Daniel Gustafsson <dan...@yesql.se> writes: > Or we could revisit Tom's proposal in the thread that implemented the feature: > to have timestamped directory names to get around this very problem? I think > we should be able to figure out a way to make it easy enough for the BF code > to > figure out (and clean up).
How about inserting an additional level of subdirectory?
pg_upgrade_output.d/20220603122528/foo.log
Then code doing "rm -rf pg_upgrade_output.d" needs no changes.
regards, tom lane
