On Mon, Jan 22, 2018 at 2:07 AM, Steve Atkins <steve@blighty.com> wrote:
> On Jan 21, 2018, at 6:44 PM, Vincenzo Romano <vincenzo.romano@notorand.it> wrote: > > 2018-01-21 19:31 GMT+01:00 Francisco Olarte <folarte@peoplecall.com>: >> On Sun, Jan 21, 2018 at 1:27 PM, Michael Paquier >> <michael.paquier@gmail.com> wrote: >>> On Fri, Jan 19, 2018 at 03:40:01PM +0000, Raymond O'Donnell wrote: >> ... >>>> How do you define "full"? > > The only possible and meaningful case, IMHO, as stated by David > earlier, is "file system full".
If your filesystem is full you're pretty much off the air. It's something that should never happen on a production system. So ... any automation around "the filesystem is full" is going to be much the same as "the server is dead". You're unlikely to be able to do anything useful from the app at that point, let alone from a trigger function.
Well, I'll agree that PostgreSQL is likely "off the air". On my system (Linux/Intel), I use quotas to restrict a user's use of disk space. Yes, even the PostgreSQL user. I also, at times, use a separate filesystem for a database and use a TABLESPACE for the tables within it. This is easy to do with Linux because I can create a filesystem in a regular disk file. It doesn't perform as well as "native", but my system is not a highly used, performance oriented, system. Use of a separate filesystem and tablespaces is, I think, a decent way to control disk usage for "something" so that if "something" goes "insane", it can't really impact "others" very much. Of course, others may reasonably disagree with me on this. Each cat, his own rat.
If the answer involves handling the case where the file system is full we're not answering a useful question, and the original poster probably needs to clarify. > Which is communicated by Postgres with the "Class 53 — Insufficient > Resources" error codes. > Please refer to official documentation like: > > https://www.postgresql.org/docs/10/static/errcodes-appendix.html > > For specific programming languages more details need to be checked. >
Cheers, Steve
--
I have a theory that it's impossible to prove anything, but I can't prove it.