About VLANs - TCP-IP

This is a discussion on About VLANs - TCP-IP ; Hi, I am Vikrant, doing training in embedded Techonology. So my question is , 1: What the difference between a layer 2 switch which is supporting only port-based vlan or it support tag (Q) based vlan. 2: If a layer ...

+ Reply to Thread
Page 1 of 2 1 2 LastLast
Results 1 to 20 of 25

Thread: About VLANs

  1. About VLANs

    Hi,
    I am Vikrant, doing training in embedded Techonology.

    So my question is ,

    1: What the difference between a layer 2 switch which is
    supporting only port-based vlan or it support tag (Q) based vlan.
    2: If a layer 2 switch is using only port based vlan then
    is it not supporting IEEE 802.1q compatibility.

    3: Please tell me in detalis about a single instance
    Spanning Tree Protocol with multiple vlans.


    In waiting of u;r reply...
    Thanks....


  2. Re: About VLANs

    On May 23, 3:44*am, vicky wrote:

    > * * * * * 1: *What the difference between a layer 2 switch which is
    > supporting only port-based vlan *or it support tag (Q) based vlan.


    This is probably a discussion more appropriate for
    comp.dcom.lans.ethernet, however ...

    If a switch supports the 802.1Q standard, then the switch can carry
    tagged frames from multiple VLANs in a single trunk link. These are
    then usually sent untagged out of edge ports, however. The switch only
    sends frames from a given VLAN to each edge port, according to static
    configuration.

    VLANs can also be determined by IP address. In this case, switch ports
    can be configured to bridge IP packets from certain IP subnets, but
    only route IP packets from IP subnets not in the VLAN. So I suppose
    this variant of VLAN would make the topic okay for this news group.

    > * * * * * 2: * If a layer 2 switch is using only port based vlan then
    > is it not supporting IEEE 802.1q compatibility.


    True. Not all VLANs are 802.1Q VLANs.

    > * * * * * 3: *Please tell me in detalis about *a single instance
    > Spanning Tree Protocol with multiple vlans.


    Originally, it was always done with a single spanning tree, as far as
    I know. Since the spanning tree links all the switches together
    through the spanning tree ports, if all of the spanning tree ports
    support tagged frames, then the switches in the mesh can participate
    in any VLAN running in that mesh.

    Bert

  3. Re: About VLANs

    On May 24, 3:14*am, Albert Manfredi wrote:
    > On May 23, 3:44*am, vicky wrote:
    >
    > > * * * * * 1: *What the difference between a layer 2 switch which is
    > > supporting only port-based vlan *or it support tag (Q) based vlan.

    >
    > This is probably a discussion more appropriate for
    > comp.dcom.lans.ethernet, however ...
    >
    > If a switch supports the 802.1Q standard, then the switch can carry
    > tagged frames from multiple VLANs in a single trunk link. These are
    > then usually sent untagged out of edge ports, however. The switch only
    > sends frames from a given VLAN to each edge port, according to static
    > configuration.
    >
    > VLANs can also be determined by IP address. In this case, switch ports
    > can be configured to bridge IP packets from certain IP subnets, but
    > only route IP packets from IP subnets not in the VLAN. So I suppose
    > this variant of VLAN would make the topic okay for this news group.
    >
    > > * * * * * 2: * If a layer 2 switch is using only port based vlan then
    > > is it not supporting IEEE 802.1q compatibility.

    >
    > True. Not all VLANs are 802.1Q VLANs.
    >
    > > * * * * * 3: *Please tell me in detalis about *a single instance
    > > Spanning Tree Protocol with multiple vlans.

    >
    > Originally, it was always done with a single spanning tree, as far as
    > I know. Since the spanning tree links all the switches together
    > through the spanning tree ports, if all of the spanning tree ports
    > support tagged frames, then the switches in the mesh can participate
    > in any VLAN running in that mesh.
    >
    > Bert


    -------------------------------------------------------------------------------------------------------

    Hello Mr. Albert
    Firstly thanks ....

    By u'r statement
    If a layer 2 switch is using only port based vlan then
    is it not supporting IEEE 802.1q compatibility.

    I want to ask one thing what happen if a layer 2 managed switch is
    supporting port-based vlan , and also a network is established by more
    than one layer 2 switches (managed - all supporting port based vlan)
    and each switch connests to other (connection is done by trunk port
    (vlan trunking)) then

    in this scenerio is both port based and ieee 802.1q is using .

  4. Re: About VLANs

    On May 24, 10:36*am, vicky wrote:
    > On May 24, 3:14*am, Albert Manfredi wrote:
    >
    >
    >

    Mr Albert

    If a layer 2 switch is using IEEE 802.1q compatibility, then is it
    support both ieee 802.1q and port-based vlan compatibility ???

  5. Re: About VLANs

    In article <838652e9-4d74-4c22-960a-d268bc1bc1a0@k37g2000hsf.googlegroups.com>,
    Albert Manfredi wrote:
    >On May 23, 3:44*am, vicky wrote:


    >> * * * * * 3: *Please tell me in detalis about *a single instance
    >> Spanning Tree Protocol with multiple vlans.


    >Originally, it was always done with a single spanning tree, as far as
    >I know.


    Right. Which promptly didn't work.

    >Since the spanning tree links all the switches together
    >through the spanning tree ports,


    Okay so far

    >if all of the spanning tree ports
    >support tagged frames,


    Conditional statement ahead

    >then the switches in the mesh can participate
    >in any VLAN running in that mesh.


    False conclusion.

    The fact that a switch port supports 802.1Q (or other) tagged frames
    does not mean that all PVIDs configured anywhere in the switch
    mess are configured to be allowed to travel through a particular
    port.

    Suppose I take two switches. On the first port of each switch, I
    configure the port to allow PVID 1-512. On the second port of
    each switch, I configure the port to allow PVID 513-1024.
    Now I connect the respective ports on the switches.

    If a single-instance spanning tree protocol is running, the
    spanning tree algorithm would detect that the switches were connected
    twice, and would disable one of the connections to preserve spanning
    tree properties. Unfortunately this would leave the vlans associated
    with the other port unable to talk. The solution with traditional
    single-instance spanning tree is "Don't Do That!" -- make sure that
    all of the links between two switches are enabled for all of the
    PVID that need to talk between those two switches, and let the spanning
    tree choose one of the ports to do all the talking over. Unfortunately
    spanning tree is intended to protect against -accidental-
    misconfiguration so saying "Don't Do That!" is not a good enough
    solution. Especially if the reason you want to run multiple links
    is for load distribution.

    There is a current draft, 802.1aq, for shortest path bridging for
    virtual bridged local area networks (local and metropolitan area)
    It is still under active consideration; I do not know how close
    it is to concensus.

  6. Re: About VLANs

    In article <1785887f-d24d-48a5-908d-f4571df45ce6@i36g2000prf.googlegroups.com>,
    vicky wrote:

    >If a layer 2 switch is using IEEE 802.1q compatibility, then is it
    >support both ieee 802.1q and port-based vlan compatibility ???


    It is allowed to; it is not required to.

    When port-based vlans are used, distribution of a packet to
    a destination port within the VLAN is based upon standard bridging
    based upon examination of MACs. If the ports that are in the
    port-based vlan are not set up to strip off 802.1q tags (and
    normally they would not be), then the packets carried over the
    port-based VLAN could include 802.1Q tags, thus allowing
    802.1Q based VLANs to be transparently carried over the port-based
    VLAN. I have configured such a situation myself (the reasons in
    my case had to do with media conversion).

  7. Re: About VLANs

    On May 24, 1:00*pm, rober...@hushmail.com (Walter Roberson) wrote:
    > In article <1785887f-d24d-48a5-908d-f4571df45...@i36g2000prf.googlegroups.com>,
    >
    > vicky * wrote:
    > >If a layer 2 switch is using *IEEE 802.1q compatibility, then is it
    > >support both ieee 802.1q and port-based vlan compatibility ???

    >
    > It is allowed to; it is not required to.
    >
    > When port-based vlans are used, distribution of a packet to
    > a destination port within the VLAN is based upon standard bridging
    > based upon examination of MACs. If the ports that are in the
    > port-based vlan are not set up to strip off 802.1q tags (and
    > normally they would not be), then the packets carried over the
    > port-based VLAN could include 802.1Q tags, thus allowing
    > 802.1Q based VLANs to be transparently carried over the port-based
    > VLAN. *I have configured such a situation myself (the reasons in
    > my case had to do with media conversion).


    -----------------------------------

    If layer 2 managed switch(s1) is in network and the by stp a loop free
    graph is created , then in the switch s1(if it not root bridge) one
    port is root port and then if two other ports from which the pc's are
    connected are became designated ports or not ?

    Please tell me....

  8. Re: About VLANs

    On May 24, 1:00*pm, rober...@hushmail.com (Walter Roberson) wrote:
    > In article <1785887f-d24d-48a5-908d-f4571df45...@i36g2000prf.googlegroups.com>,
    >
    > vicky * wrote:
    > >If a layer 2 switch is using *IEEE 802.1q compatibility, then is it
    > >support both ieee 802.1q and port-based vlan compatibility ???

    >
    > It is allowed to; it is not required to.
    >
    > When port-based vlans are used, distribution of a packet to
    > a destination port within the VLAN is based upon standard bridging
    > based upon examination of MACs. If the ports that are in the
    > port-based vlan are not set up to strip off 802.1q tags (and
    > normally they would not be), then the packets carried over the
    > port-based VLAN could include 802.1Q tags, thus allowing
    > 802.1Q based VLANs to be transparently carried over the port-based
    > VLAN. *I have configured such a situation myself (the reasons in
    > my case had to do with media conversion).


    ------------------------------------------------------------------------------------

    Is the number of vlans a switch supports is depends on , switch's
    controller (means in some switch mannual it 's mentioned that this
    switch contains 32 vlans)

    Thanks in advance....

  9. Re: About VLANs

    On May 24, 1:00*pm, rober...@hushmail.com (Walter Roberson) wrote:
    > In article <1785887f-d24d-48a5-908d-f4571df45...@i36g2000prf.googlegroups.com>,
    >
    > vicky * wrote:
    > >If a layer 2 switch is using *IEEE 802.1q compatibility, then is it
    > >support both ieee 802.1q and port-based vlan compatibility ???

    >
    > It is allowed to; it is not required to.
    >
    > When port-based vlans are used, distribution of a packet to
    > a destination port within the VLAN is based upon standard bridging
    > based upon examination of MACs. If the ports that are in the
    > port-based vlan are not set up to strip off 802.1q tags (and
    > normally they would not be), then the packets carried over the
    > port-based VLAN could include 802.1Q tags, thus allowing
    > 802.1Q based VLANs to be transparently carried over the port-based
    > VLAN. *I have configured such a situation myself (the reasons in
    > my case had to do with media conversion).


    -----------------------------------------------------------


    Is the number of vlans a switch supports is depends on , switch's
    controller (means in some switch mannual it 's mentioned that this
    switch contains 32 vlans)

    Please tell me....

  10. Re: About VLANs

    In article <9873d42b-f03a-4dd3-93eb-90dd941f81a0@g16g2000pri.googlegroups.com>,
    vicky wrote:

    >Is the number of vlans a switch supports is depends on , switch's
    >controller (means in some switch mannual it 's mentioned that this
    >switch contains 32 vlans)


    Well, let's see:

    ftp://ftp.hp.com/pub/networking/soft...hap02-VLAN.pdf

    Table 2-1 VLAN Maximums

    Series 2600 Switches Up to 253
    Series 2600-PWR Switches Up to 253
    Series 4100gl Switches Up to 30
    Series 6108 Up to 30
    Series 2800 Switches Up to 256


    Clearly, then:

    NO, the number of VLANs supported by a switch is *independant* of the
    switch controller. The number of supported VLANs is limited by FCC
    regulations based upon the model number of the switch.

  11. Re: About VLANs

    In article <9cce1a62-c4ab-4859-abd4-3456431ea488@i18g2000prn.googlegroups.com>,
    vicky wrote:

    >If layer 2 managed switch(s1) is in network and the by stp a loop free
    >graph is created , then in the switch s1(if it not root bridge) one
    >port is root port and then if two other ports from which the pc's are
    >connected are became designated ports or not ?


    http://en.wikipedia.org/wiki/Spanning_tree_protocol

    "Any active port that is not a root port or a designated port
    is a blocked port."


    We can read that backwards and thus see that any active port that
    is not a root port or a blocked port is a designated port.

  12. Re: About VLANs

    On May 24, 3:47*am, rober...@hushmail.com (Walter Roberson) wrote:

    > Albert Manfredi * wrote:
    >
    > >if all of the spanning tree ports
    > >support tagged frames,
    > >then the switches in the mesh can participate
    > >in any VLAN running in that mesh.

    >
    > False conclusion.
    >
    > The fact that a switch port supports 802.1Q (or other) tagged frames
    > does not mean that all PVIDs configured anywhere in the switch
    > mess are configured to be allowed to travel through a particular
    > port.
    >
    > Suppose I take two switches. On the first port of each switch, I
    > configure the port to allow PVID 1-512. On the second port of
    > each switch, I configure the port to allow PVID 513-1024.
    > Now I connect the respective ports on the switches.


    Obviously, you have configured the switches wrong for a case of
    multiple VLANs running over a single spanning tree.

    I did not say that that multiple VLANs *had* to operate over a single
    spanning tree. I said that if you have a single spanning tree, and if
    the spanning tree ports in the mesh are VLAN-aware, then you will be
    able to set up all the edge ports in the mesh to be members of any of
    the VLANs.

    > If a single-instance spanning tree protocol is running, the
    > spanning tree algorithm would detect that the switches were connected
    > twice, and would disable one of the connections to preserve spanning
    > tree properties.


    Not if the single spanning tree knows nothing about VLANs.

    It's true that when you have multiple VLANs on a given mesh, a
    spanning tree for each independent VLAN could be different, more
    efficient, than the single spanning tree. That's why MSTP is described
    in 802.1Q.

    Bert

  13. Re: About VLANs

    On May 24, 2:02*pm, rober...@hushmail.com (Walter Roberson) wrote:

    > vicky * wrote:
    > >If layer 2 managed switch(s1) is in network and the by stp a loop free
    > >graph is created , then in the switch s1(if it not root bridge) *one
    > >port is root port *and then if two other ports from which the pc's are
    > >connected are became *designated ports *or not ?

    >
    > http://en.wikipedia.org/wiki/Spanning_tree_protocol
    >
    > * "Any active port that is not a root port or a designated port
    > * is a blocked port."
    >
    > We can read that backwards and thus see that any active port that
    > is not a root port or a blocked port is a designated port.


    The designated port is the port of a switch on a LAN that provides the
    lowest cost path back to the root switch from that LAN. The root port
    is the port on a switch that is "closest" to the root bridge. However,
    aside form these active ports of the spanning tree, edge ports also
    forward frames. Edge ports are ports through which no BPDUs travel,
    but through which hosts are connected.

    Bert

  14. Re: About VLANs

    On May 25, 11:12*am, Albert Manfredi wrote:
    > On May 24, 2:02*pm, rober...@hushmail.com (Walter Roberson) wrote:
    >
    > > vicky * wrote:
    > > >If layer 2 managed switch(s1) is in network and the by stp a loop free
    > > >graph is created , then in the switch s1(if it not root bridge) *one
    > > >port is root port *and then if two other ports from which the pc's are
    > > >connected are became *designated ports *or not ?

    >
    > >http://en.wikipedia.org/wiki/Spanning_tree_protocol

    >
    > > * "Any active port that is not a root port or a designated port
    > > * is a blocked port."

    >
    > > We can read that backwards and thus see that any active port that
    > > is not a root port or a blocked port is a designated port.

    >
    > The designated port is the port of a switch on a LAN that provides the
    > lowest cost path back to the root switch from that LAN. The root port
    > is the port on a switch that is "closest" to the root bridge. However,
    > aside form these active ports of the spanning tree, edge ports also
    > forward frames. Edge ports are ports through which no BPDUs travel,
    > but through which hosts are connected.
    >
    > Bert


    -------------------------------------------

    Can U please tell me , about that....
    Is in a layer 2 managed switch one can provide bothe of the types
    spanning tree capability like
    IEEE 802.1D and IEEE 802.1w (RSTP)
    means at one time which is want we open either one of spanning tree
    protocol.

    Tell me please....

  15. Re: About VLANs

    Albert Manfredi wrote:
    > On May 24, 2:02 pm, rober...@hushmail.com (Walter Roberson) wrote:
    >
    >> vicky wrote:
    >>> If layer 2 managed switch(s1) is in network and the by stp a loop free
    >>> graph is created , then in the switch s1(if it not root bridge) one
    >>> port is root port and then if two other ports from which the pc's are
    >>> connected are became designated ports or not ?

    >> http://en.wikipedia.org/wiki/Spanning_tree_protocol
    >>
    >> "Any active port that is not a root port or a designated port
    >> is a blocked port."
    >>
    >> We can read that backwards and thus see that any active port that
    >> is not a root port or a blocked port is a designated port.

    >
    > The designated port is the port of a switch on a LAN that provides the
    > lowest cost path back to the root switch from that LAN. The root port
    > is the port on a switch that is "closest" to the root bridge. However,
    > aside form these active ports of the spanning tree, edge ports also
    > forward frames. Edge ports are ports through which no BPDUs travel,


    With "spanning-tree portfast" enabled on edge ports, no BPDUs are seen.
    Without the command, BPDUs are seen on the edge ports. Confirmed with
    Wireshark.

    > but through which hosts are connected.
    >
    > Bert


    Best Regards,
    News Reader

  16. Re: About VLANs

    In article <3b35a30b-368a-477a-8c6b-92adf1d6d2e6@a32g2000prf.googlegroups.com>,
    vicky wrote:

    >Can U please tell me , about that....
    >Is in a layer 2 managed switch one can provide bothe of the types
    >spanning tree capability like
    >IEEE 802.1D and IEEE 802.1w (RSTP)
    >means at one time which is want we open either one of spanning tree
    >protocol.


    >Tell me please....


    Sorry, I was not able to understand your question.

    RSTP uses the same BPDU format as legacy 802.1D, but it marks the
    BPDU as being type 2, version 2.

    There are mechanisms in RSTP to detect and interoperate with
    legacy 802.1D.

    http://www.cisco.com/warp/public/473/146.html


  17. Re: About VLANs

    On May 25, 11:26*pm, rober...@hushmail.com (Walter Roberson) wrote:
    > In article <3b35a30b-368a-477a-8c6b-92adf1d6d...@a32g2000prf.googlegroups.com>,
    >
    > vicky * wrote:
    > >Can U please tell me , about that....
    > >Is in a layer 2 managed switch one can provide bothe of the types
    > >spanning tree capability like
    > >IEEE 802.1D and IEEE 802.1w (RSTP)
    > >means at one time which is want we open either one of spanning tree
    > >protocol.
    > >Tell me please....

    >
    > Sorry, I was not able to understand your question.
    >
    > RSTP uses the same BPDU format as legacy 802.1D, but it marks the
    > BPDU as being type 2, version 2.
    >
    > There are mechanisms in RSTP to detect and interoperate with
    > legacy 802.1D.
    >
    > http://www.cisco.com/warp/public/473/146.html


    ------------------------------------------

    Please Sirs,
    Please explain one point...

    In every switch controller , this is mentioned , "
    this controller supports 512 (any no. vary by controller) VLAN table
    entries.

    So please tell me wat it means , is it means 512
    different enteries support ....
    or it means 512 different vlans (out of 4094) ,
    support by controller .

  18. Re: About VLANs

    In article <8edd2cf2-5cce-4451-b1b7-e293afe9beac@j33g2000pri.googlegroups.com>,
    vicky wrote:

    > In every switch controller , this is mentioned , "
    >this controller supports 512 (any no. vary by controller)


    Did we settle the other day whether the differences were due
    not to the controller but rather FCC regulations placed
    upon model numbers?

    > VLAN table
    >entries.


    > So please tell me wat it means , is it means 512
    >different enteries support ....
    > or it means 512 different vlans (out of 4094) ,
    >support by controller .


    That will depend upon the software. On some systems it would mean
    that any 512 different VLANs could be -active- simultaneously,
    but on other systems, it would mean that any 512 different VLANs
    could be -configured- simultaneously, even if they are not active.

    For systems that support dynamic VLANs (learned through a protocol
    such as Cisco's VTP) then it will likely be -active- VLANs.

  19. Re: About VLANs

    On May 25, 9:49*am, News Reader wrote:
    > Albert Manfredi wrote:


    > > The designated port is the port of a switch on a LAN that provides the
    > > lowest cost path back to the root switch from that LAN. The root port
    > > is the port on a switch that is "closest" to the root bridge. However,
    > > aside form these active ports of the spanning tree, edge ports also
    > > forward frames. Edge ports are ports through which no BPDUs travel,

    >
    > With "spanning-tree portfast" enabled on edge ports, no BPDUs are seen.
    > Without the command, BPDUs are seen on the edge ports. Confirmed with
    > Wireshark.


    You're right. I should have been more careful in that definition.

    Switch ports can be administratively configured as edge ports, which
    allows RSTP to set them immediately to forwarding state and to cease
    transmitting any BPDUs. However, RSTP can also auto-detect edge ports,
    although 802.1D says this is an optional feature. It's in 14.8.2.1.3.

    So in this case, although a switch transmits BPDUs on an edge port, it
    will receive no BPDUs at that port. That's how it can automatically
    detect the existence of an edge port, and set it to forwarding as soon
    as the determination is made.

    Bert

  20. Re: About VLANs

    On May 27, 3:32*am, Albert Manfredi wrote:
    > On May 25, 9:49*am, News Reader wrote:
    >
    > > Albert Manfredi wrote:
    > > > The designated port is the port of a switch on a LAN that provides the
    > > > lowest cost path back to the root switch from that LAN. The root port
    > > > is the port on a switch that is "closest" to the root bridge. However,
    > > > aside form these active ports of the spanning tree, edge ports also
    > > > forward frames. Edge ports are ports through which no BPDUs travel,

    >
    > > With "spanning-tree portfast" enabled on edge ports, no BPDUs are seen.
    > > Without the command, BPDUs are seen on the edge ports. Confirmed with
    > > Wireshark.

    >
    > You're right. I should have been more careful in that definition.
    >
    > Switch ports can be administratively configured as edge ports, which
    > allows RSTP to set them immediately to forwarding state and to cease
    > transmitting any BPDUs. However, RSTP can also auto-detect edge ports,
    > although 802.1D says this is an optional feature. It's in 14.8.2.1.3.
    >
    > So in this case, although a switch transmits BPDUs on an edge port, it
    > will receive no BPDUs at that port. That's how it can automatically
    > detect the existence of an edge port, and set it to forwarding as soon
    > as the determination is made.
    >
    > Bert


    ---------------------------------------------

    Hi.
    One thing i want to ask .....
    Is it possible to support for multiple vlan in a single
    port.....

+ Reply to Thread
Page 1 of 2 1 2 LastLast