On Sat, Aug 8, 2020 at 10:59 AM Don Seiler <don@seiler.us> wrote:
Do you think it would be worth trying to delete (move/rename) the 000000010000142E0000005B file and see if it changes things?
In my PG10 test cluster, I see a ".partial" file getting archived as well. In this case I would expect to see a 000000010000142E0000005B.partial.lz4 file in the archive but I don't. Although my test cluster used streaming replication where as this was done without streaming (just WAL archive/restore/recovery), and this one is PG 9.6 (I have both 9.6 and 10 here). I'll re-run my tests without streaming and both 9.6 and 10.
