Restart the engine within the docker container when appropriate.

Restart the engine within the docker container when appropriate.

It's possible for the engine to exit successfully right now (when a
connection is lost). The current state of start_engine.sh will cause
the Docker container to exit in this case, which isn't desired behavior.

This patch changes the behavior so that the engine will continuously be
restarted within the Docker container so long as it exits cleanly. If the
engine exits with a nonzero code (or if stunnel bombs out), the Docker
container will be torn down.

BUG=598933
Committed: https://crrev.com/7e8f69e15bd7e054e4c3cbb5e6b70870adb1951e
Cr-Commit-Position: refs/heads/master@{#384139}

[marcinjb, 2016-03-30 00:43:39]

Description was changed from

==========
Restart the engine within the docker container when appropriate.

It's possible for the engine to exit successfully right now (when a
connection is lost). The current state of start_engine.sh will cause
the Docker container to exit in this case, which isn't desired behavior.

This patch changes the behavior so that the engine will continuously be
restarted within the Docker container so long as it exits cleanly. If the
engine exits with a nonzero code (or if stunnel bombs out), the Docker
container will be torn down.

BUG=
==========

to

==========
Restart the engine within the docker container when appropriate.

It's possible for the engine to exit successfully right now (when a
connection is lost). The current state of start_engine.sh will cause
the Docker container to exit in this case, which isn't desired behavior.

This patch changes the behavior so that the engine will continuously be
restarted within the Docker container so long as it exits cleanly. If the
engine exits with a nonzero code (or if stunnel bombs out), the Docker
container will be torn down.

BUG=598933
==========

[marcinjb, 2016-03-30 22:45:34]

marcinjb@chromium.org changed reviewers:
+ kmarshall@chromium.org, sriramsr@chromium.org

[marcinjb, 2016-03-30 22:58:11]

FYI: I tested this locally.  I ran a docker container with stunnel and the like,
and had the linux client connect, disconnect, and reconnect.  The container was
still intact after this.

[Kevin M, 2016-03-30 22:59:26]

lgtm

[Sriram, 2016-03-30 23:04:08]

On 2016/03/30 22:59:26, Kevin M wrote:
> lgtm

LGTM

[Sriram, 2016-03-30 23:04:11]

On 2016/03/30 22:59:26, Kevin M wrote:
> lgtm

LGTM

[marcinjb, 2016-03-30 23:07:17]

The CQ bit was checked by marcinjb@chromium.org

[commit-bot: I haz the power, 2016-03-30 23:07:39]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1844733003/1
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1844733003/1

[commit-bot: I haz the power, 2016-03-30 23:13:39]

Description was changed from

==========
Restart the engine within the docker container when appropriate.

It's possible for the engine to exit successfully right now (when a
connection is lost). The current state of start_engine.sh will cause
the Docker container to exit in this case, which isn't desired behavior.

This patch changes the behavior so that the engine will continuously be
restarted within the Docker container so long as it exits cleanly. If the
engine exits with a nonzero code (or if stunnel bombs out), the Docker
container will be torn down.

BUG=598933
==========

to

==========
Restart the engine within the docker container when appropriate.

It's possible for the engine to exit successfully right now (when a
connection is lost). The current state of start_engine.sh will cause
the Docker container to exit in this case, which isn't desired behavior.

This patch changes the behavior so that the engine will continuously be
restarted within the Docker container so long as it exits cleanly. If the
engine exits with a nonzero code (or if stunnel bombs out), the Docker
container will be torn down.

BUG=598933
==========

[commit-bot: I haz the power, 2016-03-30 23:13:40]

Committed patchset #1 (id:1)

[commit-bot: I haz the power, 2016-03-30 23:15:06]

Description was changed from

==========
Restart the engine within the docker container when appropriate.

It's possible for the engine to exit successfully right now (when a
connection is lost). The current state of start_engine.sh will cause
the Docker container to exit in this case, which isn't desired behavior.

This patch changes the behavior so that the engine will continuously be
restarted within the Docker container so long as it exits cleanly. If the
engine exits with a nonzero code (or if stunnel bombs out), the Docker
container will be torn down.

BUG=598933
==========

to

==========
Restart the engine within the docker container when appropriate.

It's possible for the engine to exit successfully right now (when a
connection is lost). The current state of start_engine.sh will cause
the Docker container to exit in this case, which isn't desired behavior.

This patch changes the behavior so that the engine will continuously be
restarted within the Docker container so long as it exits cleanly. If the
engine exits with a nonzero code (or if stunnel bombs out), the Docker
container will be torn down.

BUG=598933

Committed: https://crrev.com/7e8f69e15bd7e054e4c3cbb5e6b70870adb1951e
Cr-Commit-Position: refs/heads/master@{#384139}
==========

[commit-bot: I haz the power, 2016-03-30 23:15:07]

Patchset 1 (id:??) landed as
https://crrev.com/7e8f69e15bd7e054e4c3cbb5e6b70870adb1951e
Cr-Commit-Position: refs/heads/master@{#384139}