[
https://issues.apache.org/jira/browse/THRIFT-1689?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13459615#comment-13459615
]
Hudson commented on THRIFT-1689:
--------------------------------
Integrated in Thrift #531 (See [https://builds.apache.org/job/Thrift/531/])
THRIFT-1689: don't exit(-1) in TNonblockingServer
Client: c++
Patch: Brian Fallik
use standard exit(1) instead (Revision 1388019)
Result = SUCCESS
henrique : http://svn.apache.org/viewvc/?view=rev&rev=1388019
Files :
* /thrift/trunk/lib/cpp/src/thrift/server/TNonblockingServer.cpp
> don't exit(-1) in TNonblockingServer
> ------------------------------------
>
> Key: THRIFT-1689
> URL: https://issues.apache.org/jira/browse/THRIFT-1689
> Project: Thrift
> Issue Type: Bug
> Components: C++ - Library
> Affects Versions: 0.8
> Reporter: Brian Fallik
> Assignee: Brian Fallik
> Priority: Minor
> Fix For: 0.9
>
> Attachments: 1689.patch
>
>
> TNonblockingServer calls exit(-1) when it encounters a std::bad_alloc
> exception. The -1 argument to exit() is problematic. Programs typically use
> non-zero exit codes since negative values are reserved for signals. On
> somewhat modern Linux (CentOS 6.0), exit(-1) causes the process to yield
> return code 255. which BASH claims is out of range:
> http://tldp.org/LDP/abs/html/exitcodes.html
> I don't see any advantage to exit(-1) in this case. If the intent is to exit
> with an error indication, 1 (or the EXIT_FAILURE macro) should do fine. If
> the intent is to signal program termination, possibly with a stack trace,
> abrt() would be better.
> Attached is a patch for the simplest change from '-1' to '1'.
> Thanks.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
