Most complex applications have twist and turns and not all of the apply evenly 
across the installations. The CVSS score is use to summarize the overall risk, 
exploitability, and how commonly used the vulnerable is deployed. It is an 
imperfect scale but it is the one the industry has settled on. It is very 
annoying that you have chose to ignore it. 

Darryl Baker, GSEC, GCLD (he/him/his) 
Sr. System Administrator 
Distributed Application Platform Services 
Northwestern University 
4th Floor 
2020 Ridge Avenue 
Evanston, IL 60208-0801 
[email protected] <mailto:[email protected]> 
(847) 467-6674 <tel:+18474676674> 




On 8/28/26, 3:45 PM, "Christopher Schultz" <[email protected] 
<mailto:[email protected]>> wrote:


Darryl,


On 8/26/26 4:14 PM, Darryl Baker wrote:
> I am trying to figure out when a vulnerability justifies and out of cycle
> update to our Tomcat installations. After determining if we are vulnerable
> then determining the risk but our security group focuses on the CVSS
> score. With the Tomcat reports not including one I have little to go
> by. Is there a rough mapping of your “Severity” and a range of CVSS
> scores? Today I need to know is “Severity: important” a high enough risk
> to justify an out of cycle update of our Tomcat installations?
To answer your question directly: no such mapping exists.


This is precisely because the level of severity will vary wildly between 
users. For example, if you don't use HTTP/2 at all than it doesn't 
matter if the CVSS score is a perfect-10.


The Tomcat Security Team has tried to capture the overall severity for 
"most users". Some security announcements represented as "Low" by the 
Tomcat Security Team might end up being super important for you. To take 
a recent example: CVE-2026-65183. If you are using UNIX domain sockets 
in an adversarial environment, then the fix for CVE-2026-65183 is far 
more important to you than the "Low" severity we have assigned. But 
since most users are (a) not using UNIX domain sockets and (b) aren't in 
adversarial environments, the Tomcat Security Team has given this CVE a 
"low" rating.


(After writing this and then reading Reg's reply, I find that I'm just 
saying the exact same thing. Hopefully, this repetition is actually 
helpful.)


Another way to look at this would be "how painful is it to actually do a 
Tomcat upgrade"? I don't know what your patch/upgrade cycles look like, 
but if you invest some time and energy into making Tomcat upgrades 
cheaper or less onerous, then maybe it doesn't matter what the CVSS 
scores, etc. are. Just upgrade because it's easier than fighting with 
your Security Overlords. :)


-chris




---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected] 
<mailto:[email protected]>
For additional commands, e-mail: [email protected] 
<mailto:[email protected]>





Reply via email to