Proxied https connection reuse by sun.net.www.http.HttpClient can send CONNECT to the destination server

[Andreas Rieber]

Hi Chris,

i have also checked this problem and yes: the proxy header can end on 
the server side. The fix looks right. I have rewritten the test so it is 
in the right place and using existing certificate. I guess Steven would 
like to see this fix going in somehow ;-)

I did run a full jtreg test on ubuntu, jdk 9 dev and jdk 9, where this 
test fails.

Bug:
https://bugs.openjdk.java.net/browse/JDK-8025710

Webrev:
http://cr.openjdk.java.net/~arieber/8025710/webrev.00/

thanks
Andreas


On 24.12.2013 15:01, Chris Hegarty wrote:
> Andreas, Steven,
>
> I updated the bug and assigned it to myself ( to remind me to sponsor 
> the change ). I'll need to look at the changes in a little more 
> detail, but at first glance they look right to me. I will also look to 
> see if the test can be reduced a little, and add appropriate jtreg 
> tags ( to allow it to be run in our test environment ).
>
> Thanks,
> -Chris
>
> On 24 Dec 2013, at 08:44, Andreas Rieber <rieberandreas at gmail.com 
> <mailto:rieberandreas at gmail.com>> wrote:
>
>> Hi Steven,
>>
>> On 24.12.2013 00:41, Steven Lawrance wrote:
>>> Hi net-dev,
>>>
>>> I filed this on September 26 at bugs.sun.com <http://bugs.sun.com>, 
>>> which assigned bug ID 9007104 to it, but that bug still does not 
>>> appear to be visible. I'm emailing in case if that system 
>>> malfunctioned. I now see that OpenJDK has a Bugzilla system, but I 
>>> don't have an account to file this bug.
>>
>> Your bug is visible in new JBS at:
>> https://bugs.openjdk.java.net/browse/JDK-8025710
>>
>> - Andreas
>>
>> cut
>> ...
>>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.openjdk.java.net/pipermail/net-dev/attachments/20140415/fec63b78/attachment.html>