[jira] [Commented] (APLO-266) 1.2 with CRLF, sometimes gets: IOException: Unable to parse header line [\u0013]

["Guy Allard (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/APLO-266?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13485119#comment-13485119 ] 

Guy Allard commented on APLO-266:
---------------------------------

One comment and question.  In that test code, if you uncomment:

p [ "msg", msg ]

and review the results, you will see a MESSAGE read from the wire.  The headers in that message are 'mixed mode'.  In other words some headers have '\n' line ends, some have '\r\n' line ends.

Is that right?  Should I expect that as the general case?

Thanks, Guy

                
> 1.2 with CRLF, sometimes gets: IOException: Unable to parse header line [\u0013]
> --------------------------------------------------------------------------------
>
>                 Key: APLO-266
>                 URL: https://issues.apache.org/jira/browse/APLO-266
>             Project: ActiveMQ Apollo
>          Issue Type: Bug
>          Components: apollo-stomp
>         Environment: Ubuntu 11.10
> Ruby (various versions)
> AMD Phenom(tm) II X6 1055T Processor
> 16GB RAM
> apache-apollo-99-trunk-20120929.031419-121-unix-distro.tar.gz
>            Reporter: Guy Allard
>         Attachments: letest2.rb
>
>
> I am testing Stomp 1.2 functionality that uses CRLF line ends.
> Under some conditions, sending a frame with \r\n line ends results in, e.g.:
> 2012-10-26 12:25:07,258 | WARN  | java.io.IOException: Unable to parse header line [\u0013] | org.apache.activemq.apollo.broker.Broker | hawtdispatch-DEFAULT-1
> This does not happen all the time.
> I do have a somewhat lengthy (but straightforward) program that recreates this.  I will attach it shortly.

--
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