Hi,
I just saw a pg_upgrade failure on my aio branch [1]. Not sure what caused it
yet. The reason I'm writing in this thread is that I looked at the
regress_log_* for the failure, and found it to be 14.95MiB (which crashed the
browser on my phone...).
https://api.cirrus-ci.com/v1/artifact/task/5167740683026432/log/src/bin/pg_upgrade/tmp_check/log/regress_log_002_pg_upgrade
That seems way beyond reasonable.
regress_log_002_pg_upgrade.log includes all of 002_pg_upgrade_old_node.log and
002_pg_upgrade_new_node.log. The old node's log includes all pg_dump queries.
Followed by many MB of diff due to
=== diff of /Users/admin/pgsql/src/bin/pg_upgrade/tmp_check/tmp_test_Q7GQ/dump1.sql and
/Users/admin/pgsql/src/bin/pg_upgrade/tmp_check/tmp_test_Q7GQ/dump2.sql
=== stdout ===
Greetings,
Andres Freund
[1] https://cirrus-ci.com/task/5167740683026432