Compare commits

..
Author SHA1 Message Date
clawbot a0277879d7 dirwatch, lookup: one test below the synctest comment ran outside a bubble
check / check (push) Failing after 2m6s
In both test files the comment said every test below it runs in a
synctest bubble, but one did not. That test now sits above the comment.

Model: opus-5-5
2026-11-19 20:39:16 +00:00
clawbot 2083584ee0 README: list internal/dirwatch under Packages
The list named every package but the new one.

Model: opus-5-5
2026-11-19 20:39:16 +00:00
clawbot fe61d6b066 dirwatch: one loop that watches a directory, for rules and lookup (closes #429)
rules and lookup each had the same loop: watch the directory, start a
timer at once, restart it at each change that counts, read again when it
fires. It now lives once in internal/dirwatch, with its quiet time and
its tests. Each package keeps its own log messages, its readAgain, and
its test for which changes count. internal/state keeps its own copy.

Model: opus-5-5
2026-11-19 20:39:01 +00:00
15 changed files with 650 additions and 1017 deletions

No files matched your search

+2
View File
@@ -906,6 +906,8 @@ stops the running process (see [`docs/state-files.md`](docs/state-files.md)).
`internal/coreruleset`, `internal/geojs`, `internal/lookup`, `internal/coreruleset`, `internal/geojs`, `internal/lookup`,
`internal/reputation`, `internal/routewatch`, `internal/anomaly` and `internal/reputation`, `internal/routewatch`, `internal/anomaly` and
`internal/alerts`: one check or source each. `internal/alerts`: one check or source each.
- `internal/dirwatch`: watches a directory, so that `internal/rules` and
`internal/lookup` read their files again when they change.
- `internal/state`: the state and cache files. - `internal/state`: the state and cache files.
- `internal/requestlog`: the request log. - `internal/requestlog`: the request log.
- `internal/processlog`: the process's own messages. - `internal/processlog`: the process's own messages.
+3
View File
@@ -93,6 +93,9 @@ never as a time.
imports no other package of this repository: `waf` gives it what it needs imports no other package of this repository: `waf` gives it what it needs
to count and report failures. to count and report failures.
- `internal/lookup`: looks clients up in the lookup database. - `internal/lookup`: looks clients up in the lookup database.
- `internal/dirwatch`: watches a directory, and calls a function once it has
gone a short time without a change. `internal/rules` and `internal/lookup`
read their files again this way.
- `internal/reputation`: fetches the lists, asks the DNS blocklists and - `internal/reputation`: fetches the lists, asks the DNS blocklists and
AbuseIPDB in the background, and keeps their answers. It also fetches AbuseIPDB in the background, and keeps their answers. It also fetches
AbuseIPDB's blacklist, which `internal/waf` bans. AbuseIPDB's blacklist, which `internal/waf` bans.
+90
View File
@@ -0,0 +1,90 @@
// Package dirwatch watches a directory, and calls a function once the
// directory has gone a short time without a change after one. The rule
// files and the lookup database are read again this way.
package dirwatch
import (
"context"
"time"
"github.com/fsnotify/fsnotify"
)
// quietTime is how long the directory must go without a change before the
// function is called, so that a file still being written, such as one
// saved in place, appended to or copied in with scp, is read only once
// whole.
const quietTime = 2 * time.Second
// Watcher watches one directory. The directory is watched, rather than a
// file in it, so that a file replaced by a rename is seen.
type Watcher struct {
watcher *fsnotify.Watcher
}
// New starts watching dir, or returns why it cannot.
func New(dir string) (*Watcher, error) {
watcher, err := fsnotify.NewWatcher()
if err != nil {
return nil, err
}
err = watcher.Add(dir)
if err != nil {
_ = watcher.Close()
return nil, err
}
return &Watcher{watcher: watcher}, nil
}
// Close stops watching the directory.
func (w *Watcher) Close() error {
return w.watcher.Close()
}
// CallWhenQuiet calls changed each time the directory has gone quietTime
// without a change, after a change to a path for which counts is true, until
// ctx is done. It also calls failed with each error the watch reports. The
// wait starts at once, as if for a change, so that a change made before the
// watch began is taken in too.
func (w *Watcher) CallWhenQuiet(
ctx context.Context,
counts func(path string) bool,
changed func(),
failed func(error),
) {
callWhenQuiet(
ctx, w.watcher.Events, w.watcher.Errors, counts, changed, failed,
)
}
// callWhenQuiet is CallWhenQuiet, for the changes from events and the
// errors from errs.
func callWhenQuiet(
ctx context.Context,
events <-chan fsnotify.Event,
errs <-chan error,
counts func(path string) bool,
changed func(),
failed func(error),
) {
quiet := time.NewTimer(quietTime)
defer quiet.Stop()
for {
select {
case <-ctx.Done():
return
case event := <-events:
if counts(event.Name) {
quiet.Reset(quietTime)
}
case <-quiet.C:
changed()
case err := <-errs:
failed(err)
}
}
}
+167
View File
@@ -0,0 +1,167 @@
package dirwatch
import (
"context"
"errors"
"path/filepath"
"sync/atomic"
"testing"
"testing/synctest"
"time"
"github.com/fsnotify/fsnotify"
)
// errWatch is an error the watch reports.
var errWatch = errors.New("too many changes")
func TestDirectoryThatCannotBeWatchedIsAnError(t *testing.T) {
t.Parallel()
_, err := New(filepath.Join(t.TempDir(), "missing"))
if err == nil {
t.Error("a missing directory is watched, want an error")
}
}
// The tests below run callWhenQuiet in a synctest bubble, where time is a
// clock of the test's own: time.Sleep moves it on at once, and
// synctest.Wait returns once callWhenQuiet waits again, so that every call
// due by then is made. The test sends the changes itself, as the watch of
// a directory cannot run in a bubble.
func TestSecondChangeStartsTheWaitAgain(t *testing.T) {
t.Parallel()
synctest.Test(t, func(t *testing.T) {
loop := start(t, everyPath)
loop.changes <- fsnotify.Event{Name: "50-app.rules", Op: fsnotify.Write}
time.Sleep(quietTime - time.Nanosecond)
synctest.Wait()
loop.wantCalls(t, 0)
// The second part starts the wait again.
loop.changes <- fsnotify.Event{Name: "50-app.rules", Op: fsnotify.Write}
time.Sleep(quietTime - time.Nanosecond)
synctest.Wait()
loop.wantCalls(t, 0)
time.Sleep(time.Nanosecond)
synctest.Wait()
loop.wantCalls(t, 1)
})
}
func TestCalledOnceQuietAfterTheStart(t *testing.T) {
t.Parallel()
synctest.Test(t, func(t *testing.T) {
// An edit saved before the watch began sends no change, and is
// taken in by this call.
loop := start(t, everyPath)
time.Sleep(quietTime)
synctest.Wait()
loop.wantCalls(t, 1)
// Long after, it has been called only once.
time.Sleep(time.Hour)
synctest.Wait()
loop.wantCalls(t, 1)
})
}
func TestChangeThatDoesNotCountIsNoChange(t *testing.T) {
t.Parallel()
synctest.Test(t, func(t *testing.T) {
loop := start(t, func(path string) bool {
return path == "ipinfo_lite.mmdb"
})
// The wait that starts with the watch ends with a call.
time.Sleep(quietTime)
synctest.Wait()
loop.changes <- fsnotify.Event{Name: "other.mmdb", Op: fsnotify.Create}
time.Sleep(quietTime)
synctest.Wait()
loop.wantCalls(t, 1)
loop.changes <- fsnotify.Event{Name: "ipinfo_lite.mmdb", Op: fsnotify.Write}
time.Sleep(quietTime)
synctest.Wait()
loop.wantCalls(t, 2)
})
}
func TestErrorsOfTheWatchHandedOn(t *testing.T) {
t.Parallel()
synctest.Test(t, func(t *testing.T) {
loop := start(t, everyPath)
loop.errs <- errWatch
synctest.Wait()
if !errors.Is(loop.failure, errWatch) {
t.Errorf("failed with %v, want %v", loop.failure, errWatch)
}
})
}
// loop is a callWhenQuiet that a test runs.
type loop struct {
// changes and errs send it changes and errors.
changes chan fsnotify.Event
errs chan error
// calls counts its calls of changed, and failure is the last error it
// handed to failed.
calls atomic.Int32
failure error
}
// start runs callWhenQuiet, with counts, until the test ends.
func start(t *testing.T, counts func(path string) bool) *loop {
t.Helper()
l := &loop{changes: make(chan fsnotify.Event), errs: make(chan error)}
ctx, stop := context.WithCancel(t.Context())
stopped := make(chan struct{})
go func() {
callWhenQuiet(ctx, l.changes, l.errs, counts,
func() { l.calls.Add(1) },
func(err error) { l.failure = err },
)
close(stopped)
}()
t.Cleanup(func() {
stop()
<-stopped
})
return l
}
// everyPath counts a change to any path.
func everyPath(string) bool {
return true
}
// wantCalls checks how many times changed has been called.
func (l *loop) wantCalls(t *testing.T, want int32) {
t.Helper()
got := l.calls.Load()
if got != want {
t.Errorf("changed called %d times, want %d", got, want)
}
}
+18 -49
View File
@@ -13,18 +13,13 @@ import (
"sync" "sync"
"time" "time"
"github.com/fsnotify/fsnotify"
"github.com/oschwald/maxminddb-golang/v2" "github.com/oschwald/maxminddb-golang/v2"
"sneak.berlin/go/smallwebwaf/internal/alerts" "sneak.berlin/go/smallwebwaf/internal/alerts"
"sneak.berlin/go/smallwebwaf/internal/dirwatch"
"sneak.berlin/go/smallwebwaf/internal/geojs" "sneak.berlin/go/smallwebwaf/internal/geojs"
) )
// quietTime is how long the lookup database must go without a change
// before it is read again, so that a file still being copied in is read
// only once whole.
const quietTime = 2 * time.Second
// FileParams are what OpenFile needs. // FileParams are what OpenFile needs.
type FileParams struct { type FileParams struct {
// Path is the lookup database, the IPinfo Lite file in its .mmdb form // Path is the lookup database, the IPinfo Lite file in its .mmdb form
@@ -122,20 +117,13 @@ func (f *File) ReadFailures() int {
} }
// Watch watches the directory of the lookup database until ctx is done, // Watch watches the directory of the lookup database until ctx is done,
// and reads the file again once it has gone without a change for // and reads the file again once it has gone without a change for a short
// quietTime, after it is replaced, written or removed, and after Watch // time, after it is replaced, written or removed, and after Watch starts
// starts watching. If the directory cannot be watched, that is logged, and // watching, as dirwatch describes. A change to another file in its
// the database read at start stays in use. // directory does not count. If the directory cannot be watched, that is
// logged, and the database read at start stays in use.
func (f *File) Watch(ctx context.Context) { func (f *File) Watch(ctx context.Context) {
watcher, err := fsnotify.NewWatcher() watcher, err := dirwatch.New(filepath.Dir(f.params.Path))
if err == nil {
defer func() {
_ = watcher.Close()
}()
err = watcher.Add(filepath.Dir(f.params.Path))
}
if err != nil { if err != nil {
f.params.ProcessLog.Error( f.params.ProcessLog.Error(
"cannot watch the lookup database for replacements", "cannot watch the lookup database for replacements",
@@ -146,41 +134,22 @@ func (f *File) Watch(ctx context.Context) {
return return
} }
defer func() {
_ = watcher.Close()
}()
f.params.ProcessLog.Info("watching the lookup database for replacements", f.params.ProcessLog.Info("watching the lookup database for replacements",
"file", f.params.Path) "file", f.params.Path)
f.readAfterChanges(ctx, watcher.Events, watcher.Errors) watcher.CallWhenQuiet(ctx, f.isDatabase, f.readAgain, func(err error) {
f.params.ProcessLog.Warn("watching the lookup database failed",
"error", err.Error())
})
} }
// readAfterChanges reads the lookup database again once quietTime has // isDatabase reports whether path names the lookup database.
// passed without a change to it from events, until ctx is done, and logs func (f *File) isDatabase(path string) bool {
// the errors from errs. A change to another file in its directory does not return filepath.Clean(path) == filepath.Clean(f.params.Path)
// count. The wait starts at once, as if for a change, so that a file
// replaced after OpenFile read it, and before its directory was watched,
// is read too.
func (f *File) readAfterChanges(
ctx context.Context, events <-chan fsnotify.Event, errs <-chan error,
) {
path := filepath.Clean(f.params.Path)
quiet := time.NewTimer(quietTime)
defer quiet.Stop()
for {
select {
case <-ctx.Done():
return
case event := <-events:
if filepath.Clean(event.Name) == path {
quiet.Reset(quietTime)
}
case <-quiet.C:
f.readAgain()
case err := <-errs:
f.params.ProcessLog.Warn("watching the lookup database failed",
"error", err.Error())
}
}
} }
// readAgain reads the lookup database again, in place of the one in use, // readAgain reads the lookup database again, in place of the one in use,
+35 -141
View File
@@ -1,7 +1,6 @@
package lookup package lookup
import ( import (
"context"
"log/slog" "log/slog"
"net/netip" "net/netip"
"net/url" "net/url"
@@ -12,7 +11,6 @@ import (
"testing/synctest" "testing/synctest"
"time" "time"
"github.com/fsnotify/fsnotify"
"github.com/maxmind/mmdbwriter/mmdbtype" "github.com/maxmind/mmdbwriter/mmdbtype"
"sneak.berlin/go/smallwebwaf/internal/alerts" "sneak.berlin/go/smallwebwaf/internal/alerts"
@@ -133,66 +131,48 @@ func TestFileThatCannotBeReadIsAnError(t *testing.T) {
} }
} }
// The tests below run readAfterChanges in a synctest bubble, where time is func TestChangeOfAnotherFileInTheDirectoryIsNoReplacement(t *testing.T) {
// a clock of the test's own: time.Sleep moves it on at once, and t.Parallel()
// synctest.Wait returns once readAfterChanges waits again, so that every
// reading due by then is done. The test sends the changes itself, as the
// watch of a directory cannot run in a bubble.
func TestReplacementCopiedOverTheFileInTwoPartsIsReadOnlyWhole( dir := t.TempDir()
t *testing.T, path := filepath.Join(dir, "ipinfo_lite.mmdb")
) { writeDatabase(t, path, "DE")
f := openFile(t, path, newQueue())
for changed, want := range map[string]bool{
path: true,
dir + "/./ipinfo_lite.mmdb": true,
filepath.Join(dir, "other.mmdb"): false,
} {
if f.isDatabase(changed) != want {
t.Errorf(
"a change to %s counts: %t, want %t",
changed,
!want,
want,
)
}
}
}
// The tests below run in a synctest bubble, where time is a clock of the
// test's own, so that the time a file is read can be checked. The wait for
// the directory to go quiet before a reading is tested in dirwatch.
func TestReplacementIsReadAgain(t *testing.T) {
t.Parallel() t.Parallel()
synctest.Test(t, func(t *testing.T) { synctest.Test(t, func(t *testing.T) {
dir := t.TempDir() path := filepath.Join(t.TempDir(), "ipinfo_lite.mmdb")
path := filepath.Join(dir, "ipinfo_lite.mmdb")
writeDatabase(t, path, "DE") writeDatabase(t, path, "DE")
queue := newQueue() queue := newQueue()
f := openFile(t, path, queue) f := openFile(t, path, queue)
changes := watch(t, f)
other := filepath.Join(dir, "replacement.mmdb") writeDatabase(t, path, "KP")
writeDatabase(t, other, "KP") time.Sleep(time.Hour)
f.readAgain()
//nolint:gosec // a file the test wrote
replacement, err := os.ReadFile(other)
if err != nil {
t.Fatalf("read %s: %v", other, err)
}
// The file in use is overwritten in place, and keeps giving what
// it gave. Its first part alone is not a .mmdb file.
file, err := os.Create(path) //nolint:gosec // a file the test wrote
if err != nil {
t.Fatalf("create %s: %v", path, err)
}
defer func() {
_ = file.Close()
}()
half := len(replacement) / 2
write(t, file, replacement[:half])
changes <- fsnotify.Event{Name: path, Op: fsnotify.Write}
time.Sleep(quietTime - time.Nanosecond)
synctest.Wait()
wantCountry(t, f, "DE")
// The second part starts the wait again.
write(t, file, replacement[half:])
changes <- fsnotify.Event{Name: path, Op: fsnotify.Write}
time.Sleep(quietTime - time.Nanosecond)
synctest.Wait()
wantCountry(t, f, "DE")
time.Sleep(time.Nanosecond)
synctest.Wait()
wantCountry(t, f, "KP") wantCountry(t, f, "KP")
if !f.LastRead().Equal(time.Now()) || f.ReadFailures() != 0 { if !f.LastRead().Equal(time.Now()) || f.ReadFailures() != 0 {
@@ -219,15 +199,10 @@ func TestReplacementThatCannotBeReadLeavesTheFileInUseWithOneAlert(
queue := newQueue() queue := newQueue()
f := openFile(t, path, queue) f := openFile(t, path, queue)
read := f.LastRead() read := f.LastRead()
changes := watch(t, f)
writeFile(t, path, "not a lookup database\n") writeFile(t, path, "not a lookup database\n")
changes <- fsnotify.Event{Name: path, Op: fsnotify.Write}
// Long after, the replacement has been read once.
time.Sleep(time.Hour) time.Sleep(time.Hour)
synctest.Wait() f.readAgain()
wantCountry(t, f, "DE") wantCountry(t, f, "DE")
if !f.LastRead().Equal(read) || f.ReadFailures() != 1 { if !f.LastRead().Equal(read) || f.ReadFailures() != 1 {
@@ -240,7 +215,7 @@ func TestReplacementThatCannotBeReadLeavesTheFileInUseWithOneAlert(
} }
wantAlerts(t, queue, alerts.Alert{ wantAlerts(t, queue, alerts.Alert{
Time: read.Add(quietTime), Time: read.Add(time.Hour),
Event: alerts.EventFileError, Event: alerts.EventFileError,
Reason: "the lookup database cannot be read, and the one read before stays in use", Reason: "the lookup database cannot be read, and the one read before stays in use",
Detail: map[string]any{ Detail: map[string]any{
@@ -252,55 +227,6 @@ func TestReplacementThatCannotBeReadLeavesTheFileInUseWithOneAlert(
}) })
} }
func TestChangeOfAnotherFileInTheDirectoryIsNoReplacement(t *testing.T) {
t.Parallel()
synctest.Test(t, func(t *testing.T) {
dir := t.TempDir()
path := filepath.Join(dir, "ipinfo_lite.mmdb")
writeDatabase(t, path, "DE")
f := openFile(t, path, newQueue())
changes := watch(t, f)
// The wait that starts with the watch ends with a reading.
time.Sleep(quietTime)
synctest.Wait()
writeDatabase(t, path, "KP")
changes <- fsnotify.Event{Name: filepath.Join(dir, "other.mmdb"), Op: fsnotify.Create}
time.Sleep(quietTime)
synctest.Wait()
wantCountry(t, f, "DE")
changes <- fsnotify.Event{Name: path, Op: fsnotify.Write}
time.Sleep(quietTime)
synctest.Wait()
wantCountry(t, f, "KP")
})
}
func TestReplacementSavedBeforeTheWatchStartsIsRead(t *testing.T) {
t.Parallel()
synctest.Test(t, func(t *testing.T) {
path := filepath.Join(t.TempDir(), "ipinfo_lite.mmdb")
writeDatabase(t, path, "DE")
f := openFile(t, path, newQueue())
// Saved after OpenFile read the file, and before its directory was
// watched, so that no change is seen for it.
writeDatabase(t, path, "KP")
watch(t, f)
time.Sleep(quietTime)
synctest.Wait()
wantCountry(t, f, "KP")
})
}
// newQueue returns a queue of alerts for a webhook that is never sent // newQueue returns a queue of alerts for a webhook that is never sent
// them, so that they wait in it for the test to look at. // them, so that they wait in it for the test to look at.
func newQueue() *alerts.Queue { func newQueue() *alerts.Queue {
@@ -344,28 +270,6 @@ func openFile(t *testing.T, path string, queue *alerts.Queue) *File {
return f return f
} }
// watch runs f's readAfterChanges until the test ends, and returns the
// channel that sends it changes.
func watch(t *testing.T, f *File) chan<- fsnotify.Event {
t.Helper()
changes := make(chan fsnotify.Event)
ctx, stop := context.WithCancel(t.Context())
stopped := make(chan struct{})
go func() {
f.readAfterChanges(ctx, changes, nil)
close(stopped)
}()
t.Cleanup(func() {
stop()
<-stopped
})
return changes
}
// wantCountry checks the country f gives testClient. // wantCountry checks the country f gives testClient.
func wantCountry(t *testing.T, f *File, want string) { func wantCountry(t *testing.T, f *File, want string) {
t.Helper() t.Helper()
@@ -401,13 +305,3 @@ func writeFile(t *testing.T, path, content string) {
t.Fatalf("write %s: %v", path, err) t.Fatalf("write %s: %v", path, err)
} }
} }
// write writes data to the end of file.
func write(t *testing.T, file *os.File, data []byte) {
t.Helper()
_, err := file.Write(data)
if err != nil {
t.Fatalf("write: %v", err)
}
}
+268 -393
View File
@@ -15,33 +15,12 @@ import (
"sneak.berlin/go/smallwebwaf/internal/alerts" "sneak.berlin/go/smallwebwaf/internal/alerts"
"sneak.berlin/go/smallwebwaf/internal/bans" "sneak.berlin/go/smallwebwaf/internal/bans"
"sneak.berlin/go/smallwebwaf/internal/ratelimit" "sneak.berlin/go/smallwebwaf/internal/ratelimit"
"sneak.berlin/go/smallwebwaf/internal/remotelog"
"sneak.berlin/go/smallwebwaf/internal/reputation" "sneak.berlin/go/smallwebwaf/internal/reputation"
"sneak.berlin/go/smallwebwaf/internal/requestlog" "sneak.berlin/go/smallwebwaf/internal/requestlog"
"sneak.berlin/go/smallwebwaf/internal/rules"
) )
// Params are what New builds the metrics from. Each part is nil while it is
// off, and a part that is nil adds no metric of its own.
type Params struct {
// BusiestCount is how many countries and how many AS numbers get series
// of their own (SWW_WAF_METRICS_BUSIEST_COUNTRIES_AND_ASNS).
BusiestCount int
// InstanceName is the label instance every metric carries
// (SWW_WAF_INSTANCE_NAME).
InstanceName string
// Now returns the time at which Bans counts the bans active.
Now func() time.Time
Bans banLedger
Clients sized
Rules sized
Lists fetchedLists
DNSBL dnsblZones
AbuseIPDB abuseIPDBChecker
Alerts alertQueue
LookupFile lookupDatabase
RemoteLog remoteLogSender
}
// Metrics are sww's metrics. They are safe for concurrent use. // Metrics are sww's metrics. They are safe for concurrent use.
type Metrics struct { type Metrics struct {
// registry gives every metric registered with it the label instance. // registry gives every metric registered with it the label instance.
@@ -57,10 +36,12 @@ type Metrics struct {
rateLimitHits *prometheus.CounterVec rateLimitHits *prometheus.CounterVec
offences *prometheus.CounterVec offences *prometheus.CounterVec
coreRuleSetMatches *prometheus.CounterVec coreRuleSetMatches *prometheus.CounterVec
ruleMatches *prometheus.CounterVec // ruleMatches are made by AddRules, and reputationHits by
reputationHits *prometheus.CounterVec // AddReputation.
countries *busiest ruleMatches *prometheus.CounterVec
asns *busiest reputationHits *prometheus.CounterVec
countries *busiest
asns *busiest
// GeoJSRequests are the requests to GeoJS, and GeoJSFailures those // GeoJSRequests are the requests to GeoJS, and GeoJSFailures those
// that failed. RequestsWithoutGeoJSAnswer are the requests that needed // that failed. RequestsWithoutGeoJSAnswer are the requests that needed
@@ -78,18 +59,20 @@ type Metrics struct {
stateFileEditsSetAside *prometheus.CounterVec stateFileEditsSetAside *prometheus.CounterVec
} }
// New returns the metrics, with the Go runtime's and the process's own, and // New returns the metrics, with the Go runtime's and the process's own.
// those read from each part in params as the metrics are asked for. // busiestCount is how many countries and how many AS numbers get series of
// their own (SWW_WAF_METRICS_BUSIEST_COUNTRIES_AND_ASNS). Every metric
// carries instanceName (SWW_WAF_INSTANCE_NAME) as its label instance.
// //
//nolint:funlen // a few lines for each metric, a list that grows with them //nolint:funlen // a few lines for each metric, a list that grows with them
func New(params Params) *Metrics { func New(busiestCount int, instanceName string) *Metrics {
byStatus := []string{"status_class", "action"} byStatus := []string{"status_class", "action"}
byFile := []string{"file"} byFile := []string{"file"}
registry := prometheus.NewRegistry() registry := prometheus.NewRegistry()
m := &Metrics{ m := &Metrics{
registry: prometheus.WrapRegistererWith( registry: prometheus.WrapRegistererWith(
prometheus.Labels{"instance": params.InstanceName}, registry), prometheus.Labels{"instance": instanceName}, registry),
handler: promhttp.HandlerFor(registry, promhttp.HandlerOpts{}), handler: promhttp.HandlerFor(registry, promhttp.HandlerOpts{}),
inFlight: prometheus.NewGauge(prometheus.GaugeOpts{ inFlight: prometheus.NewGauge(prometheus.GaugeOpts{
Name: "sww_requests_in_flight", Name: "sww_requests_in_flight",
@@ -130,19 +113,8 @@ func New(params Params) *Metrics {
"and the rule's id.", "and the rule's id.",
[]string{"mode", "rule_id"}, []string{"mode", "rule_id"},
), ),
ruleMatches: counterVec( countries: newCountries(busiestCount),
"sww_rule_matches_total", asns: newASNs(busiestCount),
"Requests that matched a rule of the rule files, by its id and action.",
[]string{"rule_id", "action"},
),
reputationHits: counterVec(
"sww_reputation_hits_total",
"Requests whose client a blocklist, a DNSBL zone or AbuseIPDB lists, "+
"by the list's URL, the zone, or abuseipdb.",
[]string{sourceLabel},
),
countries: newCountries(params.BusiestCount),
asns: newASNs(params.BusiestCount),
GeoJSRequests: prometheus.NewCounter(prometheus.CounterOpts{ GeoJSRequests: prometheus.NewCounter(prometheus.CounterOpts{
Name: "sww_geojs_requests_total", Name: "sww_geojs_requests_total",
Help: "Requests to GeoJS.", Help: "Requests to GeoJS.",
@@ -194,7 +166,6 @@ func New(params Params) *Metrics {
m.inFlight, m.requests, m.requestBytes, m.responseBytes, m.inFlight, m.requests, m.requestBytes, m.responseBytes,
m.requestDuration, m.upstreamDuration, m.requestDuration, m.upstreamDuration,
m.rateLimitHits, m.offences, m.coreRuleSetMatches, m.rateLimitHits, m.offences, m.coreRuleSetMatches,
m.ruleMatches, m.reputationHits,
m.countries, m.asns, m.countries, m.asns,
m.GeoJSRequests, m.GeoJSFailures, m.RequestsWithoutGeoJSAnswer, m.GeoJSRequests, m.GeoJSFailures, m.RequestsWithoutGeoJSAnswer,
m.stateFileWrites, m.stateFileWriteFailures, m.stateFileWrites, m.stateFileWriteFailures,
@@ -202,45 +173,261 @@ func New(params Params) *Metrics {
m.stateFileEditsTakenIn, m.stateFileEditsSetAside, m.stateFileEditsTakenIn, m.stateFileEditsSetAside,
) )
if params.Bans != nil {
m.addBans(params.Bans, params.Now)
}
if params.Clients != nil {
m.addClients(params.Clients)
}
if params.Rules != nil {
m.addRules(params.Rules)
}
if params.Lists != nil {
m.addLists(params.Lists)
}
if params.DNSBL != nil {
m.addDNSBL(params.DNSBL)
}
if params.AbuseIPDB != nil {
m.addAbuseIPDB(params.AbuseIPDB)
}
if params.Alerts != nil {
m.addAlerts(params.Alerts)
}
if params.LookupFile != nil {
m.addLookupFile(params.LookupFile)
}
if params.RemoteLog != nil {
m.addRemoteLog(params.RemoteLog)
}
return m return m
} }
// AddBansAndClients adds the metrics read from the ledger and the table
// of clients as the metrics are asked for: the bans made since the start,
// by cause, the bans active and permanent at now, and the clients in the
// table.
func (m *Metrics) AddBansAndClients(
ledger *bans.Ledger, limiter *ratelimit.Limiter, now func() time.Time,
) {
for _, cause := range []string{
bans.CauseLimit, bans.CauseAttack, bans.CauseAdmin, bans.CauseAbuseIPDB,
} {
m.registry.MustRegister(
prometheus.NewCounterFunc(prometheus.CounterOpts{
Name: "sww_bans_made_total",
Help: "Bans made, by cause.",
ConstLabels: prometheus.Labels{"cause": cause},
}, func() float64 {
return float64(ledger.Made(cause))
}),
)
}
m.registry.MustRegister(
prometheus.NewGaugeFunc(prometheus.GaugeOpts{
Name: "sww_active_bans",
Help: "Bans active now, the permanent ones included.",
}, func() float64 {
active, _ := ledger.Count(now())
return float64(active)
}),
prometheus.NewGaugeFunc(prometheus.GaugeOpts{
Name: "sww_permanent_bans",
Help: "Permanent bans not lifted.",
}, func() float64 {
_, permanent := ledger.Count(now())
return float64(permanent)
}),
prometheus.NewGaugeFunc(prometheus.GaugeOpts{
Name: "sww_tracked_clients",
Help: "Clients in the table of clients.",
}, func() float64 {
return float64(limiter.Len())
}),
)
}
// AddRules adds the metrics of the rule files: the requests that matched
// each rule, which RuleMatched counts, and the rules loaded from
// ruleFiles, read as the metrics are asked for. It is called once, before
// RuleMatched.
func (m *Metrics) AddRules(ruleFiles *rules.Files) {
m.ruleMatches = counterVec(
"sww_rule_matches_total",
"Requests that matched a rule of the rule files, by its id and action.",
[]string{"rule_id", "action"},
)
m.registry.MustRegister(m.ruleMatches,
prometheus.NewGaugeFunc(prometheus.GaugeOpts{
Name: "sww_rules_loaded",
Help: "Rules loaded from the rule files.",
}, func() float64 {
return float64(ruleFiles.Len())
}))
}
// AddRemoteLog adds the metrics of sending the log lines to
// SWW_WAF_LOG_REMOTE_URL, read from remote as the metrics are asked for: the
// lines sent, those dropped, and those waiting in the buffer.
func (m *Metrics) AddRemoteLog(remote *remotelog.Sender) {
m.registry.MustRegister(
prometheus.NewCounterFunc(prometheus.CounterOpts{
Name: "sww_remote_log_lines_sent_total",
Help: "Log lines sent to SWW_WAF_LOG_REMOTE_URL.",
}, func() float64 {
return float64(remote.Sent())
}),
prometheus.NewCounterFunc(prometheus.CounterOpts{
Name: "sww_remote_log_lines_dropped_total",
Help: "Log lines dropped: the oldest in a full buffer, and those " +
"whose sending failed.",
}, func() float64 {
return float64(remote.Dropped())
}),
prometheus.NewGaugeFunc(prometheus.GaugeOpts{
Name: "sww_remote_log_lines_waiting",
Help: "Log lines in the buffer, waiting to be sent.",
}, func() float64 {
return float64(remote.Depth())
}),
)
}
// AddLookupFile adds the metrics of the lookup database, read as the
// metrics are asked for: when the file in use was read, which lastRead
// returns, and the replacements of it that could not be read, which
// readFailures returns.
func (m *Metrics) AddLookupFile(
lastRead func() time.Time,
readFailures func() int,
) {
m.registry.MustRegister(
prometheus.NewGaugeFunc(prometheus.GaugeOpts{
Name: "sww_lookup_database_last_read_timestamp_seconds",
Help: "When the lookup database in use was read, in seconds since 1970.",
}, func() float64 {
return float64(lastRead().Unix())
}),
prometheus.NewCounterFunc(prometheus.CounterOpts{
Name: "sww_lookup_database_read_failures_total",
Help: "Replacements of the lookup database that could not be read.",
}, func() float64 {
return float64(readFailures())
}),
)
}
// sourceLabel is the label of the reputation metrics: a list's URL, a
// DNSBL zone, its key masked, or abuseipdb.
const sourceLabel = "source"
// AddReputation adds the metrics of the lists fetched from URLs and of the
// DNSBL zones, by source, each list's URL or each zone, its key masked as
// reputation.MaskZoneKey masks it: the requests whose client a blocklist, a
// zone's verdict or AbuseIPDB's score lists, which ReputationHit counts,
// and, read from lists and dnsbl as the metrics are asked for, for a list,
// the fetches that failed and when the copy in use was fetched, and for a
// zone, the queries made and those that failed. It is called once, before
// ReputationHit.
func (m *Metrics) AddReputation(
lists *reputation.Lists,
dnsbl *reputation.DNSBL,
) {
m.reputationHits = counterVec(
"sww_reputation_hits_total",
"Requests whose client a blocklist, a DNSBL zone or AbuseIPDB lists, "+
"by the list's URL, the zone, or abuseipdb.",
[]string{sourceLabel},
)
m.registry.MustRegister(m.reputationHits)
for _, zone := range dnsbl.Zones() {
source := prometheus.Labels{
sourceLabel: reputation.MaskZoneKey(zone),
}
m.addReputationQueries(
source,
func() int { return dnsbl.Queries(zone) },
)
m.addReputationFailures(
source,
func() int { return dnsbl.Failures(zone) },
)
}
for _, listURL := range lists.URLs() {
source := prometheus.Labels{sourceLabel: listURL}
m.addReputationFailures(
source,
func() int { return lists.Failures(listURL) },
)
m.registry.MustRegister(
prometheus.NewGaugeFunc(prometheus.GaugeOpts{
Name: "sww_reputation_last_fetch_timestamp_seconds",
Help: "When the copy of the list in use was fetched, in seconds since " +
"1970, or 0 while there is none.",
ConstLabels: source,
}, func() float64 {
fetched := lists.Fetched(listURL)
if fetched.IsZero() {
return 0
}
return float64(fetched.Unix())
}),
)
}
}
// AddAbuseIPDB adds the metrics of AbuseIPDB, with the source abuseipdb,
// read from abuseIPDB as the metrics are asked for: the checks made, those
// that failed, and how many checks the day's budget has left. It is
// called once, after AddReputation, while SWW_WAF_ABUSEIPDB_KEY is set.
func (m *Metrics) AddAbuseIPDB(abuseIPDB *reputation.AbuseIPDB) {
source := prometheus.Labels{sourceLabel: reputation.AbuseIPDBSource}
m.addReputationQueries(source, abuseIPDB.Checked)
m.addReputationFailures(source, abuseIPDB.Failures)
m.registry.MustRegister(
prometheus.NewGaugeFunc(prometheus.GaugeOpts{
Name: "sww_reputation_daily_budget_remaining",
Help: "Checks of the day's SWW_WAF_ABUSEIPDB_DAILY_BUDGET not yet spent.",
ConstLabels: source,
}, func() float64 {
return float64(abuseIPDB.BudgetLeft())
}),
)
}
// ReputationHit counts a request whose client source lists: a blocklist,
// by its URL, a DNSBL zone, its key masked, or AbuseIPDB, abuseipdb.
func (m *Metrics) ReputationHit(source string) {
m.reputationHits.WithLabelValues(source).Inc()
}
// AddAlerts adds the metrics of the alerts sent to each destination set,
// read from queue as the metrics are asked for, by destination: the
// alerts sent, the requests to the destination that failed, the alerts
// held back, which are the same for every destination, and those
// dropped. With no destination set, it adds none.
func (m *Metrics) AddAlerts(queue *alerts.Queue) {
for _, name := range queue.DestinationsSet() {
destination := prometheus.Labels{"destination": name}
m.registry.MustRegister(
prometheus.NewCounterFunc(prometheus.CounterOpts{
Name: "sww_alerts_sent_total",
Help: "Alerts the destination took.",
ConstLabels: destination,
}, func() float64 {
return float64(queue.Counts(name).Sent)
}),
prometheus.NewCounterFunc(prometheus.CounterOpts{
Name: "sww_alert_send_failures_total",
Help: "Requests to the destination that failed.",
ConstLabels: destination,
}, func() float64 {
return float64(queue.Counts(name).SendFailures)
}),
prometheus.NewCounterFunc(prometheus.CounterOpts{
Name: "sww_alerts_suppressed_total",
Help: "Alerts held back: repeats within SWW_WAF_ALERT_COOLDOWN, and " +
"alerts past SWW_WAF_ALERT_MAX_PER_HOUR, for the hour's summary.",
ConstLabels: destination,
}, func() float64 {
return float64(queue.Suppressed())
}),
prometheus.NewCounterFunc(prometheus.CounterOpts{
Name: "sww_alerts_dropped_total",
Help: "Alerts dropped, the oldest first, from a full queue, and " +
"alerts given up as the destination refused them.",
ConstLabels: destination,
}, func() float64 {
return float64(queue.Counts(name).Dropped)
}),
)
}
}
// ServeHTTP answers with the metrics in the Prometheus text format. // ServeHTTP answers with the metrics in the Prometheus text format.
func (m *Metrics) ServeHTTP(w http.ResponseWriter, r *http.Request) { func (m *Metrics) ServeHTTP(w http.ResponseWriter, r *http.Request) {
m.handler.ServeHTTP(w, r) m.handler.ServeHTTP(w, r)
@@ -310,12 +497,6 @@ func (m *Metrics) CoreRuleSetMatched(mode string, id int) {
m.coreRuleSetMatches.WithLabelValues(mode, strconv.Itoa(id)).Inc() m.coreRuleSetMatches.WithLabelValues(mode, strconv.Itoa(id)).Inc()
} }
// ReputationHit counts a request whose client source lists: a blocklist,
// by its URL, a DNSBL zone, its key masked, or AbuseIPDB, abuseipdb.
func (m *Metrics) ReputationHit(source string) {
m.reputationHits.WithLabelValues(source).Inc()
}
// StateFileWritten counts a write of the state or cache file name, of size // StateFileWritten counts a write of the state or cache file name, of size
// bytes, that ended with err. // bytes, that ended with err.
func (m *Metrics) StateFileWritten(name string, size int, err error) { func (m *Metrics) StateFileWritten(name string, size int, err error) {
@@ -347,312 +528,6 @@ func (m *Metrics) StateFileEditSetAside(name string) {
m.stateFileEditsSetAside.WithLabelValues(name).Inc() m.stateFileEditsSetAside.WithLabelValues(name).Inc()
} }
// banLedger is what the metrics read from the ledger of bans.
type banLedger interface {
// Made returns the bans made since the start with cause.
Made(cause string) int
// Count returns the bans active at now, the permanent ones included,
// and the permanent ones.
Count(now time.Time) (active, permanent int)
}
// addBans adds the metrics read from ledger as the metrics are asked for:
// the bans made since the start, by cause, and the bans active and
// permanent at now.
func (m *Metrics) addBans(ledger banLedger, now func() time.Time) {
for _, cause := range []string{
bans.CauseLimit, bans.CauseAttack, bans.CauseAdmin, bans.CauseAbuseIPDB,
} {
m.registry.MustRegister(
prometheus.NewCounterFunc(prometheus.CounterOpts{
Name: "sww_bans_made_total",
Help: "Bans made, by cause.",
ConstLabels: prometheus.Labels{"cause": cause},
}, func() float64 {
return float64(ledger.Made(cause))
}),
)
}
m.registry.MustRegister(
prometheus.NewGaugeFunc(prometheus.GaugeOpts{
Name: "sww_active_bans",
Help: "Bans active now, the permanent ones included.",
}, func() float64 {
active, _ := ledger.Count(now())
return float64(active)
}),
prometheus.NewGaugeFunc(prometheus.GaugeOpts{
Name: "sww_permanent_bans",
Help: "Permanent bans not lifted.",
}, func() float64 {
_, permanent := ledger.Count(now())
return float64(permanent)
}),
)
}
// sized is what the metrics read from the table of clients and from the
// rule files.
type sized interface {
// Len returns how many clients the table holds, or how many rules were
// loaded from the rule files.
Len() int
}
// addClients adds the metric of the clients in table, read as the metrics
// are asked for.
func (m *Metrics) addClients(table sized) {
m.registry.MustRegister(
prometheus.NewGaugeFunc(prometheus.GaugeOpts{
Name: "sww_tracked_clients",
Help: "Clients in the table of clients.",
}, func() float64 {
return float64(table.Len())
}),
)
}
// addRules adds the metric of the rules loaded from files, read as the
// metrics are asked for.
func (m *Metrics) addRules(files sized) {
m.registry.MustRegister(
prometheus.NewGaugeFunc(prometheus.GaugeOpts{
Name: "sww_rules_loaded",
Help: "Rules loaded from the rule files.",
}, func() float64 {
return float64(files.Len())
}))
}
// remoteLogSender is what the metrics read from what sends the log lines to
// SWW_WAF_LOG_REMOTE_URL.
type remoteLogSender interface {
// Sent returns the lines sent.
Sent() int64
// Dropped returns the lines dropped.
Dropped() int64
// Depth returns the lines waiting in the buffer.
Depth() int
}
// addRemoteLog adds the metrics of sending the log lines to
// SWW_WAF_LOG_REMOTE_URL, read from remote as the metrics are asked for: the
// lines sent, those dropped, and those waiting in the buffer.
func (m *Metrics) addRemoteLog(remote remoteLogSender) {
m.registry.MustRegister(
prometheus.NewCounterFunc(prometheus.CounterOpts{
Name: "sww_remote_log_lines_sent_total",
Help: "Log lines sent to SWW_WAF_LOG_REMOTE_URL.",
}, func() float64 {
return float64(remote.Sent())
}),
prometheus.NewCounterFunc(prometheus.CounterOpts{
Name: "sww_remote_log_lines_dropped_total",
Help: "Log lines dropped: the oldest in a full buffer, and those " +
"whose sending failed.",
}, func() float64 {
return float64(remote.Dropped())
}),
prometheus.NewGaugeFunc(prometheus.GaugeOpts{
Name: "sww_remote_log_lines_waiting",
Help: "Log lines in the buffer, waiting to be sent.",
}, func() float64 {
return float64(remote.Depth())
}),
)
}
// lookupDatabase is what the metrics read from the lookup database.
type lookupDatabase interface {
// LastRead returns when the file in use was read.
LastRead() time.Time
// ReadFailures returns the replacements of the file that could not be
// read.
ReadFailures() int
}
// addLookupFile adds the metrics of the lookup database, read from file as
// the metrics are asked for: when the file in use was read, and the
// replacements of it that could not be read.
func (m *Metrics) addLookupFile(file lookupDatabase) {
m.registry.MustRegister(
prometheus.NewGaugeFunc(prometheus.GaugeOpts{
Name: "sww_lookup_database_last_read_timestamp_seconds",
Help: "When the lookup database in use was read, in seconds since 1970.",
}, func() float64 {
return float64(file.LastRead().Unix())
}),
prometheus.NewCounterFunc(prometheus.CounterOpts{
Name: "sww_lookup_database_read_failures_total",
Help: "Replacements of the lookup database that could not be read.",
}, func() float64 {
return float64(file.ReadFailures())
}),
)
}
// sourceLabel is the label of the reputation metrics: a list's URL, a
// DNSBL zone, its key masked, or abuseipdb.
const sourceLabel = "source"
// dnsblZones is what the metrics read from what asks the DNSBL zones.
type dnsblZones interface {
// Zones returns the zones asked, their keys unmasked.
Zones() []string
// Queries returns the queries made to zone.
Queries(zone string) int
// Failures returns the queries to zone that failed.
Failures(zone string) int
}
// addDNSBL adds the metrics of each zone dnsbl asks, by source, the zone,
// its key masked as reputation.MaskZoneKey masks it, read as the metrics are
// asked for: the queries made and those that failed.
func (m *Metrics) addDNSBL(dnsbl dnsblZones) {
for _, zone := range dnsbl.Zones() {
source := prometheus.Labels{
sourceLabel: reputation.MaskZoneKey(zone),
}
m.addReputationQueries(
source,
func() int { return dnsbl.Queries(zone) },
)
m.addReputationFailures(
source,
func() int { return dnsbl.Failures(zone) },
)
}
}
// fetchedLists is what the metrics read from the lists fetched from URLs.
type fetchedLists interface {
// URLs returns the lists' URLs.
URLs() []string
// Failures returns the fetches of the list at listURL that failed.
Failures(listURL string) int
// Fetched returns when the copy in use of the list at listURL was
// fetched, or the zero time while there is none.
Fetched(listURL string) time.Time
}
// addLists adds the metrics of each list fetched from a URL, by source, the
// list's URL, read from lists as the metrics are asked for: the fetches
// that failed and when the copy in use was fetched.
func (m *Metrics) addLists(lists fetchedLists) {
for _, listURL := range lists.URLs() {
source := prometheus.Labels{sourceLabel: listURL}
m.addReputationFailures(
source,
func() int { return lists.Failures(listURL) },
)
m.registry.MustRegister(
prometheus.NewGaugeFunc(prometheus.GaugeOpts{
Name: "sww_reputation_last_fetch_timestamp_seconds",
Help: "When the copy of the list in use was fetched, in seconds since " +
"1970, or 0 while there is none.",
ConstLabels: source,
}, func() float64 {
fetched := lists.Fetched(listURL)
if fetched.IsZero() {
return 0
}
return float64(fetched.Unix())
}),
)
}
}
// abuseIPDBChecker is what the metrics read from what checks clients with
// AbuseIPDB.
type abuseIPDBChecker interface {
// Checked returns the checks made.
Checked() int
// Failures returns the checks that failed.
Failures() int
// BudgetLeft returns how many checks the day's budget has left.
BudgetLeft() int
}
// addAbuseIPDB adds the metrics of AbuseIPDB, with the source abuseipdb,
// read from abuseIPDB as the metrics are asked for: the checks made, those
// that failed, and how many checks the day's budget has left.
func (m *Metrics) addAbuseIPDB(abuseIPDB abuseIPDBChecker) {
source := prometheus.Labels{sourceLabel: reputation.AbuseIPDBSource}
m.addReputationQueries(source, abuseIPDB.Checked)
m.addReputationFailures(source, abuseIPDB.Failures)
m.registry.MustRegister(
prometheus.NewGaugeFunc(prometheus.GaugeOpts{
Name: "sww_reputation_daily_budget_remaining",
Help: "Checks of the day's SWW_WAF_ABUSEIPDB_DAILY_BUDGET not yet spent.",
ConstLabels: source,
}, func() float64 {
return float64(abuseIPDB.BudgetLeft())
}),
)
}
// alertQueue is what the metrics read from the queue of alerts.
type alertQueue interface {
// DestinationsSet returns the names of the destinations set.
DestinationsSet() []string
// Counts returns the alerts sent to the destination name, the requests
// to it that failed, and the alerts dropped.
Counts(name string) alerts.Counts
// Suppressed returns the alerts held back.
Suppressed() int64
}
// addAlerts adds the metrics of the alerts sent to each destination set,
// read from queue as the metrics are asked for, by destination: the
// alerts sent, the requests to the destination that failed, the alerts
// held back, which are the same for every destination, and those
// dropped. With no destination set, it adds none.
func (m *Metrics) addAlerts(queue alertQueue) {
for _, name := range queue.DestinationsSet() {
destination := prometheus.Labels{"destination": name}
m.registry.MustRegister(
prometheus.NewCounterFunc(prometheus.CounterOpts{
Name: "sww_alerts_sent_total",
Help: "Alerts the destination took.",
ConstLabels: destination,
}, func() float64 {
return float64(queue.Counts(name).Sent)
}),
prometheus.NewCounterFunc(prometheus.CounterOpts{
Name: "sww_alert_send_failures_total",
Help: "Requests to the destination that failed.",
ConstLabels: destination,
}, func() float64 {
return float64(queue.Counts(name).SendFailures)
}),
prometheus.NewCounterFunc(prometheus.CounterOpts{
Name: "sww_alerts_suppressed_total",
Help: "Alerts held back: repeats within SWW_WAF_ALERT_COOLDOWN, and " +
"alerts past SWW_WAF_ALERT_MAX_PER_HOUR, for the hour's summary.",
ConstLabels: destination,
}, func() float64 {
return float64(queue.Suppressed())
}),
prometheus.NewCounterFunc(prometheus.CounterOpts{
Name: "sww_alerts_dropped_total",
Help: "Alerts dropped, the oldest first, from a full queue, and " +
"alerts given up as the destination refused them.",
ConstLabels: destination,
}, func() float64 {
return float64(queue.Counts(name).Dropped)
}),
)
}
}
// count counts a request that has ended, as RequestEnded and RequestReported // count counts a request that has ended, as RequestEnded and RequestReported
// describe. // describe.
func (m *Metrics) count( func (m *Metrics) count(
-160
View File
@@ -1,160 +0,0 @@
package metrics_test
import (
"net/http"
"net/http/httptest"
"strings"
"testing"
"time"
"sneak.berlin/go/smallwebwaf/internal/alerts"
"sneak.berlin/go/smallwebwaf/internal/metrics"
)
const (
listURL = "https://lists.example/block.txt"
zone = "dnsbl.example"
destination = "webhook"
)
// Each stand-in answers with a number of its own, so that each metric
// shows which part it was read from.
type ledgerStandIn struct{}
func (ledgerStandIn) Made(string) int { return 1 }
func (ledgerStandIn) Count(time.Time) (int, int) { return 2, 1 }
type clientTableStandIn struct{}
func (clientTableStandIn) Len() int { return 3 }
type ruleFilesStandIn struct{}
func (ruleFilesStandIn) Len() int { return 4 }
type listsStandIn struct{}
func (listsStandIn) URLs() []string { return []string{listURL} }
func (listsStandIn) Failures(string) int { return 5 }
func (listsStandIn) Fetched(string) time.Time { return time.Time{} }
type dnsblStandIn struct{}
func (dnsblStandIn) Zones() []string { return []string{zone} }
func (dnsblStandIn) Queries(string) int { return 6 }
func (dnsblStandIn) Failures(string) int { return 0 }
type abuseIPDBStandIn struct{}
func (abuseIPDBStandIn) Checked() int { return 0 }
func (abuseIPDBStandIn) Failures() int { return 0 }
func (abuseIPDBStandIn) BudgetLeft() int { return 7 }
type alertQueueStandIn struct{}
func (alertQueueStandIn) DestinationsSet() []string {
return []string{destination}
}
func (alertQueueStandIn) Counts(string) alerts.Counts {
return alerts.Counts{Sent: 8}
}
func (alertQueueStandIn) Suppressed() int64 { return 0 }
type lookupDatabaseStandIn struct{}
func (lookupDatabaseStandIn) LastRead() time.Time { return time.Time{} }
func (lookupDatabaseStandIn) ReadFailures() int { return 9 }
type remoteLogStandIn struct{}
func (remoteLogStandIn) Sent() int64 { return 10 }
func (remoteLogStandIn) Dropped() int64 { return 0 }
func (remoteLogStandIn) Depth() int { return 0 }
func TestMetricsAreReadFromEachPart(t *testing.T) {
t.Parallel()
m := metrics.New(metrics.Params{
BusiestCount: 1,
InstanceName: "app",
Now: func() time.Time { return time.Time{} },
Bans: ledgerStandIn{},
Clients: clientTableStandIn{},
Rules: ruleFilesStandIn{},
Lists: listsStandIn{},
DNSBL: dnsblStandIn{},
AbuseIPDB: abuseIPDBStandIn{},
Alerts: alertQueueStandIn{},
LookupFile: lookupDatabaseStandIn{},
RemoteLog: remoteLogStandIn{},
})
scraped := scrape(t, m)
for _, line := range []string{
`sww_bans_made_total{cause="limit",instance="app"} 1`,
`sww_active_bans{instance="app"} 2`,
`sww_tracked_clients{instance="app"} 3`,
`sww_rules_loaded{instance="app"} 4`,
`sww_reputation_failures_total{instance="app",source="` + listURL + `"} 5`,
`sww_reputation_queries_total{instance="app",source="` + zone + `"} 6`,
`sww_reputation_daily_budget_remaining{instance="app",source="abuseipdb"} 7`,
`sww_alerts_sent_total{destination="` + destination + `",instance="app"} 8`,
`sww_lookup_database_read_failures_total{instance="app"} 9`,
`sww_remote_log_lines_sent_total{instance="app"} 10`,
} {
if !strings.Contains(scraped, "\n"+line+"\n") {
t.Errorf("metrics\n%s\nwant\n%s", scraped, line)
}
}
}
func TestPartThatIsOffAddsNoMetric(t *testing.T) {
t.Parallel()
scraped := scrape(t, metrics.New(metrics.Params{
BusiestCount: 1,
InstanceName: "app",
}))
for _, name := range []string{
"sww_bans_made_total",
"sww_tracked_clients",
"sww_rules_loaded",
"sww_reputation_",
"sww_alert",
"sww_lookup_database_",
"sww_remote_log_",
} {
if strings.Contains(scraped, name) {
t.Errorf("metrics\n%s\nshow %s", scraped, name)
}
}
}
// scrape returns the metrics m serves.
func scrape(t *testing.T, m *metrics.Metrics) string {
t.Helper()
scraped := httptest.NewRecorder()
m.ServeHTTP(
scraped,
httptest.NewRequestWithContext(
t.Context(), http.MethodGet, "/", http.NoBody,
),
)
return scraped.Body.String()
}
+12 -9
View File
@@ -445,9 +445,12 @@ func TestKeyIsSentInTheKeyHeaderAndNeverShown(t *testing.T) {
scores: map[string]int64{suspect: 100}, scores: map[string]int64{suspect: 100},
} }
checker := newAbuseIPDB(abuseIPDB, p) checker := newAbuseIPDB(abuseIPDB, p)
m := metrics.New(metrics.Params{ m := metrics.New(1, "app")
BusiestCount: 1, InstanceName: instanceName, AbuseIPDB: checker, m.AddReputation(
}) reputation.New(params()),
reputation.NewDNSBL(dnsblParams()),
)
m.AddAbuseIPDB(checker)
// One check that AbuseIPDB answers, and one it refuses with an answer // One check that AbuseIPDB answers, and one it refuses with an answer
// that names the key. // that names the key.
@@ -500,9 +503,12 @@ func TestMetricsCountTheChecksTheFailuresAndTheBudgetLeft(t *testing.T) {
p := abuseIPDBParams() p := abuseIPDBParams()
p.DailyBudget = 5 p.DailyBudget = 5
checker := newAbuseIPDB(abuseIPDB, p) checker := newAbuseIPDB(abuseIPDB, p)
m := metrics.New(metrics.Params{ m := metrics.New(1, "app")
BusiestCount: 1, InstanceName: instanceName, AbuseIPDB: checker, m.AddReputation(
}) reputation.New(params()),
reputation.NewDNSBL(dnsblParams()),
)
m.AddAbuseIPDB(checker)
// One check that AbuseIPDB answers, and one that fails. // One check that AbuseIPDB answers, and one that fails.
hitFrom(t, checker, suspect, true) hitFrom(t, checker, suspect, true)
@@ -805,9 +811,6 @@ func wantBudgetLeft(t *testing.T, checker *reputation.AbuseIPDB, want int) {
} }
} }
// instanceName is the label instance of the metrics the tests build.
const instanceName = "app"
// scrapeMetrics returns the metrics m serves. // scrapeMetrics returns the metrics m serves.
func scrapeMetrics(t *testing.T, m *metrics.Metrics) string { func scrapeMetrics(t *testing.T, m *metrics.Metrics) string {
t.Helper() t.Helper()
+4 -6
View File
@@ -358,9 +358,8 @@ func TestMetricsCountEachZonesQueriesAndThoseThatFailed(t *testing.T) {
"99.2.0.192." + otherZone + ".": {rcode: serverFailure}, "99.2.0.192." + otherZone + ".": {rcode: serverFailure},
}} }}
dnsbl := newDNSBL(resolver, dnsblParams(zone, otherZone)) dnsbl := newDNSBL(resolver, dnsblParams(zone, otherZone))
m := metrics.New(metrics.Params{ m := metrics.New(1, "app")
BusiestCount: 1, InstanceName: instanceName, DNSBL: dnsbl, m.AddReputation(reputation.New(params()), dnsbl)
})
wantZones(t, dnsbl, listed) wantZones(t, dnsbl, listed)
synctest.Wait() synctest.Wait()
@@ -405,9 +404,8 @@ func TestZoneKeyIsMaskedInTheVerdictsTheFailuresAndTheMetrics(t *testing.T) {
"99.2.0.192." + keyed + ".": {addrs: []string{listing}}, "99.2.0.192." + keyed + ".": {addrs: []string{listing}},
"100.2.0.192." + keyed + ".": {rcode: serverFailure}, "100.2.0.192." + keyed + ".": {rcode: serverFailure},
}}, p) }}, p)
m := metrics.New(metrics.Params{ m := metrics.New(1, "app")
BusiestCount: 1, InstanceName: instanceName, DNSBL: dnsbl, m.AddReputation(reputation.New(params()), dnsbl)
})
// Both clients are asked about before either answer comes, so that // Both clients are asked about before either answer comes, so that
// the failure does not keep the zone from the other query. // the failure does not keep the zone from the other query.
+19 -48
View File
@@ -18,12 +18,11 @@ import (
"slices" "slices"
"strings" "strings"
"sync/atomic" "sync/atomic"
"time"
"github.com/fsnotify/fsnotify"
"golang.org/x/net/http/httpguts" "golang.org/x/net/http/httpguts"
"sneak.berlin/go/smallwebwaf/internal/alerts" "sneak.berlin/go/smallwebwaf/internal/alerts"
"sneak.berlin/go/smallwebwaf/internal/dirwatch"
) )
// The actions a rule takes when it matches. // The actions a rule takes when it matches.
@@ -40,12 +39,6 @@ const (
// extension ends the name of every rule file. // extension ends the name of every rule file.
const extension = ".rules" const extension = ".rules"
// quietTime is how long SWW_WAF_RULE_FILES_DIR must go without a change before
// the rule files are read again, so that a file still being written, such
// as one saved in place, appended to or copied in with scp, is read only
// once whole.
const quietTime = 2 * time.Second
// headerTarget starts the target that is one request header, // headerTarget starts the target that is one request header,
// header:<Name>. // header:<Name>.
const headerTarget = "header:" const headerTarget = "header:"
@@ -176,26 +169,18 @@ func (f *Files) Len() int {
} }
// Watch watches Dir until ctx is done, and reads the rule files again // Watch watches Dir until ctx is done, and reads the rule files again
// once Dir has had no change for quietTime, after one is edited, added or // once Dir has had no change for a short time, after one is edited, added
// removed, and after Watch starts watching. If they then hold an error, // or removed, and after Watch starts watching, as dirwatch describes. If
// the rules stay as they were, the error is logged with its file and // they then hold an error, the rules stay as they were, the error is
// line, and the files are read again after the next change. If Dir cannot // logged with its file and line, and the files are read again after the
// be watched, that is logged, and the rules stay as they were loaded. // next change. If Dir cannot be watched, that is logged, and the rules
// While Enabled is false, Watch returns at once. // stay as they were loaded. While Enabled is false, Watch returns at once.
func (f *Files) Watch(ctx context.Context) { func (f *Files) Watch(ctx context.Context) {
if !f.params.Enabled { if !f.params.Enabled {
return return
} }
watcher, err := fsnotify.NewWatcher() watcher, err := dirwatch.New(f.params.Dir)
if err == nil {
defer func() {
_ = watcher.Close()
}()
err = watcher.Add(f.params.Dir)
}
if err != nil { if err != nil {
f.params.ProcessLog.Error("cannot watch the rule files for edits", f.params.ProcessLog.Error("cannot watch the rule files for edits",
"error", err.Error()) "error", err.Error())
@@ -203,36 +188,22 @@ func (f *Files) Watch(ctx context.Context) {
return return
} }
defer func() {
_ = watcher.Close()
}()
f.params.ProcessLog.Info("watching the rule files for edits", f.params.ProcessLog.Info("watching the rule files for edits",
"directory", f.params.Dir) "directory", f.params.Dir)
f.readAfterChanges(ctx, watcher.Events, watcher.Errors) // A change to any file in Dir counts.
} watcher.CallWhenQuiet(ctx,
func(string) bool { return true },
// readAfterChanges reads the rule files again once quietTime has passed f.readAgain,
// without a change from events, until ctx is done, and logs the errors func(err error) {
// from errs. The wait starts at once, as if for a change, so that an edit
// saved after Load read the files, and before Dir was watched, is taken
// in too.
func (f *Files) readAfterChanges(
ctx context.Context, events <-chan fsnotify.Event, errs <-chan error,
) {
quiet := time.NewTimer(quietTime)
defer quiet.Stop()
for {
select {
case <-ctx.Done():
return
case <-events:
quiet.Reset(quietTime)
case <-quiet.C:
f.readAgain()
case err := <-errs:
f.params.ProcessLog.Warn("watching the rule files failed", f.params.ProcessLog.Warn("watching the rule files failed",
"error", err.Error()) "error", err.Error())
} },
} )
} }
// readAgain reads the rule files again, in place of the rules loaded, or // readAgain reads the rule files again, in place of the rules loaded, or
-165
View File
@@ -1,165 +0,0 @@
package rules
import (
"context"
"log/slog"
"net/http"
"net/http/httptest"
"os"
"path/filepath"
"slices"
"testing"
"testing/synctest"
"time"
"github.com/fsnotify/fsnotify"
"sneak.berlin/go/smallwebwaf/internal/alerts"
)
// The tests below run readAfterChanges in a synctest bubble, where time is
// a clock of the test's own: time.Sleep moves it on at once, and
// synctest.Wait returns once readAfterChanges waits again, so that every
// reading due by then is done. The test sends the changes itself, as the
// watch of a directory cannot run in a bubble.
func TestFileWrittenInTwoPartsTakenInOnlyWhole(t *testing.T) {
t.Parallel()
synctest.Test(t, func(t *testing.T) {
dir := t.TempDir()
path := filepath.Join(dir, "50-app.rules")
writeFile(t, path, "first path block ^/first\n")
files := load(t, dir)
changes := run(t, files)
file, err := os.Create(path) //nolint:gosec // a file the test wrote
if err != nil {
t.Fatalf("create: %v", err)
}
defer func() {
_ = file.Close()
}()
// The first part ends in the middle of a ban rule's regex, which,
// read then, would ban every request.
write(t, file, "first path block ^/first\nprobe path ban ^/")
changes <- fsnotify.Event{Name: path, Op: fsnotify.Write}
time.Sleep(quietTime - time.Nanosecond)
synctest.Wait()
wantMatched(t, files, "/anything")
// The second part starts the wait again.
write(t, file, `\.env$`+"\n")
changes <- fsnotify.Event{Name: path, Op: fsnotify.Write}
time.Sleep(quietTime - time.Nanosecond)
synctest.Wait()
wantMatched(t, files, "/.env")
time.Sleep(time.Nanosecond)
synctest.Wait()
wantMatched(t, files, "/.env", "probe")
wantMatched(t, files, "/anything")
})
}
func TestEditSavedBeforeTheWatchStartsTakenIn(t *testing.T) {
t.Parallel()
synctest.Test(t, func(t *testing.T) {
dir := t.TempDir()
path := filepath.Join(dir, "50-app.rules")
writeFile(t, path, "first path block ^/first\n")
files := load(t, dir)
// Saved after Load read the files, and before the directory was
// watched, so that no change is seen for it.
writeFile(t, path, "first path block ^/edited\n")
run(t, files)
time.Sleep(quietTime)
synctest.Wait()
wantMatched(t, files, "/edited", "first")
})
}
// load loads the rules in dir.
func load(t *testing.T, dir string) *Files {
t.Helper()
files, err := Load(Params{
Dir: dir, Enabled: true, ProcessLog: slog.New(slog.DiscardHandler),
Alerts: alerts.New(alerts.Params{}),
})
if err != nil {
t.Fatalf("load: %v", err)
}
return files
}
// run runs files' readAfterChanges until the test ends, and returns the
// channel that sends it changes.
func run(t *testing.T, files *Files) chan<- fsnotify.Event {
t.Helper()
changes := make(chan fsnotify.Event)
ctx, stop := context.WithCancel(t.Context())
stopped := make(chan struct{})
go func() {
files.readAfterChanges(ctx, changes, nil)
close(stopped)
}()
t.Cleanup(func() {
stop()
<-stopped
})
return changes
}
// writeFile writes content to the file at path.
func writeFile(t *testing.T, path, content string) {
t.Helper()
err := os.WriteFile(path, []byte(content), 0o600)
if err != nil {
t.Fatalf("write %s: %v", path, err)
}
}
// write writes text to the end of file.
func write(t *testing.T, file *os.File, text string) {
t.Helper()
_, err := file.WriteString(text)
if err != nil {
t.Fatalf("write: %v", err)
}
}
// wantMatched checks the ids of the rules that a GET request for path
// matches, in order.
func wantMatched(t *testing.T, files *Files, path string, want ...string) {
t.Helper()
r := httptest.NewRequestWithContext(t.Context(), http.MethodGet,
"http://app.example"+path, nil)
matched := files.Match(r)
got := make([]string, 0, len(matched))
for _, rule := range matched {
got = append(got, rule.ID)
}
if !slices.Equal(got, want) {
t.Errorf("%s matched %v, want %v", path, got, want)
}
}
+1 -1
View File
@@ -2430,7 +2430,7 @@ func midnight() time.Time {
// counts, with thresholds fill does not reach. // counts, with thresholds fill does not reach.
func newParams(dir string) state.Params { func newParams(dir string) state.Params {
discard := slog.New(slog.DiscardHandler) discard := slog.New(slog.DiscardHandler)
m := metrics.New(metrics.Params{BusiestCount: 1, InstanceName: "app"}) m := metrics.New(1, "app")
queue := alerts.New(alerts.Params{ queue := alerts.New(alerts.Params{
WebhookURL: &url.URL{Scheme: "https", Host: "alerts.example"}, WebhookURL: &url.URL{Scheme: "https", Host: "alerts.example"},
Events: alerts.Events(), Events: alerts.Events(),
+11 -18
View File
@@ -392,33 +392,26 @@ func newMetrics(
alertQueue *alerts.Queue, lookupFile *lookup.File, alertQueue *alerts.Queue, lookupFile *lookup.File,
remote *remotelog.Sender, remote *remotelog.Sender,
) *metrics.Metrics { ) *metrics.Metrics {
params := metrics.Params{ m := metrics.New(cfg.MetricsBusiestCountriesAndASNs, cfg.InstanceName)
BusiestCount: cfg.MetricsBusiestCountriesAndASNs, m.AddBansAndClients(ledger, limiter, utcNow)
InstanceName: cfg.InstanceName, m.AddRules(ruleFiles)
Now: utcNow, m.AddReputation(lists, dnsbl)
Bans: ledger,
Clients: limiter, if cfg.AbuseIPDBKey != "" {
Rules: ruleFiles, m.AddAbuseIPDB(abuseIPDB)
Lists: lists,
DNSBL: dnsbl,
Alerts: alertQueue,
} }
// A nil pointer in an interface field is not a nil interface, so a part m.AddAlerts(alertQueue)
// that is off is left out rather than set to nil.
if cfg.AbuseIPDBKey != "" {
params.AbuseIPDB = abuseIPDB
}
if lookupFile != nil { if lookupFile != nil {
params.LookupFile = lookupFile m.AddLookupFile(lookupFile.LastRead, lookupFile.ReadFailures)
} }
if remote != nil { if remote != nil {
params.RemoteLog = remote m.AddRemoteLog(remote)
} }
return metrics.New(params) return m
} }
// newGeoJS returns what looks clients up at GeoJS, none looked up yet. Each // newGeoJS returns what looks clients up at GeoJS, none looked up yet. Each
+20 -27
View File
@@ -548,11 +548,16 @@ func addCoreRuleSet(t *testing.T, params *waf.Params) {
params.CoreRuleSet = coreRuleSet params.CoreRuleSet = coreRuleSet
} }
// addCounters adds to params the table of clients, the ledger and the // addCounters adds to params the metrics, the table of clients, the ledger
// anomaly counters, as sww builds them from params' settings. // and the anomaly counters, as sww builds them from params' settings. The
// metrics read the ledger, the table of clients and the rules.
func addCounters(params *waf.Params) { func addCounters(params *waf.Params) {
cfg := params.Config cfg := params.Config
params.Metrics = metrics.New(
cfg.MetricsBusiestCountriesAndASNs,
cfg.InstanceName,
)
params.Limiter = ratelimit.New(ratelimit.Params{ params.Limiter = ratelimit.New(ratelimit.Params{
Limits: ratelimit.Limits{ Limits: ratelimit.Limits{
PerMinute: cfg.RateLimitPerMinute, PerMinute: cfg.RateLimitPerMinute,
@@ -582,18 +587,26 @@ func addCounters(params *waf.Params) {
NamedNetblocks: cfg.AnomalyWatchedNets, NamedNetblocks: cfg.AnomalyWatchedNets,
Alerts: params.Alerts, Alerts: params.Alerts,
}) })
params.Metrics.AddBansAndClients(
params.Ledger,
params.Limiter,
params.Now,
)
params.Metrics.AddRules(params.Rules)
} }
// addOutsideSources adds to params the lists fetched from URLs, the DNSBL // addOutsideSources adds to params the lists fetched from URLs, the DNSBL
// zones' verdicts, AbuseIPDB's scores from abuseIPDBCheckURL and its // zones' verdicts, AbuseIPDB's scores from abuseIPDBCheckURL and its
// blacklist from abuseIPDBBlacklistURL, routewatch's answers while its // blacklist from abuseIPDBBlacklistURL, routewatch's answers while its
// setting is set, the metrics newMetrics returns, and GeoJS at geojsURL, as // setting is set, and GeoJS at geojsURL, as sww builds
// sww builds them from params' settings and the parts addCounters added. // them from params' settings and the parts addCounters added. The metrics
// read the lists, the zones' verdicts, and AbuseIPDB's while its key is set.
func addOutsideSources( func addOutsideSources(
params *waf.Params, geojsURL, abuseIPDBCheckURL string, params *waf.Params, geojsURL, abuseIPDBCheckURL string,
) { ) {
cfg := params.Config cfg := params.Config
limiter, ledger := params.Limiter, params.Ledger m, limiter, ledger := params.Metrics, params.Limiter, params.Ledger
processLog, alertQueue := params.ProcessLog, params.Alerts processLog, alertQueue := params.ProcessLog, params.Alerts
params.Lists = reputation.New(reputation.Params{ params.Lists = reputation.New(reputation.Params{
@@ -642,9 +655,6 @@ func addOutsideSources(
}) })
} }
m := newMetrics(params)
params.Metrics = m
params.GeoJS = geojs.New(geojs.Params{ params.GeoJS = geojs.New(geojs.Params{
URL: geojsURL, URL: geojsURL,
Answered: func(answer geojs.Answer) { Answered: func(answer geojs.Answer) {
@@ -657,29 +667,12 @@ func addOutsideSources(
}, },
WentWithoutAnswer: m.RequestsWithoutGeoJSAnswer.Inc, WentWithoutAnswer: m.RequestsWithoutGeoJSAnswer.Inc,
}) })
}
// newMetrics returns the metrics, as sww builds them from params' settings, m.AddReputation(params.Lists, params.DNSBL)
// read from the ledger, the table of clients, the rules, the lists, the
// zones' verdicts, and AbuseIPDB's while its key is set.
func newMetrics(params *waf.Params) *metrics.Metrics {
cfg := params.Config
metricsParams := metrics.Params{
BusiestCount: cfg.MetricsBusiestCountriesAndASNs,
InstanceName: cfg.InstanceName,
Now: params.Now,
Bans: params.Ledger,
Clients: params.Limiter,
Rules: params.Rules,
Lists: params.Lists,
DNSBL: params.DNSBL,
}
if cfg.AbuseIPDBKey != "" { if cfg.AbuseIPDBKey != "" {
metricsParams.AbuseIPDB = params.AbuseIPDB m.AddAbuseIPDB(params.AbuseIPDB)
} }
return metrics.New(metricsParams)
} }
// newClient returns an HTTP client that sends requests as they are made, // newClient returns an HTTP client that sends requests as they are made,