[JIRA] Issue Comment Edited: (SVC-5002) Linking rules for large objects broken under server 1.32

Maggie Darwin (JIRA) no-reply at lindenlab.cascadeo.com
Thu Nov 26 06:57:07 PST 2009


    [ http://jira.secondlife.com/browse/SVC-5002?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=148941#action_148941 ] 

Maggie Darwin edited comment on SVC-5002 at 11/26/09 6:55 AM:
--------------------------------------------------------------

@Darling -- 

I agree that SEC- fixes should not be published until after they are deployed. (I do believe they should be published once the fix is deployed long enough to be confident the deployment release is a keeper.)

However, I've heard you speak unhappily about vulnerabilities reported as SEC issues and then languishing unpatched. 

If http://www.cert.org/kb/vul_disclosure.html were followed perhaps that would happen less often. It's an industry standard best practice on disclosure of vulnerability reports.

And yes, the disabled scripts defect was caught on the beta grid, and that indeed is what the beta grid is for. But the fact that it got that far is disturbing, and speaks to the completeness of test coverage before beta. Beta is supposed to be the *last* defense against defects, not the first. 

Because if it's the first one, it is the only one. Not exactly "defense in depth". 



      was (Author: Maggie Darwin):
    @Darling -- 

I agree that SEC- fixes should not be published until after they are deployed. (I do believe they should be published once the fix is deployed long enough to be confident the deployment release is a keeper.)

However, I've heard you speak unhappily about vulnerabilities reported as SEC issues and then languishing unpatched. 

If http://www.cert.org/kb/vul_disclosure.html were followed perhaps that would happen less often. It's an industry standard best practice on disclosure of vulnerability reports.

And yes, the disabled scripts defect was caught on the beta grid, and that indeed is what the beta grid is for. But the fact that it got that far is disturbing, and speaks to the completeness of test coverage before beta. Beta is supposed to be the *last* defense against defects, not the first. 

Because if it's the first on, it is the only one. Not exactly "defense in depth". 


  
> Linking rules for large objects broken under server 1.32
> --------------------------------------------------------
>
>                 Key: SVC-5002
>                 URL: http://jira.secondlife.com/browse/SVC-5002
>             Project: 2. Second Life Service - SVC
>          Issue Type: Bug
>    Affects Versions: 1.32 Server
>         Environment: You are at 266381.7, 296564.6, 40.9 in Gallifrey located at sim4042.agni.lindenlab.com (216.82.52.200:13001)
> Second Life Server 1.32.0.136920
>            Reporter: Cheshyr Pontchartrain
>            Priority: Showstopper
>         Attachments: Server1.32BreaksRezzeers.tga
>
>
> After the rolling restart for server 1.32.0.136920, I've received reports of broken products that I had created.  I was able to confirm this on two HORIZONS holodeck scenes.  Both objects included megaprims, and in both cases an object which used to work perfectly unlinks when rezzed.  This occurs even if I rez the object manually (not via the holodeck.)  I have since replicated this bug with normal prims, and updated the JIRA title to reflect this.
> This is new to version 1.32, as the products in question were sold and worked fine for years.

-- 
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators: http://jira.secondlife.com/secure/Administrators.jspa
-
For more information on JIRA, see: http://www.atlassian.com/software/jira

        


More information about the Jira-notify mailing list