summaryrefslogtreecommitdiff
path: root/pkg/domain/infra/tunnel/containers.go
diff options
context:
space:
mode:
authorValentin Rothberg <rothberg@redhat.com>2020-09-09 16:35:38 +0200
committerValentin Rothberg <rothberg@redhat.com>2020-09-10 10:06:48 +0200
commit5c44ebf96e0a90a83cb115fa6fb7013b53dcebd8 (patch)
treeb4919c9c776ece0b8fcbf7d20201fca379c9e623 /pkg/domain/infra/tunnel/containers.go
parent6b1a1fcc5cb92a9fd5800b0d1af44f26093a8153 (diff)
downloadpodman-5c44ebf96e0a90a83cb115fa6fb7013b53dcebd8.tar.gz
podman-5c44ebf96e0a90a83cb115fa6fb7013b53dcebd8.tar.bz2
podman-5c44ebf96e0a90a83cb115fa6fb7013b53dcebd8.zip
remote run: consult events for exit code
After attaching to a container, we wait for the container to finish and return its exit code. Waiting for the container may not always succeed, for instance, when the container has been force removed by another process. In such case, we have to look at the *last* container-exit event. Also refactor the `ContainerRun` method a bit to return early on errors and de-spaghetti the code. Enable the remote-disabled system test. Fixes: #7117 Signed-off-by: Valentin Rothberg <rothberg@redhat.com>
Diffstat (limited to 'pkg/domain/infra/tunnel/containers.go')
-rw-r--r--pkg/domain/infra/tunnel/containers.go93
1 files changed, 76 insertions, 17 deletions
diff --git a/pkg/domain/infra/tunnel/containers.go b/pkg/domain/infra/tunnel/containers.go
index cc919561f..062b38a70 100644
--- a/pkg/domain/infra/tunnel/containers.go
+++ b/pkg/domain/infra/tunnel/containers.go
@@ -8,11 +8,13 @@ import (
"os"
"strconv"
"strings"
+ "sync"
"time"
"github.com/containers/common/pkg/config"
"github.com/containers/image/v5/docker/reference"
"github.com/containers/podman/v2/libpod/define"
+ "github.com/containers/podman/v2/libpod/events"
"github.com/containers/podman/v2/pkg/api/handlers"
"github.com/containers/podman/v2/pkg/bindings"
"github.com/containers/podman/v2/pkg/bindings/containers"
@@ -507,33 +509,90 @@ func (ic *ContainerEngine) ContainerRun(ctx context.Context, opts entities.Conta
for _, w := range con.Warnings {
fmt.Fprintf(os.Stderr, "%s\n", w)
}
+
report := entities.ContainerRunReport{Id: con.ID}
- // Attach
- if !opts.Detach {
- err = startAndAttach(ic, con.ID, &opts.DetachKeys, opts.InputStream, opts.OutputStream, opts.ErrorStream)
- if err == nil {
- exitCode, err := containers.Wait(ic.ClientCxt, con.ID, nil)
- if err == nil {
- report.ExitCode = int(exitCode)
- }
+
+ if opts.Detach {
+ // Detach and return early
+ err := containers.Start(ic.ClientCxt, con.ID, nil)
+ if err != nil {
+ report.ExitCode = define.ExitCode(err)
}
- } else {
- err = containers.Start(ic.ClientCxt, con.ID, nil)
+ return &report, err
}
- if err != nil {
+
+ // Attach
+ if err := startAndAttach(ic, con.ID, &opts.DetachKeys, opts.InputStream, opts.OutputStream, opts.ErrorStream); err != nil {
report.ExitCode = define.ExitCode(err)
+ if opts.Rm {
+ if rmErr := containers.Remove(ic.ClientCxt, con.ID, bindings.PFalse, bindings.PTrue); rmErr != nil {
+ logrus.Debugf("unable to remove container %s after failing to start and attach to it", con.ID)
+ }
+ }
+ return &report, err
}
+
if opts.Rm {
- if err := containers.Remove(ic.ClientCxt, con.ID, bindings.PFalse, bindings.PTrue); err != nil {
- if errors.Cause(err) == define.ErrNoSuchCtr ||
- errors.Cause(err) == define.ErrCtrRemoved {
- logrus.Warnf("Container %s does not exist: %v", con.ID, err)
- } else {
- logrus.Errorf("Error removing container %s: %v", con.ID, err)
+ // Defer the removal, so we can return early if needed and
+ // de-spaghetti the code.
+ defer func() {
+ if err := containers.Remove(ic.ClientCxt, con.ID, bindings.PFalse, bindings.PTrue); err != nil {
+ if errors.Cause(err) == define.ErrNoSuchCtr ||
+ errors.Cause(err) == define.ErrCtrRemoved {
+ logrus.Warnf("Container %s does not exist: %v", con.ID, err)
+ } else {
+ logrus.Errorf("Error removing container %s: %v", con.ID, err)
+ }
}
+ }()
+ }
+
+ // Wait
+ exitCode, waitErr := containers.Wait(ic.ClientCxt, con.ID, nil)
+ if waitErr == nil {
+ report.ExitCode = int(exitCode)
+ return &report, nil
+ }
+
+ // Determine why the wait failed. If the container doesn't exist,
+ // consult the events.
+ if !strings.Contains(waitErr.Error(), define.ErrNoSuchCtr.Error()) {
+ return &report, waitErr
+ }
+
+ // Events
+ eventsChannel := make(chan *events.Event)
+ eventOptions := entities.EventsOptions{
+ EventChan: eventsChannel,
+ Filter: []string{
+ "type=container",
+ fmt.Sprintf("container=%s", con.ID),
+ fmt.Sprintf("event=%s", events.Exited),
+ },
+ }
+
+ var lastEvent *events.Event
+ var mutex sync.Mutex
+ mutex.Lock()
+ // Read the events.
+ go func() {
+ for e := range eventsChannel {
+ lastEvent = e
}
+ mutex.Unlock()
+ }()
+
+ eventsErr := ic.Events(ctx, eventOptions)
+
+ // Wait for all events to be read
+ mutex.Lock()
+ if eventsErr != nil || lastEvent == nil {
+ logrus.Errorf("Cannot get exit code: %v", err)
+ report.ExitCode = define.ExecErrorCodeNotFound
+ return &report, nil // compat with local client
}
+ report.ExitCode = lastEvent.ContainerExitCode
return &report, err
}