Enabling VLAN tag capture for Broadcom Ethernet Adapter

There is a registry key under HKLM\SYSTEM\CurrentControlSet\control\Class\{4D36E972-E325-11CE-BFC1-08002bE10318}\00xx that can be set to  cause the driver and chip not to strip the 802.1Q headers. In order to set that key, you need to find the right instance of the driver in Registry Editor and set that key for it. You  can do this by doing following:
1.    Run the Registry Editor (regedit)
2.    Hit CTRL+Home to go to the top of the registry
3.    Hit F3 to bring up a search
4.     Search for  TxCoalescingTicks - actually a search for txcoal will do

This should take you to something like:
HKLM\SYSTEM\CurrentControlSet\control\Class\{4D36E972-E325-11CE-BFC1-08002bE10318}\00xx

5.    Right-click on a bit of white space near the TxCoalescingTicks field.
6.    Enter "PreserveVlanInfoInRxPacket" –
a.    Then highlight the text PreserveVlanInfoInRxPacket and copy it to the clipboard – you’ll need it again later.
7.    Give the newly created field (PreserveVlanInfoInRxPacket) the value "1".
8.    Hit F3 to repeat the search – goto step 5 again.  Repeat until you reach the beginning again.

9.    Reboot the PC

REF:www.cisco.com/en/US/prod/voicesw/custcosw/ps5693/ps14/prod_system_requirements0900aecd800e3149.pdf
Also: http://wiki.wireshark.org/CaptureSetup/VLAN
--------------------------------------------------------------------------------------------------------------
某台伺服器上要跑一個軟體用pcap抓封包，系統用的是Windows，但是Windows下的驅動過濾了8021q的協議header，導致軟體不能正常工作。 Google了一下， Wireshark的Wiki說要改註冊表
但是經過測試，他給出的方法不能用，至少在現在的驅動上不行​​。 文章中提示的TxCoalescingTicks在註冊表裡面根本找不到!!

在驅動文件bxvbda.sys裡找到了“keep_vlan_tag”的字樣，於是順藤摸瓜發現了這個位置：
HKEY_LOCAL_MACHINE\SYSTEM\ControlSet001\Control\Class\{4D36E97D-E325-11CE-BFC1-08002BE10318}\0051 最後這個數字可能不太相同，不過可以搜索“*ReceiveBuffers”試試看。
找到這個鍵值以後就好說了，添加一個REG_SZ 名稱為keep_vlan_tag 值為1的項，重啟。 就可以在Wireshark裡抓到VLAN TAG了。

就是這樣。 驅動版本是5.2.14.0 網卡是Broadcom BCM5708C NetXtreme II

ARP Proxy

        ARP代理就是我們通常說的ARP Proxy，它實際是通過使用一個主機(通常為router)來作為指定的設備對另一設備作出ARP請求的response。

    

    假設有網絡拓撲如上圖所示，PC1/PC2組成的局域網1通過路由器與PC3/PC4組成的局域網2連接在一起。現在PC1要給PC4發一份Packet，具體過程如下(假定Router已開啟ARP Proxy功能)：

        1).PC1首先要判斷PC4是否與自己在同一個網絡。因為PC1的IP掩碼為16位，所有IP地址在128.18.0.1與128.18.255.255之間(包含128.18.0.1和128.18.255.255)的PC都被認為與PC1屬於同一個網絡。

       2).PC4的IP地址為128.18.170.11，在上述IP地址範圍之內，因此PC1認為有可直達PC4的路由，因此，PC1會直接發送ARP Request Packet，請求PC4的MAC地址。ARP Request packet的源IP為PC1的IP，源MAC為PC1的MAC，目的IP為PC4的IP，目的MAC為全0(表示不知道對方MAC)。

       3).ARP Request packet在以太網中被封裝成二層 packet時，幀頭中的源MAC為PC1的MAC地址，目的MAC為全1的廣播地址。

       4).PC4的IP地址為128.18.170.11，IP地址掩碼為24位，所以在PC4看來，IP地址為128.18.178.100的PC1，與它不在同一個網絡，所以網絡中，沒有從PC1到PC4的直達路由。ARP請求 packet從PC1發出後，會由Router接收，Router查看自己的路由表，發現從自己的另外一個端口可以到達PC4，於是代替PC4給PC1回復一個ARP Reply unicast packet。

   雖然Router能請求到PC4的MAC，但是它不會將PC4的MAC直接告訴PC1。而是將自身的MAC提供給PC1，告訴PC1，可以接收它的數據報，在PC1看來，它收到的MAC就是PC4的MAC(其實是PC4的MAC)。

      5).PC1發出數據 packet，源MAC和IP都是自己的MAC和IP，目的IP是PC1的IP地址，目的MAC是Router的MAC。Router收到來自PC1的 packet後，會將目的MAC更換成PC4的MAC，目的IP不變，仍然是PC4的IP地址，源IP也不變，仍然是PC1的IP地址，而源MAC則更換成自己與PC4相連的那個端口MAC地址。然後把重新封裝後的 packet發給PC4，從而完成ARP代理功能。

   備註：Router的高速buffer中會存放與Router的各出接口相連的PC MAC與IP信息，若事先沒有，Router會定期發出ARP廣播 packet請求各個PC的MAC。

      ARP代理的前提條件：

      1).網絡上的主機沒有設置Default Gateway；

      2).網絡上的主機不支持動態路由協議；

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

Proxy ARP with Linux

www.internetsolver.com | Other Linux Documents 

---

By David Weis with Internet Solver, LLC

AKA: firewalling a subnet with a single IP address

Why and How

Proxy ARP basically means that a particular machine (such as a firewall) will respond to ARP requests for hosts other than itself. This can be used to make a firewall mostly disappear from the machines on a network.

For an example, say you have a /28 subnet from your ISP that is routed through a Cisco router. Your router appears at the IP of x.x.x.97 with a network address of x.x.x.96 and a broadcast address of x.x.x.111. This leaves a usable chunk of 14 addresses for your hosts.

If you wanted to firewall these hosts from the internet without using proxy arp, you would need to either subnet your addresses and lose two more addresses for the new network and broadcast, plus half of your remaining IP's would be in the non-firewalled half.

Another method would be to have the firewall do port forwarding between all of the addresses to non-routed IP's (192.168.x.x) for your servers. Done properly, this would be okay. It isn't as transparent and may break some protocols like active FTP unless the firewall will compensate.

By using Proxy ARP, you can set up your machines in a DMZ to separate them from your client machines. This is also the least invasive method to set up, since you can keep the same IP's on all of the servers as you had when things weren't firewalled.

To set up the network, you will need a machine with at least two NIC's, three if you want to also masquerade client machines for outgoing access. Some variations on how I configured this are surely possible, but this is how I know how to do it and know that it does work. You will need a 2.4 series kernel, theiproute2 utility, and a recent iptables userspace program.

Doing it

You will need to set up the machine with the software mentioned above. Be sure to compile netfilter into the kernel by selecting yes for "Network Packet Filtering" under the "Networking Options" section. All of the pieces associated with netfilter are listed under "Netfilter Configuration" further down the list. It is probably easier to build each of the options into the kernel than use modules, there should be +- 22 choices to turn on.

Build your kernel, install it, and boot to make sure it functions. You should see some lines like

ip_conntrack (2046 buckets, 16368 max)
ip_tables: (c)2000 Netfilter core team

in the kernel boot messages (use dmesg if they went by too fast).

After you have your kernel running, build and install iproute2 and iptables. Instructions for doing so are in the packages themselves. Your distribution may have included them, but they are probably older ones. Grab the newest ones to make sure you have the right versions. Test them by running ip and iptablesand see if they print something. Don't continue until they do.

After you have the above steps done, you will need to configure your network cards. This step should be done off of the network since you may end up with some conflicting addresses. Give two NIC's identical IP addresses, subnet masks, and gateways. The IP you choose needs to be an unused address on your network. In my case, I used x.x.x.98, since my router is at x.x.x.97. You could actually use about any address on the wire that isn't in use.

There is an example configuration available for download below that uses three NIC's, one for an internal 192.168.x.x network for client machines. If you want to create the file yourself, here are the steps. First off, enable Proxy ARP in the kernel. We'll assume your cards are eth0 and eth1.

echo 1 > /proc/sys/net/ipv4/conf/eth0/proxy_arp
echo 1 > /proc/sys/net/ipv4/conf/eth1/proxy_arp

Next, you will tell the kernel how to get to each of the two networks. When you are done, there will be a crossover cable from one NIC on your firewall to your router and the other NIC will be connected to the port on your hub/switch that the router was previously plugged in to. The kernel now thinks that the same machines are on each wire, which definitely won't work. We'll set it straight with these commands:

ip route del x.x.x.96/28 dev eth0
ip route del x.x.x.96/28 dev eth1
ip route add x.x.x.97 dev eth0
ip route add x.x.x.96/28 dev eth1

What we've done is first say that no network is reachable via either NIC, then say that the router (x.x.x.97) is connected to eth0 and the rest of your servers are hooked to eth1. I would suggest using this setup (router on eth0, servers on eth1) since the downloadable example assumes that.

We'll now use the ip command to verify our work. Running ip addr should yield something like this:

1: lo:  mtu 3904 qdisc noqueue 
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
    inet 127.0.0.1/8 brd 127.255.255.255 scope host lo
2: eth0:  mtu 1500 qdisc pfifo_fast qlen 100
    link/ether 00:90:27:3f:23:9c brd ff:ff:ff:ff:ff:ff
    inet x.x.x.98/28 brd x.x.x.111 scope global eth0
3: eth1:  mtu 1500 qdisc pfifo_fast qlen 100
    link/ether 00:90:27:3f:4d:6c brd ff:ff:ff:ff:ff:ff
    inet x.x.x.98/28 brd x.x.x.111 scope global eth1

and ip route should show this

x.x.x.98 dev eth0  scope link 
x.x.x.98 dev eth1  scope link 
x.x.x.97 dev eth0  scope link 
x.x.x.96/28 dev eth1  scope link 
127.0.0.0/8 dev lo  scope link 
default via x.x.x.97 dev eth0 

After the routing tables and IP addresses look okay, you can turn on IP forwarding in the kernel by doing this:

echo 1 > /proc/sys/net/ipv4/ip_forward

At this point, you could take your new firewall, hook up the ethernet cables correctly, and have identical functionality (hopefully) to what you have now with no firewalling. Of course, this machine doesn't really do any firewalling, grab the IP tables howto from the link below and set up your tables. You can also use the example script provided below.

Troubleshooting

First, verify that you can ping all of the machines from the firewall, including the router. The main problem you can run into is that the router will probably cache the ARP entries for the machines for quite a while. The easiest way to solve that is to kill the power to your router and plug it back in. Try to ping again.

Let me know if you have any problems, there is a link to my email address at the bottom of the page, along with a place that you can post questions directly on this page.

Downloads

• Example incorporating firewall and proxy arp

Other Sources of Information

• Linux 2.4 Advanced Routing & Traffic Control
• Netfilter Home

Other Stuff

This is a reply to a question on comp.os.linux.networking where I first explained how to do some of this. Above is a cleaned up response. It's here as a source for my copy and paste. :-)

What you want is called proxy-arp. You would set the two NICS to both
have the same real IP address, such as .55 in your second diagram. Then
you do this
echo 1 > /proc/sys/net/ipv4/conf/eth0/proxy_arp
echo 1 > /proc/sys/net/ipv4/conf/eth1/proxy_arp

You will need to use the 2.4 kernel and get iproute2, then try these
commands

ip route del x.x.x.32/27 dev eth0
ip route del x.x.x.32/27 dev eth1
ip route add x.x.x.33 dev eth0
ip route add x.x.x.32/27 dev eth1

This tells the kernel that the router is the only thing reachable via
eth0 and the rest of the network is on eth1 (change as needed). This
will make the linux machine respond for the machines on either side.
Finally, do this

echo 1 > /proc/sys/net/ipv4/ip_forward

to enable routing. Either wait a while for the arp caches to expire or
restart your router. At that point you should be able to get back and
forth between the router and the other servers on the network. If you
look at your arp cache on a server, it will show the mac address of the
router as the mac address of eth1 on your linux firewall.

After you have this layer working, you can add your rules.

Good luck!
dave

"Norman D. Megill" wrote:
>
> I want to set up an IP filter in the following setup.  Right now every
> machine is connected directly to the internet with static (real) IP's in
> subnet X.Y.Z.32/27.  X.Y.Z.35-X.Y.Z.54 are DHCP'd with an NT server and
> the others are hard-configured in various machines.  There is currently
> no firewalling.
>
> Current setup:
>                                                          internet
>    other X.Y.Z.32/27 ------------------------------------ISP feed
>      NT/W98 machines      |                            X.Y.Z.32/27
>                           |                            gw=X.Y.Z.33
>                           |
>                           |
>                     NT web server
>                       X.Y.Z.60
>
> The management of the Windows-only office is more paranoid about Linux
> than about security, and furthermore wants no work disruption or risk
> caused by changing IP setups on various machines.  To demo an initial
> firewall, it must be "transparent" so that if it has a problem the
> internal network cable can be instantly unplugged from Linux and hooked
> back to the ISP feed to restore the current configuration.  My idea is
> that once we get it working we can slowly start to tighten things up,
> move the machines to masq'd/forwarded 192.168.x.x's, etc.
>
> Without getting into the security problems that will be initially
> present with this "firewall", if I don't get a "transparent" mode to
> work there is not going to be any firewall at all, or at best it might
> be an NT machine for management comfort.  The current setup has been in
> place for a few years with no apparent security problems, and "if it
> ain't broke don't fix it".
>
> Because both sides of the firewall are on the same subnet, I have not
> been able to get it to work after experimenting with all kinds of
> routing and ip setups.  I am starting to look at bridging, which may be
> the only solution, but the fact that it (I think) sets the NICs to
> promiscuous mode seems against the spirit of the firewall.  But before I
> give up I'll ask about it here.
>
> I am using RedHat 7.0 with kernel 2.2.16 and RH security updates.
>
> Experiment 1:
>                                   eth1       eth0
>    other X.Y.Z.32/27 ----------------- Linux -----------ISP feed
>      NT/W98 machines      |     X.Y.Z.55   X.Y.Z.56    X.Y.Z.32/27
>                           |                            gw=X.Y.Z.33
>                           |
>                           |
>                     NT web server
>                       X.Y.Z.60
>
> Experiment #1 doesn't work because a packet from the internet to
> the X.Y.Z.60 server makes the ISP gateway think that X.Y.Z.60 is
> on its local cable, and it sits there doing "arp who-has" for X.Y.Z.60.
> Of course Linux never responds because it only looks for packets
> to X.Y.Z.56.
>
> Experiment 2:  Same as Experiment 1 but added X.Y.Z.33 to eth1:0 and
> added X.Y.Z.60 to eth0:0
>
> Experiment #2 allows Linux to see the gateway arp's, but it seems to
> think the packets are for itself and I have found no way to
> transparently forward them to the eth1 side.
>
> Now, it seems that what I want should be theoretical possible, but I
> can't figure out how to make Linux can do it (without promiscuous
> bridging).  Adding to my puzzlement is that Figure 3-2 on
>
>    http://www.bb-zone.com/FWHowTo/chapter3.html
>
> shows the exact setup I want.  The author seems to say that the
> FW_ROUTER variable just bypasses the spoofing filter, but that would not
> solve the problem (since for now I have ipchains completely open).  He
> references "the SuSE firewall script," but there is no FW_ROUTER
> variable in SuSE's firewals-2.6-33.rpm if that's the script he means.
>
> Thanks for any advice.

Windows 7-Intel 網卡支援 VLAN Tag

前言

因為需要測試該 Switch 所出來的 Tag Port 到底能否正常運作，在 Windows 7 中所使用的 Intel 82566DM-2 網卡可以在更新驅動及安裝進階工具後即可測試 VLAN Tag 功能。

實作環境

• Windows 7 旗艦版 x64
• Intel 82566DM-2 Gigabit Network Connection

安裝及設定

1. 至 [Intel 下載中心 - Intel® 82566 Gigabit 乙太網路 PHY] 下載，因為本次實作採用的是 Windows 7 x64 所以下載 PROWinx64.exe。
2. 安裝後至【裝置管理員】 查看網卡內容，會多了一堆頁籤可以選擇，其中【VLAN】頁籤就是本次實作的主角。
3. 點選【VLAN】後按下【新增】，在彈出視窗中填入【VLAN ID (即 Tag VID)】而 VLAN 名稱會自已填好，例如 VLAN ID 為 110 時則 VLAN 名稱會為 VLAN110。
4. 設定完成後按下【確定】，系統會自動新增一片網卡，網卡的結尾名稱就是剛才的 VLAN 名稱，例如 【Intel 82566DM-2 Gigabit Network Connection - VLAN:VLAN110】。
5. 此時便可至該網卡內設定屬於該 Tag 網段內的固定 IP 位址即可驗證 Tag Port 設定是否正確。

Windows 7-Realtek 網卡支援 VLAN Tag

前言

因為需要測試該 Switch 所出來的 Tag Port 到底能否正常運作，在 Windows 7 中所使用的 Realtek RTL8169 網卡可以在更新驅動及下載 Diagnostic 工具後即可測試。

實作環境

• Windows 7 專業版
• Realtek RTL8169 網卡

安裝及設定

1. 下載 [REALTEK - Windows Diagnostic Program] 工具後進行安裝
2. 安裝後查看 Realtek 網卡內容應該會多了三個驅動，分別是 【Realtek Teaming、VLAN、NDIS Protocol Driver】
3. 開啟【Realtek Ethernet Diagnostic Utility】後選擇【虛擬區域網路】，按下【增加】，填入【VLAN ID (即 Tag VID)】
4. 當增加後系統會新增一片網卡，在該網卡內設定屬於該 Tag 網段內的固定 IP 位址即可驗證 Tag Port 設定是否正確

  

參考資料 不自量力的 Weithenn - Windows 7-Realtek 網卡支援 VLAN Tag

DHCPv4 Option Summary

Option Name Code ISC Option Name Microsoft Option Name Descritpion All subnets are local 27 all-subnets-local All subnets are local 0 specifies that the client should assume that some IP subnets have smaller MTUs; 1 specifies that all subnets have same MTU as subnet to which the client is connected. ARP cache timeout 35 arp-cache-timeout ARP cache Timeout Timeout (in seconds) for ARP cache entries Autoconfiguration 116 autoconfiguration 0 specifies that the client should not perform link-local IP address autoconfiguration; 1 specifies that the client should perform autoconfiguration Boot file size 13 boot-size Boot file size Size of the client bootfile in 512-byte blocks Boot file name 67 bootfile-name Boot file name Name of the bootfile to use when the file field is used to carry options Broadcast address 28 Broadcast-address Broadcast address The broadcast address for the subnet to which the client is attached. Client identifier 61 Dhcp-client-identifier Client’s unique identifier Cookie server 8 Cookie-servers Cookie servers Cookie servers Default IP TTL 23 Default-ip-ttl Default Time-to-Live Default TTL client should use for outgoing  datagrams Dhcp authentication option 90 authentication Option used to authenticate DHCP packets transmitted by client and servers DHCP message type 53 Dhcp-message-type Identifies the type of the DHCP message Domain name servers 6 Domain-name-servers Dns server List of DNS server IP addresses Domain name 15 Domain-name Domain name Default name for DNS name resolution. End  255 Option-end Indicates end of options in field Ethernet encapsulation 36 Ieee802-3-encapsulation Ethernet encapsulation 0 specifies Ethernet version 2 encapsulation; 1 specifies IEEE 802.3 Extensions path 18 Extensions-path Extensions path The name of the file containing additional options to be interpreted according to RFC2132 format Finger server 73 Finger-server Finger servers List of finger server IP addresses Host name 12 Host-name Host name Client hostname Impress server 10 Impress-servers Impress servers Imagen impress printer servers Interface MTU 26 Interface-mtu MTU option Value of MTU that the client should use for this interface Length of DHCP lease 51 Dhcp-lease-time Lease duration (expressed in seconds) for the assigned IP address IP forwarding 19 Ip-forwarding IP Layer Forwarding 0 specifies that datagram forwarding between interfaces is to be disabled; 1 specifies that forwarding is to be enabled IRC Server 74 Irc-server Internet relay chat (IRC) servers List of internet relay chat server IP address Log server 7 Log-servers Log servers MIT-LCS log servers LPR server 9 Lpr-servers LPR Servers LPDP servers Mask supplier 30 Mask-supplier Mask Supplier Option 0 specifies that the client should not respond to ICMP subnet mask request messages; 1 specifies that the client should respond. Maximum datagram 22 Max-dgram-reassembly Max DG Reassembly Size Maximum size of datagram that reassembly the client should expect to reassemble Maximum DHCP 57 Dhcp-max-message-size Maximum DHCP message size message size accepted by the client Merit dump file 14 Merit-dump Merit dump file Name of file for memory dump Message 56 Dhcp-message Message from the server to be displayed to the user by the client Mobile IP home agents 68 Mobile-ip-home-agents Mobile IP home agents List of mobile IP home agents Name server 5 Ien116-name-servers Name Servers IEN 116 name servers Name Service 117 Name-service-search-order Ordered list of name service types that the client should use initial netware directory server (NDS) for client NDS Server 85 Nds-servers List of NDS servers NDS tree name 86 Nds-tree-name NDS tree name for the client to use. NetBIOS over TCP/IP 45 Netbios-dd-server NetBIOS over TCP/IP NBDD List of NetBIOS Datagram data-Distribution(NBDD) servers NetBIOS over TCP/IP name server 44 Netbios-name-servers WINS/NBNS servers List of NetBIOS Name Server(NBNS) IP addresses.

Option Name Code ISC Option Name Microsoft Option Name Descritpion NetBIOS over TCP/IP type 46 Netbios-node-type WINS/NBT type Client’s NetBIOS over TCP/IP node NetBIOS over TCP/IP scope 47 Netbios-scope NetBIOS scope ID Client’s NetBIOS over TCP/IP scope NetWare/IP domain 62 Nwip-domain Name of NetWare/IP domain for client NetWare/IP information 63 Nwip.nsq-broadcast nwip.preferred-dss nwip.nearest-nwip-server nwip.autoretries nwip.autoretry-secs nwip.nwip-1-1 nwip.primary-dss NetWare/IP parameters NIS domain Name 40 Nis-domain NIS Domain Name Client’s Network Information Service (NIS) domain NIS Servers 41 Nis-servers NIS Servers List of NIS servers NIS+ domain Name 64 Nisplus-domain NIS+ Domain Name Client’s Network information Server+(NIS+) domain NIS+ servers 65 Nisplus-servers Nis+ servers List of NIS+ server IP addresses NNTP server 71 nntp-server Network News Transport Protocol (NNTP) servers List of NNTP servers Non-local source routing 20 Non-local-source-routing Non-local source routing 0 specifies that forwarding of datagrams with nonlocal source routes is to be disallowed; 1 specifies that forwarding of sub datagrams is to be allowed NTP servers 42 Ntp-servers NTP server Lsit of network Time Protocol (NTP) servers Option overload 52 Dhcp-option-overload Specifies whether file and sname fields are used to carry options Pad 0 Pad Carries no data Parameters request list 55 Dhcp-parameter-request-list List of options requested by the client Path MTU aging 24 Path-mtu-aging-seconds Path MTU Aging Timeout Timeout(in seconds) for aging PMTU values Path MTU plateau table 25 Path-mtu-plateau-table Path MTU Plateau Table List of MTU sizes for PMTU discovery. Perform mask discovery 29 Perform-mask-discovery Perform mask discovery 0 specifies that the client should not perform ICMP subnet mask discovery; 1 specifies that the client should perform subnet mask discovery Policy filter 21 Policy-filter Policy Filter Masks List of policy filters for nonlocal source routing. POP3 server 70 Pop-server (pop3) server IP Post office protocol (pop3) servers List of post office protocol 3 addresses Requested IP address 50 Dhcp-requested-address IP address requested by the client. Relay agent information the relay option 82 Relay-agent-information Additional information supplied by agent Resource location servers 11 Resource-location-servers servers Resource location servers Resource location protocol Root path 17 Root-path Root path Name of the client’s root disk. Router solicitation for router 32 Router-solicitation-address Router solicitation The address the client should use address discovery Routers 3 routers router IP addresses of default routers to use on the client’s network segment Server identifier 54 Dhcp-server-identifier The IP address of the DHCP server SLP Directory Agent 78 Slp-directory-agent The service location Protocol service scope SMTP server 69 Smtp-server Simple mail transport protocol (SMTP) servers List of simple mail transport protocol (SMTP) servers Static route 33 Static-routes Static route option List of static routes; each static route includes a destination Streettalk directory assistance server 76 Streettalk-directory-assistance -server Streettalk directory assistance (STDA) servers List of STDA servers Subnet mask 1 Subnet-mask The subnet mask for the network segment to which the client is connected. Subnet selection option 118 Subnet-selection The subnet number to use in selecting the client IP address. Swap server 16 Swap-server Swap server The swap server for the client T1 58 Dhcp-renewal-time The time, in seconds ,until the client should renew the lease on the assigned address. T2 59 Dhcp-rebinding-time The time, in seconds, until the client should rebind the lease on the assigned address. TCP default TTL 37 Default-tcp-ttl Default TTL option The default TTL for TCP segments TCP keepalive garbage 39 Tcp-keepalive-garbage Keepalive Garbage 0 specifies that the client should not send a garbage byte with a TCP keepalive message; 1 specifies that the client should send a garbage byte TCP keepalive interval 38 Tcp-keepalive-interval Keepalive interval The time, in seconds, to wait before sending a keepalive message on a TCP connection TFTP server name 66 Tftp-server-name Boot server Host name The name of the TFTP serve r to use when the Sname field is used to carry options Time offset 2 Time-offset Time offset The time offset, in seconds, from coordinated universal time (UTC) for the client’s network segment Time server 4 Time-servers Time server The time server’s IP addresses Trailer 34 Trailer-encapsulation Trailer encapsulation 0 specifies that the client should not negotiate the use of trailers through ARP; 1 specifies that the client should negotiate the use of trailers. UAP servers 98 Uap-servers List of URLs specifying User authentication protocol servers User class 77 User class User class information Vendor class identifier 60 Vendor-class-identifier Vendor identification Vendor specific information 43 Vendor-encapsulated-options Vendor specific info Vendor-specific information WWW server 72 www-server World wide web(WWW) servers List of world web server IP addresses X window system display managers 49 x-display-managers X window system display List of X window system display manager X window system font server 48 Font-servers X window system font List of x window system font servers