|
Home > Archive > alt.certification.cisco > August 2006 > CIDR or VSLM
You are viewing an archived Text-only version of the thread.
To view this thread in it's original format and/or if you want to reply to
this thread please [click here]
|
|
| Jason Jones 2003-09-13, 3:31 am |
| Hey all.
Im about to take my 640-607 exam. however im having a problem understanding
subnetting. My main problem is that i dont understand the difference between
a CIDR and VSLM. Where i work, we do our subnetting totally diffent that
what all the Cisco study material teaches.
For example: at Bellsouth.net
68.157.156.6/29 would equal=
68.157.156.6= Network Address (on ISP end)
68.157.156.7= Router Address
68.157.156.8 - .12= host address
68.157.156.13= broadcast address.
Then on top of that, we put the Router address on the LAN side of the router
and customer has to turn NAT off. The WAN side of the router is doing PPPoE
and actually pulls a dynamic ip address (which we tell the customer to just
ignore!)
From there, the ip addresses are accessable (pingable) from the internet.
I actually understand this but i think i need to clear this out of my head
before i take the test. I keep getting the questions wrong on ever practice
test i take.
THANKS IN ADVANCE FOR ANY HELP, SUGGESTIONS, or ADVICE
| |
|
| First of all, if you have 68.157.156.x/29, the network address can't be
68.157.156.6 This is because in the fourth octet, you subnet mask is
11111000. So, when you start incrementing the least significant bit of your
network number, you whole fourth octet will be incrementing by 8.
mask .11111 000
fourth octet .00000 000 = .0
.00001 000 = .8
.00010 000 = .16
.00011 000 = .24
. .................................
.11111 000 = .248
So, your example is incorrect.
There's a much easier way to calculate network numbers. You don't have to do
it in binary (like I did above). It's a lot faster in decimals. Read Wendell
Odom's Certification Guide. He explains it very well there.
As for VLSM vs. CIDR, I've had the same question for a long time. Basically,
they are similar. I think that the distinction is in that VLSM is used to
variably subnet classful networks. The difference between VLSM and regular
subnetting is that classful routing protocols such as RIP v.1 and IGRP don't
send masks in their routing updates. Therefore, they can only infer the
subnet from their own interfaces. When sending from one subnet to another
subnet of the same classful network, they assume that the length of the mask
is the length configured on their interfaces. Therefore, it must be the same
length on all routers in the routing domain. When sending updates from a
subnet of one classful network to a subnet of another classful network,
classful routing protocols summarize subnet numbers to the classful
boundaries of the respective networks (they assume the classful mask
length). In case of VLSM, subnets with different mask lengths can be
assigned throughout the routing domain because classless routing protocols
send mask along with the network address in their routing updates. So, a
point to point link can have a mask of /30 (allowing only 2 usable IP
addresses on the link), whereas an Ethernet subnet can have a shorter subnet
mask (i.e. /28, allowing 14 usable addresses). So, VLSM allows to do away
with the same-length mask within a classful network boundary.
CIDR allows to do away with classful boundaries all together by getting rid
of default classful masks. It's used by ISPs for a better management of
their address space. For example, if one company needs 4000 IP addresses,
instead of giving it a Class-B network (which has 65, 024 usable IP
addresses), an ISP can assign it 16 Class-C networks (which will give it
4064 IP addresses. Thus, almost 61, 000 IP addresses will be saved. In order
to keep routing updates and routing tables to a minimum, the ISP will
implement CIDR. Cisco books say that CIDR is used to aggregate Class-3
addresses into a classless address. For example, let's assume the ISP
assigned the following networks to the company:
200.200.0.0/24
200.200.1.0/24
200.200.2.0/24
..............................
200.200.15.0/24
The third octet of the above networks looks like this:
mask 255.255.1111 1111.0
network 200.200.0000 0000.0
200.200.0000 0001.0
200.200.0000 0010.0
..............................................................
200.200.0000 1111.0
As you can see, the last four bits of the third octet have incremented
through the complete range of 0000 -- 1111. So, instead of advertising all
16 Class-C networks to the Internet, the ISP can advertise one aggregated
address: 200.200.0.0/20. The third octet will look like this now:
mask 255.255.1111 0000.0
network 200.200.0000 0000.0
Within this network number/mask you are allowed to increment the last four
bits of the third octet, which will yield all 16 networks assigned to the
company. This uses the principle of modifying mask length across classful
boundaries, which is called supernetting. It's the same principle as when
changing the subnet mask length within the classful boundary, which is used
in subnetting. Supernettnig helps combine several Class-3 networks into a
classless network, which is the concept of CIDR. Classless Inter-Domain
Routing means that one ISP (routing domain) will send routes to classless
networks into another routing domain (another ISP). Sixteen classful
networks are combined into one classless network through the process called
Route Aggregation. On the other hand, several subnets are combined into a
larger subnet within a classful boundary through the process called Route
Summarization. CIDR is the concept that says, "Not only can you summarize
subnets within a classful network into larger subnets (super-subnets), but
you can also aggregate several Class-3 networks into a supernet (a classless
network)."
Grey
"Jason Jones" <jjones9@bellsouth.net> wrote in message
news:vpz8b.1638$4j3.1421@bignews6.bellsouth.net...
> Hey all.
> Im about to take my 640-607 exam. however im having a problem
understanding
> subnetting. My main problem is that i dont understand the difference
between
> a CIDR and VSLM. Where i work, we do our subnetting totally diffent that
> what all the Cisco study material teaches.
> For example: at Bellsouth.net
> 68.157.156.6/29 would equal=
> 68.157.156.6= Network Address (on ISP end)
> 68.157.156.7= Router Address
> 68.157.156.8 - .12= host address
> 68.157.156.13= broadcast address.
>
> Then on top of that, we put the Router address on the LAN side of the
router
> and customer has to turn NAT off. The WAN side of the router is doing
PPPoE
> and actually pulls a dynamic ip address (which we tell the customer to
just
> ignore!)
> From there, the ip addresses are accessable (pingable) from the internet.
> I actually understand this but i think i need to clear this out of my head
> before i take the test. I keep getting the questions wrong on ever
practice
> test i take.
>
> THANKS IN ADVANCE FOR ANY HELP, SUGGESTIONS, or ADVICE
>
>
| |
| Scooby 2003-09-13, 4:26 pm |
| In simple terms, VLSM is used to break a network (classful) into smaller
networks. CIDR works in the other direction. When you take a list of
contiguous networks and summarize them together for routing purposes, that
is CIDR.
So, the two terms really work together. Say I assign you a block of IP's
for your network - 10.1.1.0/24. You decide to subnet these ip's for your
network. Personally, I don't care and there is no reason for you to send me
updates of every single network that you create. So, VLSM is you creating
the subnets, but CIDR is you sending me a summary route for the whole block
and not worry about individual updates.
"Grey" <bbb@ccc.com> wrote in message
news:lM2dnfvuAOBWlf6iXTWJkA@co
mcast.com...
> First of all, if you have 68.157.156.x/29, the network address can't be
> 68.157.156.6 This is because in the fourth octet, you subnet mask is
> 11111000. So, when you start incrementing the least significant bit of
your
> network number, you whole fourth octet will be incrementing by 8.
>
> mask .11111 000
> fourth octet .00000 000 = .0
> .00001 000 = .8
> .00010 000 = .16
> .00011 000 = .24
> . .................................
> .11111 000 = .248
>
> So, your example is incorrect.
>
> There's a much easier way to calculate network numbers. You don't have to
do
> it in binary (like I did above). It's a lot faster in decimals. Read
Wendell
> Odom's Certification Guide. He explains it very well there.
>
> As for VLSM vs. CIDR, I've had the same question for a long time.
Basically,
> they are similar. I think that the distinction is in that VLSM is used to
> variably subnet classful networks. The difference between VLSM and regular
> subnetting is that classful routing protocols such as RIP v.1 and IGRP
don't
> send masks in their routing updates. Therefore, they can only infer the
> subnet from their own interfaces. When sending from one subnet to another
> subnet of the same classful network, they assume that the length of the
mask
> is the length configured on their interfaces. Therefore, it must be the
same
> length on all routers in the routing domain. When sending updates from a
> subnet of one classful network to a subnet of another classful network,
> classful routing protocols summarize subnet numbers to the classful
> boundaries of the respective networks (they assume the classful mask
> length). In case of VLSM, subnets with different mask lengths can be
> assigned throughout the routing domain because classless routing protocols
> send mask along with the network address in their routing updates. So, a
> point to point link can have a mask of /30 (allowing only 2 usable IP
> addresses on the link), whereas an Ethernet subnet can have a shorter
subnet
> mask (i.e. /28, allowing 14 usable addresses). So, VLSM allows to do away
> with the same-length mask within a classful network boundary.
>
> CIDR allows to do away with classful boundaries all together by getting
rid
> of default classful masks. It's used by ISPs for a better management of
> their address space. For example, if one company needs 4000 IP addresses,
> instead of giving it a Class-B network (which has 65, 024 usable IP
> addresses), an ISP can assign it 16 Class-C networks (which will give it
> 4064 IP addresses. Thus, almost 61, 000 IP addresses will be saved. In
order
> to keep routing updates and routing tables to a minimum, the ISP will
> implement CIDR. Cisco books say that CIDR is used to aggregate Class-3
> addresses into a classless address. For example, let's assume the ISP
> assigned the following networks to the company:
> 200.200.0.0/24
> 200.200.1.0/24
> 200.200.2.0/24
> .............................
> 200.200.15.0/24
>
> The third octet of the above networks looks like this:
> mask 255.255.1111 1111.0
> network 200.200.0000 0000.0
> 200.200.0000 0001.0
> 200.200.0000 0010.0
> .............................................................
> 200.200.0000 1111.0
>
> As you can see, the last four bits of the third octet have incremented
> through the complete range of 0000 -- 1111. So, instead of advertising
all
> 16 Class-C networks to the Internet, the ISP can advertise one aggregated
> address: 200.200.0.0/20. The third octet will look like this now:
>
> mask 255.255.1111 0000.0
> network 200.200.0000 0000.0
>
> Within this network number/mask you are allowed to increment the last four
> bits of the third octet, which will yield all 16 networks assigned to the
> company. This uses the principle of modifying mask length across classful
> boundaries, which is called supernetting. It's the same principle as when
> changing the subnet mask length within the classful boundary, which is
used
> in subnetting. Supernettnig helps combine several Class-3 networks into a
> classless network, which is the concept of CIDR. Classless Inter-Domain
> Routing means that one ISP (routing domain) will send routes to classless
> networks into another routing domain (another ISP). Sixteen classful
> networks are combined into one classless network through the process
called
> Route Aggregation. On the other hand, several subnets are combined into a
> larger subnet within a classful boundary through the process called Route
> Summarization. CIDR is the concept that says, "Not only can you summarize
> subnets within a classful network into larger subnets (super-subnets), but
> you can also aggregate several Class-3 networks into a supernet (a
classless
> network)."
>
> Grey
>
> "Jason Jones" <jjones9@bellsouth.net> wrote in message
> news:vpz8b.1638$4j3.1421@bignews6.bellsouth.net...
> > Hey all.
> > Im about to take my 640-607 exam. however im having a problem
> understanding
> > subnetting. My main problem is that i dont understand the difference
> between
> > a CIDR and VSLM. Where i work, we do our subnetting totally diffent that
> > what all the Cisco study material teaches.
> > For example: at Bellsouth.net
> > 68.157.156.6/29 would equal=
> > 68.157.156.6= Network Address (on ISP end)
> > 68.157.156.7= Router Address
> > 68.157.156.8 - .12= host address
> > 68.157.156.13= broadcast address.
> >
> > Then on top of that, we put the Router address on the LAN side of the
> router
> > and customer has to turn NAT off. The WAN side of the router is doing
> PPPoE
> > and actually pulls a dynamic ip address (which we tell the customer to
> just
> > ignore!)
> > From there, the ip addresses are accessable (pingable) from the
internet.
> > I actually understand this but i think i need to clear this out of my
head
> > before i take the test. I keep getting the questions wrong on ever
> practice
> > test i take.
> >
> > THANKS IN ADVANCE FOR ANY HELP, SUGGESTIONS, or ADVICE
> >
> >
>
>
| |
| Scooby 2003-09-13, 4:26 pm |
| "Scooby" <mmscooby1@earthlink.net> wrote in message
news:KwK8b.6995$NM1.1139@newsread2.news.atl.earthlink.net...
> In simple terms, VLSM is used to break a network (classful) into smaller
> networks. CIDR works in the other direction. When you take a list of
> contiguous networks and summarize them together for routing purposes, that
> is CIDR.
>
> So, the two terms really work together. Say I assign you a block of IP's
> for your network - 10.1.1.0/24. You decide to subnet these ip's for your
> network. Personally, I don't care and there is no reason for you to send
me
> updates of every single network that you create. So, VLSM is you creating
> the subnets, but CIDR is you sending me a summary route for the whole
block
> and not worry about individual updates.
>
Let me add to that cause I'm sure that the thought will be raised that we
summarized it right back to a class C. Keep the example of VLSM the same.
If I then send out that I have the networks in 10.1.0.0/20 because I have
other 24 bit networks as well, that becomes a better description of CIDR.
>
>
> "Grey" <bbb@ccc.com> wrote in message
> news:lM2dnfvuAOBWlf6iXTWJkA@co
mcast.com...
> > First of all, if you have 68.157.156.x/29, the network address can't be
> > 68.157.156.6 This is because in the fourth octet, you subnet mask is
> > 11111000. So, when you start incrementing the least significant bit of
> your
> > network number, you whole fourth octet will be incrementing by 8.
> >
> > mask .11111 000
> > fourth octet .00000 000 = .0
> > .00001 000 = .8
> > .00010 000 = .16
> > .00011 000 = .24
> > . .................................
> > .11111 000 = .248
> >
> > So, your example is incorrect.
> >
> > There's a much easier way to calculate network numbers. You don't have
to
> do
> > it in binary (like I did above). It's a lot faster in decimals. Read
> Wendell
> > Odom's Certification Guide. He explains it very well there.
> >
> > As for VLSM vs. CIDR, I've had the same question for a long time.
> Basically,
> > they are similar. I think that the distinction is in that VLSM is used
to
> > variably subnet classful networks. The difference between VLSM and
regular
> > subnetting is that classful routing protocols such as RIP v.1 and IGRP
> don't
> > send masks in their routing updates. Therefore, they can only infer the
> > subnet from their own interfaces. When sending from one subnet to
another
> > subnet of the same classful network, they assume that the length of the
> mask
> > is the length configured on their interfaces. Therefore, it must be the
> same
> > length on all routers in the routing domain. When sending updates from a
> > subnet of one classful network to a subnet of another classful network,
> > classful routing protocols summarize subnet numbers to the classful
> > boundaries of the respective networks (they assume the classful mask
> > length). In case of VLSM, subnets with different mask lengths can be
> > assigned throughout the routing domain because classless routing
protocols
> > send mask along with the network address in their routing updates. So, a
> > point to point link can have a mask of /30 (allowing only 2 usable IP
> > addresses on the link), whereas an Ethernet subnet can have a shorter
> subnet
> > mask (i.e. /28, allowing 14 usable addresses). So, VLSM allows to do
away
> > with the same-length mask within a classful network boundary.
> >
> > CIDR allows to do away with classful boundaries all together by getting
> rid
> > of default classful masks. It's used by ISPs for a better management of
> > their address space. For example, if one company needs 4000 IP
addresses,
> > instead of giving it a Class-B network (which has 65, 024 usable IP
> > addresses), an ISP can assign it 16 Class-C networks (which will give it
> > 4064 IP addresses. Thus, almost 61, 000 IP addresses will be saved. In
> order
> > to keep routing updates and routing tables to a minimum, the ISP will
> > implement CIDR. Cisco books say that CIDR is used to aggregate Class-3
> > addresses into a classless address. For example, let's assume the ISP
> > assigned the following networks to the company:
> > 200.200.0.0/24
> > 200.200.1.0/24
> > 200.200.2.0/24
> > .............................
> > 200.200.15.0/24
> >
> > The third octet of the above networks looks like this:
> > mask 255.255.1111 1111.0
> > network 200.200.0000 0000.0
> > 200.200.0000 0001.0
> > 200.200.0000 0010.0
> > .............................................................
> > 200.200.0000 1111.0
> >
> > As you can see, the last four bits of the third octet have incremented
> > through the complete range of 0000 -- 1111. So, instead of advertising
> all
> > 16 Class-C networks to the Internet, the ISP can advertise one
aggregated
> > address: 200.200.0.0/20. The third octet will look like this now:
> >
> > mask 255.255.1111 0000.0
> > network 200.200.0000 0000.0
> >
> > Within this network number/mask you are allowed to increment the last
four
> > bits of the third octet, which will yield all 16 networks assigned to
the
> > company. This uses the principle of modifying mask length across
classful
> > boundaries, which is called supernetting. It's the same principle as
when
> > changing the subnet mask length within the classful boundary, which is
> used
> > in subnetting. Supernettnig helps combine several Class-3 networks into
a
> > classless network, which is the concept of CIDR. Classless Inter-Domain
> > Routing means that one ISP (routing domain) will send routes to
classless
> > networks into another routing domain (another ISP). Sixteen classful
> > networks are combined into one classless network through the process
> called
> > Route Aggregation. On the other hand, several subnets are combined into
a
> > larger subnet within a classful boundary through the process called
Route
> > Summarization. CIDR is the concept that says, "Not only can you
summarize
> > subnets within a classful network into larger subnets (super-subnets),
but
> > you can also aggregate several Class-3 networks into a supernet (a
> classless
> > network)."
> >
> > Grey
> >
> > "Jason Jones" <jjones9@bellsouth.net> wrote in message
> > news:vpz8b.1638$4j3.1421@bignews6.bellsouth.net...
> > > Hey all.
> > > Im about to take my 640-607 exam. however im having a problem
> > understanding
> > > subnetting. My main problem is that i dont understand the difference
> > between
> > > a CIDR and VSLM. Where i work, we do our subnetting totally diffent
that[
color=darkred]
> > > what all the Cisco study material teaches.
> > > For example: at Bellsouth.net
> > > 68.157.156.6/29 would equal=
> > > 68.157.156.6= Network Address (on ISP end)
> > > 68.157.156.7= Router Address
> > > 68.157.156.8 - .12= host address
> > > 68.157.156.13= broadcast address.
> > >
> > > Then on top of that, we put the Router address on the LAN side of the
> > router
> > > and customer has to turn NAT off. The WAN side of the router is doing
> > PPPoE
> > > and actually pulls a dynamic ip address (which we tell the customer to
> > just
> > > ignore!)
> > > From there, the ip addresses are accessable (pingable) from the
> internet.
> > > I actually understand this but i think i need to clear this out of my
> head
> > > before i take the test. I keep getting the questions wrong on ever
> > practice
> > > test i take.
> > >
> > > THANKS IN ADVANCE FOR ANY HELP, SUGGESTIONS, or ADVICE
> > >
> > >
> >
> >
>
>[/color]
| |
| Jamajo 2003-09-14, 11:28 pm |
| Thank you all. im still absorbing your answers, but i think i understand
what both of you are saying.
Now, which one (VSLM or CIDR) do i have to worry about on the test (CCNA).
"Jason Jones" <jjones9@bellsouth.net> wrote in message
news:vpz8b.1638$4j3.1421@bignews6.bellsouth.net...
> Hey all.
> Im about to take my 640-607 exam. however im having a problem
understanding
> subnetting. My main problem is that i dont understand the difference
between
> a CIDR and VSLM. Where i work, we do our subnetting totally diffent that
> what all the Cisco study material teaches.
> For example: at Bellsouth.net
> 68.157.156.6/29 would equal=
> 68.157.156.6= Network Address (on ISP end)
> 68.157.156.7= Router Address
> 68.157.156.8 - .12= host address
> 68.157.156.13= broadcast address.
>
> Then on top of that, we put the Router address on the LAN side of the
router
> and customer has to turn NAT off. The WAN side of the router is doing
PPPoE
> and actually pulls a dynamic ip address (which we tell the customer to
just
> ignore!)
> From there, the ip addresses are accessable (pingable) from the internet.
> I actually understand this but i think i need to clear this out of my head
> before i take the test. I keep getting the questions wrong on ever
practice
> test i take.
>
> THANKS IN ADVANCE FOR ANY HELP, SUGGESTIONS, or ADVICE
>
>
| |
|
| It's not VSLM, it's VLSM (Variable Length Subnet Masking). VLSM is still
subnetting, but assigning variable-length subnets to different interfaces.
VLSM will definitely be on CCNA. If you know how to subnet, you know how to
do VLSM. The only thing you need to remember is that VLSM is only possible
in classless routing protocols, such as RIP v.2. I'm not sure which routing
protocols are on CCNA now. VLSM is also possible with OSPF, IS-IS, EIGRP,
and BGP, but those are probably not part of CCNA curriculum. You also need
to understand that classful routing protocols, such as RIP v.1 and IGRP
auto-summarize subnets to the classful boundaries when an update is sent
about subnets of one classful network into a subnet of another classful
network. With classless routing protocols, auto-summarization can be (or is
by default) turned off. In RIP v.2 auto-summarization is still enabled by
default, but you can turn it off (*no auto-summary* router configuration
command). Once auto-summarization is turned off in a classless routing
protocol, you can manually configure summarization at any bit position in
the subnet mask. This is done with the *summary-address* command in most
routing protocols. VLSM and route summarization are NOT the same thing, but
they are both possible in classless routing protocols due to the fact that
classless routing protocols exchange not only network numbers but also masks
unlike classful routing protocols, which only exchange network numbers.
CIDR will most likely not be on CCNA. At least, I doubt you will be required
to configure route aggregation. However, CIDR is discussed in Wendell Odom's
book, and you may get a question or two about what CIDR is used for.
If you understand the concepts of VLSM and CIDR, you should be okay.
However, If you are still concerned about the depth of knowledge required
for this test, you can purchase a practice test from www.boson.com and see
what kind of questions may be on the actual test. Also, Wendell Odom's book
comes with a CD-ROM, which has a testing engine and a lot of good questions
to practice with. By the way, have you looked on
www.cisco.com/go/certification to see what topics are covered on the new
CCNA exam? All of the above will help you get more confident in your
readiness for the exam.
Grey
"Jamajo" <jjones9@bellsouth.net> wrote in message
news:A_99b.3833$me.3360@bignews5.bellsouth.net...
> Thank you all. im still absorbing your answers, but i think i understand
> what both of you are saying.
> Now, which one (VSLM or CIDR) do i have to worry about on the test (CCNA).
> "Jason Jones" <jjones9@bellsouth.net> wrote in message
> news:vpz8b.1638$4j3.1421@bignews6.bellsouth.net...
> > Hey all.
> > Im about to take my 640-607 exam. however im having a problem
> understanding
> > subnetting. My main problem is that i dont understand the difference
> between
> > a CIDR and VSLM. Where i work, we do our subnetting totally diffent that
> > what all the Cisco study material teaches.
> > For example: at Bellsouth.net
> > 68.157.156.6/29 would equal=
> > 68.157.156.6= Network Address (on ISP end)
> > 68.157.156.7= Router Address
> > 68.157.156.8 - .12= host address
> > 68.157.156.13= broadcast address.
> >
> > Then on top of that, we put the Router address on the LAN side of the
> router
> > and customer has to turn NAT off. The WAN side of the router is doing
> PPPoE
> > and actually pulls a dynamic ip address (which we tell the customer to
> just
> > ignore!)
> > From there, the ip addresses are accessable (pingable) from the
internet.
> > I actually understand this but i think i need to clear this out of my
head
> > before i take the test. I keep getting the questions wrong on ever
> practice
> > test i take.
> >
> > THANKS IN ADVANCE FOR ANY HELP, SUGGESTIONS, or ADVICE
> >
> >
>
>
| |
| zillah 2006-08-24, 2:20 am |
| Might be this one will be useful as well :
code:
http://www.experts-exchange.com/Networking/Q_21400371.html
|
|
|
|
|