Hey folks <@U03T9R61SHK> <@UGRJA9TEH> could you pl...
# pinot-dev
j
Hey folks @Chris Peck @Xiang Fu could you please take a look at https://github.com/apache/pinot/issues/14017 https://github.com/apache/pinot/pull/14016 and let me know your thoughts? They are causing backward incompatibility for us
c
thanks for catching that! my thinking was that a lot of Pinot configs' effects/combinations are opaque (at least to me) without reading the code. The goal was to provide some hint to the user if configs don't make sense but that doesn't work w/ the current pattern
👍 1
j
Thanks for confirming! Spent some time to understand the root cause. I think this is a convoluted case.
for https://github.com/apache/pinot/issues/14017 @Xiang Fu you can skip this for now, I think this one is probably due to some internal verisoning issue
x
ack
do you have some incompatible schemas?
j
minimum piece, but this seems to pass in my local oss unit test, I'll dig more and keep you posted
Copy code
{
  "metricFieldSpecs": [
    {
      "dataType": "INT",
      "name": "ActualElapsedTime"
    }
  ],
  "timeFieldSpec": {
    "incomingGranularitySpec": {
      "dataType": "INT",
      "name": "DaysSinceEpoch",
      "timeType": "DAYS"
    }
  },
  "dimensionFieldSpecs": [
    {
      "dataType": "INT",
      "name": "TotalAddGTime"
    }
  ],
  "schemaName": "mytable"
}
x
sure, the above PRs also passed the tests, so pretty weird