ELASTIC_SEARCH.md

Elasticsearch queries

These have been tested with Postman, the API will probably use an Elasticsearch JS client. With the URLs shown below,

List indices

The root route defaults to the service index instead of the span index, which is what we usually need. So we list them to see how they are called since their name changes depending on the date of deployment.

Request URL: http://localhost:9200/_cat/indices

Response example:

green open jaeger-span-2024-09-12    NHSuiiRfR-2elT0mJecMpw 1 0 393743  0 19.1mb 19.1mb 19.1mb
green open jaeger-service-2024-09-12 Uvmk-GWtRW2Uh-owFHS2kA 1 0     15 52 14.4kb 14.4kb 14.4kb

List traces

The jaeger-span-<date> index only serves spans. But we can get trace related data by querying root spans, those whose references field is an empty array, meaning they have no parent span.

Request URL: http://localhost:9200/jaeger-span-2024-09-12/_search

Request body:

{
  "query": {
    "bool": {
      "must_not": [
        {
          "nested": {
            "path": "references",
            "query": {
              "exists": {
                "field": "references"
              }
            }
          }
        }
      ]
    }
  }
}

The returned response would be an array of spans, one span per trace on the DB.

Get trace

Request URL: http://localhost:9200/jaeger-span-2024-09-12/_search

Request body:

{
  "query": {
    "match": {
      "traceID": {
        "query": "3872e1704986f71e0fd51a5d4b7c3be6"
      }
    }
  }
}

The returned response would be an array with all of the spans that belong to the queried traceID.

Match traces by top-level fields

These fields can be queried the same as traceID. For example, operationName:

Request URL: http://localhost:9200/jaeger-span-2024-09-12/_search

Request body:

{
  "query": {
    "match": {
      "operationName": {
        "query": "abci_query"
      }
    }
  }
}

Match traces by tags

Get traces on a given height

Request URL: http://localhost:9200/jaeger-span-2024-09-12/_search

Request body:

{
  "query": {
    "nested": {
      "path": "tags",
      "query": {
        "bool": {
          "must": [
            {
              "match": {
                "tags.key": "height"
              }
            },
            {
              "match": {
                "tags.value": "772"
              }
            }
          ]
        }
      }
    }
  }
}

Get traces with a given tx_hash

Request URL: http://localhost:9200/jaeger-span-2024-09-12/_search

Request body:

{
  "query": {
    "nested": {
      "path": "tags",
      "query": {
        "bool": {
          "must": [
            {
              "match": {
                "tags.key": "tx_hash"
              }
            },
            {
              "match": {
                "tags.value": "51B60C806E4F685C8A7FC79CBB72BF860B566B354F2C57B023407DD7AD633CB7"
              }
            }
          ]
        }
      }
    }
  }
}

Get traces where a concrete signer appears, using wildcards

Request URL: http://localhost:9200/jaeger-span-2024-09-12/_search

Request body:

{
  "query": {
    "nested": {
      "path": "tags",
      "query": {
        "bool": {
          "must": [
            {
              "match": {
                "tags.key": "tx"
              }
            },
            {
              "wildcard": {
                "tags.value": "*signer: layer1pkptre7fdkl6gfrzlesjjvhxhlc3r4gmt53rug*"
              }
            }
          ]
        }
      }
    }
  }
}

Match traces by combining top-level fields and tags

This example query would return spans with operationName = execute_tx and height = 176.

Request URL: http://localhost:9200/jaeger-span-2024-09-12/_search

Request body:

{
  "query": {
    "bool": {
      "must": [
        {
          "match": {
            "operationName": {
              "query": "execute_tx"
            }
          }
        },
        {
          "nested": {
            "path": "tags",
            "query": {
              "bool": {
                "must": [
                  {
                    "match": {
                      "tags.key": "height"
                    }
                  },
                  {
                    "match": {
                      "tags.value": "176"
                    }
                  }
                ]
              }
            }
          }
        }
      ]
    }
  }
}