[ 
https://issues.apache.org/jira/browse/WICKET-6685?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Martin Grigorov updated WICKET-6685:
------------------------------------
    Description: 
Testerd on 8.5.0.

The destroy method of Session has added some clean-up calls, e.q. metaData = 
null.

The destroy method is also called by replaceSession method. That means, that 
replaceSession deletes metadata. But metadata are used in 
KeyInSessionSunJceCryptFactory to store the crypt key. So now in Wicket 8 
calling replaceSession (quite common security practise) means that all links 
generated before get broken.

I don't think this was the intention...

  was:
Testerd on 8.0.5.

The destroy method od Session has added some clean-up calls, e.q. metaData = 
null.

The destroy method is also called by replaceSession method. That means, that 
replaceSession deletes metadata. But metadata are used in 
KeyInSessionSunJceCryptFactory to store the crypt key. So now in Wicket 8 
calling replaceSession (quite common security practise) means that all links 
generated before get broken.

I don't think this was the intention...


> Session#destroy (used in replaceSession) deletes metadata
> ---------------------------------------------------------
>
>                 Key: WICKET-6685
>                 URL: https://issues.apache.org/jira/browse/WICKET-6685
>             Project: Wicket
>          Issue Type: Bug
>          Components: wicket
>    Affects Versions: 8.0.0
>         Environment: Windows 8 / JDK 8
>            Reporter: David Rain
>            Priority: Major
>
> Testerd on 8.5.0.
> The destroy method of Session has added some clean-up calls, e.q. metaData = 
> null.
> The destroy method is also called by replaceSession method. That means, that 
> replaceSession deletes metadata. But metadata are used in 
> KeyInSessionSunJceCryptFactory to store the crypt key. So now in Wicket 8 
> calling replaceSession (quite common security practise) means that all links 
> generated before get broken.
> I don't think this was the intention...



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

Reply via email to