Re: Server Backup: pg_dump vs pg_dumpall
От | Ben Chobot |
---|---|
Тема | Re: Server Backup: pg_dump vs pg_dumpall |
Дата | |
Msg-id | 4A64F1BE.50202@silentmedia.com обсуждение исходный текст |
Ответ на | Server Backup: pg_dump vs pg_dumpall (APseudoUtopia <apseudoutopia@gmail.com>) |
Список | pgsql-general |
APseudoUtopia wrote: > I'm writing a backup script. Right now, I only have one database on my > postgresql server. I'm deciding if I should use pg_dump or pg_dumpall > when backing up the server. As far as I can tell, pg_dumpall cannot > compress the dumps automatically and it only dumps data in the > standard SQL text file format. This means that I would not be able to > use pg_restore to selectively restore the database, correct? > > What is *NOT* included in a pg_dump that IS in pg_dumpall (Other than > all the databases)? Things like user-defined functions and datatypes? > Roles? Views? > > I was leaning towards pg_dumpall, but then I realized that it only > dumps in the standard SQL text file format, and it cannot be > compressed automatically. Things that span all databases in your cluster (like your roles) do not get saved with pg_dump. So yes, if you only have one database, pg_dumpall is probably what you should be using. You can always pipe the output of pg_dumpall to your compression program of choice.
В списке pgsql-general по дате отправления: