Tim*_*chy 16 ip debian tcpdump ntp network-interface
所以大学 IT 安全团队和我一直在不停地讨论这个问题……任何人对此都有任何想法：
我最近为我的实验室在一台专用计算机（英特尔 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）上运行 Debian 8.6 设置了一个小型文件服务器。Debian 安装没有任何问题，当时我还安装了 Samba、NTP、ZFS 和 python。事情似乎运行良好，所以我让它在实验室的角落里坐了几个星期。
大约两周前，我收到了来自 IT 团队的一封电子邮件，说我的服务器已被“入侵”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https 所述： //www.us-cert.gov/ncas/alerts/TA14-013A）。他们指出的标志是端口 123 上的大量 NTPv2 流量。奇怪的是，他们确定来自 ( *.*.*.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 上生成了此流量（如 tcpdump 所揭示的）。
这就是奇怪的开始。我首先浏览了针对 CVE-2013-5211 推荐的“修复”（更新 NTP 过去版本 4.2.7 以及禁用 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
但这对交通也没有影响。我终于尝试从系统中清除 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数周。
在这条道路上遇到死胡同后，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（通过连接以太网电缆成功 ping 确认，并且在拔下电缆时都不可用），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
在 /etc/network/interfaces 中没有对 *.233 的引用，所以我看不出这个 IP 分配是从哪里来的。
所以，我有两个可能相关的问题，我希望有人可以帮助我：1）如何消除从我的服务器喷出的 NTP 流量，让 IT 摆脱困境？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，伙计们:)
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地址不匹配外，MAC ID也不匹配。这真的让我三思而后行，是否真的来自我的机器。但是：（1）从网络上拔下我的服务器会使流量消失；(2) 移动到不同的网络端口，流量随之而来；(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。今天（但是是的，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，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)
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 工具访问：
apt install ipmitool
安装后，您可以像这样将命令传递给 BMC（如果它在端口 1 上）：
ipmitool lan set 1 ipaddr 192.168.1.211
这是使用 IPMITool 配置 IPMI 网络的资源。 man ipmitool如果你被卡住了，这总是一个很好的阅读......
如果需要，您可以重置 BMC。
| 归档时间： |
|
| 查看次数： |
477 次 |
| 最近记录： |