[monitor-query] Batch queries will occasionally fail to deserialize due to an incorrectly formatted response body

[richardpark-msft]

The service (randomly) will return an improperly formatted body for a batch request.

Example:

The request we sent:

{
    "url": "https://api.loganalytics.io/v1/$batch",
    "method": "POST",
    "headers": {
      "_headersMap": {
        "content-type": {
          "name": "content-type",
          "value": "application/json"
        },
        "accept": {
          "name": "accept",
          "value": "application/json"
        },
        "user-agent": {
          "name": "user-agent",
          "value": "core-http/1.2.6 Node/v12.15.0 OS/(x64-Windows_NT-10.0.19043)"
        },
        "x-ms-client-request-id": {
          "name": "x-ms-client-request-id",
          "value": "c292e40f-cabd-4acb-988c-f364f06c85f0"
        },
        "authorization": {
          "name": "authorization",
          "value": "Bearer <proper bearer token was here>"
        },
        "cookie": {
          "name": "Cookie",
          "value": ""
        }
      }
    },
    "body": "{\"requests\":[{\"id\":\"0\",\"body\":{\"query\":\"print \\\"hello\\\", true, make_datetime(\\\"2000-01-02 03:04:05Z\\\"), toint(100), long(101), 102.1, dynamic({ \\\"hello\\\": \\\"world\\\" })\\n      | project \\n          stringcolumn=print_0, \\n          boolcolumn=print_1,\\n          datecolumn=print_2,\\n          intcolumn=print_3,\\n          longcolumn=print_4,\\n          realcolumn=print_5,\\n          dynamiccolumn=print_6\\n      \",\"timespan\":\"PT5M\"},\"path\":\"/query\",\"method\":\"POST\",\"workspace\":\"1c1ef545-f240-4740-987a-52cdb50ad087\"}]}",
  }

The response, with the body seeming to be double-serialized:

{"responses":[{"id":"0","status":200,"headers":{"Age":"0","request-context":"appId=cid-v1:70941e4f-7e8f-40b7-b730-183893db0297"},"body":"{\"tables\":[{\"name\":\"PrimaryResult\",\"columns\":[{\"name\":\"stringcolumn\",\"type\":\"string\"},{\"name\":\"boolcolumn\",\"type\":\"bool\"},{\"name\":\"datecolumn\",\"type\":\"datetime\"},{\"name\":\"intcolumn\",\"type\":\"int\"},{\"name\":\"longcolumn\",\"type\":\"long\"},{\"name\":\"realcolumn\",\"type\":\"real\"},{\"name\":\"dynamiccolumn\",\"type\":\"dynamic\"}],\"rows\":[[\"hello\",true,\"2000-01-02T03:04:05Z\",100,101,102.1,\"{\\\"hello\\\":\\\"world\\\"}\"]]}]}"}]}

In contrast, here the body of a response when things succeed:

{"responses":[{"id":"0","status":200,"headers":{"Age":"60","request-context":"appId=cid-v1:70941e4f-7e8f-40b7-b730-183893db0297"},"body":{"tables":[{"name":"PrimaryResult","columns":[{"name":"stringcolumn","type":"string"},{"name":"boolcolumn","type":"bool"},{"name":"datecolumn","type":"datetime"},{"name":"intcolumn","type":"int"},{"name":"longcolumn","type":"long"},{"name":"realcolumn","type":"real"},{"name":"dynamiccolumn","type":"dynamic"}],"rows":[["hello",true,"2000-01-02T03:04:05Z",100,101,102.1,"{\"hello\":\"world\"}"]]}]}}]}

The first one appears to have double-encoded the 'body' field, essentially, so it doesn't deserialize properly in the client.

[richardpark-msft]

CC: @srnagar, @pakrym , @rakshith91

[richardpark-msft]

Same issue, but with the .net SDK: Azure/azure-sdk-for-net#21655

[richardpark-msft]

Adding @KarishmaGhiya, as this is an active bug that will require changes later.

[rakshith91]

Yay!! Finally able to reproduce this in python

[richardpark-msft]

There is a workaround currently in main from #15718. Remove once this is solved from the service side.

[KarishmaGhiya]

Service team is aware, but no ETA on the fix. Moving to backlog

[github-actions]

Hi @richardpark-msft, we deeply appreciate your input into this project. Regrettably, this issue has remained inactive for over 2 years, leading us to the decision to close it. We've implemented this policy to maintain the relevance of our issue queue and facilitate easier navigation for new contributors. If you still believe this topic requires attention, please feel free to create a new issue, referencing this one. Thank you for your understanding and ongoing support.