Re: Re: [COMMITTERS] Can not create more than 32766 databases in ufs file system.

From: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
To: Andrew Dunstan <andrew(at)dunslane(dot)net>
Cc: Stephen Frost <sfrost(at)snowman(dot)net>, Mark Mielke <mark(at)mark(dot)mielke(dot)cc>, pgsql-hackers(at)postgresql(dot)org, fulan Peng <fulanpeng(at)gmail(dot)com>
Subject: Re: Re: [COMMITTERS] Can not create more than 32766 databases in ufs file system.
Date: 2009-09-13 01:43:21
Message-ID: 22500.1252806201@sss.pgh.pa.us
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-committers pgsql-hackers

Andrew Dunstan <andrew(at)dunslane(dot)net> writes:
> Tom Lane wrote:
>> So the question I would ask goes more like "do you really need 32K
>> databases in one installation? Have you considered using schemas
>> instead?" Databases are, by design, pretty heavyweight objects.

> That's a fair question. OTOH, devising a scheme to get around it would
> not be terribly difficult, would it? I can imagine a scheme where the
> subdir for a database was lo/hi for some division of the database oid. I
> guess it could make matters ugly for pg_migrator, though.

As I said earlier, the number-of-subdirectories issue is not the
important thing. The OP was already up to 160GB worth of system
catalogs before his filesystem wimped out, and would be needing
terabytes if he wanted to go significantly past the filesystem limit.
So there is no point in devising some clever workaround for the
limitations of one filesystem unless you want to reconsider our system
catalog representation --- and that will carry actual user-visible
functional costs; it's not just a cute hack somewhere in the guts of
the system.

regards, tom lane

In response to

Browse pgsql-committers by date

  From Date Subject
Next Message User Andreas 2009-09-13 11:19:02 docbot - docbot: - incorporate changes from running docbot
Previous Message Tom Lane 2009-09-13 01:34:09 Re: pgsql: Add Unicode support in PL/Python PL/Python now accepts Unicode

Browse pgsql-hackers by date

  From Date Subject
Next Message daveg 2009-09-13 02:14:24 Re: Ragged CSV import
Previous Message Tom Lane 2009-09-13 01:34:09 Re: pgsql: Add Unicode support in PL/Python PL/Python now accepts Unicode