Follow @Openwall on Twitter for new release announcements and other news
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Fri, 21 Aug 2009 09:28:08 +0200
From: "Matthias Andree" <>
Cc: "Steven M. Christey" <>
Subject: Re: "umbrella" CVE names (was: CVE request: fetchmail <= 6.3.10 SSL
 certificate NUL prefix verification bypass)

Am 05.08.2009, 20:13 Uhr, schrieb Steven M. Christey  

> So use CVE-2009-2666 for fetchmail (I'll fill it in later) and Tomas,  
> even
> if it results in dozens of CVEs, I suspect this is how we should go.

Following up an earlier question of mine (umbrella CVE for a class of  
problems, here: weak X.509 name verification that terminates early on  
embedded NUL bytes):

Mandriva Security (I think it was them - if I recall correctly) wrote in  
their fetchmail security advisory something along the lines of  
"CVE-2009-2666, [...] related to CVE-2009-2408" (with some more details).  
This is probably the best way around this problem of how do we assign and  
organize: We have the individual CVE name for the fetchmail weakness (so  
it can be tracked), and we also have as reference the CVE name of the  
first published issue that sort of founded a problem class, by instance of  
Mozilla NSS.

CVE-2009-2408 here turns into some dual-use: (1) to track the  
library/Mozilla application bug, (2) to name the problem class.

Perhaps this should/could be considered a pragmatic solution to the  
"umbrella CVE" problem I posed earlier.

Matthias Andree

Powered by blists - more mailing lists

Please check out the Open Source Software Security Wiki, which is counterpart to this mailing list.

Confused about mailing lists and their use? Read about mailing lists on Wikipedia and check out these guidelines on proper formatting of your messages.