因此，大学 IT 安全团队和我一直在不停地讨论这个问题...任何人对此有任何想法：
我最近为我的实验室设置了一个小型文件服务器，在专用计算机上运行 Debian 8.6（Intel Avoton C2550 处理器——如果需要，很乐意提供更多硬件信息，但我认为没有必要）。 Debian安装没有任何问题，当时我还安装了Samba、NTP、ZFS和python。一切似乎运行良好，所以我让它在实验室的角落里运行了几个星期。
大约两周前，我收到了 IT 团队发来的一封电子邮件，其中称我的服务器已“受损”，并且容易被用于 NTP 放大/DDoS 攻击（使用 CVE-2013-5211 的 NTP 放大攻击，如中所述）https://www.us-cert.gov/ncas/alerts/TA14-013A）。他们指出的迹象是端口 123 上有大量 NTPv2 流量。奇怪的是，他们识别出的来自 ( *.*.*.233) 的 IP 地址与我的服务器配置并通过 ifconfig ( *.*.*.77) 报告的 IP 地址不同。尽管如此，一些基本的故障排除表明我的计算机确实在端口 123 上生成此流量（如 tcpdump 所示）。
这就是奇怪的地方开始的地方。我首先执行了针对 CVE-2013-5211 推荐的“修复”（既更新了 NTP 4.2.7 版本，又禁用了 monlist 功能）。两者都没有阻止交通流量。然后我尝试通过 IP 表阻止 UDP 123 端口：
$ /sbin/iptables -A INPUT -o eth0 -p udp --destination-port 123 -j DROP
$ /sbin/iptables -A OUTPUT -o eth0 -p udp --destination-port 123 -j DROP
但这也对交通没有影响。我最终尝试从系统中清除 NTP，但这对流量也没有影响。截至今天下午，nmap 仍在报告：
Starting Nmap 5.51 ( http://nmap.org ) at 2016-12-19 16:15 EST
Nmap scan report for *.233
Host is up (0.0010s latency).
PORT STATE SERVICE
123/udp open ntp
| ntp-monlist:
| Public Servers (2)
| 50.116.52.97 132.163.4.101
| Public Clients (39)
| 54.90.159.15 185.35.62.119 185.35.62.233 185.35.63.86
| 54.197.89.98 185.35.62.142 185.35.62.250 185.35.63.108
| 128.197.24.176 185.35.62.144 185.35.62.251 185.35.63.128
| 180.97.106.37 185.35.62.152 185.35.63.15 185.35.63.145
| 185.35.62.27 185.35.62.159 185.35.63.27 185.35.63.146
| 185.35.62.52 185.35.62.176 185.35.63.30 185.35.63.167
| 185.35.62.65 185.35.62.186 185.35.63.34 185.35.63.180
| 185.35.62.97 185.35.62.194 185.35.63.38 185.35.63.183
| 185.35.62.106 185.35.62.209 185.35.63.39 185.35.63.185
|_ 185.35.62.117 185.35.62.212 185.35.63.43
这很奇怪，因为 NTP 已经从系统中清除了几周。
在这条路上遇到了死胡同之后，我开始思考整个 IP 地址不匹配问题。我的计算机似乎同时位于 *.233 和 *.77 IP 上（通过在连接以太网电缆的情况下成功 ping 通这两个 IP 来确认，并且在拔掉电缆后两者均不可用），但 *.233 从未出现在 ifconfig 中：
Link encap:Ethernet HWaddr d0:XX:XX:51:78:XX
inet addr:*.77 Bcast:*.255 Mask:255.255.255.0
inet6 addr: X::X:X:X:787a/64 Scope:Link
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:23023571 errors:0 dropped:1362 overruns:0 frame:0
TX packets:364849 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:7441732389 (6.9 GiB) TX bytes:44699444 (42.6 MiB)
Memory:df300000-df37ffff
/etc/network/interfaces 中没有对 *.233 的引用，因此我看不到此 IP 分配来自何处。
所以，我有两个可能相关的问题，希望有人能帮助我：1）如何消除从我的服务器喷出的 NTP 流量，让 IT 摆脱我的困扰？ 2) 我的服务器所在的第二个 IP 地址是怎么回事？我该如何删除它？
谢谢大家:)
更新：根据要求：
$iptables -L -v -n
Chain INPUT (policy ACCEPT 57 packets, 6540 bytes)
pkts bytes target prot opt in out source destination
Chain FORWARD (policy ACCEPT 0 packets, 0 bytes)
pkts bytes target prot opt in out source destination
Chain OUTPUT (policy ACCEPT 27 packets, 2076 bytes)
pkts bytes target prot opt in out source destination
和$ip addr ls
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
inet 127.0.0.1/8 scope host lo
valid_lft forever preferred_lft forever
inet6 ::1/128 scope host
valid_lft forever preferred_lft forever
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq state UP group default qlen 1000
link/ether d0:50:99:51:78:7a brd ff:ff:ff:ff:ff:ff
inet *.77/24 brd *.255 scope global eth0
valid_lft forever preferred_lft forever
inet *.167/24 brd *.255 scope global secondary dynamic eth0
valid_lft 24612sec preferred_lft 24612sec
inet6 X::X:X:X:787a/64 scope link
valid_lft forever preferred_lft forever
3: eth1: <NO-CARRIER,BROADCAST,MULTICAST,UP> mtu 1500 qdisc mq state DOWN group default qlen 1000
link/ether d0:50:99:51:78:7b brd ff:ff:ff:ff:ff:ff
更新 2：我没有提到，除了 IP 地址不匹配之外，MAC ID 也不匹配。这确实让我重新思考流量是否确实来自我的机器。但是：（1）将我的服务器从网络上拔掉会使流量消失； (2) 移动到不同的网络端口以及随后的流量； (3)tcpdump port 123显示异常流量：
13:24:33.329514 IP cumm024-0701-dhcp-233.bu.edu.ntp > 183.61.254.77.44300: NTPv2, Reserved, length 440
13:24:33.329666 IP cumm024-0701-dhcp-233.bu.edu.ntp > 183.61.254.77.44300: NTPv2, Reserved, length 440
13:24:33.329777 IP cumm024-0701-dhcp-233.bu.edu.ntp > 183.61.254.77.44300: NTPv2, Reserved, length 296
更新3：
$ss -uapn 'sport = :123'
State Recv-Q Send-Q Local Address:Port Peer Address:Port
（即什么都没有）
$sudo cat /proc/net/dev
Inter-| Receive | Transmit
face |bytes packets errs drop fifo frame compressed multicast|bytes packets errs drop fifo colls carrier compressed
lo: 327357 5455 0 0 0 0 0 0 327357 5455 0 0 0 0 0 0
eth1: 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
eth0: 13642399917 36270491 0 6522 0 0 0 2721337 45098276 368537 0 0 0 0 0 0
更新 4：这些数据包是几天前的典型数据包。今天（但是，仍然很高）：
20:19:37.011762 IP cumm024-0701-dhcp-233.bu.edu.ntp > 103.56.63.147.26656: NTPv2, Reserved, length 152
20:19:37.011900 IP cumm024-0701-dhcp-233.bu.edu.ntp > 202.83.122.78.58066: NTPv2, Reserved, length 152
20:19:37.012036 IP cumm024-0701-dhcp-233.bu.edu.ntp > 103.56.63.147.17665: NTPv2, Reserved, length 152
20:19:37.014539 IP cumm024-0701-dhcp-233.bu.edu.ntp > 202.83.122.78.27945: NTPv2, Reserved, length 152
20:19:37.015482 IP cumm024-0701-dhcp-233.bu.edu.ntp > 202.83.122.78.42426: NTPv2, Reserved, length 152
20:19:37.015644 IP cumm024-0701-dhcp-233.bu.edu.ntp > 103.56.63.147.16086: NTPv2, Reserved, length 152
$ sudo ss -uapn '( sport = :42426 or dport = :42426 )'
State Recv-Q Send-Q Local Address:Port Peer Address:Port
是的，我可以 ping *.233 IP：
$ping 128.197.112.233
PING 128.197.112.233 (128.197.112.233) 56(84) bytes of data.
64 bytes from 128.197.112.233: icmp_seq=1 ttl=64 time=0.278 ms
64 bytes from 128.197.112.233: icmp_seq=2 ttl=64 time=0.282 ms
64 bytes from 128.197.112.233: icmp_seq=3 ttl=64 time=0.320 ms
不，MAC 不匹配 我的硬件 MAC 地址是：d0:50:99:51:78:7a 流量与 MAC 关联：bc:5f:f4:fe:a1:00
更新 5：根据要求，针对 *.233 进行端口扫描：
Starting Nmap 6.00 ( http://nmap.org ) at 2016-12-20 20:38 EET
NSE: Loaded 17 scripts for scanning.
Initiating SYN Stealth Scan at 20:38
Scanning cumm024-0701-dhcp-233.bu.edu (128.197.112.233) [1024 ports]
Discovered open port 22/tcp on 128.197.112.233
Completed SYN Stealth Scan at 20:38, 9.79s elapsed (1024 total ports)
Initiating Service scan at 20:38
Scanning 1 service on cumm024-0701-dhcp-233.bu.edu (128.197.112.233)
Completed Service scan at 20:38, 0.37s elapsed (1 service on 1 host)
Initiating OS detection (try #1) against cumm024-0701-dhcp-233.bu.edu (128.197.112.233)
Initiating Traceroute at 20:38
Completed Traceroute at 20:38, 0.10s elapsed
NSE: Script scanning 128.197.112.233.
[+] Nmap scan report for cumm024-0701-dhcp-233.bu.edu (128.197.112.233)
Host is up (0.083s latency).
Not shown: 1013 filtered ports
PORT STATE SERVICE VERSION
21/tcp closed ftp
22/tcp open ssh OpenSSH 5.5p1 Debian 6+squeeze1 (protocol 2.0)
23/tcp closed telnet
25/tcp closed smtp
43/tcp closed whois
80/tcp closed http
105/tcp closed unknown
113/tcp closed ident
210/tcp closed z39.50
443/tcp closed https
554/tcp closed rtsp
Device type: general purpose
Running: Linux 2.6.X
OS CPE: cpe:/o:linux:kernel:2.6
OS details: DD-WRT v24-sp2 (Linux 2.6.19)
Uptime guess: 45.708 days (since Sat Nov 5 03:39:36 2016)
Network Distance: 9 hops
TCP Sequence Prediction: Difficulty=204 (Good luck!)
IP ID Sequence Generation: All zeros
Service Info: OS: Linux; CPE: cpe:/o:linux:kernel
TRACEROUTE (using port 25/tcp)
HOP RTT ADDRESS
1 0.95 ms router1-lon.linode.com (212.111.33.229)
2 0.70 ms 109.74.207.0
3 1.09 ms be4464.ccr21.lon01.atlas.cogentco.com (204.68.252.85)
4 1.00 ms be2871.ccr42.lon13.atlas.cogentco.com (154.54.58.185)
5 63.45 ms be2983.ccr22.bos01.atlas.cogentco.com (154.54.1.178)
6 63.60 ms TrusteesOfBostonUniversity.demarc.cogentco.com (38.112.23.118)
7 63.55 ms comm595-core-res01-gi2-3-cumm111-bdr-gw01-gi1-2.bu.edu (128.197.254.125)
8 63.61 ms cumm024-dist-aca01-gi5-2-comm595-core-aca01-gi2-2.bu.edu (128.197.254.206)
9 90.28 ms cumm024-0701-dhcp-233.bu.edu (128.197.112.233)
OS and Service detection performed. Please report any incorrect results at http://nmap.org/submit/ .
Nmap done: 1 IP address (1 host up) scanned in 20.73 seconds
Raw packets sent: 557 (25.462KB) | Rcvd: 97 (8.560KB)
在 UDP 上：
Starting Nmap 6.00 ( http://nmap.org ) at 2016-12-20 20:44 EET
NSE: Loaded 17 scripts for scanning.
Initiating Ping Scan at 20:44
Scanning 128.197.112.233 [4 ports]
Completed Ping Scan at 20:44, 1.10s elapsed (1 total hosts)
Initiating UDP Scan at 20:44
Scanning cumm024-0701-dhcp-233.bu.edu (128.197.112.233) [1024 ports]
Completed UDP Scan at 20:44, 6.31s elapsed (1024 total ports)
Initiating Service scan at 20:44
Scanning 1024 services on cumm024-0701-dhcp-233.bu.edu (128.197.112.233)
Service scan Timing: About 0.39% done
Service scan Timing: About 3.12% done; ETC: 22:12 (1:25:46 remaining)
Service scan Timing: About 6.05% done; ETC: 21:53 (1:04:39 remaining)
Service scan Timing: About 8.98% done; ETC: 21:46 (0:56:03 remaining)
Discovered open port 123/udp on 128.197.112.233
Discovered open|filtered port 123/udp on cumm024-0701-dhcp-233.bu.edu (128.197.112.233) is actually open
Completed Service scan at 21:31, 2833.50s elapsed (1024 services on 1 host)
Initiating OS detection (try #1) against cumm024-0701-dhcp-233.bu.edu (128.197.112.233)
Retrying OS detection (try #2) against cumm024-0701-dhcp-233.bu.edu (128.197.112.233)
NSE: Script scanning 128.197.112.233.
Initiating NSE at 21:31
Completed NSE at 21:31, 10.02s elapsed
[+] Nmap scan report for cumm024-0701-dhcp-233.bu.edu (128.197.112.233)
Host is up (0.089s latency).
Not shown: 1023 open|filtered ports
PORT STATE SERVICE VERSION
123/udp open ntp?
1 service unrecognized despite returning data. If you know the service/version, please submit the following fingerprint at http://www.insecure.org/cgi-bin/servicefp-submit.cgi :
SF-Port123-UDP:V=6.00%I=7%D=12/20%Time=58597D5C%P=x86_64-unknown-linux-gnu
SF:%r(NTPRequest,30,"\xe4\x02\x04\xee\0\0\x8a\xff\0:t\xd9\x84\xa3\x04e\xdb
SF:\xcaeEX\xdbC'\xc5O#Kq\xb1R\xf3\xdc\x03\xfb\xb8\+>U\xab\xdc\x03\xfb\xb8\
SF:+T\xd1\xe9")%r(Citrix,C,"\xde\xc0\x010\x02\0\xa8\xe3\0\0\0\0");
Too many fingerprints match this host to give specific OS details
Network Distance: 9 hops
OS and Service detection performed. Please report any incorrect results at http://nmap.org/submit/ .
Nmap done: 1 IP address (1 host up) scanned in 2863.89 seconds
Raw packets sent: 175 (6.720KB) | Rcvd: 50 (10.088KB)
答案1
这是带有 IPMI 的服务器级机器。导致问题的“幽灵”NTP 服务器在系统上的 BMC 处理器上运行，而不是在主 CPU 上运行。
答案2
正如已经说过的，您的 IPMI BMC（可能）是问题所在。如果您无法访问 IPMI 界面的 Web 界面或 ssh 界面，您可以尝试使用 Debian 安装上的 IPMI Tool 进行访问：
apt install ipmitool
安装后，您可以像这样将命令传递到 BMC（如果它位于端口 1）：
ipmitool lan set 1 ipaddr 192.168.1.211
这是一个资源使用 IPMITool 进行 IPMI 网络配置。 man ipmitool如果你遇到困难，这总是一本好书……
您可以重置BMC如果你需要。