On 10/19/23 10:56, Robert Haas wrote:
On Thu, Oct 19, 2023 at 10:43 AM David Steele <da...@pgmasters.net> wrote:
What I meant here (but said badly) is that in the case of snapshot
backups, the backup_label and tablespace_map will likely need to be
stored somewhere off the server since they can't be part of the
snapshot, perhaps in a key store. In that case the backup software would
still need to read the files from wherever we stored then and correctly
handle them when storing elsewhere. If you were moving the files to say,
S3, a similar thing needs to happen. In general, I think a locally
mounted filesystem is very unlikely to be the final destination for
these files, and if it is then probably pg_basebackup is your friend.
I mean, writing those tiny little files locally and then uploading
them should be fine in a case like that. It still reduces the surface
for mistakes. And you could also have --backup-label='| whatever' or
something if you wanted. The point is that right now we're asking
people to pull this information out of a query result, and that means
people are trying to do it by calling out to psql, and that is a GREAT
way to screw up the escaping or the newlines or whatever. I don't
think the mistakes people are making here are being made by people
using Perl and DBD::Pg, or Python and psycopg2, or C and libpq.
They're being made by people who are trying to shell script their way
through it, which entails using psql, which makes screwing it up a
breeze.
OK, I see what you mean and I agree. Better documentation might be the
answer here, but I doubt that psql is a good tool for starting/stopping
backup and not sure we want to encourage it.
Regards,
-David
