Darren Reed <darrenr+postgres@fastmail.net> writes:
>>> How do I know which files those are?
>>
>> find $PGDATA -name pg_internal.init
> Doesn't exist.
Sucks to be you, then :-(
I'm curious though exactly where the failure is, because there's not
much in RelationCacheInitializePhase2 that looks like it could crash,
other than the init-file reading (I assume load_relcache_init_file
got inlined into RelationCacheInitializePhase2). Do you want to try
rebuilding the backend with --enable-debug so we can get a better
traceback?
regards, tom lane