VNC Viewer Vulnerability CVE-2008-4770

2008-11-26 Thread jnw
A vulnerability has been reported in a core VNC Viewer component's 
validation of server-supplied RFB protocol data. This issue only affects the 
VNC Viewer component, VNC Servers are not affected.

VNC Free Edition Viewer users should upgrade to version 4.1.3, and providers 
of software based on the VNC Free Edition open-source codebase should patch 
it to version 4.1.3.

For more information please visit 
http://www.realvnc.com/products/upgrade.html

Regards,

The RealVNC Team
___
VNC-List mailing list
VNC-List@realvnc.com
To remove yourself from the list visit:
http://www.realvnc.com/mailman/listinfo/vnc-list


RE: screen not correctly rendered via

2008-12-01 Thread jnw
Dieter,

It sounds like your graphics card has a second output enabled.  You should be
able to switch that off via the Display Properties dialog.

Regards,

--
Wez @ RealVNC Ltd



> -Original Message-
> From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On
> Behalf Of Dieter Blaas
> Sent: 01 December 2008 16:20
> To: vnc-list@realvnc.com
> Subject: screen not correctly rendered via
>
> Hi,
>I remotly operate a PC running WinXP SVP3 and RealVNC
> server free edition 4.1.2. The PC has a NVIDIA GeForce 2
> MX/MX400 card with the laterst drivers installed.
>
> Locally everything is fine, however remotely the screen is
> rendered with the normal hight but roughly the double
> broadness. There is no distorsion of windows or images when
> displayed on this screen. On the bottom right there is a black
> square about 1/5 of the dimension of the screen that does not
> go away whatever I do. In full screen mode the VNC screen
> looks like a small strip (narrow and very long).
>
> How do I get a normal aspect of the remote screen? This is the
> first time I have seen this phenomenon.
>
> Thanks for hints,
> Dieter
> ---
> -
> Dieter Blaas,
> Max F. Perutz Laboratories
> Medical University of Vienna,
> Inst. Med. Biochem., Vienna Biocenter (VBC),
> Dr. Bohr Gasse 9/3,
> A-1030 Vienna, Austria,
> Tel: 0043 1 4277 61630,
> Fax: 0043 1 4277 9616,
> e-mail: [EMAIL PROTECTED]
> ---
> -
> ___
> VNC-List mailing list
> VNC-List@realvnc.com
> To remove yourself from the list visit:
> http://www.realvnc.com/mailman/listinfo/vnc-list
___
VNC-List mailing list
VNC-List@realvnc.com
To remove yourself from the list visit:
http://www.realvnc.com/mailman/listinfo/vnc-list


RE: Request for allocation of new security type code for SASL auth

2008-12-15 Thread jnw
Hi Daniel,

Please use security type number 20.

Thanks,

--
Wez @ RealVNC Ltd



> -Original Message-
> From: vnc-list-ad...@realvnc.com [mailto:vnc-list-ad...@realvnc.com] On
> Behalf Of Daniel P. Berrange
> Sent: 14 December 2008 16:54
> To: vnc-list@realvnc.com
> Subject: Request for allocation of new security type code for SASL auth
> 
> I have defined a mapping of the SASL authentication scheme into the RFB
> authentication protocol. Please could you allocate an official security
> type code for use with this auth scheme, under the name "SASL".
> 
> There are a number of reasons why it is desirable to support SASL in
> the RFB protocol
> 
>  - It is an official internet RFC with API support across all OS
> platforms,
>providing for both authentication and session data encryption.
> 
>  - The SASL protocol is independant of authentication mechanism
> allowing
>new mechanisms to be plugged in without any application code changes
> 
>  - The DIGEST-MD5 SASL mechanism will provide a simple
> username+password
>based alternative to the existing "VNC" security type, but with
> stronger
>security, and session data encryption.
> 
>  - The GSSAPI SASL mechanism will provider integration with Kerberos
> for
>Single-Sign-On capabilities, again also having session data
> encryption
> 
> There are a variety of other plugins that will be useful to many people
> such as passwd auth against a SQL database,  auth against LDAP, one-
> time
> passwords, and NTLM.
> 
> I intend to implement client side support for this "SASL" security type
> in the GTK-VNC client, the GNOME VINO server (remote desktop), and the
> QEMU built-in VNC server (OS virtualization).
> 
> My current security type protocol spec documentation follows, though I
> may
> well refine details of it as I finish the implementations. I currently
> have GTK-VNC and QEMU operational, without session encryption.
> 
> NB, the protocol description that follows, just picks security code 100
> out of the hat. This will obviously be updated to reflect the actual
> security type code officially allocated.
> 
> Daniel
> 
> 6.2.100 SASL Authentication
> ===
> 
> SASL authentication is to be used, optionally with a encryption
> capability for the protocol data to be negotiated between the
> client and server. The author recommends that an SSF layer always
> be requested during negotiation unless it is known that the
> connection is secured by a prior authentication exchange such as
> TLS / VeNCrypt, or externally via a SSH tunnel.
> 
> The SASL negotiation is a multi-step protocol, initiated by the
> server. The precise number of steps required for the complete
> negotiation is determined by the SASL mechanism chosen for auth.
> Compliant implementations must expect an arbitrary number of
> steps.
> 
> The SASL protocol is defined in RFC . Recommended libraries
> for implementation on either client or server end are Cyrus-SASL
> and gSASL.
> 
> 
> 6.2.100.1 SASL server initialization message
> 
> 
> The server initializes the SASL authentication process by sending
> the list of mechanisms it is prepared to accept from the client.
> 
>  # of bytes   | Type | Description
>  -+--+
>   4   | u32  | mechlist-length
>   mechlist-length | u8 array | mechlist-string
>  -
> 
> The "mechlist-string" is a list of SASL mechanism names, each
> separated by a comma. The "mechlist-length" is the number of
> characters in the string. The trailing '\0' is not transmitted
> on the wire. Some example "mechlist-string" values are
> 
> "DIGEST-MD5,GSSAPI"
> "ANONYMOUS,KERBEROS_V4,DIGEST-MD5"
> 
> The "mechlist-string" is typically generated on the server by
> a call to the function:
> 
> err = sasl_listmech(vs->saslconn,
> NULL,
> "",  /* Prefix */
> ",", /* Separator */
> "",  /* Suffix */
> &mechlist,
> NULL,
> NULL);
> 
> The library used for the SASL implementation will usually
> provide a means for the adminsitrator to configure exactly
> what mechanisms are enabled.
> 
> 
> 6.2.100.2 SASL client start message
> ---
> 
> Upon receiving the list of mechanisms allowed by the server,
> the client may choose a specific mechanism from the list, or
> allow the SASL library to make the choice.
> 
> The client shall call "sasl_client_start" providing either
> a specific mechanism name, or the full list supported by the
> server. The client shall normally supply a list of prompts
> it is prepared to support for collecting credentials from
> the user.
> 
> If the "sasl_client_start" call is successfull, the returned
> "clientout" data will need to be sent to the se