You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@mesos.apache.org by Vinod Kone <vi...@gmail.com> on 2013/05/01 05:29:02 UTC

Review Request: Fixed process isolator to safely setup launcher's environment.

-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/10879/
-----------------------------------------------------------

Review request for mesos, Benjamin Hindman and Ben Mahler.


Description
-------

This hopefully reduces (eliminates?) the chance that a process isolator deadlocks on ostringstream between a fork and exec.


This addresses bug MESOS-393.
    https://issues.apache.org/jira/browse/MESOS-393


Diffs
-----

  src/launcher/launcher.hpp 9f0e247978d1609c3ad564ade2b528d3e04a9b9f 
  src/launcher/launcher.cpp 11cffc915f2e9218aae8ed6b9d19ed82a4e2ea66 
  src/launcher/main.cpp 57d671e7ee6e88429ead0a257b3545676a896ffd 
  src/slave/process_isolator.cpp 6938fbcca9d384bb014ff4dd52a13763c1f8397a 

Diff: https://reviews.apache.org/r/10879/diff/


Testing
-------

make check on OSX and CentOS.


Thanks,

Vinod Kone

Re: Review Request: Fixed process isolator to safely setup launcher's environment.

Posted by Vinod Kone <vi...@gmail.com>.
-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/10879/#review20036
-----------------------------------------------------------



src/slave/process_isolator.cpp
<https://reviews.apache.org/r/10879/#comment41293>

    indentation


- Vinod Kone


On May 1, 2013, 3:29 a.m., Vinod Kone wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/10879/
> -----------------------------------------------------------
> 
> (Updated May 1, 2013, 3:29 a.m.)
> 
> 
> Review request for mesos, Benjamin Hindman and Ben Mahler.
> 
> 
> Description
> -------
> 
> This hopefully reduces (eliminates?) the chance that a process isolator deadlocks on ostringstream between a fork and exec.
> 
> 
> This addresses bug MESOS-393.
>     https://issues.apache.org/jira/browse/MESOS-393
> 
> 
> Diffs
> -----
> 
>   src/launcher/launcher.hpp 9f0e247978d1609c3ad564ade2b528d3e04a9b9f 
>   src/launcher/launcher.cpp 11cffc915f2e9218aae8ed6b9d19ed82a4e2ea66 
>   src/launcher/main.cpp 57d671e7ee6e88429ead0a257b3545676a896ffd 
>   src/slave/process_isolator.cpp 6938fbcca9d384bb014ff4dd52a13763c1f8397a 
> 
> Diff: https://reviews.apache.org/r/10879/diff/
> 
> 
> Testing
> -------
> 
> make check on OSX and CentOS.
> 
> 
> Thanks,
> 
> Vinod Kone
> 
>

Re: Review Request: Fixed process isolator to safely setup launcher's environment.

Posted by Benjamin Hindman <be...@berkeley.edu>.
-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/10879/#review20035
-----------------------------------------------------------

Ship it!


Ship It!

- Benjamin Hindman


On May 1, 2013, 3:29 a.m., Vinod Kone wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/10879/
> -----------------------------------------------------------
> 
> (Updated May 1, 2013, 3:29 a.m.)
> 
> 
> Review request for mesos, Benjamin Hindman and Ben Mahler.
> 
> 
> Description
> -------
> 
> This hopefully reduces (eliminates?) the chance that a process isolator deadlocks on ostringstream between a fork and exec.
> 
> 
> This addresses bug MESOS-393.
>     https://issues.apache.org/jira/browse/MESOS-393
> 
> 
> Diffs
> -----
> 
>   src/launcher/launcher.hpp 9f0e247978d1609c3ad564ade2b528d3e04a9b9f 
>   src/launcher/launcher.cpp 11cffc915f2e9218aae8ed6b9d19ed82a4e2ea66 
>   src/launcher/main.cpp 57d671e7ee6e88429ead0a257b3545676a896ffd 
>   src/slave/process_isolator.cpp 6938fbcca9d384bb014ff4dd52a13763c1f8397a 
> 
> Diff: https://reviews.apache.org/r/10879/diff/
> 
> 
> Testing
> -------
> 
> make check on OSX and CentOS.
> 
> 
> Thanks,
> 
> Vinod Kone
> 
>

Re: Review Request: Fixed process isolator to safely setup launcher's environment.

Posted by Benjamin Hindman <be...@berkeley.edu>.

> On May 2, 2013, 1:54 a.m., Ben Mahler wrote:
> > Any chance of a similar fix for the CgroupIsolator?
> 
> Vinod Kone wrote:
>     It's a bit more tricky, because the launcher.setup() has to happen before exec() and it has lot of C++ code.

But to be clear, we did see this in production with CgroupsIsolator correct?


- Benjamin


-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/10879/#review20044
-----------------------------------------------------------


On May 1, 2013, 3:29 a.m., Vinod Kone wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/10879/
> -----------------------------------------------------------
> 
> (Updated May 1, 2013, 3:29 a.m.)
> 
> 
> Review request for mesos, Benjamin Hindman and Ben Mahler.
> 
> 
> Description
> -------
> 
> This hopefully reduces (eliminates?) the chance that a process isolator deadlocks on ostringstream between a fork and exec.
> 
> 
> This addresses bug MESOS-393.
>     https://issues.apache.org/jira/browse/MESOS-393
> 
> 
> Diffs
> -----
> 
>   src/launcher/launcher.hpp 9f0e247978d1609c3ad564ade2b528d3e04a9b9f 
>   src/launcher/launcher.cpp 11cffc915f2e9218aae8ed6b9d19ed82a4e2ea66 
>   src/launcher/main.cpp 57d671e7ee6e88429ead0a257b3545676a896ffd 
>   src/slave/process_isolator.cpp 6938fbcca9d384bb014ff4dd52a13763c1f8397a 
> 
> Diff: https://reviews.apache.org/r/10879/diff/
> 
> 
> Testing
> -------
> 
> make check on OSX and CentOS.
> 
> 
> Thanks,
> 
> Vinod Kone
> 
>

Re: Review Request: Fixed process isolator to safely setup launcher's environment.

Posted by Vinod Kone <vi...@gmail.com>.

> On May 2, 2013, 1:54 a.m., Ben Mahler wrote:
> > Any chance of a similar fix for the CgroupIsolator?

It's a bit more tricky, because the launcher.setup() has to happen before exec() and it has lot of C++ code.


- Vinod


-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/10879/#review20044
-----------------------------------------------------------


On May 1, 2013, 3:29 a.m., Vinod Kone wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/10879/
> -----------------------------------------------------------
> 
> (Updated May 1, 2013, 3:29 a.m.)
> 
> 
> Review request for mesos, Benjamin Hindman and Ben Mahler.
> 
> 
> Description
> -------
> 
> This hopefully reduces (eliminates?) the chance that a process isolator deadlocks on ostringstream between a fork and exec.
> 
> 
> This addresses bug MESOS-393.
>     https://issues.apache.org/jira/browse/MESOS-393
> 
> 
> Diffs
> -----
> 
>   src/launcher/launcher.hpp 9f0e247978d1609c3ad564ade2b528d3e04a9b9f 
>   src/launcher/launcher.cpp 11cffc915f2e9218aae8ed6b9d19ed82a4e2ea66 
>   src/launcher/main.cpp 57d671e7ee6e88429ead0a257b3545676a896ffd 
>   src/slave/process_isolator.cpp 6938fbcca9d384bb014ff4dd52a13763c1f8397a 
> 
> Diff: https://reviews.apache.org/r/10879/diff/
> 
> 
> Testing
> -------
> 
> make check on OSX and CentOS.
> 
> 
> Thanks,
> 
> Vinod Kone
> 
>

Re: Review Request: Fixed process isolator to safely setup launcher's environment.

Posted by Vinod Kone <vi...@gmail.com>.

> On May 2, 2013, 1:54 a.m., Ben Mahler wrote:
> > Any chance of a similar fix for the CgroupIsolator?
> 
> Vinod Kone wrote:
>     It's a bit more tricky, because the launcher.setup() has to happen before exec() and it has lot of C++ code.
> 
> Benjamin Hindman wrote:
>     But to be clear, we did see this in production with CgroupsIsolator correct?

yup.


- Vinod


-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/10879/#review20044
-----------------------------------------------------------


On May 1, 2013, 3:29 a.m., Vinod Kone wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/10879/
> -----------------------------------------------------------
> 
> (Updated May 1, 2013, 3:29 a.m.)
> 
> 
> Review request for mesos, Benjamin Hindman and Ben Mahler.
> 
> 
> Description
> -------
> 
> This hopefully reduces (eliminates?) the chance that a process isolator deadlocks on ostringstream between a fork and exec.
> 
> 
> This addresses bug MESOS-393.
>     https://issues.apache.org/jira/browse/MESOS-393
> 
> 
> Diffs
> -----
> 
>   src/launcher/launcher.hpp 9f0e247978d1609c3ad564ade2b528d3e04a9b9f 
>   src/launcher/launcher.cpp 11cffc915f2e9218aae8ed6b9d19ed82a4e2ea66 
>   src/launcher/main.cpp 57d671e7ee6e88429ead0a257b3545676a896ffd 
>   src/slave/process_isolator.cpp 6938fbcca9d384bb014ff4dd52a13763c1f8397a 
> 
> Diff: https://reviews.apache.org/r/10879/diff/
> 
> 
> Testing
> -------
> 
> make check on OSX and CentOS.
> 
> 
> Thanks,
> 
> Vinod Kone
> 
>

Re: Review Request: Fixed process isolator to safely setup launcher's environment.

Posted by Ben Mahler <be...@gmail.com>.
-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/10879/#review20044
-----------------------------------------------------------

Ship it!


Any chance of a similar fix for the CgroupIsolator?

- Ben Mahler


On May 1, 2013, 3:29 a.m., Vinod Kone wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/10879/
> -----------------------------------------------------------
> 
> (Updated May 1, 2013, 3:29 a.m.)
> 
> 
> Review request for mesos, Benjamin Hindman and Ben Mahler.
> 
> 
> Description
> -------
> 
> This hopefully reduces (eliminates?) the chance that a process isolator deadlocks on ostringstream between a fork and exec.
> 
> 
> This addresses bug MESOS-393.
>     https://issues.apache.org/jira/browse/MESOS-393
> 
> 
> Diffs
> -----
> 
>   src/launcher/launcher.hpp 9f0e247978d1609c3ad564ade2b528d3e04a9b9f 
>   src/launcher/launcher.cpp 11cffc915f2e9218aae8ed6b9d19ed82a4e2ea66 
>   src/launcher/main.cpp 57d671e7ee6e88429ead0a257b3545676a896ffd 
>   src/slave/process_isolator.cpp 6938fbcca9d384bb014ff4dd52a13763c1f8397a 
> 
> Diff: https://reviews.apache.org/r/10879/diff/
> 
> 
> Testing
> -------
> 
> make check on OSX and CentOS.
> 
> 
> Thanks,
> 
> Vinod Kone
> 
>