> On July 20, 2016, 9:39 p.m., Adam B wrote:
> > Won't this break upgrades for any user already using these environment 
> > variables?!?
> 
> Adam B wrote:
>     Nevermind.. premature panic. Now I see
>     ```
>       // To be backward compatible, for each environment variable prefixed
>       // by SSL_, we generate the corresponding LIBPROCESS_SSL_ version.
>     ```
>     So even if we use the `SSL_` variety, libprocess will use the values, but 
> subprocesses will not see them?

No, it won't. see the doc.


- Jie


-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/50257/#review143010
-----------------------------------------------------------


On July 20, 2016, 8:34 p.m., Jie Yu wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/50257/
> -----------------------------------------------------------
> 
> (Updated July 20, 2016, 8:34 p.m.)
> 
> 
> Review request for mesos, Benjamin Mahler, Joris Van Remoortere, Joseph Wu, 
> and Vinod Kone.
> 
> 
> Bugs: MESOS-5863
>     https://issues.apache.org/jira/browse/MESOS-5863
> 
> 
> Repository: mesos
> 
> 
> Description
> -------
> 
> Used LIBPROCESS_SSL_ instead of SSL_ as the prefix for ssl support.
> 
> 
> Diffs
> -----
> 
>   3rdparty/libprocess/include/process/ssl/gtest.hpp 
> a929cc90fc9ee975f3635957518ced4eb70bdfeb 
>   3rdparty/libprocess/src/openssl.hpp 
> 68f88970610293107b8349c216c34a32d5df9105 
>   3rdparty/libprocess/src/openssl.cpp 
> 63916ff66b4daa29120b7e6b12b329b68f746694 
>   3rdparty/libprocess/src/process.cpp 
> 9661386afd4fddd1877d55941fa403afc9230280 
>   3rdparty/libprocess/src/tests/ssl_client.cpp 
> 6b43cfc77681b5aca76da2638443155f682cd92f 
>   3rdparty/libprocess/src/tests/ssl_tests.cpp 
> 72432ec8f91eb4d628df73e19edbaceb8a2c0fb9 
> 
> Diff: https://reviews.apache.org/r/50257/diff/
> 
> 
> Testing
> -------
> 
> make check
> 
> 
> Thanks,
> 
> Jie Yu
> 
>

Reply via email to