mirror of
https://github.com/akvorado/akvorado.git
synced 2025-12-11 22:14:02 +01:00
756e4a8fbd
The concurrency of this library is easier to handle than Sarama. Notably, it is more compatible with the new model of "almost share nothing" we use for the inlet and the outlet. The lock for workers in outlet is removed. We can now use sync.Pool to allocate slice of bytes in inlet. It may also be more performant. In the future, we may want to commit only when pushing data to ClickHouse. However, this does not seem easy when there is a rebalance. In case of rebalance, we need to do something when a partition is revoked to avoid duplicating data. For example, we could flush the current batch to ClickHouse. Have a look at the `example/mark_offsets/main.go` file in franz-go repository for a possible approach. In the meantime, we rely on autocommit. Another contender could be https://github.com/segmentio/kafka-go. Also see https://github.com/twmb/franz-go/pull/1064.
108 lines
2.2 KiB
Go
108 lines
2.2 KiB
Go
// SPDX-FileCopyrightText: 2022 Free Mobile
|
|
// SPDX-License-Identifier: AGPL-3.0-only
|
|
|
|
package flow
|
|
|
|
import (
|
|
"bytes"
|
|
"fmt"
|
|
"path"
|
|
"runtime"
|
|
"sync"
|
|
"testing"
|
|
"time"
|
|
|
|
"akvorado/common/daemon"
|
|
"akvorado/common/helpers"
|
|
"akvorado/common/httpserver"
|
|
kafkaCommon "akvorado/common/kafka"
|
|
"akvorado/common/pb"
|
|
"akvorado/common/reporter"
|
|
"akvorado/inlet/flow/input/file"
|
|
"akvorado/inlet/kafka"
|
|
|
|
"github.com/twmb/franz-go/pkg/kgo"
|
|
)
|
|
|
|
func TestFlow(t *testing.T) {
|
|
_, src, _, _ := runtime.Caller(0)
|
|
base := path.Join(path.Dir(src), "input", "file", "testdata")
|
|
paths := []string{
|
|
path.Join(base, "file1.txt"),
|
|
path.Join(base, "file2.txt"),
|
|
}
|
|
|
|
inputs := []InputConfiguration{
|
|
{
|
|
Config: &file.Configuration{
|
|
Paths: paths,
|
|
MaxFlows: 100,
|
|
},
|
|
},
|
|
}
|
|
|
|
r := reporter.NewMock(t)
|
|
config := DefaultConfiguration()
|
|
config.Inputs = inputs
|
|
|
|
producer, cluster := kafka.NewMock(t, r, kafka.DefaultConfiguration())
|
|
defer cluster.Close()
|
|
|
|
// Use the new helper to intercept messages
|
|
var mu sync.Mutex
|
|
helloCount := 0
|
|
byeCount := 0
|
|
totalCount := 0
|
|
done := make(chan bool)
|
|
|
|
kafkaCommon.InterceptMessages(t, cluster, func(record *kgo.Record) {
|
|
mu.Lock()
|
|
defer mu.Unlock()
|
|
|
|
// Check topic
|
|
expectedTopic := fmt.Sprintf("flows-v%d", pb.Version)
|
|
if record.Topic != expectedTopic {
|
|
t.Errorf("Expected topic %s, got %s", expectedTopic, record.Topic)
|
|
return
|
|
}
|
|
|
|
// Count messages based on content
|
|
if bytes.Contains(record.Value, []byte("hello world!")) {
|
|
helloCount++
|
|
} else if bytes.Contains(record.Value, []byte("bye bye")) {
|
|
byeCount++
|
|
}
|
|
|
|
totalCount++
|
|
if totalCount >= 100 {
|
|
close(done)
|
|
}
|
|
})
|
|
|
|
c, err := New(r, config, Dependencies{
|
|
Daemon: daemon.NewMock(t),
|
|
HTTP: httpserver.NewMock(t, r),
|
|
Kafka: producer,
|
|
})
|
|
if err != nil {
|
|
t.Fatalf("New() error:\n%+v", err)
|
|
}
|
|
helpers.StartStop(t, c)
|
|
|
|
// Wait for flows
|
|
select {
|
|
case <-done:
|
|
// Check that we got the expected number of each message type
|
|
mu.Lock()
|
|
if helloCount != 50 {
|
|
t.Errorf("Expected 50 'hello world!' messages, got %d", helloCount)
|
|
}
|
|
if byeCount != 50 {
|
|
t.Errorf("Expected 50 'bye bye' messages, got %d", byeCount)
|
|
}
|
|
mu.Unlock()
|
|
case <-time.After(time.Second):
|
|
t.Fatalf("flows not received")
|
|
}
|
|
}
|