Bug 325798

Summary:	Xvnc uses all the CPU it can after a machine has been portscanned
Product:	[openSUSE] openSUSE 10.3	Reporter:	Michael Wolf <maw>
Component:	X11 Applications	Assignee:	Reinhard Max <max>
Status:	RESOLVED WONTFIX	QA Contact:	Stefan Dirsch <sndirsch>
Severity:	Normal
Priority:	P5 - None
Version:	Beta 3
Target Milestone:	---
Hardware:	Other
OS:	openSUSE 10.3
Whiteboard:
Found By:	---	Services Priority:
Business Priority:		Blocker:	---
Marketing QA Status:	---	IT Deployment:	---

Description Michael Wolf 2007-09-17 20:07:30 UTC

Xvnc uses all the CPU it can after a machine has been portscanned.

Steps to reproduce:

1) Install and activate Xvnc (I have xorg-x11-Xvnc)

2) From a different box, run nmap against the machine where Xvnc is enabled

3) On the target machine, watch top -- Xvnc will show up as one of the top consumers of CPU

Comment 1 Michael Wolf 2007-09-17 20:08:33 UTC

Woops, forgot to mention that this doesn't happen if you do "nmap localhost".

However, a similar problem exists on 10.2 -- AFAICT it is the same, with the following exception: the problem _does_ occur if you do "nmap localhost".

Comment 2 Stefan Dirsch 2007-09-17 21:07:10 UTC

What do you mean with activate Xvnc? Run it manually or enable remote configuration with YaST2?

Comment 3 Michael Wolf 2007-09-17 21:12:45 UTC

(In reply to comment #2 from Stefan Dirsch)
> What do you mean with activate Xvnc? Run it manually or enable remote
> configuration with YaST2?

The latter - configuring remote administration via yast, and poking a hole in the firewall if necessary.

Comment 4 Stefan Dirsch 2007-09-17 21:41:35 UTC

I can reproduce this issue. It doesn not happen, when you run Xvnc manually with the same options. Maybe Reinhard has an idea.

Comment 5 Reinhard Max 2008-08-13 16:50:34 UTC

Whoops, this one is still sitting in my bug list.

Having no 10.3 installations left here, I just tried it on 11.0, first with the Xvnc and xinetd packages from 11.0 and then with the ones from 10.3 and couldn't reproduce it.

As I think this isn't emportant enough for finding and fixing it in 10.3 at this point, I'll close it with WONTFIX.

Please re-open if you manage to reproduce it in 11.0 and provide more details, e.g. on the architecture of the involved machines and on the options you gave to nmap.