subnet mask - TCP-IP

This is a discussion on subnet mask - TCP-IP ; There is a network as follows: DHCP server - Subnet A - 10.10.10.65 - Router - 10.10.10.93 - Subnet B 5 Client computers in Subnet A: IP address: 10.10.10.66 - 10.10.10.70 Subnet mask: 255.255.255.192 Default gateway: 10.10.10.65 Server in Subnet ...

+ Reply to Thread
Page 1 of 2 1 2 LastLast
Results 1 to 20 of 22

Thread: subnet mask

  1. subnet mask

    There is a network as follows:


    DHCP server - Subnet A - 10.10.10.65 - Router - 10.10.10.93 - Subnet B

    5 Client computers in Subnet A:
    IP address: 10.10.10.66 - 10.10.10.70
    Subnet mask: 255.255.255.192
    Default gateway: 10.10.10.65


    Server in Subnet B:
    IP address: 10.10.10.94
    Subnet mask: 255.255.255.240
    Default gateway: 10.10.10.93


    Clients cannot connect to the server in Subnet B unless the subnet mask of
    the server is changed to 255.255.255.192

    Why is that? Isn't it the router's job to forward the packets from one
    subnet to the other?


    Thanks!



  2. Re: subnet mask

    On Jul 3, 12:37*am, "eager" wrote:
    > There is a network as follows:
    >
    > DHCP server - Subnet A - 10.10.10.65 - Router - 10.10.10.93 - Subnet B
    >
    > 5 Client computers in Subnet A:
    > IP address: 10.10.10.66 - 10.10.10.70
    > Subnet mask: 255.255.255.192
    > Default gateway: 10.10.10.65
    >
    > Server in Subnet B:
    > IP address: 10.10.10.94
    > Subnet mask: 255.255.255.240
    > Default gateway: 10.10.10.93
    >
    > Clients cannot connect to the server in Subnet B unless the subnet mask of
    > the server is changed to 255.255.255.192
    >
    > Why is that? Isn't it the router's job to forward the packets from one
    > subnet to the other?



    While it's hard to say without more information, your first problem is
    that your two subnets overlap. IOW, subnet B is addresses
    10.10.10.92-10.10.10.107, which is wholly within subnet A
    (10.10.10.64-10.10.10.127). That basically needs to be fixed.

    That you're getting any communication between the clients on A and the
    server on B suggests that you also have the two subnets on a single
    physical network. When you change the subnet mask on B to
    255.255.255.192, you're basically including the clients, which, if
    your on the same physical LAN, allows them to talk. While there are
    situations where having more than one subnet on a single LAN is
    (debatably) useful or desirable, it's more likely to be a network
    design error, or a misconstrued (and incorrect) attempt to partition
    the network for security reasons.

  3. Re: subnet mask

    eager wrote:
    > There is a network as follows:
    >
    >
    > DHCP server - Subnet A - 10.10.10.65 - Router - 10.10.10.93 - Subnet B
    >
    > 5 Client computers in Subnet A:
    > IP address: 10.10.10.66 - 10.10.10.70
    > Subnet mask: 255.255.255.192
    > Default gateway: 10.10.10.65
    >
    >
    > Server in Subnet B:
    > IP address: 10.10.10.94
    > Subnet mask: 255.255.255.240
    > Default gateway: 10.10.10.93
    >
    >
    > Clients cannot connect to the server in Subnet B unless the subnet mask of
    > the server is changed to 255.255.255.192
    >
    > Why is that? Isn't it the router's job to forward the packets from one
    > subnet to the other?
    >
    >
    > Thanks!
    >
    >


    As indicated by the other respondent, your key issue is the address
    overlap resulting from your erroneous network mask assignments.

    It's important to understand that the first routing decision is made by
    the "host". A host applies its network mask to its own IP address and
    the destination host IP address, then performs a comparison to determine
    whether the destination host is directly reachable, or whether the
    packet needs to be forwarded to the router.

    When host addresses are mis-assigned due to erroneous masks, a host may
    erroneously conclude that a destination host is directly reachable, and
    attempt to resolve its IP address into a MAC address via ARP (which
    fails), rather than forwarding the packet to the router. This could
    occur on the initiator or the responder side depending on the erroneous
    configuration.

    The selection of masks should be driven by the potential number of hosts
    to be assigned to each network.

    Typically, the first or last assignable host address available for each
    network will be used for the gateway(s) which can lessen confusion for
    those unfamiliar with your addressing scheme.

    Best Regards,
    News Reader

  4. Re: subnet mask

    In article ,
    News Reader wrote:

    > eager wrote:
    > > There is a network as follows:
    > >
    > >
    > > DHCP server - Subnet A - 10.10.10.65 - Router - 10.10.10.93 - Subnet B
    > >
    > > 5 Client computers in Subnet A:
    > > IP address: 10.10.10.66 - 10.10.10.70
    > > Subnet mask: 255.255.255.192
    > > Default gateway: 10.10.10.65
    > >
    > >
    > > Server in Subnet B:
    > > IP address: 10.10.10.94
    > > Subnet mask: 255.255.255.240
    > > Default gateway: 10.10.10.93
    > >
    > >
    > > Clients cannot connect to the server in Subnet B unless the subnet mask of
    > > the server is changed to 255.255.255.192
    > >
    > > Why is that? Isn't it the router's job to forward the packets from one
    > > subnet to the other?
    > >
    > >
    > > Thanks!
    > >
    > >

    >
    > As indicated by the other respondent, your key issue is the address
    > overlap resulting from your erroneous network mask assignments.
    >
    > It's important to understand that the first routing decision is made by
    > the "host". A host applies its network mask to its own IP address and
    > the destination host IP address, then performs a comparison to determine
    > whether the destination host is directly reachable, or whether the
    > packet needs to be forwarded to the router.
    >
    > When host addresses are mis-assigned due to erroneous masks, a host may
    > erroneously conclude that a destination host is directly reachable, and
    > attempt to resolve its IP address into a MAC address via ARP (which
    > fails), rather than forwarding the packet to the router. This could
    > occur on the initiator or the responder side depending on the erroneous
    > configuration.


    If there's some reason you can't fix the subnet masks on Subnet A, you
    can work around this if the router supports Proxy ARP. When the server
    ARPs for addresses in Subnet B, the router will respond to these ARP
    queries.

    --
    Barry Margolin, barmar@alum.mit.edu
    Arlington, MA
    *** PLEASE don't copy me on replies, I'll read them in the group ***

  5. Re: subnet mask


    wrote in message
    news:5bed34cb-ae13-4e58-90cb-73d5cc6274d1@x41g2000hsb.googlegroups.com...
    On Jul 3, 12:37 am, "eager" wrote:
    > There is a network as follows:
    >
    > DHCP server - Subnet A - 10.10.10.65 - Router - 10.10.10.93 - Subnet B
    >
    > 5 Client computers in Subnet A:
    > IP address: 10.10.10.66 - 10.10.10.70
    > Subnet mask: 255.255.255.192
    > Default gateway: 10.10.10.65
    >
    > Server in Subnet B:
    > IP address: 10.10.10.94
    > Subnet mask: 255.255.255.240
    > Default gateway: 10.10.10.93
    >
    > Clients cannot connect to the server in Subnet B unless the subnet mask of
    > the server is changed to 255.255.255.192
    >
    > Why is that? Isn't it the router's job to forward the packets from one
    > subnet to the other?



    While it's hard to say without more information,

    there is no more INFO, it's a question from 70-291 exam.

    your first problem is that your two subnets overlap.

    What do you mean by " two subnets overlap"? - they are in totally different
    broadcast domains.

    IOW, subnet B is addresses 10.10.10.92-10.10.10.107,

    No idea where you got that from, because based on the server INFO, the IP
    range of the subnet B should be 10.10.10.80 - 10.10.10.95
    That IP range does not overlap with the scope 10.10.10.66 - 10.10.10.70 of
    the clients in subnet A.
    Since the gateway is configured properly on both subnets, according to the
    routing table I do not see why there will be no connectivity between the
    clients and the server.

    which is wholly within subnet A (10.10.10.64-10.10.10.127). That basically
    needs to be fixed.

    I agree that the IP range of subnet A is 10.10.10.64-10.10.10.127, but there
    is nothing else on subnet A except for the clients that are configured with
    dynamic IP addresses from the DHCP server scope.

    That you're getting any communication between the clients on A and the
    server on B suggests that you also have the two subnets on a single physical
    network.

    Again, they are two different broadcast domains.

    When you change the subnet mask on B to 255.255.255.192, you're basically
    including the clients, which, if your on the same physical LAN, allows them
    to talk.

    Now you are contradicting yourself because in this situation the subnets
    overlap even better!!

    While there are situations where having more than one subnet on a single LAN
    is (debatably) useful or desirable, it's more likely to be a network design
    error, or a misconstrued (and incorrect) attempt to partition the network
    for security reasons.

    What do you mean, are you against subnetting?



  6. Re: subnet mask


    "News Reader" wrote in message
    news:zB7bk.13586$Ec.12497@read2.cgocable.net...
    > eager wrote:
    >> There is a network as follows:
    >>
    >>
    >> DHCP server - Subnet A - 10.10.10.65 - Router - 10.10.10.93 - Subnet B
    >>
    >> 5 Client computers in Subnet A:
    >> IP address: 10.10.10.66 - 10.10.10.70
    >> Subnet mask: 255.255.255.192
    >> Default gateway: 10.10.10.65
    >>
    >>
    >> Server in Subnet B:
    >> IP address: 10.10.10.94
    >> Subnet mask: 255.255.255.240
    >> Default gateway: 10.10.10.93
    >>
    >>
    >> Clients cannot connect to the server in Subnet B unless the subnet mask
    >> of the server is changed to 255.255.255.192
    >>
    >> Why is that? Isn't it the router's job to forward the packets from one
    >> subnet to the other?
    >>
    >>
    >> Thanks!


    The clients get the ip addresses from the DHCP server and IP address:
    10.10.10.66 - 10.10.10.70 are the scope.


    >
    > As indicated by the other respondent, your key issue is the address
    > overlap resulting from your erroneous network mask assignments.


    Are you saying there is no overlap after changin the subnet mask for subnet
    B?

    >
    > It's important to understand that the first routing decision is made by
    > the "host".


    What do you mean by "the host", the client PC or the router?
    From what I know, the only Host that makes routing decision is the router
    and the routing table that it creates based on the routing protocol that it
    uses.

    A host applies its network mask to its own IP address and the destination
    host IP address, then performs a comparison to determine
    > whether the destination host is directly reachable, or whether the packet
    > needs to be forwarded to the router.


    I disagree. The router makes the decisions based on the routing table, not
    the host. The host simply broadcasts and the other host within the broadcast
    domain that "hears" its name, responds.

    >
    > When host addresses are mis-assigned due to erroneous masks, a host may
    > erroneously conclude that a destination host is directly reachable, and
    > attempt to resolve its IP address into a MAC address via ARP (which
    > fails), rather than forwarding the packet to the router. This could occur
    > on the initiator or the responder side depending on the erroneous
    > configuration.
    >
    > The selection of masks should be driven by the potential number of hosts
    > to be assigned to each network.
    >
    > Typically, the first or last assignable host address available for each
    > network will be used for the gateway(s) which can lessen confusion for
    > those unfamiliar with your addressing scheme.
    >
    > Best Regards,
    > News Reader




  7. Re: subnet mask

    In article <0egbk.54388$Jx.52325@pd7urf1no>, "eager"
    wrote:

    > wrote in message
    > news:5bed34cb-ae13-4e58-90cb-73d5cc6274d1@x41g2000hsb.googlegroups.com...
    > On Jul 3, 12:37 am, "eager" wrote:
    > > There is a network as follows:
    > >
    > > DHCP server - Subnet A - 10.10.10.65 - Router - 10.10.10.93 - Subnet B
    > >
    > > 5 Client computers in Subnet A:
    > > IP address: 10.10.10.66 - 10.10.10.70
    > > Subnet mask: 255.255.255.192
    > > Default gateway: 10.10.10.65
    > >
    > > Server in Subnet B:
    > > IP address: 10.10.10.94
    > > Subnet mask: 255.255.255.240
    > > Default gateway: 10.10.10.93
    > >
    > > Clients cannot connect to the server in Subnet B unless the subnet mask of
    > > the server is changed to 255.255.255.192
    > >
    > > Why is that? Isn't it the router's job to forward the packets from one
    > > subnet to the other?

    >
    >
    > While it's hard to say without more information,
    >
    > there is no more INFO, it's a question from 70-291 exam.
    >
    > your first problem is that your two subnets overlap.
    >
    > What do you mean by " two subnets overlap"? - they are in totally different
    > broadcast domains.
    >
    > IOW, subnet B is addresses 10.10.10.92-10.10.10.107,
    >
    > No idea where you got that from, because based on the server INFO, the IP
    > range of the subnet B should be 10.10.10.80 - 10.10.10.95
    > That IP range does not overlap with the scope 10.10.10.66 - 10.10.10.70 of
    > the clients in subnet A.


    Subnet A's subnet mask is 255.255.255.192, which means it's a block of
    64 addresses. So the range is 10.10.10.64 - 10.10.10.127. Subnet B's
    address range 10.10.10.80 - 10.10.10.95 overlaps that.

    P.S. Please use proper prefixing of quotes so that your replies can be
    distinguished from the message you're replying to. It's really hard
    reading your message because of this.

    --
    Barry Margolin, barmar@alum.mit.edu
    Arlington, MA
    *** PLEASE don't copy me on replies, I'll read them in the group ***

  8. Re: subnet mask


    "Barry Margolin" wrote in message
    news:barmar-A3EC0D.23221703072008@newsgroups.comcast.net...
    > In article <0egbk.54388$Jx.52325@pd7urf1no>, "eager"
    > wrote:
    >
    >> wrote in message
    >> news:5bed34cb-ae13-4e58-90cb-73d5cc6274d1@x41g2000hsb.googlegroups.com...
    >> On Jul 3, 12:37 am, "eager" wrote:
    >> > There is a network as follows:
    >> >
    >> > DHCP server - Subnet A - 10.10.10.65 - Router - 10.10.10.93 - Subnet B
    >> >
    >> > 5 Client computers in Subnet A:
    >> > IP address: 10.10.10.66 - 10.10.10.70
    >> > Subnet mask: 255.255.255.192
    >> > Default gateway: 10.10.10.65
    >> >
    >> > Server in Subnet B:
    >> > IP address: 10.10.10.94
    >> > Subnet mask: 255.255.255.240
    >> > Default gateway: 10.10.10.93
    >> >
    >> > Clients cannot connect to the server in Subnet B unless the subnet mask
    >> > of
    >> > the server is changed to 255.255.255.192
    >> >
    >> > Why is that? Isn't it the router's job to forward the packets from one
    >> > subnet to the other?

    >>
    >>
    >> While it's hard to say without more information,
    >>
    >> there is no more INFO, it's a question from 70-291 exam.
    >>
    >> your first problem is that your two subnets overlap.
    >>
    >> What do you mean by " two subnets overlap"? - they are in totally
    >> different
    >> broadcast domains.
    >>
    >> IOW, subnet B is addresses 10.10.10.92-10.10.10.107,
    >>
    >> No idea where you got that from, because based on the server INFO, the IP
    >> range of the subnet B should be 10.10.10.80 - 10.10.10.95
    >> That IP range does not overlap with the scope 10.10.10.66 - 10.10.10.70
    >> of
    >> the clients in subnet A.

    >
    > Subnet A's subnet mask is 255.255.255.192, which means it's a block of
    > 64 addresses. So the range is 10.10.10.64 - 10.10.10.127. Subnet B's
    > address range 10.10.10.80 - 10.10.10.95 overlaps that.
    >
    > P.S. Please use proper prefixing of quotes so that your replies can be
    > distinguished from the message you're replying to. It's really hard
    > reading your message because of this.



    Sorry, I will keep it in mind.
    Anyway, this overlapping theory seems nonsense to me. As I mentioned it,
    after "fixing" the subnet, there is even a better overlap





  9. Re: subnet mask

    eager wrote:
    > "News Reader" wrote in message
    > news:zB7bk.13586$Ec.12497@read2.cgocable.net...
    >> eager wrote:
    >>> There is a network as follows:
    >>>
    >>>
    >>> DHCP server - Subnet A - 10.10.10.65 - Router - 10.10.10.93 - Subnet B
    >>>
    >>> 5 Client computers in Subnet A:
    >>> IP address: 10.10.10.66 - 10.10.10.70
    >>> Subnet mask: 255.255.255.192
    >>> Default gateway: 10.10.10.65
    >>>
    >>>
    >>> Server in Subnet B:
    >>> IP address: 10.10.10.94
    >>> Subnet mask: 255.255.255.240
    >>> Default gateway: 10.10.10.93
    >>>
    >>>
    >>> Clients cannot connect to the server in Subnet B unless the subnet mask
    >>> of the server is changed to 255.255.255.192
    >>>
    >>> Why is that? Isn't it the router's job to forward the packets from one
    >>> subnet to the other?
    >>>
    >>>
    >>> Thanks!

    >
    > The clients get the ip addresses from the DHCP server and IP address:
    > 10.10.10.66 - 10.10.10.70 are the scope.


    This issue has nothing to do with DHCP scopes.

    An interface is configured with an IP address and a network mask. The
    mask is used to identify which "portion" of the address defines the
    "Network ID", and which "portion" uniquely identifies that
    "host/interface" within that network.

    The combination of the Network ID and the mask define the range of IP
    addresses that are considered to be directly connected (i.e.: on the
    same network).

    >
    >
    >> As indicated by the other respondent, your key issue is the address
    >> overlap resulting from your erroneous network mask assignments.

    >
    > Are you saying there is no overlap after changin the subnet mask for subnet
    > B?
    >
    >> It's important to understand that the first routing decision is made by
    >> the "host".

    >
    > What do you mean by "the host", the client PC or the router?
    > From what I know, the only Host that makes routing decision is the router
    > and the routing table that it creates based on the routing protocol that it
    > uses.


    By "host", I mean a PC.

    Are you not aware that a host (PC) has a routing table, and that a host
    only forwards packets to the router if the host determines that it can
    not reach the destination host directly?

    If you are using a Windows PC, type "route print" at the cmd prompt, and
    press enter.

    >
    > A host applies its network mask to its own IP address and the destination
    > host IP address, then performs a comparison to determine
    >> whether the destination host is directly reachable, or whether the packet
    >> needs to be forwarded to the router.

    >
    > I disagree. The router makes the decisions based on the routing table, not
    > the host. The host simply broadcasts and the other host within the broadcast
    > domain that "hears" its name, responds.
    >


    The router separates networks, and broadcast domains.

    If a host concludes that the destination host is on the same network
    (determined by the mask comparison), it will send an ARP broadcast
    (containing the destination host's IP address) that will be contained on
    that side of the router. The destination host will respond with an ARP
    Reply containing its MAC address. The host will cache this IP-to-MAC
    mapping in its ARP table and forward the packet to the destination host
    (framed with the newly resolved destination MAC address), without the
    router being involved in any way.

    If a host concludes that the destination host is NOT on the same network
    (determined by the mask comparison), it knows that it will need the
    services of a router. It will check its ARP table to see if the router's
    MAC address is known. If known, it will forward the packet to the router
    (framed with the router's MAC address). If it is not known, it will send
    an ARP broadcast (containing the configured gateway IP address). When
    the router responds, the host will then know the MAC address required to
    forward the packet to the router.

    >> When host addresses are mis-assigned due to erroneous masks, a host may
    >> erroneously conclude that a destination host is directly reachable, and
    >> attempt to resolve its IP address into a MAC address via ARP (which
    >> fails), rather than forwarding the packet to the router. This could occur
    >> on the initiator or the responder side depending on the erroneous
    >> configuration.
    >>
    >> Best Regards,
    >> News Reader

    >
    >


    Best Regards,
    News Reader

  10. Re: subnet mask


    "News Reader" wrote in message
    news:Iqhbk.14045$Mc.2768@read1.cgocable.net...
    > eager wrote:
    >> "News Reader" wrote in message
    >> news:zB7bk.13586$Ec.12497@read2.cgocable.net...
    >>> eager wrote:
    >>>> There is a network as follows:
    >>>>
    >>>>
    >>>> DHCP server - Subnet A - 10.10.10.65 - Router - 10.10.10.93 - Subnet B
    >>>>
    >>>> 5 Client computers in Subnet A:
    >>>> IP address: 10.10.10.66 - 10.10.10.70
    >>>> Subnet mask: 255.255.255.192
    >>>> Default gateway: 10.10.10.65
    >>>>
    >>>>
    >>>> Server in Subnet B:
    >>>> IP address: 10.10.10.94
    >>>> Subnet mask: 255.255.255.240
    >>>> Default gateway: 10.10.10.93
    >>>>
    >>>>
    >>>> Clients cannot connect to the server in Subnet B unless the subnet mask
    >>>> of the server is changed to 255.255.255.192
    >>>>
    >>>> Why is that? Isn't it the router's job to forward the packets from one
    >>>> subnet to the other?
    >>>>
    >>>>
    >>>> Thanks!

    >>
    >> The clients get the ip addresses from the DHCP server and IP address:
    >> 10.10.10.66 - 10.10.10.70 are the scope.

    >
    > This issue has nothing to do with DHCP scopes.
    >
    > An interface is configured with an IP address and a network mask. The mask
    > is used to identify which "portion" of the address defines the "Network
    > ID", and which "portion" uniquely identifies that "host/interface" within
    > that network.
    >
    > The combination of the Network ID and the mask define the range of IP
    > addresses that are considered to be directly connected (i.e.: on the same
    > network).
    >
    >>
    >>
    >>> As indicated by the other respondent, your key issue is the address
    >>> overlap resulting from your erroneous network mask assignments.

    >>
    >> Are you saying there is no overlap after changin the subnet mask for
    >> subnet B?
    >>
    >>> It's important to understand that the first routing decision is made by
    >>> the "host".

    >>
    >> What do you mean by "the host", the client PC or the router?
    >> From what I know, the only Host that makes routing decision is the router
    >> and the routing table that it creates based on the routing protocol that
    >> it uses.

    >
    > By "host", I mean a PC.
    >
    > Are you not aware that a host (PC) has a routing table, and that a host
    > only forwards packets to the router if the host determines that it can not
    > reach the destination host directly?
    >
    > If you are using a Windows PC, type "route print" at the cmd prompt, and
    > press enter.
    >
    >>
    >> A host applies its network mask to its own IP address and the destination
    >> host IP address, then performs a comparison to determine
    >>> whether the destination host is directly reachable, or whether the
    >>> packet needs to be forwarded to the router.

    >>
    >> I disagree. The router makes the decisions based on the routing table,
    >> not the host. The host simply broadcasts and the other host within the
    >> broadcast domain that "hears" its name, responds.
    >>

    >
    > The router separates networks, and broadcast domains.
    >
    > If a host concludes that the destination host is on the same network
    > (determined by the mask comparison), it will send an ARP broadcast
    > (containing the destination host's IP address) that will be contained on
    > that side of the router. The destination host will respond with an ARP
    > Reply containing its MAC address. The host will cache this IP-to-MAC
    > mapping in its ARP table and forward the packet to the destination host
    > (framed with the newly resolved destination MAC address), without the
    > router being involved in any way.
    >
    > If a host concludes that the destination host is NOT on the same network
    > (determined by the mask comparison), it knows that it will need the
    > services of a router. It will check its ARP table to see if the router's
    > MAC address is known. If known, it will forward the packet to the router
    > (framed with the router's MAC address). If it is not known, it will send
    > an ARP broadcast (containing the configured gateway IP address). When the
    > router responds, the host will then know the MAC address required to
    > forward the packet to the router.
    >




    Thank you for your reply, but I am afraid you are missing the whole point
    here:
    After "fixing" the problem by changing the subnet mask, there is a better
    overlapping of the subnets.
    What would you say to that?

    Anyway, I am aware of the "route print" as I am aware of "arp -a", but lets
    not confuse routing with routes and routing tables with broadcasting and
    unicasting ...

    All I need to know is why making both subnet masks the same will establish
    the connectivity.

    Thanks




  11. Re: subnet mask

    eager wrote:
    >
    > Sorry, I will keep it in mind.
    > Anyway, this overlapping theory seems nonsense to me. As I mentioned it,
    > after "fixing" the subnet, there is even a better overlap
    >
    >


    Perhaps you have fixed the wrong subnet, and or used the wrong mask.
    Please read this entire post before responding.


    You initially stated (03/07/2008 1:37 AM):

    "Clients cannot connect to the server in Subnet B unless the subnet mask
    of the server is changed to 255.255.255.192"


    Robert offered an explanation (03/07/2008 2:29 AM):

    "That you're getting any communication between the clients on A and the
    server on B suggests that you also have the two subnets on a single
    physical network." ... more.

    .... but did NOT say you were to change the subnet B mask to
    "255.255.255.192".


    Barry indicated (03/07/2008 10:44 PM):

    "If there's some reason you can't fix the subnet masks on Subnet A, ...."


    Have you corrected "Subnet A", and used an appropriate mask?

    e.g:

    Subnet A with a mask of 255.255.255.240:

    Network ID: 10.10.10.64
    Host range: 10.10.10.65 - 10.10.10.78
    Broadcast address: 10.10.10.79


    Subnet B with a mask of 255.255.255.240:

    Network ID: 10.10.10.80
    Host range: 10.10.10.81 - 10.10.10.94
    Broadcast address: 10.10.10.95


    Best Regards,
    News Reader

  12. Re: subnet mask


    "News Reader" wrote in message
    news:r2ibk.14050$Mc.4272@read1.cgocable.net...
    > eager wrote:
    >>
    >> Sorry, I will keep it in mind.
    >> Anyway, this overlapping theory seems nonsense to me. As I mentioned it,
    >> after "fixing" the subnet, there is even a better overlap
    >>
    >>

    >
    > Perhaps you have fixed the wrong subnet, and or used the wrong mask.
    > Please read this entire post before responding.
    >
    >
    > You initially stated (03/07/2008 1:37 AM):
    >
    > "Clients cannot connect to the server in Subnet B unless the subnet mask
    > of the server is changed to 255.255.255.192"
    >
    >
    > Robert offered an explanation (03/07/2008 2:29 AM):
    >
    > "That you're getting any communication between the clients on A and the
    > server on B suggests that you also have the two subnets on a single
    > physical network." ... more.
    >
    > ... but did NOT say you were to change the subnet B mask to
    > "255.255.255.192".
    >
    >
    > Barry indicated (03/07/2008 10:44 PM):
    >
    > "If there's some reason you can't fix the subnet masks on Subnet A, ...."
    >
    >
    > Have you corrected "Subnet A", and used an appropriate mask?
    >
    > e.g:
    >
    > Subnet A with a mask of 255.255.255.240:
    >
    > Network ID: 10.10.10.64
    > Host range: 10.10.10.65 - 10.10.10.78
    > Broadcast address: 10.10.10.79
    >
    >
    > Subnet B with a mask of 255.255.255.240:
    >
    > Network ID: 10.10.10.80
    > Host range: 10.10.10.81 - 10.10.10.94
    > Broadcast address: 10.10.10.95
    >
    >
    > Best Regards,
    > News Reader



    Please, read my OP again. It's a question and the correct answer from a
    70-291 exam. You may find it online too.
    Sorry, but your theory of overlapping is not convincing.



  13. Re: subnet mask

    eager wrote:
    > "News Reader" wrote in message
    > news:Iqhbk.14045$Mc.2768@read1.cgocable.net...
    >> eager wrote:
    >>> "News Reader" wrote in message
    >>> news:zB7bk.13586$Ec.12497@read2.cgocable.net...
    >>>> eager wrote:
    >>>>> There is a network as follows:
    >>>>>
    >>>>>
    >>>>> DHCP server - Subnet A - 10.10.10.65 - Router - 10.10.10.93 - Subnet B
    >>>>>
    >>>>> 5 Client computers in Subnet A:
    >>>>> IP address: 10.10.10.66 - 10.10.10.70
    >>>>> Subnet mask: 255.255.255.192
    >>>>> Default gateway: 10.10.10.65
    >>>>>
    >>>>>
    >>>>> Server in Subnet B:
    >>>>> IP address: 10.10.10.94
    >>>>> Subnet mask: 255.255.255.240
    >>>>> Default gateway: 10.10.10.93
    >>>>>
    >>>>>
    >>>>> Clients cannot connect to the server in Subnet B unless the subnet mask
    >>>>> of the server is changed to 255.255.255.192
    >>>>>
    >>>>> Why is that? Isn't it the router's job to forward the packets from one
    >>>>> subnet to the other?
    >>>>>
    >>>>>
    >>>>> Thanks!
    >>> The clients get the ip addresses from the DHCP server and IP address:
    >>> 10.10.10.66 - 10.10.10.70 are the scope.

    >> This issue has nothing to do with DHCP scopes.
    >>
    >> An interface is configured with an IP address and a network mask. The mask
    >> is used to identify which "portion" of the address defines the "Network
    >> ID", and which "portion" uniquely identifies that "host/interface" within
    >> that network.
    >>
    >> The combination of the Network ID and the mask define the range of IP
    >> addresses that are considered to be directly connected (i.e.: on the same
    >> network).
    >>
    >>>
    >>>> As indicated by the other respondent, your key issue is the address
    >>>> overlap resulting from your erroneous network mask assignments.
    >>> Are you saying there is no overlap after changin the subnet mask for
    >>> subnet B?
    >>>
    >>>> It's important to understand that the first routing decision is made by
    >>>> the "host".
    >>> What do you mean by "the host", the client PC or the router?
    >>> From what I know, the only Host that makes routing decision is the router
    >>> and the routing table that it creates based on the routing protocol that
    >>> it uses.

    >> By "host", I mean a PC.
    >>
    >> Are you not aware that a host (PC) has a routing table, and that a host
    >> only forwards packets to the router if the host determines that it can not
    >> reach the destination host directly?
    >>
    >> If you are using a Windows PC, type "route print" at the cmd prompt, and
    >> press enter.
    >>
    >>> A host applies its network mask to its own IP address and the destination
    >>> host IP address, then performs a comparison to determine
    >>>> whether the destination host is directly reachable, or whether the
    >>>> packet needs to be forwarded to the router.
    >>> I disagree. The router makes the decisions based on the routing table,
    >>> not the host. The host simply broadcasts and the other host within the
    >>> broadcast domain that "hears" its name, responds.
    >>>

    >> The router separates networks, and broadcast domains.
    >>
    >> If a host concludes that the destination host is on the same network
    >> (determined by the mask comparison), it will send an ARP broadcast
    >> (containing the destination host's IP address) that will be contained on
    >> that side of the router. The destination host will respond with an ARP
    >> Reply containing its MAC address. The host will cache this IP-to-MAC
    >> mapping in its ARP table and forward the packet to the destination host
    >> (framed with the newly resolved destination MAC address), without the
    >> router being involved in any way.
    >>
    >> If a host concludes that the destination host is NOT on the same network
    >> (determined by the mask comparison), it knows that it will need the
    >> services of a router. It will check its ARP table to see if the router's
    >> MAC address is known. If known, it will forward the packet to the router
    >> (framed with the router's MAC address). If it is not known, it will send
    >> an ARP broadcast (containing the configured gateway IP address). When the
    >> router responds, the host will then know the MAC address required to
    >> forward the packet to the router.
    >>

    >
    >
    >
    > Thank you for your reply, but I am afraid you are missing the whole point
    > here:


    No, I was addressing a lack of understanding displayed in your post of
    03/07/2008 11:01 PM, and confined my response to such.

    > After "fixing" the problem by changing the subnet mask, there is a better
    > overlapping of the subnets.
    > What would you say to that?


    Read my other post (12:54 AM) which indicates you probably did not
    address the correct subnet, and probably not with the correct mask.

    >
    > Anyway, I am aware of the "route print" as I am aware of "arp -a", but lets
    > not confuse routing with routes and routing tables with broadcasting and
    > unicasting ...
    >
    > All I need to know is why making both subnet masks the same will establish
    > the connectivity.
    >
    > Thanks
    >
    >
    >


    Best Regards,
    News Reader

  14. Re: subnet mask

    eager wrote:
    > "News Reader" wrote in message
    > news:r2ibk.14050$Mc.4272@read1.cgocable.net...
    >> eager wrote:
    >>> Sorry, I will keep it in mind.
    >>> Anyway, this overlapping theory seems nonsense to me. As I mentioned it,
    >>> after "fixing" the subnet, there is even a better overlap
    >>>
    >>>

    >> Perhaps you have fixed the wrong subnet, and or used the wrong mask.
    >> Please read this entire post before responding.
    >>
    >>
    >> You initially stated (03/07/2008 1:37 AM):
    >>
    >> "Clients cannot connect to the server in Subnet B unless the subnet mask
    >> of the server is changed to 255.255.255.192"
    >>
    >>
    >> Robert offered an explanation (03/07/2008 2:29 AM):
    >>
    >> "That you're getting any communication between the clients on A and the
    >> server on B suggests that you also have the two subnets on a single
    >> physical network." ... more.
    >>
    >> ... but did NOT say you were to change the subnet B mask to
    >> "255.255.255.192".
    >>
    >>
    >> Barry indicated (03/07/2008 10:44 PM):
    >>
    >> "If there's some reason you can't fix the subnet masks on Subnet A, ...."
    >>
    >>
    >> Have you corrected "Subnet A", and used an appropriate mask?
    >>
    >> e.g:
    >>
    >> Subnet A with a mask of 255.255.255.240:
    >>
    >> Network ID: 10.10.10.64
    >> Host range: 10.10.10.65 - 10.10.10.78
    >> Broadcast address: 10.10.10.79
    >>
    >>
    >> Subnet B with a mask of 255.255.255.240:
    >>
    >> Network ID: 10.10.10.80
    >> Host range: 10.10.10.81 - 10.10.10.94
    >> Broadcast address: 10.10.10.95
    >>
    >>
    >> Best Regards,
    >> News Reader

    >
    >
    > Please, read my OP again. It's a question and the correct answer from a
    > 70-291 exam. You may find it online too.
    > Sorry, but your theory of overlapping is not convincing.
    >
    >


    Your OP did NOT convey that it was an exam question. Most of us probably
    thought you were in need of assistance in defining non-overlapping
    subnets in a real world scenario, and have offered advice based on that
    belief.

    Some of your responses have been rather ignorant.

    I've wasted enough time trying to assist you.

    Good luck with you understanding.

    Best Regards,
    News Reader

  15. Re: subnet mask

    On Friday 4 July 2008 07:22, News Reader wrote:

    > Some of your responses have been rather ignorant.
    >
    > I've wasted enough time trying to assist you.


    Yes, as far as I'm concerned the OP is a troll. I plonked him long ago.


  16. Re: subnet mask


    "pk" wrote in message news:g4kjv4$l2r$1@aioe.org...
    > On Friday 4 July 2008 07:22, News Reader wrote:
    >
    >> Some of your responses have been rather ignorant.
    >>
    >> I've wasted enough time trying to assist you.

    >
    > Yes, as far as I'm concerned the OP is a troll. I plonked him long ago.
    >


    troll??
    I just asked a question and was hoping for a convincing answer (from a
    knowledgeable person).
    FYI, I am very polite in my posts and I do not use expressions like the ones
    above.



  17. Re: subnet mask

    In article , "eager"
    wrote:

    > There is a network as follows:
    >
    >
    > DHCP server - Subnet A - 10.10.10.65 - Router - 10.10.10.93 - Subnet B
    >
    > 5 Client computers in Subnet A:
    > IP address: 10.10.10.66 - 10.10.10.70
    > Subnet mask: 255.255.255.192
    > Default gateway: 10.10.10.65
    >
    >
    > Server in Subnet B:
    > IP address: 10.10.10.94
    > Subnet mask: 255.255.255.240
    > Default gateway: 10.10.10.93
    >
    >
    > Clients cannot connect to the server in Subnet B unless the subnet mask of
    > the server is changed to 255.255.255.192


    Something's very confusing here. At the very top you have the DHCP
    server in Subnet A, but now you say the server is in Subnet B. Which is
    it?

    And the solution of changing the subnet mask on the server in Subnet B
    would only make sense if the device in between was a hub or switch, not
    a router. For a router, the fix should be to correct the subnet mask of
    Subnet A, as I said earlier.

    I can't figure out why that change fixes things, and I've been managing
    TCP/IP networks (including a nationwide ISP backbone) for 20 years.

    --
    Barry Margolin, barmar@alum.mit.edu
    Arlington, MA
    *** PLEASE don't copy me on replies, I'll read them in the group ***

  18. Re: subnet mask

    <--snip-->

    I have been following this for a while silently. However, this appears to
    be train wreck. It seams as you do not understand the relation ship between
    IP addressing, subnet masks and how it pertains to routing.
    >
    >
    > Thank you for your reply, but I am afraid you are missing the whole point
    > here:
    > After "fixing" the problem by changing the subnet mask, there is a better
    > overlapping of the subnets.
    > What would you say to that?
    >


    I would say that overlapping of subnets is bad. I don't know a how there
    can be better or any gray are for this. That is sort of like saying that
    binary has three values (0,1 and 2). That is not the case. If there is any
    overlap in subnet, you better know what you are doing to address it properly
    with nat. Also, there are concerns like proxy arp that change the behavior
    of inconsistent mask on the router and hosts.

    > Anyway, I am aware of the "route print" as I am aware of "arp -a", but
    > lets not confuse routing with routes and routing tables with broadcasting
    > and unicasting ...


    Actually, these are all related. Route print shows the routing table. Arp
    also lends clues to if you are doing proxy arp etc. Route print is the
    local routing table. This defines the broadcast. Unicast is anything that
    falls within this range as well as anything that goes out non multicast
    destination. However, in some cases this can be converted to a directed
    broadcast by destination router. You cannot tell that though. All of these
    are the concepts that go together to understand how a frame is built, and
    what the destination mac address is (the real destination, or the router)
    >
    > All I need to know is why making both subnet masks the same will establish
    > the connectivity.
    >
    > Thanks
    >


    What you need to understand is how IP functions. If you understand this,
    this is a very simple thing to troubleshoot and understand for you. You can
    use wireshark. Understanding that the communication is bidirectional, does
    each host arp for the destination ip or the gateway? Check in the reverse
    direction as well. If it arps for the destination, the router may respond
    with a proxy arp. The frame can be build with a destination IP of the
    destination and the destination mac of the router. Or if everthing is
    properly configured, the hosts should arp for the default gateways and the
    packets would be built as such. In nearly every case like this, the issue
    is only in one direction.

    There have been many people took time to answer the questions to the best
    of anyones abilities. They are correct in response based on what you have
    told them. I would suggest that if you are interested in understanding this
    that you do two things. 1). Get a copy of TCP/IP unleashed by Tim Parker.
    This book is great. 2). The next thing is to start looking at how this
    stuff works in Wireshark.

    TCP/IP alone is not very complicated once you understand it. Well, I guess
    that can be said for anything.



  19. Re: subnet mask


    "Barry Margolin" wrote in message
    news:barmar-E16E9D.10363204072008@newsgroups.comcast.net...
    > In article , "eager"
    > wrote:
    >
    >> There is a network as follows:
    >>
    >>
    >> DHCP server - Subnet A - 10.10.10.65 - Router - 10.10.10.93 - Subnet B
    >>
    >> 5 Client computers in Subnet A:
    >> IP address: 10.10.10.66 - 10.10.10.70
    >> Subnet mask: 255.255.255.192
    >> Default gateway: 10.10.10.65
    >>
    >>
    >> Server in Subnet B:
    >> IP address: 10.10.10.94
    >> Subnet mask: 255.255.255.240
    >> Default gateway: 10.10.10.93
    >>
    >>
    >> Clients cannot connect to the server in Subnet B unless the subnet mask
    >> of
    >> the server is changed to 255.255.255.192

    >
    > Something's very confusing here. At the very top you have the DHCP
    > server in Subnet A, but now you say the server is in Subnet B. Which is
    > it?
    >
    > And the solution of changing the subnet mask on the server in Subnet B
    > would only make sense if the device in between was a hub or switch, not
    > a router. For a router, the fix should be to correct the subnet mask of
    > Subnet A, as I said earlier.
    >
    > I can't figure out why that change fixes things, and I've been managing
    > TCP/IP networks (including a nationwide ISP backbone) for 20 years.
    >



    Barry,

    The DHCP server in subnet A is to lease IP addresses to 5 Client computers
    in Subnet A.
    The server in subnet B offers services to 5 clients in subnet A.

    I totally agree, this does not make any sense to me, that's why I posted it
    here.
    It's an actual question from Microsoft 70-291 exam.



  20. Re: subnet mask

    In article <1bzbk.55672$Jx.21863@pd7urf1no>, "eager"
    wrote:

    > It's an actual question from Microsoft 70-291 exam.


    You keep saying that as if "70-291" is supposed to mean something to us.

    --
    Barry Margolin, barmar@alum.mit.edu
    Arlington, MA
    *** PLEASE don't copy me on replies, I'll read them in the group ***

+ Reply to Thread
Page 1 of 2 1 2 LastLast