -targets gen_targets.http -output \"$abs_output\" \"''${_arg_leftovers[@]}\""
;;
jwt-rsa)
export PGRST_JWT_CACHE_MAX_ENTRIES="0"
${genRsaMaterials} --rsa="$_arg_testdir"/gen_jwk.json --private-key="$_arg_testdir"/gen_private.json
export PGRST_JWT_SECRET="@$_arg_testdir/gen_jwk.json"
${genTargets} --method "$_arg_method" --rsa="$_arg_testdir"/gen_jwk.json --private-key="$_arg_testdir"/gen_private.json "$_arg_testdir"/gen_targets.http
# shellcheck disable=SC2145
${withTools.withPg} -f "$_arg_testdir"/fixtures.sql \
${withTools.withPgrst} -m "$_arg_monitor" \
sh -c "cd \"$_arg_testdir\" && \
${runner} -lazy -targets gen_targets.http -output \"$abs_output\" \"''${_arg_leftovers[@]}\""
;;
jwt-rsa-cache)
${genRsaMaterials} --rsa="$_arg_testdir"/gen_jwk.json --private-key="$_arg_testdir"/gen_private.json
export PGRST_JWT_SECRET="@$_arg_testdir/gen_jwk.json"
${genTargets} --method "$_arg_method" --rsa="$_arg_testdir"/gen_jwk.json --private-key="$_arg_testdir"/gen_private.json "$_arg_testdir"/gen_targets.http
# shellcheck disable=SC2145
${withTools.withPg} -f "$_arg_testdir"/fixtures.sql \
${withTools.withPgrst} -m "$_arg_monitor" \
sh -c "cd \"$_arg_testdir\" && \
${runner} -lazy -targets gen_targets.http -output \"$abs_output\" \"''${_arg_leftovers[@]}\""
;;
jwt-rsa-cache-worst)
${genRsaMaterials} --rsa="$_arg_testdir"/gen_jwk.json --private-key="$_arg_testdir"/gen_private.json
export PGRST_JWT_SECRET="@$_arg_testdir/gen_jwk.json"
${libfaketime}/bin/faketime '2000-01-01 00:00:00' ${genTargets} --method "$_arg_method" --worst --rsa="$_arg_testdir"/gen_jwk.json --private-key="$_arg_testdir"/gen_private.json "$_arg_testdir"/gen_targets.http
# shellcheck disable=SC2145
${withTools.withPg} -f "$_arg_testdir"/fixtures.sql \
${withTools.withPgrst} --faketime '2000-01-01 00:00:00' -m "$_arg_monitor" \
sh -c "cd \"$_arg_testdir\" && \
${runner} -lazy -targets gen_targets.http -output \"$abs_output\" \"''${_arg_leftovers[@]}\""
;;
mixed)
# shellcheck disable=SC2145
${withTools.withPg} -f "$_arg_testdir"/fixtures.sql \
${withTools.withPgrst} -m "$_arg_monitor" \
sh -c "cd \"$_arg_testdir\" && \
${runner} -targets targets.http -output \"$abs_output\" \"''${_arg_leftovers[@]}\""
;;
# here we sleep purposefully to check how much memory does the schema cache consume in the final report
errors)
# shellcheck disable=SC2145
${withTools.withPg} -f "$_arg_testdir"/errors.sql \
${withTools.withPgrst} --timeout 2 --sleep 5 -m "$_arg_monitor" \
sh -c "cd \"$_arg_testdir\" && \
${runner} -targets errors.http -output \"$abs_output\" \"''${_arg_leftovers[@]}\""
;;
esac
${vegeta}/bin/vegeta report -type=text "$_arg_output"
if [ "$_arg_kind" != "errors" ]; then
# fail in case 401 happened on jwt loadtests
unauthorized_count="$(${vegeta}/bin/vegeta report -type=json "$_arg_output" \
| ${jq}/bin/jq -r '.status_codes["401"] // 0')"
if [ "$unauthorized_count" -gt 0 ]; then
last_unauthorized_body="$(${vegeta}/bin/vegeta encode "$_arg_output" \
| ${jq}/bin/jq -rn '
reduce inputs as $item (null;
if $item.code == 401 then $item else . end
)
| if . == null then
empty
else
(.body | @base64d)
end
')"
echo "loadtest failed: found $unauthorized_count 401 Unauthorized responses" >&2
if [ -n "$last_unauthorized_body" ]; then
printf '%s\n' "Last 401 response body:" >&2
printf '%s\n' "$last_unauthorized_body" >&2
fi
exit 1
fi
fi
'';
loadtestAgainst =
let
name = "postgrest-loadtest-against";
in
checkedShellScript
{
inherit name;
docs =
''
Run the vegeta loadtest against every target branch and HEAD:
- once on the every <target-#> branch
- once in the current worktree
'';
args = [
"ARG_POSITIONAL_INF([target], [Commit-ish reference to compare with], 1)"
"ARG_OPTIONAL_SINGLE([kind], [k], [Kind of loadtest], [mixed])"
];
positionalCompletion =
''
if test "$prev" == "${name}"; then
__gitcomp_nl "$(__git_refs)"
fi
'';
workingDir = "/";
}
''
# run loadtest for every target
for tgt in "''${_arg_target[@]}"; do
cat << EOF
Running "$_arg_kind" loadtest on "$tgt"...
EOF
# Runs the test files from the current working tree
# to make sure both tests are run with the same files.
# Save the results in the current working tree, too,
# otherwise they'd be lost in the temporary working tree
# created by withTools.withGit.
${withTools.withGit} "$tgt" ${loadtest} -k "$_arg_kind" -m "$PWD/loadtest/$tgt.csv" --output "$PWD/loadtest/$tgt.bin" --testdir "$PWD/test/load"
cat << EOF
Done running on "$tgt".
EOF
done
# run loadtest once on HEAD
cat << EOF
Running "$_arg_kind" loadtest on HEAD...
EOF
${loadtest} -k "$_arg_kind" -m "$PWD/loadtest/head.csv" --output "$PWD/loadtest/head.bin" --testdir "$PWD/test/load"
cat << EOF
Done running on HEAD.
EOF
'';
reporter =
checkedShellScript
{
name = "postgrest-loadtest-reporter";
docs = "Create a named json report for a single result file.";
args = [
"ARG_POSITIONAL_SINGLE([file], [Filename of result to create report for])"
"ARG_LEFTOVERS([additional vegeta arguments])"
];
workingDir = "/";
}
''
${vegeta}/bin/vegeta encode "$_arg_file" \
| ${jq}/bin/jq --slurp 'map(select(.url != "")) | group_by("\(.code) \(.method) \(.url)") | map({("\(.[0].code) \(.[0].method) \(.[0].url)" | sub("http://postgrest";"")): map(.latency) | min / 10e3 }) | .[]' \
| ${jq}/bin/jq --arg branch "$(basename "$_arg_file" .bin)" '. + {branch: $branch}'
'';
toMarkdown =
writers.writePython3 "postgrest-loadtest-to-markdown"
{
libraries = [ python3Packages.pandas python3Packages.tabulate ];
}
''
import sys
import pandas as pd
pd.read_json(sys.stdin) \
.rename(columns={'latency': 'min latency [μs]'}) \
.set_index('min latency [μs]') \
.drop(['branch']) \
.convert_dtypes() \
.to_markdown(sys.stdout, floatfmt='.1f')
'';
report =
checkedShellScript
{
name = "postgrest-loadtest-report";
docs = "Create a report of all loadtest reports as markdown.";
args = [
"ARG_OPTIONAL_SINGLE([group], [g], [Marker to group results])"
];
workingDir = "/";
}
''
marker=''${_arg_group:+"($_arg_group)"}
echo -e "## Loadtest results $marker\n"
find loadtest -type f -iname '*.bin' -exec ${reporter} {} \; \
| ${jq}/bin/jq '[paths(scalars) as $path | {latency: $path | join("."), (.branch): getpath($path)}]' \
| ${jq}/bin/jq --slurp 'flatten | group_by(.latency) | map(add)' \
| ${toMarkdown}
echo -e "\n\n## Loadtest elapsed seconds vs CPU/MEM usage $marker\n"
find loadtest -type f -iname '*.csv' \
| sort -m \
| ${mergeMonitorResults}
'';
genTargets =
writers.writePython3 "postgrest-gen-loadtest-targets"
{
libraries = [ python3Packages.pyjwt python3Packages.jwcrypto ];
doCheck = false; # postgrest-style conflicts with this
}
(builtins.readFile ./generate_targets.py);
genRsaMaterials =
writers.writePython3 "postgrest-gen-rsa-materials"
{
libraries = [ python3Packages.jwcrypto ];
doCheck = false; # postgrest-style conflicts with this
}
(builtins.readFile ./gen_rsa_materials.py);
mergeMonitorResults =
writers.writePython3 "postgrest-merge-monitor-results"
{
libraries = [ python3Packages.pandas python3Packages.tabulate ];
}
(builtins.readFile ./merge_monitor_result.py);
in
buildToolbox {
name = "postgrest-loadtest";
tools = { inherit loadtest loadtestAgainst report; };
}