How to deal with bigger loops in ElasticSearch?

Elastic Stack Elasticsearch
1

Hello everyone,

I have a painless script that calculates user sessions. The script works fine but when I have more than 10k entry in the painless script loop throws this error:

    "caused_by": {
            "type": "painless_error",
            "reason": "painless_error: The maximum number of statements that can be executed in a loop has been reached."
          }

How we can deal with bigger data sizes with a painless script?

And here is my query:

    {
      "size": 0,
      "query": {
        "bool": {
          "filter": [
            {
              "range": {
                "time": {
                  "lt": "2020-07-01T10:32:43.868Z",
                  "gt": "2020-01-01T10:31:58.836Z"
                }
              }
            }
          ]
        }
      },
      "aggs": {
        "sessions": {
          "terms": {
            "field": "user.id",
            "size": 5,
            "shard_size": 10
          },
          "aggs": {
            "session": {
              "scripted_metric": {
                "init_script": "state.times = [];",
                "map_script": "state.times.add(doc['time'].value.toInstant().toEpochMilli())",
                "combine_script": "def start_session = state.times[0]; def total_time = 0; def treshold = 60; if(state.times.length <= 1) { return treshold; } for(int i=1;i<state.times.length;i++) { def difference = Math.abs((state.times[i] - start_session) / 1000); if(difference!=null) { if(difference < treshold) { total_time = total_time + difference; } else { total_time = total_time + treshold } } start_session = state.times[i]; } return total_time;",
                "reduce_script": "return states[0]"
              }
            }
          }
        }
      }
    }

Thanks a lot.

2

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