otterc commented on pull request #34934: URL: https://github.com/apache/spark/pull/34934#issuecomment-1017905427
> Thanks @otterc, changed code as you suggested, [pan3793@1c14eb9](https://github.com/pan3793/spark/commit/1c14eb9c4d2710c6a906781d80246e8adc1628ff), and have sent logs to you via email, please take a look again. Thanks @pan3793. The new logs, however, don't have the statement on the shuffle server which logs the partition that is finalized. I see that you removed this: ``` logger.info("shuffle partition {}_{} {} {}, chunk_size={}, meta_length={}, data_length={}", msg.appId, msg.appAttemptId, msg.shuffleId, partition.reduceId, partition.indexFile.getPos() / 8 - 1, partition.metaFile.getPos(), partition.lastChunkOffset); ``` I think that the partitions that the reducer is trying to read are not finalized by the shuffle server. This log statement would have shown that. I still don't know how the reducer is getting these merged block but at least we can rule out the shuffle service. There is also a possibility that push-based shuffle has a conflict with AQE. Do you see the same issue when `spark.sql.adaptive.coalescePartitions.enabled` and `spark.sql.adaptive.fetchShuffleBlocksInBatch` are disabled? -- 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]
