Question

我有一个循环设置来输入一个包含IP列表的文件。循环将使用IP搜索文件以获取关联的MAC地址。有时，对于每个IP，将列出多个MAC。它可以是一个，也可以是七个或更多。那真是个未知数。我所知道的是，我需要能够比较每个IP报告的MAC数量，并且如果它们不匹配则会回显一些MAC。这是我的脚本输出现在的样子：

Testing 192.168.17.25
00:0f:XX:41:e7:50; 00:0f:XX:41:e7:50; 00:0f:XX:41:e7:50; 
Testing 192.168.17.26
00:0f:XX:43:7a:f0; 00:0f:XX:43:7a:f0; 00:0f:XX:43:7a:f0; 00:0f:XX:43:7a:f0;

这是我的while循环：

cat $list | while read line;

do
echo "Testing $line"
var=$(/bin/grep -A 10 -w $line $file | /bin/grep 'ethernet' | /bin/awk '{print  $3}')
echo $var

done

如何构建一个if / then进入循环内部来比较MAC地址并回显一条消息，如果它找到一个不同的消息？

更新：$ list文件是一个带IP的简单文本文件。

192.168.17.1

192.168.17.2

等...

我运行IP的$文件获取MAC是一个DHCP租约文件。

lease 192.168.17.28 {
starts 2 2016/07/12 07:36:15;
ends 3 2016/07/13 07:36:15;
tstp 3 2016/07/13 07:36:15;
tsfp 3 2016/07/13 19:36:15;
atsfp 3 2016/07/13 19:36:15;
cltt 2 2016/07/12 07:36:15;
binding state free;
hardware ethernet 84:7b:eb:24:87:ef;
uid "\001\204{\353$\207\357";

更新：@ pakistanprogrammerclub

当我运行脚本时，我得到：

[root@localhost]# ./genipmacs.sh < leasesfile > ipmaclist
awk: cmd. line:2: {ipmac[$1][$2]=1}
awk: cmd. line:2:           ^ syntax error
awk: cmd. line:6:     for(mac in ipmac[ip])
awk: cmd. line:6:                     ^ syntax error

不确定这里有什么需要。是否需要将其他内容括在引号中？

更新＃2 @ pakistanprogrammerclub

使用修改后的awk代码后，脚本按预期运行没有错误。但是，我的ipmaclist文件如下所示：

 192.168.17.25 1       00:0
 00:0    4164    192.168.17.26
 192.168.17.27    1       192.168.17.28
 00:0    4164    192.168.17.29   192.168.17.30

由于某些原因它没有获得MAC。我想我只需要在代码中调整grep和sed命令。这似乎是根本原因。在脚本之外运行那些产生上面看到的数据，然后awk尝试解析它。 Grep和sed外部脚本：

x.x.x.x x.x.x.x.  00:0
x.x.x.x.    00:0
x.x.x.x.  48:
x.x.x.x x.x.x.x  48:
x.x.x.x.   00:0

似乎是让写行只是切断了mac。

更新＃3

新的ipmaclist内容：

00:0f:94:43:fd:d0       2       x.x.x.x
x.x.x.x.   1       dc:37:14:82:a2:82
x.x.x.x.  1       00:0f:94:41:d4:d0
x.x.x.x 1       00:0f:94:41:9f:e0

租赁档案 -

x.x.x.x
x.x.x.x
x.x.x.x
x.x.x.x

更新4

新代码也可以反向运行。

00:0f:94:c1:31:20       1       192.168.17.26
00:0f:94:c1:1f:30       3       192.168.17.27   192.168.17.28  192.168.17.29
00:0f:94:66:84:f0       1       192.168.17.30

Answer 1

不要手动循环使用它们，而是考虑使用sort -u删除重复项，然后使用wc -l进行计数。

unique_mac_addrs() {
     # First, split them up to individual lines, stripping out
     # any stray spaces or blank lines, then sort and count.
     echo "$@" | tr \; $'\n' | awk '$1{print $1}' | sort -u | wc -l
}
sample1="00:0f:XX:43:7a:f0; 00:0f:XX:43:7a:f0; 00:0f:XX:43:7a:f0; 00:0f:XX:43:7a:f0;"
sample2="00:0f:XX:43:7a:f0; FF:0f:XX:43:7a:f0; EE:0f:XX:43:7a:f0; 00:0f:XX:43:7a:f0;"
unique_mac_addrs $sample1  # => 1
unique_mac_addrs $sample2  # => 3

Answer 2

这种方法使用gawk 4.0生成一个以制表符分隔的MACS列表，并从租约文件中为每个IP计数 - 称之为ipmacslist：

10.0.1.1        1       00:11:22:33:44:55
10.0.1.2        2       00:11:22:33:44:56       00:11:22:37:44:56
10.0.1.3        1       00:11:22:33:44:57

你可以在$ list中为你的IP列表grep ipmacslist然后过滤计数：

grep -F -f$list ipmacslist|awk '$2 > 1'

这是生成ipmacslist的代码 - 这是一个管道 - 把它放在一个脚本中说genipmacs.sh并在stdin上提供租约文件内容 - ./genipmacs.sh <leasefile.leases >ipmacslist

# grep ip and mac from lease file
grep -Po '^lease \K[0-9.]*|hardware ethernet \K[0-9:a-f]*'|
# join pairs of consecutive lines to get ip-mac pair on one line
sed -n 'N; s/\n/\t/;p'|
# collect unique ip-mac pairs - output ip unique macs list and count
# uses gawk 4.0 2D arrays
awk '
# bail on invalid input
!/^[0-9.]+\t[0-9:a-f]+/ {
  printf("error line %d not IP-MAC pair: \"%s\"\n", NR, $0)
  exitstatus=1
  exit exitstatus
}
{ipmac[$1][$2]=1}
END {
  if(exitstatus) exit exitstatus
  for(ip in ipmac) {
    printf("%s\t%d", ip, length(ipmac[ip]))
    for(mac in ipmac[ip])
      printf("\t%s", mac)
    printf("\n")
  }
}
'

Answer 3

Pre-gawk 4.0解决方案使用假的2D阵列 - 这很痛苦 - 请升级到gawk 4.0 - 它已经有几年了 -

# grep ip and mac from lease file
grep -Po '^lease \K[0-9.]*|hardware ethernet \K[0-9:a-f]*'|
# join pairs of consecutive lines to get ip-mac pair on one line
sed -n 'N; s/\n/\t/;p'|
# collect unique ip-mac pairs - output ip unique macs list and count
awk '
# bail on invalid input
!/^[0-9.]+\t[0-9:a-f]+/ {
  printf("error line %d not IP-MAC pair: \"%s\"\n", NR, $0)
  exitstatus=1
  exit exitstatus
}
{
  if(($1, $2) in ipmac) next
  if($1 in macs) {
    macs[$1]=macs[$1] "\t" $2
  } else {
    ips[i++]=$1
    macs[$1]=$2
  }
  ++nmacs[$1]
  ipmac[$1, $2]=1
}
END {
  if(exitstatus) exit exitstatus
  for(i=0; i < length(ips); ++i) {
    ip=ips[i]
    printf("%s\t%d", ip, nmacs[ip])
    printf("\t%s\n", macs[ip])
  }
}
'

其输出和测试租用文件如下

10.0.1.1        1       00:11:22:33:44:55
10.0.1.2        2       00:11:22:33:44:56       00:11:22:37:4f:56
10.0.1.3        2       00:11:22:33:44:57       00:11:22:a3:44:57

测试租约文件

lease 10.0.1.1 {
  starts 5 2015/05/15 01:57:17;
  ends 5 2015/05/15 02:07:17;
  tstp 5 2015/05/15 02:07:17;
  cltt 5 2015/05/15 01:57:17;
  binding state free;
  hardware ethernet 00:11:22:33:44:55;
  client-hostname "host1";
  uid "foo";
}
lease 10.0.1.2 {
  starts 5 2015/05/15 02:09:16;
  ends 5 2015/05/15 02:19:16;
  tstp 5 2015/05/15 02:19:16;
  cltt 5 2015/05/15 02:09:16;
  binding state free;
  hardware ethernet 00:11:22:33:44:56;
  client-hostname "host2";
  uid "bar";
}
lease 10.0.1.3 {
  starts 5 2015/05/15 02:16:01;
  ends 5 2015/05/15 02:26:01;
  tstp 5 2015/05/15 02:26:01;
  cltt 5 2015/05/15 02:16:01;
  binding state free;
  hardware ethernet 00:11:22:33:44:57;
  client-hostname "host3";
  uid "baz-1";
}
lease 10.0.1.3 {
  starts 5 2015/05/15 02:17:01;
  ends 5 2015/05/15 02:27:01;
  tstp 5 2015/05/15 02:27:01;
  cltt 5 2015/05/15 02:17:01;
  binding state free;
  hardware ethernet 00:11:22:a3:44:57;
  client-hostname "host3";
  uid "baz-2";
}
lease 10.0.1.2 {
  starts 5 2015/05/15 02:09:16;
  ends 5 2015/05/15 02:19:16;
  tstp 5 2015/05/15 02:19:16;
  cltt 5 2015/05/15 02:09:16;
  binding state free;
  hardware ethernet 00:11:22:37:4f:56;
  client-hostname "host2";
  uid "bar-1";
}

while循环输出多值比较和echo消息如果不同

3 个答案: