Question

我正在努力格式化收集的绘图文件，因为以后我可以将其导入到Influx数据库实例中。

文件的外观如下：

BSON

我想做的就是以一种看起来像这样的格式来获取它：

 #Date Time [CPU]User% [CPU]Nice% [CPU]Sys% [CPU]Wait% [CPU]Irq% [CPU]Soft% [CPU]Steal% [CPU]Idle% [CPU]Totl% [CPU]Intrpt/sec [CPU]Ctx/sec [CPU]Proc/sec [CPU]ProcQue [CPU]ProcRun [CPU]L-Avg1 [CPU]L-Avg5 [CPU]L-Avg15 [CPU]RunTot [CPU]BlkTot [MEM]Tot [MEM]Used [MEM]Free [MEM]Shared [MEM]Buf [MEM]Cached [MEM]Slab [MEM]Map [MEM]Anon [MEM]Commit [MEM]Locked [MEM]SwapTot [MEM]SwapUsed [MEM]SwapFree [MEM]SwapIn [MEM]SwapOut [MEM]Dirty [MEM]Clean [MEM]Laundry [MEM]Inactive [MEM]PageIn [MEM]PageOut [MEM]PageFaults [MEM]PageMajFaults [MEM]HugeTotal [MEM]HugeFree [MEM]HugeRsvd [MEM]SUnreclaim [SOCK]Used [SOCK]Tcp [SOCK]Orph [SOCK]Tw [SOCK]Alloc [SOCK]Mem [SOCK]Udp [SOCK]Raw [SOCK]Frag [SOCK]FragMem [NET]RxPktTot [NET]TxPktTot [NET]RxKBTot [NET]TxKBTot [NET]RxCmpTot [NET]RxMltTot [NET]TxCmpTot [NET]RxErrsTot [NET]TxErrsTot [DSK]ReadTot [DSK]WriteTot [DSK]OpsTot [DSK]ReadKBTot [DSK]WriteKBTot [DSK]KbTot [DSK]ReadMrgTot [DSK]WriteMrgTot [DSK]MrgTot [INODE]NumDentry [INODE]openFiles [INODE]MaxFile% [INODE]used [NFS]ReadsS [NFS]WritesS [NFS]MetaS [NFS]CommitS [NFS]Udp [NFS]Tcp [NFS]TcpConn [NFS]BadAuth [NFS]BadClient [NFS]ReadsC [NFS]WritesC [NFS]MetaC [NFS]CommitC [NFS]Retrans [NFS]AuthRef [TCP]IpErr [TCP]TcpErr [TCP]UdpErr [TCP]IcmpErr [TCP]Loss [TCP]FTrans [BUD]1Page [BUD]2Pages [BUD]4Pages [BUD]8Pages [BUD]16Pages [BUD]32Pages [BUD]64Pages [BUD]128Pages [BUD]256Pages [BUD]512Pages [BUD]1024Pages
20190228 00:01:00 12 0 3 0 0 1 0 84 16 26957 20219 14 2991 3 0.05 0.18 0.13 1 0 198339428 197144012 1195416 0 817844 34053472 1960600 76668 158641184 201414800 0 17825788 0 17825788 0 0 224 0 0 19111168 3 110 4088 0 0 0 0 94716 2885 44 0 5 1982 1808 0 0 0 0 9739 9767 30385 17320 0 0 0 0 0 0 12 13 3 110 113 0 16 16 635592 7488 0 476716 0 0 0 0 0 0 0 0 0 0 0 8 0 0 22 0  1  0  0  0  0  48963 10707 10980 1226 496 282 142 43 19 6 132
20190228 00:02:00 11 0 3 0 0 1 0 85 15 26062 18226 5 2988 3 0.02 0.14 0.12 2 0 198339428 197138128 1201300 0 817856 34054692 1960244 75468 158636064 201398036 0 17825788 0 17825788 0 0 220 0 0 19111524 0 81 960 0 0 0 0 94420 2867 42 0 7 1973 1842 0 0 0 0 9391 9405 28934 16605 0 0 0 0 0 0 9 9 0 81 81 0 11 11 635446 7232 0 476576 0 0 0 0 0 0 0 0 0 0 0 3 0 0 8 0  1  0  0  0  0  49798 10849 10995 1241 499 282 142 43 19 6 132
20190228 00:03:00 11 0 3 0 0 1 0 85 15 25750 17963 4 2980 0 0.00 0.11 0.10 2 0 198339428 197137468 1201960 0 817856 34056400 1960312 75468 158633880 201397832 0 17825788 0 17825788 0 0 320 0 0 19111712 0 75 668 0 0 0 0 94488 2869 42 0 5 1975 1916 0 0 0 0 9230 9242 28411 16243 0 0 0 0 0 0 9 9 0 75 75 0 10 10 635434 7232 0 476564 0 0 0 0 0 0 0 0 0 0 0 2 0 0 6 0  1  0  0  0  0  50029 10817 10998 1243 501 282 142 43 19 6 132
20190228 00:04:00 11 0 3 0 0 1 0 84 16 25755 17871 10 2981 5 0.08 0.11 0.10 3 0 198339428 197140864 1198564 0 817856 34058072 1960320 75468 158634508 201398088 0 17825788 0 17825788 0 0 232 0 0 19111980 0 79 2740 0 0 0 0 94488 2867 4 0 2 1973 1899 0 0 0 0 9191 9197 28247 16183 0 0 0 0 0 0 9 9 0 79 79 0 10 10 635433 7264 0 476563 0 0 0 0 0 0 0 0 0 0 0 5 0 0 12 0  1  0  0  0  0  49243 10842 10985 1245 501 282 142 43 19 6 132
20190228 00:05:00 12 0 4 0 0 1 0 83 17 26243 18319 76 2985 3 0.06 0.10 0.09 2 0 198339428 197148040 1191388 0 817856 34059808 1961420 75492 158637636 201405208 0 17825788 0 17825788 0 0 252 0 0 19112012 0 85 18686 0 0 0 0 95556 2884 43 0 6 1984 1945 0 0 0 0 9176 9173 28153 16029 0 0 0 0 0 0 10 10 0 85 85 0 12 12 635473 7328 0 476603 0 0 0 0 0 0 0 0 0 0 0 3 0 0 7 0  1  0  0  0  0  47625 10801 10979 1253 505 282 142 43 19 6 132

到目前为止，我设法将日期字符串转换为EPOCH格式。

我一直在想将第一个值“ [CPU]”用作度量，将“ User％”用作类型，主机可以从运行脚本的系统中获取。

非常感谢您的帮助，因为我真的非常了解文本编辑。

谢谢。

编辑：这是使用第一行的标题作为第二行的信息所期望得到的：

cpu_value,host=mxspacr1,instance=5,type=cpu,type_instance=softirq value=180599 1551128614916131663
cpu_value,host=mxspacr1,instance=2,type=cpu,type_instance=interrupt value=752 1551128614916112943
cpu_value,host=mxspacr1,instance=4,type=cpu,type_instance=softirq value=205697 1551128614916128446
cpu_value,host=mxspacr1,instance=7,type=cpu,type_instance=nice value=19250943 1551128614916111618
cpu_value,host=mxspacr1,instance=2,type=cpu,type_instance=softirq value=160513 1551128614916127690
cpu_value,host=mxspacr1,instance=1,type=cpu,type_instance=softirq value=178677 1551128614916127265
cpu_value,host=mxspacr1,instance=0,type=cpu,type_instance=softirq value=212274 1551128614916126586
cpu_value,host=mxspacr1,instance=6,type=cpu,type_instance=interrupt value=673 1551128614916116661
cpu_value,host=mxspacr1,instance=4,type=cpu,type_instance=interrupt value=701 1551128614916115893
cpu_value,host=mxspacr1,instance=3,type=cpu,type_instance=interrupt value=723 1551128614916115492
cpu_value,host=mxspacr1,instance=1,type=cpu,type_instance=interrupt value=756 1551128614916112550
cpu_value,host=mxspacr1,instance=6,type=cpu,type_instance=nice value=21661921 1551128614916111032
cpu_value,host=mxspacr1,instance=3,type=cpu,type_instance=nice value=18494760 1551128614916098304
cpu_value,host=mxspacr1,instance=0,type=cpu,type_instance=interrupt value=552 1551

编辑：这是我到目前为止所拥有的，我被困在这里。

cpu_value,host=mxspacr1,type=cpu,type_instance=user% value=0 1551128614916131663

这就是我得到的，但这仅适用于1列，现在我不知道如何处理其余的列，例如Nice，Sys，Wait等。

awk -v HOSTNAME="$HOSTNAME" 'BEGIN { FS="[][]"; getline; NR==1; f1=$2; f2=$3 } { RS=" "; printf f1"_measurement,host="HOSTNAME",type="f2"value="$3" ", system("date +%s -d \""$1" "$2"\"") }' mxmcaim01-20190228.tab

请帮助。

编辑。首先，谢谢您的帮助。

利用您在文本编辑方面的知识，我希望可以将其用于3个单独的文件，但是不幸的是，我不知道为什么格式不同，例如：

CPU_measurement,host=mxmcamon05,type=User% value= 1552014000
CPU_measurement,host=mxmcamon05,type=User% value= 1551960000
CPU_measurement,host=mxmcamon05,type=User% value= 1551343500
CPU_measurement,host=mxmcamon05,type=User% value= 1551997620
CPU_measurement,host=mxmcamon05,type=User% value= 1551985200
CPU_measurement,host=mxmcamon05,type=User% value= 1551938400
CPU_measurement,host=mxmcamon05,type=User% value= 1551949200
CPU_measurement,host=mxmcamon05,type=User% value= 1551938400
CPU_measurement,host=mxmcamon05,type=User% value= 1551938400
CPU_measurement,host=mxmcamon05,type=User% value= 1551945600
CPU_measurement,host=mxmcamon05,type=User% value= 1551938400

因此，我不怎么处理数组，可以使用#Date Time SlabName ObjInUse ObjInUseB ObjAll ObjAllB SlabInUse SlabInUseB SlabAll SlabAllB SlabChg SlabPct 20190228 00:01:00 nfsd_drc 0 0 0 0 0 0 0 0 0 0 20190228 00:01:00 nfsd4_delegations 0 0 0 0 0 0 0 0 0 0 20190228 00:01:00 nfsd4_stateids 0 0 0 0 0 0 0 0 0 0 20190228 00:01:00 nfsd4_files 0 0 0 0 0 0 0 0 0 0 20190228 00:01:00 nfsd4_stateowners 0 0 0 0 0 0 0 0 0 0 20190228 00:01:00 nfs_direct_cache 0 0 0 0 0 0 0 0 0 0作为类型，然后通过ObjInUse ObjInUseB ObjAll ObjAllB SlabInUse SlabInUseB SlabAll SlabAllB SlabChg SlabPct进行迭代并像{{1 }}，最后，在这种情况下，ObjInUse的值将为0，ObjInUseB = 0，ObjAll = 0，这样一个值，如下所示：

nfsd_drc

slab_value是一个硬编码的值。

谢谢。

Answer 1

目前尚不清楚instance和type_instance=interrupt的最终期望格式来自何处。否则下面的awk代码应该可以工作。注意：它不会从标记值中剥离%并以秒为单位在行尾打印时间戳（如果需要纳秒，请添加额外的零）。

gawk -v HOSTNAME="$HOSTNAME" 'NR==1 {split($0,h,/[ \t\[\]]+/,s); for(i=0;i<length(h);i++){ h[i]=tolower(h[i]); };}; NR>1 { for(j=2;j<NF;j++) {k=2*j; printf("%s_value,host=%s,type=%s,type_instance=%s value=%s %s\n", h[k], HOSTNAME, h[k], h[k+1],$(j+1), mktime(substr($1,1,4)" "substr($1,5,2)" "substr($1,7,2)" "substr($2,1,2)" "substr($2,4,2)" "substr($2,7,2)));}}' mxmcaim01-20190228.tab

使用第一行作为标题并按列迭代的AWK格式

1 个答案: