Thread: commit_delay and commit_siblings -- restart or reload?
The documentation for commit_delay and commit_siblings doesn't mention whether a reload or restart is required to set them. I recommend something like the trivial attached docs patch to make it clear(er) that a reload is required. -- Joshua Tolley / eggyknap End Point Corporation http://www.endpoint.com
Attachment
On Tue, Jun 1, 2010 at 11:51 AM, Joshua Tolley <eggyknap@gmail.com> wrote: > The documentation for commit_delay and commit_siblings doesn't mention whether > a reload or restart is required to set them. I recommend something like the > trivial attached docs patch to make it clear(er) that a reload is required. These parameters are PGC_USERSET - a reload is not the only way to change them. Of course, changing any parameter in postgresql.conf won't take effect unless you reload, but the point of that sentence, in the other places where we use it, is that you can't change those parameters using things like ALTER DATABASE ... SET, ALTER USER ... SET, just plain SET, etc. -- Robert Haas EnterpriseDB: http://www.enterprisedb.com The Enterprise Postgres Company
On Tue, Jun 01, 2010 at 03:33:24PM -0400, Robert Haas wrote: > On Tue, Jun 1, 2010 at 11:51 AM, Joshua Tolley <eggyknap@gmail.com> wrote: > > The documentation for commit_delay and commit_siblings doesn't mention whether > > a reload or restart is required to set them. I recommend something like the > > trivial attached docs patch to make it clear(er) that a reload is required. > > These parameters are PGC_USERSET - a reload is not the only way to change them. > > Of course, changing any parameter in postgresql.conf won't take effect > unless you reload, but the point of that sentence, in the other places > where we use it, is that you can't change those parameters using > things like ALTER DATABASE ... SET, ALTER USER ... SET, just plain > SET, etc. Ah, you're right. My mistake -- thanks. -- Joshua Tolley / eggyknap End Point Corporation http://www.endpoint.com