egalpin opened a new issue, #10875: URL: https://github.com/apache/pinot/issues/10875
https://github.com/apache/pinot/issues/10713 seems highly related to an issue I've found recently, where adding a new STRING column to an existing schema (existing table) while also adding it to noDictionaryColumns, then executing a segment reload, results in errors at query time. The errors emanate from the fact that the segment reload process for sealed segments exclusively sets the new column values to a singular (default) value; in doing so, the cardinality of the column is 1. When the cardinality of the column is 1, the column is inferred to be sorted. Because the segment metadata indicates that the column is sorted, the column's value reader is instantiated as an instance of `SortedIndexReaderImpl`, which exclusively works with dictionary encoding. Could a similar fix be applied as in https://github.com/apache/pinot/pull/10851 to also address this issue? -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
