On Wed, Jan 5, 2011 at 7:48 AM, David Fetter <da...@fetter.org> wrote:
> On Wed, Jan 05, 2011 at 09:04:08AM -0500, Robert Haas wrote:
>> On Tue, Jan 4, 2011 at 10:56 PM, Rob Wultsch <wult...@gmail.com> wrote:
>> > 1. Could the making a table logged be a non-exclusive lock if the
>> > ALTER is allowed to take a full checkpoint?
>> No, that doesn't solve either of the two problems I described,
>> unfortunately.

That is too bad.

>> > 2. Unlogged to logged has giant use case.
>> Agree.
>> > 3. In MySQL I have had to ALTER tables to engine BLACKHOLE because
>> > they held data that was not vital, but the server was out of IO.
>> > Going logged -> unlogged has a significant placed, I think.
>> Interesting.  So you'd change a logged table into an unlogged table
>> to cut down on I/O, and take the risk of losing the data if the
>> server went down?
> BLACKHOLE is a "storage engine" that's equivalent to /dev/null, so it
> wasn't a risk /per se/.

Exactly. It was data I could live without and by having schema
attached to /dev/null the application did not error out and die. It is
a very bad option and being able to turn off logging for a table is a
much better one.

Rob Wultsch

Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)
To make changes to your subscription:

Reply via email to