Re: Review Request 29889: Recover Docker containers when mesos slave is in a container
--- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/29889/ --- (Updated May 24, 2015, 6:26 p.m.) Review request for mesos, Benjamin Hindman, Bernd Mathiske, and Till Toenshoff. Bugs: MESOS-2115 https://issues.apache.org/jira/browse/MESOS-2115 Repository: mesos Description --- This is a one mega patch and contains many reviews that's already on rb. This review is not meant to be merged, only provided for easier review. Diffs (updated) - Dockerfile 35abf25 docs/configuration.md 54c4e31 docs/docker-containerizer.md a5438b7 src/Makefile.am 34755cf src/docker/docker.hpp 3ebbc1f src/docker/docker.cpp 3a485a2 src/docker/executor.hpp PRE-CREATION src/docker/executor.cpp PRE-CREATION src/slave/constants.hpp fd1c1ab src/slave/constants.cpp 2a99b11 src/slave/containerizer/docker.hpp ed4ee19 src/slave/containerizer/docker.cpp 408a443 src/slave/flags.hpp 5c57478 src/slave/flags.cpp b5e2518 src/tests/docker_containerizer_tests.cpp 154bf98 src/tests/docker_tests.cpp 5520c58 Diff: https://reviews.apache.org/r/29889/diff/ Testing --- make check Thanks, Timothy Chen
Re: Review Request 29889: Recover Docker containers when mesos slave is in a container
On April 28, 2015, 10:11 p.m., Bernd Mathiske wrote: Dockerfile, line 18 https://reviews.apache.org/r/29889/diff/6/?file=940077#file940077line18 It seems that docker.io could be on its own line, since it is a kinda separate topic, which seems to be the grouping criterium here? I'll revert this since it's not reall related. - Timothy --- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/29889/#review81873 --- On May 24, 2015, 6:26 p.m., Timothy Chen wrote: --- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/29889/ --- (Updated May 24, 2015, 6:26 p.m.) Review request for mesos, Benjamin Hindman, Bernd Mathiske, and Till Toenshoff. Bugs: MESOS-2115 https://issues.apache.org/jira/browse/MESOS-2115 Repository: mesos Description --- This is a one mega patch and contains many reviews that's already on rb. This review is not meant to be merged, only provided for easier review. Diffs - Dockerfile 35abf25 docs/configuration.md 54c4e31 docs/docker-containerizer.md a5438b7 src/Makefile.am 34755cf src/docker/docker.hpp 3ebbc1f src/docker/docker.cpp 3a485a2 src/docker/executor.hpp PRE-CREATION src/docker/executor.cpp PRE-CREATION src/slave/constants.hpp fd1c1ab src/slave/constants.cpp 2a99b11 src/slave/containerizer/docker.hpp ed4ee19 src/slave/containerizer/docker.cpp 408a443 src/slave/flags.hpp 5c57478 src/slave/flags.cpp b5e2518 src/tests/docker_containerizer_tests.cpp 154bf98 src/tests/docker_tests.cpp 5520c58 Diff: https://reviews.apache.org/r/29889/diff/ Testing --- make check Thanks, Timothy Chen
Re: Review Request 29889: Recover Docker containers when mesos slave is in a container
--- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/29889/ --- (Updated May 25, 2015, 5:25 a.m.) Review request for mesos, Benjamin Hindman, Bernd Mathiske, and Till Toenshoff. Bugs: MESOS-2115 https://issues.apache.org/jira/browse/MESOS-2115 Repository: mesos Description --- This is a one mega patch and contains many reviews that's already on rb. This review is not meant to be merged, only provided for easier review. Diffs (updated) - Dockerfile 35abf25 docs/configuration.md 54c4e31 docs/docker-containerizer.md a5438b7 src/Makefile.am 34755cf src/docker/docker.hpp 3ebbc1f src/docker/docker.cpp 3a485a2 src/docker/executor.hpp PRE-CREATION src/docker/executor.cpp PRE-CREATION src/slave/constants.hpp fd1c1ab src/slave/constants.cpp 2a99b11 src/slave/containerizer/docker.hpp ed4ee19 src/slave/containerizer/docker.cpp 408a443 src/slave/flags.hpp 5c57478 src/slave/flags.cpp b5e2518 src/tests/docker_containerizer_tests.cpp 154bf98 src/tests/docker_tests.cpp 5520c58 Diff: https://reviews.apache.org/r/29889/diff/ Testing --- make check Thanks, Timothy Chen
Re: Review Request 29889: Recover Docker containers when mesos slave is in a container
--- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/29889/#review85058 --- Bad patch! Reviews applied: [29327] Failed command: ./support/apply-review.sh -n -r 29327 Error: 2015-05-23 06:40:39 URL:https://reviews.apache.org/r/29327/diff/raw/ [2935/2935] - 29327.patch [1] error: patch failed: src/slave/containerizer/docker.cpp:499 error: src/slave/containerizer/docker.cpp: patch does not apply Failed to apply patch - Mesos ReviewBot On May 23, 2015, 6:14 a.m., Timothy Chen wrote: --- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/29889/ --- (Updated May 23, 2015, 6:14 a.m.) Review request for mesos, Benjamin Hindman, Bernd Mathiske, and Till Toenshoff. Bugs: MESOS-2115 https://issues.apache.org/jira/browse/MESOS-2115 Repository: mesos Description --- This is a one mega patch and contains many reviews that's already on rb. This review is not meant to be merged, only provided for easier review. Diffs - Dockerfile 35abf25 docs/configuration.md 54c4e31 docs/docker-containerizer.md a5438b7 src/Makefile.am 34755cf src/docker/docker.hpp 3ebbc1f src/docker/docker.cpp 3a485a2 src/docker/executor.hpp PRE-CREATION src/docker/executor.cpp PRE-CREATION src/slave/constants.hpp fd1c1ab src/slave/constants.cpp 2a99b11 src/slave/containerizer/docker.hpp ed4ee19 src/slave/containerizer/docker.cpp 408a443 src/slave/flags.hpp 5c57478 src/slave/flags.cpp b5e2518 src/tests/docker_containerizer_tests.cpp 154bf98 src/tests/docker_tests.cpp 5520c58 Diff: https://reviews.apache.org/r/29889/diff/ Testing --- make check Thanks, Timothy Chen
Re: Review Request 29889: Recover Docker containers when mesos slave is in a container
--- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/29889/ --- (Updated May 23, 2015, 6:14 a.m.) Review request for mesos, Benjamin Hindman, Bernd Mathiske, and Till Toenshoff. Bugs: MESOS-2115 https://issues.apache.org/jira/browse/MESOS-2115 Repository: mesos Description --- This is a one mega patch and contains many reviews that's already on rb. This review is not meant to be merged, only provided for easier review. Diffs (updated) - Dockerfile 35abf25 docs/configuration.md 54c4e31 docs/docker-containerizer.md a5438b7 src/Makefile.am 34755cf src/docker/docker.hpp 3ebbc1f src/docker/docker.cpp 3a485a2 src/docker/executor.hpp PRE-CREATION src/docker/executor.cpp PRE-CREATION src/slave/constants.hpp fd1c1ab src/slave/constants.cpp 2a99b11 src/slave/containerizer/docker.hpp ed4ee19 src/slave/containerizer/docker.cpp 408a443 src/slave/flags.hpp 5c57478 src/slave/flags.cpp b5e2518 src/tests/docker_containerizer_tests.cpp 154bf98 src/tests/docker_tests.cpp 5520c58 Diff: https://reviews.apache.org/r/29889/diff/ Testing --- make check Thanks, Timothy Chen
Re: Review Request 29889: Recover Docker containers when mesos slave is in a container
--- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/29889/#review83305 --- src/docker/docker.cpp https://reviews.apache.org/r/29889/#comment134286 1. There is no declaration in the hpp file with a comment and there is no explanation here. I would suggest using static here so nobody needs to browse the hpp file in search for a potential comment that does not exist. 2. OK, I can guess what this does, but future readers may want to skip reading the function body to find out. The name suggests running something. 3. I can't find where this function is used. It seems to me that you wanted to break out part of Docker::run() below, but have not removed this code there and replaced it with a call yet. src/docker/executor.cpp https://reviews.apache.org/r/29889/#comment134292 This is a bit tricky. You have to know that 'stop' receives a non-Nothing value once 'run' gets one. Suggestion: reduce this to only one future - 'run'. See below in line 170. src/docker/executor.cpp https://reviews.apache.org/r/29889/#comment134245 Why can't we call 'stop.onAny(...)' directly here? src/docker/executor.cpp https://reviews.apache.org/r/29889/#comment134241 s/might still/might still be src/docker/executor.cpp https://reviews.apache.org/r/29889/#comment134242 Suggestions: Break this out as a separate statement/comment pair. s/copy to remove const/mutable copy of the future src/docker/executor.cpp https://reviews.apache.org/r/29889/#comment134294 Could we use a LOCAL future 'stop' instead and discard 'run' once it is satisfied? This way 'stop' does not have to be a global and one can see here locally what is going on. You could then amend _reaped() for checking for run having been discarded. src/docker/executor.cpp https://reviews.apache.org/r/29889/#comment134247 See line 136, it seems that we do not need this function. If you were to keep it, it would be better placed above _reaped, so that one could read upper half / lower half contiguously. src/docker/executor.cpp https://reviews.apache.org/r/29889/#comment134248 Aren't we going to inspect 'stop'? What if it failed? Could we at least log something then? src/docker/executor.cpp https://reviews.apache.org/r/29889/#comment134295 s/after stopping/after attempting to stop src/docker/executor.cpp https://reviews.apache.org/r/29889/#comment134296 No default? Why? src/slave/containerizer/docker.hpp https://reviews.apache.org/r/29889/#comment134298 Which docker image? src/slave/containerizer/docker.hpp https://reviews.apache.org/r/29889/#comment134305 only? What other cases are there? src/slave/containerizer/docker.cpp https://reviews.apache.org/r/29889/#comment134303 Which of these arguments signifies the executor? 'container'? Could we rename it accordingly? Before you updated the comment, I overlooked that this might be executor-specific code! src/tests/docker_containerizer_tests.cpp https://reviews.apache.org/r/29889/#comment134310 We should generally avoid bool args, because it is hard to see at the call site what they mean. Suggestions: - Use an enum. - Keep the exists() and running() methods, factor out what's common between them, pass a closure as an arg to the subroutine for what's different. src/tests/docker_containerizer_tests.cpp https://reviews.apache.org/r/29889/#comment134311 Insert blank line above, please. In other places below, too. Maybe better: move the comment above the declaration. src/tests/docker_containerizer_tests.cpp https://reviews.apache.org/r/29889/#comment134314 What changed so that we don't need this test any more? - Bernd Mathiske On May 8, 2015, 2:40 p.m., Timothy Chen wrote: --- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/29889/ --- (Updated May 8, 2015, 2:40 p.m.) Review request for mesos, Benjamin Hindman, Bernd Mathiske, and Till Toenshoff. Bugs: MESOS-2115 https://issues.apache.org/jira/browse/MESOS-2115 Repository: mesos Description --- This is a one mega patch and contains many reviews that's already on rb. This review is not meant to be merged, only provided for easier review. Diffs - Dockerfile 35abf25 docs/configuration.md 54c4e31 docs/docker-containerizer.md a5438b7 src/Makefile.am 54271f7 src/docker/docker.hpp 3ebbc1f src/docker/docker.cpp 3a485a2 src/docker/executor.cpp PRE-CREATION src/slave/constants.hpp fd1c1ab src/slave/constants.cpp 2a99b11 src/slave/containerizer/docker.hpp
Re: Review Request 29889: Recover Docker containers when mesos slave is in a container
--- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/29889/#review82325 --- Bad patch! Reviews applied: [29327] Failed command: ./support/apply-review.sh -n -r 29327 Error: 2015-05-02 20:34:47 URL:https://reviews.apache.org/r/29327/diff/raw/ [2935/2935] - 29327.patch [1] error: patch failed: src/slave/containerizer/docker.cpp:499 error: src/slave/containerizer/docker.cpp: patch does not apply Failed to apply patch - Mesos ReviewBot On May 2, 2015, 8:22 p.m., Timothy Chen wrote: --- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/29889/ --- (Updated May 2, 2015, 8:22 p.m.) Review request for mesos, Benjamin Hindman, Bernd Mathiske, and Till Toenshoff. Bugs: MESOS-2115 https://issues.apache.org/jira/browse/MESOS-2115 Repository: mesos Description --- This is a one mega patch and contains many reviews that's already on rb. This review is not meant to be merged, only provided for easier review. Diffs - Dockerfile 35abf25 docs/configuration.md 54c4e31 docs/docker-containerizer.md a5438b7 src/Makefile.am 93c7c8a src/docker/docker.hpp 3ebbc1f src/docker/docker.cpp 3a485a2 src/docker/executor.cpp PRE-CREATION src/slave/constants.hpp fd1c1ab src/slave/constants.cpp 2a99b11 src/slave/containerizer/docker.hpp b25ec55 src/slave/containerizer/docker.cpp f9fc89a src/slave/flags.hpp d3b1ce1 src/slave/flags.cpp d0932b0 src/tests/docker_containerizer_tests.cpp c9d66b3 Diff: https://reviews.apache.org/r/29889/diff/ Testing --- make check Thanks, Timothy Chen
Re: Review Request 29889: Recover Docker containers when mesos slave is in a container
On May 1, 2015, 11:14 p.m., Bernd Mathiske wrote: src/slave/containerizer/docker.hpp, line 454 https://reviews.apache.org/r/29889/diff/8/?file=946143#file946143line454 This is a good place to explain the whole picture what each override... is, what it's purpose is and how it gets used. However, if this is deemed a generic facility, then some of the use sites need to be upgraded with docker-specific scenario descriptions. I got rid of override completely now. - Timothy --- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/29889/#review82287 --- On May 1, 2015, 9:43 p.m., Timothy Chen wrote: --- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/29889/ --- (Updated May 1, 2015, 9:43 p.m.) Review request for mesos, Benjamin Hindman, Bernd Mathiske, and Till Toenshoff. Bugs: MESOS-2115 https://issues.apache.org/jira/browse/MESOS-2115 Repository: mesos Description --- This is a one mega patch and contains many reviews that's already on rb. This review is not meant to be merged, only provided for easier review. Diffs - Dockerfile 35abf25 docs/configuration.md 54c4e31 docs/docker-containerizer.md a5438b7 src/Makefile.am 93c7c8a src/docker/docker.hpp 3ebbc1f src/docker/docker.cpp 3a485a2 src/docker/executor.cpp PRE-CREATION src/slave/constants.hpp fd1c1ab src/slave/constants.cpp 2a99b11 src/slave/containerizer/docker.hpp b25ec55 src/slave/containerizer/docker.cpp f9fc89a src/slave/flags.hpp d3b1ce1 src/slave/flags.cpp d0932b0 src/tests/docker_containerizer_tests.cpp c9d66b3 Diff: https://reviews.apache.org/r/29889/diff/ Testing --- make check Thanks, Timothy Chen
Re: Review Request 29889: Recover Docker containers when mesos slave is in a container
--- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/29889/ --- (Updated May 2, 2015, 8:22 p.m.) Review request for mesos, Benjamin Hindman, Bernd Mathiske, and Till Toenshoff. Bugs: MESOS-2115 https://issues.apache.org/jira/browse/MESOS-2115 Repository: mesos Description --- This is a one mega patch and contains many reviews that's already on rb. This review is not meant to be merged, only provided for easier review. Diffs (updated) - Dockerfile 35abf25 docs/configuration.md 54c4e31 docs/docker-containerizer.md a5438b7 src/Makefile.am 93c7c8a src/docker/docker.hpp 3ebbc1f src/docker/docker.cpp 3a485a2 src/docker/executor.cpp PRE-CREATION src/slave/constants.hpp fd1c1ab src/slave/constants.cpp 2a99b11 src/slave/containerizer/docker.hpp b25ec55 src/slave/containerizer/docker.cpp f9fc89a src/slave/flags.hpp d3b1ce1 src/slave/flags.cpp d0932b0 src/tests/docker_containerizer_tests.cpp c9d66b3 Diff: https://reviews.apache.org/r/29889/diff/ Testing --- make check Thanks, Timothy Chen
Re: Review Request 29889: Recover Docker containers when mesos slave is in a container
--- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/29889/#review82287 --- src/docker/executor.cpp https://reviews.apache.org/r/29889/#comment133009 Can we add some output here, too, indicating that the reason for shutting down is killing? src/docker/executor.cpp https://reviews.apache.org/r/29889/#comment133010 Emit some output here? src/docker/executor.cpp https://reviews.apache.org/r/29889/#comment133011 This is copied from CommandExecutor. Please add a TODO to fix both. src/docker/executor.cpp https://reviews.apache.org/r/29889/#comment133012 Owned... src/docker/executor.cpp https://reviews.apache.org/r/29889/#comment133013 cli - executable src/docker/executor.cpp https://reviews.apache.org/r/29889/#comment133014 Suggestion: The path to the container sandbox holding stdout and stderr files into which docker container logs will be redirected. src/docker/executor.cpp https://reviews.apache.org/r/29889/#comment133016 This pertains to the boolean arg. Please indicate this. src/docker/executor.cpp https://reviews.apache.org/r/29889/#comment133017 Suggestion, add something like: ..., since this has already been validated in which step before. src/slave/containerizer/docker.hpp https://reviews.apache.org/r/29889/#comment133026 This is a good place to explain the whole picture what each override... is, what it's purpose is and how it gets used. However, if this is deemed a generic facility, then some of the use sites need to be upgraded with docker-specific scenario descriptions. src/slave/containerizer/docker.cpp https://reviews.apache.org/r/29889/#comment133027 zap the TABs src/slave/containerizer/docker.cpp https://reviews.apache.org/r/29889/#comment133028 You have a constant declared for 5. src/slave/containerizer/docker.cpp https://reviews.apache.org/r/29889/#comment133029 Would it be possible to pull up the conditional branch to here and then call two different constructors depending on the condition, each of which fills out the fields differently. We would then no longer need the override... fields, would we? src/slave/flags.cpp https://reviews.apache.org/r/29889/#comment133021 There is extra text on this in the docs. Please determine how much of it pull up here as well. - Bernd Mathiske On May 1, 2015, 2:43 p.m., Timothy Chen wrote: --- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/29889/ --- (Updated May 1, 2015, 2:43 p.m.) Review request for mesos, Benjamin Hindman, Bernd Mathiske, and Till Toenshoff. Bugs: MESOS-2115 https://issues.apache.org/jira/browse/MESOS-2115 Repository: mesos Description --- This is a one mega patch and contains many reviews that's already on rb. This review is not meant to be merged, only provided for easier review. Diffs - Dockerfile 35abf25 docs/configuration.md 54c4e31 docs/docker-containerizer.md a5438b7 src/Makefile.am 93c7c8a src/docker/docker.hpp 3ebbc1f src/docker/docker.cpp 3a485a2 src/docker/executor.cpp PRE-CREATION src/slave/constants.hpp fd1c1ab src/slave/constants.cpp 2a99b11 src/slave/containerizer/docker.hpp b25ec55 src/slave/containerizer/docker.cpp f9fc89a src/slave/flags.hpp d3b1ce1 src/slave/flags.cpp d0932b0 src/tests/docker_containerizer_tests.cpp c9d66b3 Diff: https://reviews.apache.org/r/29889/diff/ Testing --- make check Thanks, Timothy Chen
Re: Review Request 29889: Recover Docker containers when mesos slave is in a container
--- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/29889/ --- (Updated April 30, 2015, 7:43 a.m.) Review request for mesos, Benjamin Hindman, Bernd Mathiske, and Till Toenshoff. Repository: mesos Description --- This is a one mega patch and contains many reviews that's already on rb. This review is not meant to be merged, only provided for easier review. Diffs (updated) - Dockerfile 35abf25 docs/configuration.md 54c4e31 docs/docker-containerizer.md a5438b7 src/Makefile.am 93c7c8a src/docker/docker.hpp 3ebbc1f src/docker/docker.cpp 3a485a2 src/docker/executor.cpp PRE-CREATION src/slave/constants.hpp fd1c1ab src/slave/constants.cpp 2a99b11 src/slave/containerizer/docker.hpp b25ec55 src/slave/containerizer/docker.cpp f9fc89a src/slave/flags.hpp d3b1ce1 src/slave/flags.cpp d0932b0 src/tests/docker_containerizer_tests.cpp c9d66b3 Diff: https://reviews.apache.org/r/29889/diff/ Testing --- make check Thanks, Timothy Chen
Re: Review Request 29889: Recover Docker containers when mesos slave is in a container
--- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/29889/#review81873 --- Dockerfile https://reviews.apache.org/r/29889/#comment132377 It seems that docker.io could be on its own line, since it is a kinda separate topic, which seems to be the grouping criterium here? docs/configuration.md https://reviews.apache.org/r/29889/#comment132378 s/CLI/Docker client CLI/ docs/configuration.md https://reviews.apache.org/r/29889/#comment132385 Suggestion how to phrase all this: The UNIX socket path to be mounted into the docker executor container to provide docker CLI access to the docker daemon. This must be the path used by the slave's docker image. docs/configuration.md https://reviews.apache.org/r/29889/#comment132387 s/a/an src/docker/docker.cpp https://reviews.apache.org/r/29889/#comment132389 s/!isSome/isNone src/docker/docker.cpp https://reviews.apache.org/r/29889/#comment132390 s/ / src/docker/docker.cpp https://reviews.apache.org/r/29889/#comment132391 s/contianer/container src/docker/docker.cpp https://reviews.apache.org/r/29889/#comment132394 Unfortunately, checkError() and _run() are so similar, but so differently named. Maybe a comment explaining that error checking is different when not detached in that stderr does not get read? src/docker/executor.cpp https://reviews.apache.org/r/29889/#comment132400 Suggestion to replace the last sentence with: The executor is set up in a way that ensures its continued execution if the slave exits. If the slave is running in a docker container this is inherently the case. Otherwise the executor creates a new process group for itself. src/docker/executor.cpp https://reviews.apache.org/r/29889/#comment132402 Shouldn't we leave a trace of what happened in the output? src/docker/executor.cpp https://reviews.apache.org/r/29889/#comment132403 Given rephrasing the class comment above, this comment may get removed. src/docker/executor.cpp https://reviews.apache.org/r/29889/#comment132406 Do we really need this variable? Could we just use task.container()? src/docker/executor.cpp https://reviews.apache.org/r/29889/#comment132410 Please comment why we are adding the executor resources here: for reporting total usage of task + executor instead of just task. However, isn't it the case that we are now actually using task.resources() + executor.resources() + executor.resources(), since the xecutor is already running? src/docker/executor.cpp https://reviews.apache.org/r/29889/#comment132412 Out of the 3 vars launched, dockerRun, killed you only need 2. src/docker/executor.cpp https://reviews.apache.org/r/29889/#comment132413 This is also virtual in the CommandExecutor, but why? - Bernd Mathiske On April 22, 2015, 3:50 p.m., Timothy Chen wrote: --- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/29889/ --- (Updated April 22, 2015, 3:50 p.m.) Review request for mesos, Benjamin Hindman, Bernd Mathiske, and Till Toenshoff. Repository: mesos Description --- This is a one mega patch and contains many reviews that's already on rb. This review is not meant to be merged, only provided for easier review. Diffs - Dockerfile 35abf25 docs/configuration.md 54c4e31 docs/docker-containerizer.md a5438b7 src/Makefile.am 93c7c8a src/docker/docker.hpp 3ebbc1f src/docker/docker.cpp 3a485a2 src/docker/executor.cpp PRE-CREATION src/slave/containerizer/docker.hpp 0d5289c src/slave/containerizer/docker.cpp f9fb078 src/slave/flags.hpp d3b1ce1 src/slave/flags.cpp d0932b0 src/tests/docker_containerizer_tests.cpp b119a17 Diff: https://reviews.apache.org/r/29889/diff/ Testing --- make check Thanks, Timothy Chen