Yeah tried to do the pg_dump, 8 hours down the line, not even halfway done. As expected, it is stuck on the way with the largest toast.
pg_dump -v -d [database] -U [user] \
-h [hostname] \
--if-exists --clean -Fc \
-f [dumpfile] \
--quote-all-identifiers -n [schemaname]
-h [hostname] \
--if-exists --clean -Fc \
-f [dumpfile] \
--quote-all-identifiers -n [schemaname]
Will try the -Fd and -j option.
On Sun, Mar 2, 2025 at 1:40 PM Ron Johnson <ronljohnsonjr@xxxxxxxxx> wrote:
On Sat, Mar 1, 2025 at 4:27 PM Tom Lane <tgl@xxxxxxxxxxxxx> wrote:Edwin UY <edwin.uy@xxxxxxxxx> writes:
> If need be, do I manually edit the dumpfile using something like vi?
That's usually the last-ditch answer. Often you can just feed the
script to psql and ignore any errors, though.Single-threaded text is pretty slow for a 500GB database.--Death to <Redacted>, and butter sauce.Don't boil me, I'm still alive.<Redacted> lobster!