Re: netbios [7:71084]

2003-06-24 Thread j k
hi pple, well the reason why i ask this is because, recently i was told by
my network manager that there is a virus which uses netbios (udp 137, tcp
138 and tcp 139) as a transport and had acrosses the WAN from a spoke site
to a hub site. And i was told to put an ACL by blocking the above port on
the fastethernet interface, well i was kind of confuse as in, i remember
that netbios arnt routable across the WAN, IF, and i mean IF there is really
such virus uses this ports, they shouldnt be able to traverse to the other
site across the WAN rite?? And when i did some debug ip packet, the udp 136
and or ofcourse the tcp138 and 139, was captured and dropped! at the
fastethernet interface and TR interface (i had place the ACL on both
fastether and TR) but when i place it on the serial, i dun see any udp 136
at all!...i jus need some clarification from people at this forum here



Message Posted at:
http://www.groupstudy.com/form/read.php?f=7i=71227t=71084
--
FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html
Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]


Re: netbios [7:71084]

2003-06-24 Thread ericbrouwers
Is NetBIOS a protocol in the sense of ISO's OSI definition?? I never really
checked it. Originally it was a programming interface on IBM PCs. I did some
network programming with NetBIOS back in 1989... yes, old man...

When I started reading commercial Cisco certification books, the authors
sometimes tried to convince me that it is a protocolWhatever, I'm not
going to
give a formal answer, but for those interested maybe give the following a
try. It's from IBM's TCP/IP Tutorial and Technical Overview, October 1998,
one of their famous redbooks (http://www.redbooks.ibm.com):

...
NetBIOS is a vendor-independant software interface (API), not a protocol.
There is no official NetBIOS specification, although in practice, the
NetBIOS version described in the IBM publication SC30-3587 LAN Technical
Reference: 802.2 and NetBIOS APIs is used as reference.
..

Have fun!
:-)
Eric Brouwers

- Original Message -
From: Priscilla Oppenheimer 
To: 
Sent: Tuesday, June 24, 2003 3:50 AM
Subject: Re: netbios [7:71084]


 - jvd wrote:
 
  OT:
  hi, i just have to say that i will never try to answer anything
  on this forum again. :-)

 Well, would that be Grumpy, Bashful, Sleepy, or Dopey to do that? :-)
 Seriously, you should keep answering. You have sent some great answers,
but
 you don't want to keep insisting something when replying to my messages.
It
 makes me very Grumpy and I'm not Bashful when wielding a keyboard (just in
 person). I know lots of books claim that NetBIOS isn't routable, but I bet
 those exact same books also classify it as a session-layer protocol. And
it
 does make a good example of a session-layer protocol. One of the few that
we
 have! And if it runs at that layer, then it is routable. I think even IBM
 said it was a session-layer protocol in some of their early documents,
which
 unfortunately, I recently tossed.

 Directed broadcasts came from out of the blue. I really don't think
Windows
 networking uses them, although maybe it does. Was the comment maybe in
 reference to the helper address suggestion that I made? You can tell a
 router to send the packets when it helps as a broadcast. That's not a
 directed broadcast, though, and will work even if router forwarding of
 directed broadcasts is disabled, which is the default these days. Instead,
 it's a broadcast sent by the router (it has the router's IP address as
 source, on behalf of some other station, to a local LAN, because the
router
 is acting as a proxy, for example, a DHCP Relay Agent.) Was that a run-on
 sentence, or what? :-)

 A directed broadcast is directed from afar into a subnet. The sender
usually
 makes classful assumptions, since it can't actually know the local
 definition of a broadcast. It's used by ping scan to send a ping to
 172.16.255.255, for example, in an attempt to ping everyone on network
 172.16.0.0. Routers don't forward those these days because of the security
 risks.

 Back to NetBIOS. It does send a lot of broadcast traffic for naming
 purposes. In an IP environment, however, a host can be configured to send
 unicast naming queries and name registrations to a WINS server. There are
 probably lots of other issues, though. It really can be quite a pain to
get
 it to work correctly when you migrate from a small LAN to a larger
 internetwork with WANs, subnetting, VLANs, etc.


 I wonder what the original poster is really trying to do and where he can
 get a good Windows networking (internetworking) design guide. Cisco used
to
 have one, but it's probably way dated now


 Well, it's late and my writing is deteriorating. Howard covers directed
 broadcasts, by the way, (and a much better description of the OSI model,
 without reference to the dwarves, as I recall, although possibly with
 reference to the deadly sins) in his CertificationZone papers. I recommend
 them.

 Priscilla


 
  once i tried to answer a question with regards to bgp and a
  1720 router and only after howard helped us out was it clear
  that the processor does play an important role. ;-)
 
  this time only after the input from priscilla is everybody
  happy about the netbios/netbeui issue. ;-)
 
  but then i think what is important is that we dig a bit deeper
  into some topics!
 
  Good work!




Message Posted at:
http://www.groupstudy.com/form/read.php?f=7i=71231t=71084
--
FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html
Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]


Re: netbios [7:71084]

2003-06-24 Thread annlee
Since the original question related to virus and certain ports, etc., here's
a good reference to keep an eye on:

http://isc.incidents.org/

You'll notice which ports have the most activity by geographical region
(there are marked differences). You can also look at the hyperlinks
associated with each of the most frequently attacked ports as well as the
links under ISC Analysis to particular exploits that are currently being
seen frequently.

This may help you with your manager, as it's not necessarily enough to
understand what's happening across the router; you may need to know what's
happening on the hosts, as well.

HTH

Annlee

j k  wrote in message
news:[EMAIL PROTECTED]
 hi pple, well the reason why i ask this is because, recently i was told by
 my network manager that there is a virus which uses netbios (udp 137, tcp
 138 and tcp 139) as a transport and had acrosses the WAN from a spoke site
 to a hub site. And i was told to put an ACL by blocking the above port on
 the fastethernet interface, well i was kind of confuse as in, i remember
 that netbios arnt routable across the WAN, IF, and i mean IF there is
really
 such virus uses this ports, they shouldnt be able to traverse to the other
 site across the WAN rite?? And when i did some debug ip packet, the udp
136
 and or ofcourse the tcp138 and 139, was captured and dropped! at the
 fastethernet interface and TR interface (i had place the ACL on both
 fastether and TR) but when i place it on the serial, i dun see any udp 136
 at all!...i jus need some clarification from people at this forum here




Message Posted at:
http://www.groupstudy.com/form/read.php?f=7i=71233t=71084
--
FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html
Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]


Re: netbios [7:71084]

2003-06-24 Thread Carroll Kong
From my /etc/services...

netbios-ns  137/tcp#NETBIOS Name Service
netbios-ns  137/udp#NETBIOS Name Service
netbios-dgm 138/tcp#NETBIOS Datagram Service
netbios-dgm 138/udp#NETBIOS Datagram Service
netbios-ssn 139/tcp#NETBIOS Session Service
netbios-ssn 139/udp#NETBIOS Session Service

I believe that these are the NetBIOS over TCP/IP instantiations so to 
speak.  While NetBIOS can easily be run over IPX/SPX or even NetBeui, 
clearly a tcp/ip port number has to relevant in that case.

[rant mode]

I cannot blame you for the confusion as Priscilla mentioned that 
quite a few people somehow believe it is not.  I think they are 
confusing it with NetBeui which techically has nothing to do with 
each other.  (yes the name Netbeui means Netbios Extended User 
Interface, but still, technically nothing to do with each other in 
terms of NetBios functionality, it can ride over other network 
transports)

I have had countless debates and arguments where people insisted they 
are bound to the hip or interchange their names like candy.

Here is an interesting excerpt of some dialog I had at a startup I 
worked at years ago.

Premise:
When dealing with two separate LANs, as defined as Layer2 domains

Is it possible to get network neighborhood to work between the 
upstairs and the basement. - VP/Sales
Sure, we just need to bind Netbios over TCP/IP and make sure we can 
route over the two different networks.  We might need to deal with 
WINS for seamless naming integration but it should work fine 
otherwise.  - Carroll
You also will need NetBeui. - Other Tech Guy
[Trying to be nice].  No, sorry [Other Tech Guy], I am pretty sure 
you will not. - Carroll
Yes you do. - Other Tech Guy
[Still trying to be nice.].  Well, I do not think you do, since 
Netbeui is a transport protocol, and Netbios rides on top of any 
protocol it wants to.  You already have TCP/IP as your transport, you 
do not need Netbeui, and on top of that, Netbeui will not cross over 
the LAN. - Carroll
You are wrong, you need Netbeui. - Other Tech Guy

Trying the wait, look there is a transport, you only need one 
angle.

But, if that was true, how come I can get a Unix box with Samba to 
work with a Windows machine.  TCP/IP is the transport there, my Unix 
box has no concept of NetBeui yet it works. - Carroll
Look, Carroll, I have been in the ISP business for over 5 years, I 
think I know what I am doing. - Other Tech Guy

Not that I could see the relevance of NetBeui in an ISP, just that he 
was clearly pushing his move aside green horn argument instead of 
trying to sensible attack the problem through theory.

Well, since the other tech guy was older than me, and supposedly 
far more experienced, they made sure Netbeui was on every machine.  
Sigh, I had other responsibilities rather than to go around proving 
him wrong.  But experiences like these is what makes me say...

-  Check the theory and make sure it sounds right.
-  Check the practice, make sure it works right.
-  I don't care about your past experiences; technology moves so fast 
it invalidates so many truisms within months.

The guy was wrong on 1, 2, and... for 3, he never had a truism to 
begin with, just a false sense of knowledge of the systems he worked 
with.

As with those logical fallacies, does not matter how smart or how 
great your past work is, people can make mistakes.  If you say 
something that is true in the now, it is true.  If you say 
something that is false in the now it is false regardless of your 
past history.

 hi pple, well the reason why i ask this is because, recently i was told by
 my network manager that there is a virus which uses netbios (udp 137, tcp
 138 and tcp 139) as a transport and had acrosses the WAN from a spoke site
 to a hub site. And i was told to put an ACL by blocking the above port on
 the fastethernet interface, well i was kind of confuse as in, i remember
 that netbios arnt routable across the WAN, IF, and i mean IF there is
really
 such virus uses this ports, they shouldnt be able to traverse to the other
 site across the WAN rite?? And when i did some debug ip packet, the udp 136
 and or ofcourse the tcp138 and 139, was captured and dropped! at the
 fastethernet interface and TR interface (i had place the ACL on both
 fastether and TR) but when i place it on the serial, i dun see any udp 136
 at all!...i jus need some clarification from people at this forum here

-Carroll Kong




Message Posted at:
http://www.groupstudy.com/form/read.php?f=7i=71235t=71084
--
FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html
Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]


Re: netbios [7:71084]

2003-06-24 Thread Howard C. Berkowitz
At 11:26 AM + 6/24/03, ericbrouwers wrote:
Is NetBIOS a protocol in the sense of ISO's OSI definition??


Not necessarily for ISO, but see RFC 1001 and 1002 for the IETF definition.

  I never really
checked it. Originally it was a programming interface on IBM PCs. I did some
network programming with NetBIOS back in 1989... yes, old man...

When I started reading commercial Cisco certification books, the authors
sometimes tried to convince me that it is a protocolWhatever, I'm not
going to
give a formal answer, but for those interested maybe give the following a
try. It's from IBM's TCP/IP Tutorial and Technical Overview, October 1998,
one of their famous redbooks (http://www.redbooks.ibm.com):

...
NetBIOS is a vendor-independant software interface (API), not a protocol.
There is no official NetBIOS specification, although in practice, the
NetBIOS version described in the IBM publication SC30-3587 LAN Technical
Reference: 802.2 and NetBIOS APIs is used as reference.
..

Have fun!
:-)
Eric Brouwers

- Original Message -
From: Priscilla Oppenheimer
To:
Sent: Tuesday, June 24, 2003 3:50 AM
Subject: Re: netbios [7:71084]


  - jvd wrote:
  
   OT:
   hi, i just have to say that i will never try to answer anything
   on this forum again. :-)

  Well, would that be Grumpy, Bashful, Sleepy, or Dopey to do that? :-)
  Seriously, you should keep answering. You have sent some great answers,
but
  you don't want to keep insisting something when replying to my messages.
It
  makes me very Grumpy and I'm not Bashful when wielding a keyboard (just
in
  person). I know lots of books claim that NetBIOS isn't routable, but I
bet
  those exact same books also classify it as a session-layer protocol. And
it
  does make a good example of a session-layer protocol. One of the few that
we
  have! And if it runs at that layer, then it is routable. I think even IBM
  said it was a session-layer protocol in some of their early documents,
which
  unfortunately, I recently tossed.

  Directed broadcasts came from out of the blue. I really don't think
Windows
  networking uses them, although maybe it does. Was the comment maybe in
  reference to the helper address suggestion that I made? You can tell a
  router to send the packets when it helps as a broadcast. That's not a
  directed broadcast, though, and will work even if router forwarding of
  directed broadcasts is disabled, which is the default these days.
Instead,
  it's a broadcast sent by the router (it has the router's IP address as
  source, on behalf of some other station, to a local LAN, because the
router
  is acting as a proxy, for example, a DHCP Relay Agent.) Was that a run-on
  sentence, or what? :-)

  A directed broadcast is directed from afar into a subnet. The sender
usually
  makes classful assumptions, since it can't actually know the local
  definition of a broadcast. It's used by ping scan to send a ping to
  172.16.255.255, for example, in an attempt to ping everyone on network
  172.16.0.0. Routers don't forward those these days because of the
security
  risks.

  Back to NetBIOS. It does send a lot of broadcast traffic for naming
  purposes. In an IP environment, however, a host can be configured to send
  unicast naming queries and name registrations to a WINS server. There are
  probably lots of other issues, though. It really can be quite a pain to
get
  it to work correctly when you migrate from a small LAN to a larger
  internetwork with WANs, subnetting, VLANs, etc.


  I wonder what the original poster is really trying to do and where he can
  get a good Windows networking (internetworking) design guide. Cisco used
to
  have one, but it's probably way dated now


  Well, it's late and my writing is deteriorating. Howard covers directed
  broadcasts, by the way, (and a much better description of the OSI model,
  without reference to the dwarves, as I recall, although possibly with
  reference to the deadly sins) in his CertificationZone papers. I
recommend
   them.

  Priscilla


  
   once i tried to answer a question with regards to bgp and a
   1720 router and only after howard helped us out was it clear
   that the processor does play an important role. ;-)
  
   this time only after the input from priscilla is everybody
   happy about the netbios/netbeui issue. ;-)
  
   but then i think what is important is that we dig a bit deeper
   into some topics!
  
   Good work!




Message Posted at:
http://www.groupstudy.com/form/read.php?f=7i=71239t=71084
--
FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html
Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]


Re: netbios [7:71084]

2003-06-24 Thread Priscilla Oppenheimer
j k wrote:
 
 hi pple, well the reason why i ask this is because, recently i
 was told by my network manager that there is a virus which uses
 netbios (udp 137, tcp 138 and tcp 139) as a transport and had
 acrosses the WAN from a spoke site to a hub site.

The NetBIOS ports are infamous targets for hackers and viruses. Yes,
definitely close them up.

 And i was
 told to put an ACL by blocking the above port on the
 fastethernet interface, well i was kind of confuse as in, i
 remember that netbios arnt routable across the WAN, IF, and i
 mean IF there is really such virus uses this ports, they
 shouldnt be able to traverse to the other site across the WAN
 rite?? 

Yes, they can traverse. They are carried in IP, so of course, they are
routable. But the packets to UDP port 137 are usually broadcasts, and so
they don't traverse without a helper address.

 And when i did some debug ip packet, the udp 136 and or
 ofcourse the tcp138 and 139, was captured and dropped! at the
 fastethernet interface and TR interface (i had place the ACL on
 both fastether and TR) but when i place it on the serial, i dun
 see any udp 136 at all!...i jus need some clarification from
 people at this forum here

136 is a typo?

If the broadcast packets using port 137 don't get through, perhaps you won't
see the 138 and 139. The session establishment won't work if the broadcasts
don't work first.

I would still block it. It can't hurt. They are infamous. And, of course it
is routable. If you've read enough networking material to have heard that
NetBIOS isn't routable then you must have some idea of what UDP and TCP do
and what they run above and one of the main jobs of that protocol!?! Sorry,
getting GRUMPY again. :-)

Priscilla

 




Message Posted at:
http://www.groupstudy.com/form/read.php?f=7i=71269t=71084
--
FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html
Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]


Re: netbios [7:71084]

2003-06-24 Thread Priscilla Oppenheimer
annlee wrote:
 
 Since the original question related to virus and certain ports,
 etc., here's
 a good reference to keep an eye on:
 
 http://isc.incidents.org/

Great! There you have it. NetBIOS port 137 at the top of the list. Since
broadcasts aren't carried across a router, the attackers don't send as
broadcasts, as a real Windows station would. But who says the attackers have
to behave like real stations? :-0

I'm sure most personal firewall default to blocking the NetBIOS ports. I
think it's a good idea to block on global firewalls too.

I wish I hadn't published my Windows troubleshooting information exclusively
with a company that essentially swallowed it and made it disappear. Here's
an excerpt from it, FYI. It was mostly written by my co-author, Joe Bardwell.

The terminology of NetBIOS communication can be confusing. This is because
the NetBIOS acronym has been used to describe more than one thing. NetBIOS
refers to the programming interface in all implementations. In the
NetBIOS/TCP environment, the term NetBIOS also refers to the portion of the
packet that carries NetBIOS commands, replies, and data. In the
NetBIOS/NetBEUI environment, the term NetBIOS refers only to the API, and
the term NetBEUI refers to the protocol. In the NetBIOS/IPX environment, the
term NetBIOS refers to both the API and to the protocol. To understand the
details of terminology use, itÂ’s worthwhile to examine the three different
frame structures for TCP, NetBEUI, and IPX.

A Windows Internet Name Service Query Carried on UDP

The NetBIOS/TCP implementation includes NetBIOS commands, replies, and data
carried on both TCP and UDP. When a station wants to determine the IP
address associated with a particular NetBIOS name, it sends a Windows
Internet Name Service (WINS) query which is carried on top of UDP. In this
case, there is no specific NetBIOS header in the packet, as seen in the
following analyzer output. The packet simply carries a NetBIOS Name Service
command directly above UDP.

 Ethernet Header
   Destination:  FF:FF:FF:FF:FF:FF  Ethernet Broadcast
   Source:   00:60:08:15:A6:9B
   Protocol Type:0x0800  IP
 IP Header - Internet Protocol Datagram
   Version:  4
   Header Length:5  (20  bytes)
   Type of Service:  %
   Precedence: Routine,Normal Delay,Normal Throughput,Normal
 Reliability
   Total Length: 78
   Identifier:   43062
   Fragmentation Flags:  %000  May Fragment   Last Fragment
   Fragment Offset:  0  (0  bytes)
   Time To Live: 128
   Protocol: 17  UDP
   Header Checksum:  0x1781
   Source IP Address:192.216.124.55
   Dest. IP Address: 192.216.124.255
   No IP Options
 UDP - User Datagram Protocol
   Source Port:  137  NETBIOS Name Service
   Destination Port: 137
   Length:   58
   Checksum: 0x8FD2
 NetBIOS Name Service - Network Basic Input/Output System
   Identification:   0x883A
   Parameter:0x0110
 Request
 Standard Query
 Recursion Desired
 Packet Was Broadcast
   Number of Questions:  1
   Number of Answers:0
   Number of Authority:  0
   Number of Additional: 0
   Query Domain Name:MIKE-PC   Server Service
   Query Type:   32  NetBIOS General Name Service
   Query Class:  1  Internet
 Frame Check Sequence:  0x59DF750B

A TCP NetBIOS Session Setup Request

After determining the IP address of a target node, a NetBIOS/TCP station
resolves the IP address to a data-link-layer address by sending an Address
Resolution Protocol (ARP) frame. (The station uses the data-link-layer
address of the Default Gateway for remote targets). Next, the station
establishes a TCP session with the target in the normal manner with a TCP
three-way handshake. Using the established TCP session, the originator must
now create a NetBIOS session. The following packet is an example of a
NetBIOS Session Setup request.

 Flags:0x00
   Status:   0x01
   Packet Length:130
 Ethernet Header
   Destination:  00:40:95:96:30:07
   Source:   00:60:08:15:A6:9B
   Protocol Type:0x0800  IP
 IP Header - Internet Protocol Datagram
   Version:  4
   Header Length:5  (20  bytes)
   Type of Service:  %
   Precedence: Routine,   Normal Delay,   Normal Throughput,   Normal
 Reliability
   Total Length: 112
   Identifier:   43830
   Fragmentation Flags:  %010  Do Not Fragment   Last Fragment
   Fragment Offset:  0  (0  bytes)
   Time To Live: 128
   Protocol: 6  TCP
   Header Checksum:  0xD53B
   Source IP Address:192.216.124.55
   Dest. IP Address: 192.216.124.45
   No IP Options
 TCP - 

Re: netbios [7:71084]

2003-06-24 Thread j k
ok great referrences, thanks everybody, keep the forum live!!!

:-)


Message Posted at:
http://www.groupstudy.com/form/read.php?f=7i=71312t=71084
--
FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html
Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]


Re: netbios [7:71084]

2003-06-23 Thread jountao
Directed broadcastes are less of a burden as far as passing brodcasts for
specific apps.
indeed netbios is not routeble, to route it u use NBT or NBIPX

Priscilla Oppenheimer  a icrit dans le message de
news: [EMAIL PROTECTED]
 koh jef wrote:
 
  hi guys,
 
  can netbios,using port 137, 138 and 139 be routed thru WAN ???

 NetBIOS uses UDP and TCP which run on top of IP, which is routable over an
 IP internetwork, including WAN links.

 Routers don't forward broadcasts though, by default. When NetBIOS runs
over
 UDP ports 137 and 138, a lot of it is broadcasts. You can use an IP helper
 address and udp forwarding on a router to get the router to forward those.
 That might not be such a good idea, though. It could make resources
 available across the WAN that you don't want to make available. It could
 require you to open ports on firewalls, resulting in security risks.

 You need take a higher-level view of what you're trying to do... Windows
 networking across an internetwork can be challenging...

 Priscilla




Message Posted at:
http://www.groupstudy.com/form/read.php?f=7i=71126t=71084
--
FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html
Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]


Re: netbios [7:71084]

2003-06-23 Thread - jvd
Hi,

I just want to make something clear. NetBIOS and NetBEUI were created by IBM
and later found its way to Microsoft networks who changed it.

- These protocols are non-routable in an IBM environment.
- NetBEUI is non-routable in TCP/IP networks.
- NetBIOS is non-routable without the help of NetBT.

NetBT is NetBIOS over TCP/IP and was conceived in RFCs 1001 and 1002 to
enable NetBIOS to be routed. Short overview here:
http://www.microsoft.com/windows2000/techinfo/reskit/en-us/default.asp?url=/windows2000/techinfo/reskit/en-us/cnet/cnad_arc_khqp.asp




Message Posted at:
http://www.groupstudy.com/form/read.php?f=7i=71150t=71084
--
FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html
Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]


Re: netbios [7:71084]

2003-06-23 Thread Priscilla Oppenheimer
jountao wrote:
 
 Directed broadcastes are less of a burden as far as passing
 brodcasts for
 specific apps.

Directed broadcasts are not allowed on modern networks and are totally
irrelevant to Windows networking anyway, which doesn't use them.

 indeed netbios is not routeble, to route it u use NBT or NBIPX

He's using NBT obviously or he wouldn't be talking about UDP and TCP port
numbers.

Priscilla


 
 Priscilla Oppenheimer  a icrit dans
 le message de
 news: [EMAIL PROTECTED]
  koh jef wrote:
  
   hi guys,
  
   can netbios,using port 137, 138 and 139 be routed thru WAN
 ???
 
  NetBIOS uses UDP and TCP which run on top of IP, which is
 routable over an
  IP internetwork, including WAN links.
 
  Routers don't forward broadcasts though, by default. When
 NetBIOS runs
 over
  UDP ports 137 and 138, a lot of it is broadcasts. You can use
 an IP helper
  address and udp forwarding on a router to get the router to
 forward those.
  That might not be such a good idea, though. It could make
 resources
  available across the WAN that you don't want to make
 available. It could
  require you to open ports on firewalls, resulting in security
 risks.
 
  You need take a higher-level view of what you're trying to
 do... Windows
  networking across an internetwork can be challenging...
 
  Priscilla
 
 




Message Posted at:
http://www.groupstudy.com/form/read.php?f=7i=71156t=71084
--
FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html
Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]


Re: netbios [7:71084]

2003-06-23 Thread Priscilla Oppenheimer
NetBIOS is a session layer protocol. Of course it's routable.

Yes, it's true that IBM invented NetBIOS.

NetBEUI was the name of IBM's software module that implemented NetBIOS. It
usually ran in conjunction with TOKREUI, by the way, which was driver
software for Token Ring.

NetBEUI didn't make any calls to a network layer. It assumed it was running
directly above LLC and made calls to LLC's interface. It also make
source-route bridging calls, by the way.

It's just semantics at this point but does point out a couple higher-level
concepts.

One, networking is more complicated than the simple statements like NetBIOS
is not routable that you will find in Networking 101 papers.

Two, it's important to understand that every layer makes calls to a layer
below and every layer offers services to a layer above. Because the service
interface of a layer, whether it be LLC or UDP, is well known (hopefully),
any upper layer can call on it. So, an implementation of a session layer
protocol such as NetBIOS can call on UDP, TCP, LLC, or IPX. That's the real
message from the infamous OSI model. It's all about service interfaces.


Priscilla


- jvd wrote:
 
 Hi,
 
 I just want to make something clear. NetBIOS and NetBEUI were
 created by IBM and later found its way to Microsoft networks
 who changed it.
 
 - These protocols are non-routable in an IBM environment.
 - NetBEUI is non-routable in TCP/IP networks.
 - NetBIOS is non-routable without the help of NetBT.
 
 NetBT is NetBIOS over TCP/IP and was conceived in RFCs 1001 and
 1002 to enable NetBIOS to be routed. Short overview here:

http://www.microsoft.com/windows2000/techinfo/reskit/en-us/default.asp?url=/windows2000/techinfo/reskit/en-us/cnet/cnad_arc_khqp.asp
 
 




Message Posted at:
http://www.groupstudy.com/form/read.php?f=7i=71161t=71084
--
FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html
Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]


Re: netbios [7:71084]

2003-06-23 Thread Howard C. Berkowitz
At 6:15 PM + 6/23/03, Priscilla Oppenheimer wrote:


Two, it's important to understand that every layer makes calls to a layer
below and every layer offers services to a layer above. Because the service
interface of a layer, whether it be LLC or UDP, is well known (hopefully),
any upper layer can call on it. So, an implementation of a session layer
protocol such as NetBIOS can call on UDP, TCP, LLC, or IPX. That's the real
message from the infamous OSI model. It's all about service interfaces.


OO!  The real message is there are seven layers named Happy, 
Sneezy, Grumpy, Bashful, Sleepy, Dopey and Doc!




Message Posted at:
http://www.groupstudy.com/form/read.php?f=7i=71171t=71084
--
FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html
Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]


Re: netbios [7:71084]

2003-06-23 Thread - jvd
hi, i just have to say that i will never try to answer anything on this
forum again. :-)

once i tried to answer a question with regards to bgp and a 1720 router and
only after howard helped us out was it clear that the processor does play an
important role. ;-)

this time only after the input from priscilla is everybody happy about the
netbios/netbeui issue. ;-)

but then i think what is important is that we dig a bit deeper into some
topics!

Good work!



Message Posted at:
http://www.groupstudy.com/form/read.php?f=7i=71179t=71084
--
FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html
Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]


Re: netbios [7:71084]

2003-06-23 Thread John Neiberger
 - jvd 6/23/03 4:06:36 PM 
hi, i just have to say that i will never try to answer anything on this
forum again. :-)

once i tried to answer a question with regards to bgp and a 1720 router
and
only after howard helped us out was it clear that the processor does play
an
important role. ;-)

this time only after the input from priscilla is everybody happy about the
netbios/netbeui issue. ;-)

but then i think what is important is that we dig a bit deeper into some
topics!

Good work!

I have two requests:

First, don't go into hiding.  Please continue to participate.  We welcome
all participants!

And second, please quote the post you're referring to when you reply.  When
you don't include a quote it is very difficult to follow what you're talking
about.

Regards,
John




Message Posted at:
http://www.groupstudy.com/form/read.php?f=7i=71186t=71084
--
FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html
Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]


Re: netbios [7:71084]

2003-06-23 Thread Bikespace
There are too many smart arses around sometimes. The whole idea of a
discussion group isn't for everyone to Provide the right answer or beware.
I think everyone would like to have a go sometimes, but there's that nagging
doubt that you're going to make a big slip up.
It's better I think when people have a go. It's amazing how many people
answer the easy password recovery type questions, but no one is listening
when the toughies come out, even though a lot of people could have a stab at
it (guilty myself).
Some people would end up on their arse if they replied face to face, the way
they do in some follow up posts.

Good on you - I will try to do my bit by making a fool of myself at every
opportunity. After about 6 pints this may well be the first of many.
:-)
Keep digging.

Bikespace.



- jvd  wrote in message
news:[EMAIL PROTECTED]
 hi, i just have to say that i will never try to answer anything on this
 forum again. :-)

 once i tried to answer a question with regards to bgp and a 1720 router
and
 only after howard helped us out was it clear that the processor does play
an
 important role. ;-)

 this time only after the input from priscilla is everybody happy about the
 netbios/netbeui issue. ;-)

 but then i think what is important is that we dig a bit deeper into some
 topics!

 Good work!




Message Posted at:
http://www.groupstudy.com/form/read.php?f=7i=71188t=71084
--
FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html
Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]


RE: netbios [7:71084]

2003-06-23 Thread Jamie Johnson
What about Donner and Blitz..., oops, different story


-Original Message-
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Behalf Of
Howard C. Berkowitz
Sent: Monday, June 23, 2003 12:29 PM
To: [EMAIL PROTECTED]
Subject: Re: netbios [7:71084]


At 6:15 PM + 6/23/03, Priscilla Oppenheimer wrote:


Two, it's important to understand that every layer makes calls to a layer
below and every layer offers services to a layer above. Because the service
interface of a layer, whether it be LLC or UDP, is well known (hopefully),
any upper layer can call on it. So, an implementation of a session layer
protocol such as NetBIOS can call on UDP, TCP, LLC, or IPX. That's the real
message from the infamous OSI model. It's all about service interfaces.


OO!  The real message is there are seven layers named Happy,
Sneezy, Grumpy, Bashful, Sleepy, Dopey and Doc!




Message Posted at:
http://www.groupstudy.com/form/read.php?f=7i=71192t=71084
--
FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html
Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]


RE: netbios [7:71084]

2003-06-23 Thread Howard C. Berkowitz
At 4:47 PM -0700 6/23/03, Jamie Johnson wrote:
What about Donner and Blitz..., oops, different story

Nahh...you need Sneezy as a multicast server.



-Original Message-
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Behalf Of
Howard C. Berkowitz
Sent: Monday, June 23, 2003 12:29 PM
To: [EMAIL PROTECTED]
Subject: Re: netbios [7:71084]


At 6:15 PM + 6/23/03, Priscilla Oppenheimer wrote:


Two, it's important to understand that every layer makes calls to a layer
below and every layer offers services to a layer above. Because the service
interface of a layer, whether it be LLC or UDP, is well known (hopefully),
any upper layer can call on it. So, an implementation of a session layer
protocol such as NetBIOS can call on UDP, TCP, LLC, or IPX. That's the real
message from the infamous OSI model. It's all about service interfaces.


OO!  The real message is there are seven layers named Happy,
Sneezy, Grumpy, Bashful, Sleepy, Dopey and Doc!




Message Posted at:
http://www.groupstudy.com/form/read.php?f=7i=71195t=71084
--
FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html
Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]


Re: netbios [7:71084]

2003-06-23 Thread Priscilla Oppenheimer
- jvd wrote:
 
 OT:
 hi, i just have to say that i will never try to answer anything
 on this forum again. :-)

Well, would that be Grumpy, Bashful, Sleepy, or Dopey to do that? :-)
Seriously, you should keep answering. You have sent some great answers, but
you don't want to keep insisting something when replying to my messages. It
makes me very Grumpy and I'm not Bashful when wielding a keyboard (just in
person). I know lots of books claim that NetBIOS isn't routable, but I bet
those exact same books also classify it as a session-layer protocol. And it
does make a good example of a session-layer protocol. One of the few that we
have! And if it runs at that layer, then it is routable. I think even IBM
said it was a session-layer protocol in some of their early documents, which
unfortunately, I recently tossed.

Directed broadcasts came from out of the blue. I really don't think Windows
networking uses them, although maybe it does. Was the comment maybe in
reference to the helper address suggestion that I made? You can tell a
router to send the packets when it helps as a broadcast. That's not a
directed broadcast, though, and will work even if router forwarding of
directed broadcasts is disabled, which is the default these days. Instead,
it's a broadcast sent by the router (it has the router's IP address as
source, on behalf of some other station, to a local LAN, because the router
is acting as a proxy, for example, a DHCP Relay Agent.) Was that a run-on
sentence, or what? :-)

A directed broadcast is directed from afar into a subnet. The sender usually
makes classful assumptions, since it can't actually know the local
definition of a broadcast. It's used by ping scan to send a ping to
172.16.255.255, for example, in an attempt to ping everyone on network
172.16.0.0. Routers don't forward those these days because of the security
risks.

Back to NetBIOS. It does send a lot of broadcast traffic for naming
purposes. In an IP environment, however, a host can be configured to send
unicast naming queries and name registrations to a WINS server. There are
probably lots of other issues, though. It really can be quite a pain to get
it to work correctly when you migrate from a small LAN to a larger
internetwork with WANs, subnetting, VLANs, etc.


I wonder what the original poster is really trying to do and where he can
get a good Windows networking (internetworking) design guide. Cisco used to
have one, but it's probably way dated now


Well, it's late and my writing is deteriorating. Howard covers directed
broadcasts, by the way, (and a much better description of the OSI model,
without reference to the dwarves, as I recall, although possibly with
reference to the deadly sins) in his CertificationZone papers. I recommend
them.

Priscilla


 
 once i tried to answer a question with regards to bgp and a
 1720 router and only after howard helped us out was it clear
 that the processor does play an important role. ;-)
 
 this time only after the input from priscilla is everybody
 happy about the netbios/netbeui issue. ;-)
 
 but then i think what is important is that we dig a bit deeper
 into some topics!
 
 Good work!
 




Message Posted at:
http://www.groupstudy.com/form/read.php?f=7i=71203t=71084
--
FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html
Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]


RE: netbios [7:71084]

2003-06-22 Thread - jvd
Hello,

NetBIOS is a non-routeable protocol. If you want to transport it over WAN
links you will need to configure bridging. Check out:
http://www.cisco.com/en/US/partner/tech/tk331/tk660/technologies_tech_
note09186a0080093d4d.shtml

Regards,



Message Posted at:
http://www.groupstudy.com/form/read.php?f=7i=71086t=71084
--
FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html
Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]


Re: netbios [7:71084]

2003-06-22 Thread Bikespace
I'd agree to NetBEUI not being routable, but surely NETBIOS over IP gotta be
routable!
What does a WINS server do for you, or LMHOST files?


Bikespace



- jvd  wrote in message
news:[EMAIL PROTECTED]
 Hello,

 NetBIOS is a non-routeable protocol. If you want to transport it over WAN
 links you will need to configure bridging. Check out:
 http://www.cisco.com/en/US/partner/tech/tk331/tk660/technologies_tech_
 note09186a0080093d4d.shtml

 Regards,




Message Posted at:
http://www.groupstudy.com/form/read.php?f=7i=71104t=71084
--
FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html
Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]


Re: netbios [7:71084]

2003-06-22 Thread Henry D.
Since your question already assumes these port ranges, it would
mean your question is really whether NetBIOS over TCP/IP can
be routed. And as such, it can, just like any other IP traffic.

koh jef  wrote in message
news:[EMAIL PROTECTED]
 hi guys,

 can netbios,using port 137, 138 and 139 be routed thru WAN ???




Message Posted at:
http://www.groupstudy.com/form/read.php?f=7i=71105t=71084
--
FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html
Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]


RE: netbios [7:71084]

2003-06-22 Thread Priscilla Oppenheimer
koh jef wrote:
 
 hi guys,
 
 can netbios,using port 137, 138 and 139 be routed thru WAN ???

NetBIOS uses UDP and TCP which run on top of IP, which is routable over an
IP internetwork, including WAN links.

Routers don't forward broadcasts though, by default. When NetBIOS runs over
UDP ports 137 and 138, a lot of it is broadcasts. You can use an IP helper
address and udp forwarding on a router to get the router to forward those.
That might not be such a good idea, though. It could make resources
available across the WAN that you don't want to make available. It could
require you to open ports on firewalls, resulting in security risks.

You need take a higher-level view of what you're trying to do... Windows
networking across an internetwork can be challenging...

Priscilla



Message Posted at:
http://www.groupstudy.com/form/read.php?f=7i=71106t=71084
--
FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html
Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]