All IP addresses of the form 192.168.x.y are supposed to be available for
private use, so I have allocated numbers such as 192.168.0.[1234] to my
various local gadgets, withthe NetMask set to 255.255.255.0 (which seems
to the the usual recommendation).

Recently, I acquired a device with default address 192.168.1.77, and my
Netgear DG 834 did not like it (it would not show up amongs the "Connected
Devices" that it claimed to be able to see).

So I changed the NetMask to 255.255.0.0, and that enabled me to access the
device from my computer, but even then the Netgear could not see it as a
"Connected Device".

But being able to connect to it enabled me to reconfigure its IP to
192.168.0.4, whereupon the Netgear was able to "see" it.

But how can the Netgear be in a state where it can deliver packets to a
device that it claims not to be able to see? Or is there a second Netmask
setting somewhere that I need to change?

And why is the usual advice to set NetMasks to 255.255.255.0 rather than
to 255.255.0.0?

--
Charles H. Lindsey ---------At Home, doing my own thing------------------------
Tel: +44 161 436 6131 Fax: +44 161 436 6133 Web: http://www.cs.man.ac.uk/~chl
Email: chl@clerew.man.ac.uk Snail: 5 Clerewood Ave, CHEADLE, SK8 3JU, U.K.
PGP: 2C15F1A9 Fingerprint: 73 6D C2 51 93 A0 01 E7 65 E8 64 7E 14 A4 AB A5

Charles Lindsey wrote:
I would hazard a guess that it isn't anything to do with the netmask but
rather the IP address range that the router uses for it's DHCP pool. For
instance, the router may be set to allocate DHCP addresses of 192.168.1.2 to
192.168.1.50, so the new device, having an address ending in 77, fell
outside the DHCP pool. Just a guess.

John

Charles Lindsey wrote:
It looks like the "Attached Devices" page on my DG834GT is using nbtscan. If
your target device doesn't respond to UDP queries on port 137, I guess it
might not be listed.

Charles Lindsey wrote:
want the 10.something network if you intend running a class B network
behind a firewall.

I am unclear as to what the 'seeing' of the netgear amounts to: with a
normal class C netmask it will neither see hear nor respond to packets
from a different class C.

Your ability to communiacte with the device on a different class C does
not mean the netgear was routing the packets: you are presumably on the
same LAN and would communicate directly. If you want top open up the
whole lot t a cass B every machine on the net including the netgear
needs its netmask adjusting.

But its a recipe for disaster, as the default settings will be for class
C operation. Far far easier to change the IP address on the offending
peice of kit.

"Charles Lindsey" <chl@clerew.man.ac.uk> wrote in message
news:K19Mqw.9Js@clerew.man.ac.uk...
The netmask defines whether an IP address is regarded as being on the
"local" network or a "foreign" network.

The "ones" in the mask define the bits in the IP address which will specify
the "local" address. Thus your example of 255.255.255.0 has the first
24-bits at "one" and the last 8 bits at "zero". This is also known as a
24-bit mask.

When your computer sends out an IP packet it compares the destination
address with the mask. If all the masked bits match its own address, it
knows the destination is on the local network. A process called ARP (Address
Resolution Protocol) is used to identify the physical (MAC) address of the
destination, and the packet is sent. If the masked bits do not match, the
destination address is assumed to be elsewhere, and the packet is sent to
the default gateway. The IP packets are said to be "routable" - they can go
through routers to other networks. This is described in Layer 3 of the OSI
model - somebody will be along in a bit with a good reference ...

Your Netgear router knows it is on a network which has the address range
192.168.0.0 through 192.168.0.255 - the mask specifies the 192.168.0
component.

The device 192.168.1.77 is therefore not on the local network and cannot be
sent IP packets. There are other protocols which do not use IP packets, and
programs which set out to identify new devices (e.g. printers) sometimes use
these.

You changed the computer's mask to 255.255.0.0 (i.e. 16-bit) This caused
the computer to be part of the address range 192.168.0.0 through
192.168.255.255 which of course includes your new device. The mask now
specifies only the 192.168 component. So the computer can send IP packets
to the new device.

The only part that Netgear router takes in delivering these packets is that
it provides a "packet switch" to carry traffic from one port to another. By
examining the MAC addresses it delivers packets only to the connected device
that will receive them. This is described in Layer 2 of the OSI model.

Older equipment used "repeaters" to achieve the "star" topology, but these
worked in the same way as the original 10base5 Ethernet in that all packets
were sent to all devices, the responsibility for recognising packets being
wholly carried out by the network interface(s) in the computer(s). The
"switch" provides a performance improvement over a repeater.

If you had also changed the netmask in the router, that would have allowed
the router to send IP packets to the new device.

Hope this helps ...

-- Graham J

On Thu, 22 May 2008 15:25:49 +0100, Graham J scribbled:

That's one hell of a long winded reply, but totally fascinating to read.
Thanks Graham J.

[snip]
Yes, whole books are written about networks ...

-- Graham

In <1211460277.19513.0@proxy01.news.clara.net> "Grant" <news@mason.sh> writes:

No, because after I had changed the device's address to 108.168.0.3, the
"Attached Devices" page could see it perfectly well, so presumably it was
then responding to port 137.

--
Charles H. Lindsey ---------At Home, doing my own thing------------------------
Tel: +44 161 436 6131 Fax: +44 161 436 6133 Web: http://www.cs.man.ac.uk/~chl
Email: chl@clerew.man.ac.uk Snail: 5 Clerewood Ave, CHEADLE, SK8 3JU, U.K.
PGP: 2C15F1A9 Fingerprint: 73 6D C2 51 93 A0 01 E7 65 E8 64 7E 14 A4 AB A5

In <1211465998.27708.0@proxy01.news.clara.net> The Natural Philosopher <a@b.c> writes:

But the whole of 192.168.X.X is a Class B subnet, and when I set the
Netmask on the Netgear to 255.255.0.0 I expected it to behave as a Class
B. And indeed it did, insofar as it was now prepared to route 192.168.1.77
correctly. The 10.something area is in fact Class A (10.0.0.0/8).

But somehow, for 'seeing' (i.e. when it probed to determine what was out
there), it was still behaving like a Class C subnet (192.168.0.0/24) even
though it was willing to route like a Class B. That is what I do not
understand.

No, the only connection between my computer and the device was via the
Netgear. I had already adjusted the Mask accordingly (before that, it was
trying to send packets address to the device up my ADSL connection).

Yes, that is what I eventually did. But I find it exceedingly odd that
anyone should supply a device that, by default, has IP 192.168.1.77.

--
Charles H. Lindsey ---------At Home, doing my own thing------------------------
Tel: +44 161 436 6131 Fax: +44 161 436 6133 Web: http://www.cs.man.ac.uk/~chl
Email: chl@clerew.man.ac.uk Snail: 5 Clerewood Ave, CHEADLE, SK8 3JU, U.K.
PGP: 2C15F1A9 Fingerprint: 73 6D C2 51 93 A0 01 E7 65 E8 64 7E 14 A4 AB A5

Charles Lindsey wrote:
Techinically it an be made to be, but by defintion, its reserved for
class c usage. Therefore manfacturers defaults will treat it as such.
And may not allow clean operation as a class B in noddy consumer level
gear tha tmakes lots of assumptions to simplify things for 'usahs'.



and when I set the
reserved and guarnteed not to be on the public internet.



What was it probing with..?


That doesn;t means anything if it contains a switch type hub. The HUB
/SWITCH is before and independent of the routing part of the netgear.



I had already adjusted the Mask accordingly (before that, it was
Well everything has got to be on some IP address.

Why not that one?

"The Natural Philosopher" <a@b.c> wrote in message
news:1211617812.6424.0@proxy00.news.clara.net...
At first sight it does seem an odd IP address to set as the default, but I
can see a certain logic. Most routers have an IP address of 192.168.x.1 or
192.168.x.254 (where x is usually 0, 1 or 2). DHCP usually allocates
addresses in ascending order starting at 2, so PCs on a network will usually
be 192.168.0.2, 3, 4 etc; some routers allocate starting at 65.

So a device with the address 192.168.1.77 is fairly unlikely to conflict
with a router or a PC, as long as you don't have a large number of PCs, and
will work without change for at least some routers which use the 192.168.1.x
subnet - but not Netgear which uses 192.168.0.x!

On Sat, 24 May 2008 08:30:18 UTC, The Natural Philosopher <a@b.c> wrote:

172.16.0.0 - 172.31.255.255

(RFC 1597)

--
Bob Eager
Use the BIG mirror service in the UK:
http://www.mirrorservice.org

Bob Eager wrote:

Mortimer wrote:
dhcpd (linux - so possibly relevant, but probably too high powered for a
router) hashes addresses and allocates sparsely across the range.

DHCP servers "should" avoid addresses already in use - dhcpd uses a ping
to check before allocating, but that relies on the devices responding
correctly :-(

--
PeeGee

The reply address is a spam trap. All mail is reported as spam.
"Nothing should be able to load itself onto a computer without the
knowledge or consent of the computer user. Software should also be able
to be removed from a computer easily."
Peter Cullen, Microsoft Chief Privacy Strategist (Computing 18 Aug 05)

Graham J formulated on Thursday :
Thanks for that explanation, I have always wondered what the mask was
for/did.

--
Regards,
Harry (M1BYT) (L)
http://www.ukradioamateur.co.uk

Harry Bloomfield wrote:
Just to be picky :-) Though your statement *is* accurate, it omitted
essential information: the first and last addresses in the range are
reserved and cannot be used for a device as they are used to denote the
"network" and "broadcast" (to all network devices) addresses.

One additional snippet: I've had a lot of problems with MS Windows
systems being allocated valid 192.168.x.0 and 192.168.x.255 addresses
with a 255.255.0.0 netmask though Linux systems work fine. Built-in
class C restrictions?

--
PeeGee

The reply address is a spam trap. All mail is reported as spam.
"Nothing should be able to load itself onto a computer without the
knowledge or consent of the computer user. Software should also be able
to be removed from a computer easily."
Peter Cullen, Microsoft Chief Privacy Strategist (Computing 18 Aug 05)

"PeeGee" <triessuk@yahoo.co.uk> wrote in message
news:g1b89f$n9f$1@registered.motzarella.org...
As somebody else commented, my explanation was somewhat long-winded. Sadly,
this still means that I missed out details such as the one you mention.

--
Graham J

PeeGee wrote:

Broadcast on a 192.168/16 net is 192.168.255.255
Brodcast on a 192.168.0/24 net is 192.168.0.255

Go figure.\

"The Natural Philosopher" <a@b.c> wrote in message
news:1211717047.12088.3@proxy01.news.clara.net...
If I understand correctly PeeGee is saying that on a properly configured
192.168.0.0/16 network a device with the address 192.168.0.255 is recognised
by Linux but not by Windows. On the face of it this is a Windows problem
....

--
Graham J

Graham J wrote:
Exactly.

--
PeeGee

The reply address is a spam trap. All mail is reported as spam.
"Nothing should be able to load itself onto a computer without the
knowledge or consent of the computer user. Software should also be able
to be removed from a computer easily."
Peter Cullen, Microsoft Chief Privacy Strategist (Computing 18 Aug 05)