Elasticsearch highlight is showing everything in different order and missing statements

I am working on elasticsearch highlight, if there is exact phrase match then the keywords are to be highlighted. The query is having following multi match query

"multi_match" : {
          "query" : "From active node's NSF trace log, ping test failed to receive ECHO from external gateway:\n\n161310 14:16:00 29024  0x7f124bfff700 Unable to receive ECHO reply for 10.32.122.225\n161310 14:16:00 29024  0x7f124bfff700 Pinging failed for 10.32.122.225 for 2 time\n161310 14:16:00 29024  0x7f124bfff700 NSF PINGING COMPLETED for the gateway",
          "fields" : [ "stmtcontent.exact" ],
          "type" : "phrase"
        }

Now the issue is the content is having a field stmtcontent as

"stmtcontent": [
          "APG cyclic reboot every hour",
          "\"AP EXTERNAL NETWORK FAULT\" alarm reported on O&M gateway and PS gateway:\n\nAlarm Identifier    Class    Category    Time\n10304:5             A2       APZ         yyyy-mm-dd hh:mm:ss\n\nObject of Reference\nA/10.32.122.225\n\nAlarm Text\nAP EXTERNAL NETWORK FAULT\nPROBLEM\nAP EXTERNAL NETWORK \n\nProblem Data\nConnection to remote gateway is down\n\n\nAlarm Identifier    Class    Category    Time\n10304:7             A2       APZ         yyyy-mm-dd hh:mm:ss\n\nObject of Reference\nA/10.32.82.129\n\nAlarm Text\nAP EXTERNAL NETWORK FAULT\nPROBLEM\nAP EXTERNAL NETWORK 2 \n\nProblem Data\nConnection to remote gateway on network 2 is down\n",
          "From active node's NSF trace log, ping test failed to receive ECHO from external gateway:\n\n161310 14:16:00 29024  0x7f124bfff700 Unable to receive ECHO reply for 10.32.122.225\n161310 14:16:00 29024  0x7f124bfff700 Pinging failed for 10.32.122.225 for 2 time\n161310 14:16:00 29024  0x7f124bfff700 NSF PINGING COMPLETED for the gateway"
       ]

I am using the following analayzer in mapping

"metis_exact_analyzer": {
 "filter": [
      "lowercase"
  ],
  "type": "custom",
  "tokenizer": "soft_tokenizer"
  }

  "tokenizer": {
      "soft_tokenizer": {
      "pattern": "(\w*[-*#+=;:\/,~_ ]*\w+)",
      "type": "pattern",
       "group": "1"
      }
  }

the exact match which I am getting is getting distorted and the results are in random. I am unable to identify the solution any help will be great. The result I am getting is :-

"stmtcontent.exact" : [ " Reference\nA/10.<span style='font-weight: bold;'>32</span>.<span style='font-weight: bold;'>122</span>.<span style='font-weight: bold;'>225</span>\n\nAlarm Text\nAP EXTERNAL NETWORK FAULT\nPROBLEM\nAP EXTERNAL NETWORK \n\nProblem Data", "\nAP EXTERNAL NETWORK FAULT\nPROBLEM\nAP EXTERNAL NETWORK<span style='font-weight: bold;'> 2</span> \n\nProblem Data\nConnection to remote<span style='font-weight: bold;'> gateway</span> on network<span style='font-weight: bold;'> 2</span> is down\n", "\"AP EXTERNAL NETWORK FAULT\" alarm reported on O&M gateway and PS<span style='font-weight: bold;'> gateway</span>:\n\nAlarm Identifier", "\nConnection to remote<span style='font-weight: bold;'> gateway</span> is down\n\n\nAlarm Identifier    Class    Category    Time\n10304:7", "             A2       APZ         yyyy-mm-dd hh:mm:ss\n\nObject of Reference\nA/10.<span style='font-weight: bold;'>32</span>.82.129\n\nAlarm Text" ]

the highlights tags which I am using are :-

"highlight" : {
"pre_tags" : [ "<span style='font-weight: bold;'>" ],
"post_tags" : [ "</span>" ],
"fields" : {
   "stmtcontent.exact" : { }
}
}

What am I missing ?? Any help will be appreciated !!!

Thanks

This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.