"mark" <dvlhntr(at)gmail(dot)com> writes:
> From: Tom Lane [mailto:tgl(at)sss(dot)pgh(dot)pa(dot)us]
>>> An exception is LockTableCommand(), which is still behaving unsafely;
>>> but that seems to require some more discussion before we change it.
> So is my adding a access exclusive lock to the parent table when dropping
> children a bad thing (tm), in terms of introducing an error, since it uses
> the locktablecommand()?
No; in 8.3 it's all right. That comment was about the then
development-tip behavior of LOCK TABLE in 8.4, which would try to
recurse to child tables. 8.3 doesn't do that, it just locks exactly the
table you specify, so it won't fail unless you're dropping and
recreating the parent too...
regards, tom lane
In response to
pgsql-novice by date
|Next:||From: Machiel Richards||Date: 2010-11-23 08:26:56|
|Subject: Postgresql on windows / solid state|
|Previous:||From: mark||Date: 2010-11-23 01:11:49|
|Subject: Re: Could not open relation with OID (table partitioning issue?) |