moby/daemon/logger/adapter_test.go

217 lines
4.4 KiB
Go
Raw Normal View History

package logger // import "github.com/docker/docker/daemon/logger"
import (
"encoding/binary"
"io"
"sync"
"testing"
"time"
"github.com/docker/docker/api/types/plugins/logdriver"
protoio "github.com/gogo/protobuf/io"
"gotest.tools/v3/assert"
is "gotest.tools/v3/assert/cmp"
)
// mockLoggingPlugin implements the loggingPlugin interface for testing purposes
// it only supports a single log stream
type mockLoggingPlugin struct {
io.WriteCloser
inStream io.Reader
logs []*logdriver.LogEntry
c *sync.Cond
err error
}
func newMockLoggingPlugin() *mockLoggingPlugin {
r, w := io.Pipe()
return &mockLoggingPlugin{
WriteCloser: w,
inStream: r,
logs: []*logdriver.LogEntry{},
c: sync.NewCond(new(sync.Mutex)),
}
}
func (l *mockLoggingPlugin) StartLogging(file string, info Info) error {
go func() {
dec := protoio.NewUint32DelimitedReader(l.inStream, binary.BigEndian, 1e6)
for {
var msg logdriver.LogEntry
if err := dec.ReadMsg(&msg); err != nil {
l.c.L.Lock()
if l.err == nil {
l.err = err
}
l.c.L.Unlock()
l.c.Broadcast()
return
}
l.c.L.Lock()
l.logs = append(l.logs, &msg)
l.c.L.Unlock()
l.c.Broadcast()
}
}()
return nil
}
func (l *mockLoggingPlugin) StopLogging(file string) error {
l.c.L.Lock()
if l.err == nil {
l.err = io.EOF
}
l.c.L.Unlock()
l.c.Broadcast()
return nil
}
func (l *mockLoggingPlugin) Capabilities() (cap Capability, err error) {
return Capability{ReadLogs: true}, nil
}
func (l *mockLoggingPlugin) ReadLogs(info Info, config ReadConfig) (io.ReadCloser, error) {
r, w := io.Pipe()
go func() {
var idx int
enc := logdriver.NewLogEntryEncoder(w)
l.c.L.Lock()
defer l.c.L.Unlock()
for {
if l.err != nil {
w.Close()
return
}
if idx >= len(l.logs) {
if !config.Follow {
w.Close()
return
}
l.c.Wait()
continue
}
if err := enc.Encode(l.logs[idx]); err != nil {
w.CloseWithError(err)
return
}
idx++
}
}()
return r, nil
}
func (l *mockLoggingPlugin) waitLen(i int) {
l.c.L.Lock()
defer l.c.L.Unlock()
for len(l.logs) < i {
l.c.Wait()
}
}
func (l *mockLoggingPlugin) check(t *testing.T) {
if l.err != nil && l.err != io.EOF {
t.Fatal(l.err)
}
}
func newMockPluginAdapter(plugin *mockLoggingPlugin) Logger {
enc := logdriver.NewLogEntryEncoder(plugin)
a := &pluginAdapterWithRead{
&pluginAdapter{
plugin: plugin,
stream: plugin,
enc: enc,
},
}
a.plugin.StartLogging("", Info{})
return a
}
func TestAdapterReadLogs(t *testing.T) {
plugin := newMockLoggingPlugin()
l := newMockPluginAdapter(plugin)
testMsg := []Message{
{Line: []byte("Are you the keymaker?"), Timestamp: time.Now()},
{Line: []byte("Follow the white rabbit"), Timestamp: time.Now()},
}
for _, msg := range testMsg {
m := msg.copy()
assert.Check(t, l.Log(m))
}
// Wait until messages are read into plugin
plugin.waitLen(len(testMsg))
lr, ok := l.(LogReader)
assert.Check(t, ok, "Logger does not implement LogReader")
lw := lr.ReadLogs(ReadConfig{})
for _, x := range testMsg {
select {
case msg := <-lw.Msg:
testMessageEqual(t, &x, msg)
case <-time.After(10 * time.Second):
t.Fatal("timeout reading logs")
}
}
select {
case _, ok := <-lw.Msg:
assert.Check(t, !ok, "expected message channel to be closed")
case <-time.After(10 * time.Second):
t.Fatal("timeout waiting for message channel to close")
}
daemon.ContainerLogs(): fix resource leak on follow When daemon.ContainerLogs() is called with options.follow=true (as in "docker logs --follow"), the "loggerutils.followLogs()" function never returns (even then the logs consumer is gone). As a result, all the resources associated with it (including an opened file descriptor for the log file being read, two FDs for a pipe, and two FDs for inotify watch) are never released. If this is repeated (such as by running "docker logs --follow" and pressing Ctrl-C a few times), this results in DoS caused by either hitting the limit of inotify watches, or the limit of opened files. The only cure is daemon restart. Apparently, what happens is: 1. logs producer (a container) is gone, calling (*LogWatcher).Close() for all its readers (daemon/logger/jsonfilelog/jsonfilelog.go:175). 2. WatchClose() is properly handled by a dedicated goroutine in followLogs(), cancelling the context. 3. Upon receiving the ctx.Done(), the code in followLogs() (daemon/logger/loggerutils/logfile.go#L626-L638) keeps to send messages _synchronously_ (which is OK for now). 4. Logs consumer is gone (Ctrl-C is pressed on a terminal running "docker logs --follow"). Method (*LogWatcher).Close() is properly called (see daemon/logs.go:114). Since it was called before and due to to once.Do(), nothing happens (which is kinda good, as otherwise it will panic on closing a closed channel). 5. A goroutine (see item 3 above) keeps sending log messages synchronously to the logWatcher.Msg channel. Since the channel reader is gone, the channel send operation blocks forever, and resource cleanup set up in defer statements at the beginning of followLogs() never happens. Alas, the fix is somewhat complicated: 1. Distinguish between close from logs producer and logs consumer. To that effect, - yet another channel is added to LogWatcher(); - {Watch,}Close() are renamed to {Watch,}ProducerGone(); - {Watch,}ConsumerGone() are added; *NOTE* that ProducerGone()/WatchProducerGone() pair is ONLY needed in order to stop ConsumerLogs(follow=true) when a container is stopped; otherwise we're not interested in it. In other words, we're only using it in followLogs(). 2. Code that was doing (logWatcher*).Close() is modified to either call ProducerGone() or ConsumerGone(), depending on the context. 3. Code that was waiting for WatchClose() is modified to wait for either ConsumerGone() or ProducerGone(), or both, depending on the context. 4. followLogs() are modified accordingly: - context cancellation is happening on WatchProducerGone(), and once it's received the FileWatcher is closed and waitRead() returns errDone on EOF (i.e. log rotation handling logic is disabled); - due to this, code that was writing synchronously to logWatcher.Msg can be and is removed as the code above it handles this case; - function returns once ConsumerGone is received, freeing all the resources -- this is the bugfix itself. While at it, 1. Let's also remove the ctx usage to simplify the code a bit. It was introduced by commit a69a59ffc7e3d ("Decouple removing the fileWatcher from reading") in order to fix a bug. The bug was actually a deadlock in fsnotify, and the fix was just a workaround. Since then the fsnofify bug has been fixed, and a new fsnotify was vendored in. For more details, please see https://github.com/moby/moby/pull/27782#issuecomment-416794490 2. Since `(*filePoller).Close()` is fixed to remove all the files being watched, there is no need to explicitly call fileWatcher.Remove(name) anymore, so get rid of the extra code. Should fix https://github.com/moby/moby/issues/37391 Signed-off-by: Kir Kolyshkin <kolyshkin@gmail.com>
2018-08-01 04:03:55 +00:00
lw.ProducerGone()
lw = lr.ReadLogs(ReadConfig{Follow: true})
for _, x := range testMsg {
select {
case msg := <-lw.Msg:
testMessageEqual(t, &x, msg)
case <-time.After(10 * time.Second):
t.Fatal("timeout reading logs")
}
}
x := Message{Line: []byte("Too infinity and beyond!"), Timestamp: time.Now()}
assert.Check(t, l.Log(x.copy()))
select {
case msg, ok := <-lw.Msg:
assert.Check(t, ok, "message channel unexpectedly closed")
testMessageEqual(t, &x, msg)
case <-time.After(10 * time.Second):
t.Fatal("timeout reading logs")
}
l.Close()
select {
case msg, ok := <-lw.Msg:
assert.Check(t, !ok, "expected message channel to be closed")
assert.Check(t, is.Nil(msg))
case <-time.After(10 * time.Second):
t.Fatal("timeout waiting for logger to close")
}
plugin.check(t)
}
func testMessageEqual(t *testing.T, a, b *Message) {
assert.Check(t, is.DeepEqual(a.Line, b.Line))
assert.Check(t, is.DeepEqual(a.Timestamp.UnixNano(), b.Timestamp.UnixNano()))
assert.Check(t, is.Equal(a.Source, b.Source))
}