History | Log In     View a printable version of the current page. Get help!  
Issue Details [XML]

Key: CHD-1940
Type: Bug Bug
Status: Closed Closed
Resolution: Fixed
Assignee: Unassigned
Reporter: Joe Geck
Votes: 0
Watchers: 0
Operations

Clone this issue
If you were logged in you would be able to see more operations.
Cerberus Helpdesk

[Tickets] Access Denied or Invalid Tickets after replying

Created: 07/Jun/10 08:32 AM   Updated: 07/Jun/10 04:37 PM
Fix Version/s: 5.0.0-pl1

Original Estimate: Unknown Remaining Estimate: Unknown Time Spent: Unknown
Environment: 5.0.0-pl1 (Build 2010052401)


 Description   
Summary:
I have seen the "invalid ticket" version myself, but have not been able to narrow down an exact cause. Now that we have gotten at least three other reports, I wanted to just file a generic bug for them to dump feedback in.

As far as I know there's two active cases going around, and both of them started with the 5.0.0-pl1 code. Meaning it just started happening after a recent SVN Update.

1) Access Denied on replies.
2) Invalid Ticket on replies.

What seems to happen for both is the URL is not properly redirected to include the ticket mask. So instead of cerb5/display/AAA-00000-000 you just get cerb5/display. As far as I know, the replies actually do go out to the requesters, so it's just an internal interface bug.

Also the .htaccess (friendly URL) does not seem to factor in.



 All   Comments   Work Log   Change History      Sort Order:
Comment by Joe Geck [07/Jun/10 08:33 AM]
Batch in Town Hall reported #1
The forums are reporting #2
http://forums.cerb4.com/showthread.php?t=2849

Comment by Jeff Standen [WGM] [07/Jun/10 02:59 PM]
There's something in common for those installs where a ticket mask isn't available during the reply. It could potentially just be a cached template.

I've committed a fail-safe anyway that will fall back to the internal ID of a ticket if no mask was provided.

Comment by Joe Geck [07/Jun/10 04:37 PM]
Verified fix itself forces the ticket ID into the URL and that clearing the cache seems to eliminate the root problem.

Closing out.
5.0.0-pl1 (Build 2010060701)