As it arrives in your log or error chain — the values are your call's own:
partition creation cannot keep up with the id sequence -- raise PartitionSize, or run a consumer so create-ahead stays ahead of the producers [SQL0056]
Every rerun of the insert drew ids past the partition the previous heal
created. The producer heals a missing partition by creating the one the
id sequence’s next value lands in and rerunning the insert — once for the
boundary itself, then once per configured retry (Retry.MaxRetries).
Reaching this means other producers advanced the sequence a whole
partition between each heal and its rerun: PartitionSize is too small
for the write rate, or nothing is running the stream’s upkeep so every
boundary is met cold.
Raise PartitionSize, or run a consumer so create-ahead stays ahead of
the producers.
raise PartitionSize, or run a consumer so create-ahead stays ahead of the producers
Looking for a different code? Search every thread.