【实验 】open flow 协议分析实践

一、实验目的

1.能够运用 wireshark 对 OpenFlow 协议数据交互过程进行抓包;

2.能够借助包解析工具,分析与解释 OpenFlow协议的数据包交互过程与机制。

二、实验环境

1.下载虚拟机软件VMware Workstation

2.在虚拟机中安装Ubuntu,并完整安装Mininet;

三、实验要求

(一)基本要求

1.搭建下图所示拓扑,完成相关 IP 配置,并实现主机与主机之间的 IP 通信。用抓包软件获取控制器与交换机之间的通信数据包。

主机

IP地址

h1

192.168.0.101/24

h2

192.168.0.102/24

h3

192.168.0.103/24

h4

192.168.0.104/24

具体步骤

  1. 在终端输入以下命令并搭建拓扑
  1. 进入root用户下
  2. mininet/examples/miniedit.py

  1. 配置网段 (点击Edit -> Preferences

网段为实验要求中的192.168.0.0/24

OK保存退出

3.配置IP地址(点击File -> Export Level 2 Script

并保存为py文件

图中mxh22.py为自定义,根据学号可自行修改

使用vi或vim编辑mxh22.py文件中下图中的信息(行数为30-33)

查看行数的命令为  :set nu

修改完成后 :wq 保存退出

3.查看抓包结果,分析OpenFlow协议中交换机与控制器的消息交互过程,画出相关交互图或流程图。

新建终端

打开wireshark,命令为sudo wireshark

如果在用户视图则直接wireshark,下图为管理员视图操作

运行sudo wireshark命令,并选择any模式进行抓包

开启另一个终端,命令行运行mxh22.py文件,运行pingall

注:顺序不可乱,先打开wireshark进行抓包,再去执行Python文件,如出现找不到包的情况则关闭重新打开

(1)Hello

注:如果抓不到hello包,需要先打开wireshark抓包再运行python文件

  • 控制器6633端口(我最高能支持OpenFlow 1.0) ---> 交换机53202端口
  • 交换机53202端口(我最高能支持OpenFlow 1.5)--- 控制器6633端口于是双方建立连接,并使用OpenFlow 1.0
  • 2)Features Request / Set Conig
  • 控制器6633端口(我需要你的特征信息) ---> 交换机53202端口
  • 控制器6633端口(请按照我给你的flag和max bytes of packet进行配置) ---> 交换机53202端口
  • flag:指示交换机如何处理 IP 分片数据包
  • max bytes of packet:当交换机无法处理到达的数据包时,向控制器发送如何处理的最大字节数,本实验中控制器发送的值是0x0080,即128字节。
  • (3)Port_Status

  • 当交换机端口发生变化时,告知控制器相应的端口状态。(4)Features Reply
  • 交换机53202端口(这是我的特征信息,请查收)--- 控制器6633端口
  • (5)Packet_in

  • 有两种情况:
    • 交换机查找流表,发现没有匹配条目时
    • 有匹配条目但是对应的action是OUTPUT=CONTROLLER时
  • 交换机53202端口(有数据包进来,请指示)--- 控制器6633端口
  • 分析抓取的数据包,可以发现是因为交换机发现此时自己并没有匹配的流表(Reason: No matching flow (table-miss flow entry) (0)),所以要问控制器如何处理
  • (6)Flow_mod

  • 分析抓取的flow_mod数据包,控制器通过6633端口向交换机53202端口、交换机53214端口下发流表项,指导数据的转发处理
  • 分析抓取的flow_mod数据包,控制器通过6633端口向交换机53202端口、交换机53214端口下发流表项,指导数据的转发处理
  • (7)Packet_out

    控制器6633端口(请按照我给你的action进行处理) ---> 交换机53202端口

  • (8)交互图

  • 3.回答问题:交换机与控制器建立通信时是使用TCP协议还是UDP协议?
  • 分析wireshark抓包,可以看出交换器与控制器建立通信时使用的是TCP协议
  • (二)进阶要求

    1.将抓包结果对照OpenFlow源码,了解OpenFlow主要消息类型对应的数据结构定义。

    (1)Hello

    struct ofp_header {

        uint8_t version;    /* OFP_VERSION. */

        uint8_t type;       /* One of the OFPT_ constants. */

        uint16_t length;    /* Length including this ofp_header. */

        uint32_t xid;       /* Transaction id associated with this packet.

                               Replies use the same id as was in the request

                               to facilitate pairing. */

    };/* OFPT_HELLO.  This message has an empty body, but implementations must

     * ignore any data included in the body, to allow for future extensions. */struct ofp_hello {

        struct ofp_header header;

    };

  • 对应
  • (2)Features Request

  • 与Hello中的数据格式相同
  • (3)Set Conig

    /* Switch configuration. */struct ofp_switch_config {

        struct ofp_header header;

        uint16_t flags;             /* OFPC_* flags. */

        uint16_t miss_send_len;     /* Max bytes of new flow that datapath should

                                       send to the controller. */

    };

  • 对应
  • (4)Port_Status

    /* A physical port has changed in the datapath */struct ofp_port_status {

        struct ofp_header header;

        uint8_t reason;          /* One of OFPPR_*. */

        uint8_t pad[7];          /* Align to 64-bits. */

        struct ofp_phy_port desc;

    };

  • 对应
  • (5)Features Reply

    /* Description of a physical port */struct ofp_phy_port {

        uint16_t port_no;

        uint8_t hw_addr[OFP_ETH_ALEN];

        char name[OFP_MAX_PORT_NAME_LEN]; /* Null-terminated */

        uint32_t config;        /* Bitmap of OFPPC_* flags. */

        uint32_t state;         /* Bitmap of OFPPS_* flags. */

        /* Bitmaps of OFPPF_* that describe features.  All bits zeroed if

         * unsupported or unavailable. */

        uint32_t curr;          /* Current features. */

        uint32_t advertised;    /* Features being advertised by the port. */

        uint32_t supported;     /* Features supported by the port. */

        uint32_t peer;          /* Features advertised by peer. */

    };/* Switch features. */struct ofp_switch_features {

        struct ofp_header header;

        uint64_t datapath_id;   /* Datapath unique ID.  The lower 48-bits are for

                                   a MAC address, while the upper 16-bits are

                                   implementer-defined. */

        uint32_t n_buffers;     /* Max packets buffered at once. */

        uint8_t n_tables;       /* Number of tables supported by datapath. */

        uint8_t pad[3];         /* Align to 64-bits. */

        /* Features. */

        uint32_t capabilities;  /* Bitmap of support "ofp_capabilities". */

        uint32_t actions;       /* Bitmap of supported "ofp_action_type"s. */

        /* Port info.*/

        struct ofp_phy_port ports[0];  /* Port definitions.  The number of ports

                                          is inferred from the length field in

                                          the header. */

    };

    对应

  • (6)Packet_in

    /* Why is this packet being sent to the controller? */enum ofp_packet_in_reason {

        OFPR_NO_MATCH,          /* No matching flow. */

        OFPR_ACTION             /* Action explicitly output to controller. */

    };

    /* Packet received on port (datapath -> controller). */struct ofp_packet_in {

        struct ofp_header header;

        uint32_t buffer_id;     /* ID assigned by datapath. */

        uint16_t total_len;     /* Full length of frame. */

        uint16_t in_port;       /* Port on which frame was received. */

        uint8_t reason;         /* Reason packet is being sent (one of OFPR_*) */

        uint8_t pad;

        uint8_t data[0];        /* Ethernet frame, halfway through 32-bit word,

                                   so the IP header is 32-bit aligned.  The

                                   amount of data is inferred from the length

                                   field in the header.  Because of padding,

                                   offsetof(struct ofp_packet_in, data) ==

                                   sizeof(struct ofp_packet_in) - 2. */

    };

  • (7)Flow_mod

    /* Fields to match against flows */struct ofp_match {

        uint32_t wildcards;        /* Wildcard fields. */

        uint16_t in_port;          /* Input switch port. */

        uint8_t dl_src[OFP_ETH_ALEN]; /* Ethernet source address. */

        uint8_t dl_dst[OFP_ETH_ALEN]; /* Ethernet destination address. */

        uint16_t dl_vlan;          /* Input VLAN id. */

        uint8_t dl_vlan_pcp;       /* Input VLAN priority. */

        uint8_t pad1[1];           /* Align to 64-bits */

        uint16_t dl_type;          /* Ethernet frame type. */

        uint8_t nw_tos;            /* IP ToS (actually DSCP field, 6 bits). */

        uint8_t nw_proto;          /* IP protocol or lower 8 bits of

                                    * ARP opcode. */

        uint8_t pad2[2];           /* Align to 64-bits */

        uint32_t nw_src;           /* IP source address. */

        uint32_t nw_dst;           /* IP destination address. */

        uint16_t tp_src;           /* TCP/UDP source port. */

        uint16_t tp_dst;           /* TCP/UDP destination port. */

    };/* Flow setup and teardown (controller -> datapath). */struct ofp_flow_mod {

        struct ofp_header header;

        struct ofp_match match;      /* Fields to match */

        uint64_t cookie;             /* Opaque controller-issued identifier. */

        /* Flow actions. */

        uint16_t command;             /* One of OFPFC_*. */

        uint16_t idle_timeout;        /* Idle time before discarding (seconds). */

        uint16_t hard_timeout;        /* Max time before discarding (seconds). */

        uint16_t priority;            /* Priority level of flow entry. */

        uint32_t buffer_id;           /* Buffered packet to apply to (or -1).

                                         Not meaningful for OFPFC_DELETE*. */

        uint16_t out_port;            /* For OFPFC_DELETE* commands, require

                                         matching entries to include this as an

                                         output port.  A value of OFPP_NONE

                                         indicates no restriction. */

        uint16_t flags;               /* One of OFPFF_*. */

        struct ofp_action_header actions[0]; /* The action length is inferred

                                                from the length field in the

                                                header. */

    };

    对应

  • (8)Packet_out

    /* Action header that is common to all actions.  The length includes the

     * header and any padding used to make the action 64-bit aligned.

     * NB: The length of an action *must* always be a multiple of eight. */struct ofp_action_header {

        uint16_t type;                  /* One of OFPAT_*. */

        uint16_t len;                   /* Length of action, including this

                                           header.  This is the length of action,

                                           including any padding to make it

                                           64-bit aligned. */

        uint8_t pad[4];

    };OFP_ASSERT(sizeof(struct ofp_action_header) == 8);

    /* Send packet (controller -> datapath). */struct ofp_packet_out {

        struct ofp_header header;

        uint32_t buffer_id;           /* ID assigned by datapath (-1 if none). */

        uint16_t in_port;             /* Packet's input port (OFPP_NONE if none). */

        uint16_t actions_len;         /* Size of action array in bytes. */

        struct ofp_action_header actions[0]; /* Actions. */

        /* uint8_t data[0]; */        /* Packet data.  The length is inferred

                                         from the length field in the header.

                                         (Only meaningful if buffer_id == -1.) */

    };

  • 四、个人总结

  • 实验难度:适中
  • 实验过程遇到的困难及解决办法:
    • 打开wireshark后创建拓扑,过滤出Openflow数据包,发现一些数据包没有找到,重新看了文件,并上网查阅资料后发现要先打开wireshark再执行Python文件,进行pingall即可找到。
  • 个人感想:本次实验是结课实验,实验难度相对于平常练习会有难度一些,需要建立拓扑后在wireshark中操作,首先要能够熟练使用wireshark,其次就是找包的时候数据很多,容易错过,需要很细心。通过阅读openflow的源码,能理清楚其中的数据结构,并和实际抓到的包的详细信息对应起来,从而理解openflow协议的数据包交互过程。除此之外,对wireshark中过滤器的使用也有了更深刻的了解,对openflow协议有了更深刻的认识,整理实验报告的过程中能够加深对整个交互过程的理解,截图较多,整理起来需要足够的耐心。实验中遇到的问题都得到了解决,对mininet的一些命令行操作也更加的熟练。
  • 本文为学院sdn结课作业,仅供参考,如有问题欢迎留言指正。
  • 35
    点赞
  • 12
    收藏
    觉得还不错? 一键收藏
  • 0
    评论
评论
添加红包

请填写红包祝福语或标题

红包个数最小为10个

红包金额最低5元

当前余额3.43前往充值 >
需支付:10.00
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明:

1.余额是钱包充值的虚拟货币,按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载,可以购买VIP、付费专栏及课程。

余额充值