code SQL0020 · recovery permanent — an unchanged retry cannot succeed; retry machinery stops immediately
the log line
As it arrives in your log or error chain — example values stand where yours will:
stream partitions remain after draining: schema "sqlstreams", stream_id 1 -- stop the stream's producers and call Client.Stream(name).Destroy again [SQL0020]
ACCEPTED ANSWER Posted: 2026-09-09
stop the stream's producers and call Client.Stream(name).Destroy again
Helpful info.
the partitions still attached to the log
SELECT partition.relname AS partition FROM pg_inherits JOIN pg_class AS partition ON partition.oid = pg_inherits.inhrelid WHERE pg_inherits.inhparent = to_regclass('{schema}.message_log_{stream_id}') ORDER BY partition.relname;
whether a producer is still writing -- run it twice
SELECT max(id) AS head, count(*) AS message_count FROM {schema}.message_log_{stream_id};
Nothing pasted yet — the queries above show their blanks.
Looking for a different code? Search every thread.