On Thu, Oct 29, 2015 at 05:23:56PM +0100, Andres Freund wrote:
> On 2015-10-29 12:54:22 +1100, Dave Chinner wrote:
> > On Thu, Oct 29, 2015 at 12:23:12AM +0100, Andres Freund wrote:
> > > By calling sync_file_range() over small ranges of pages shortly after
> > > they've been written we make it unl
On 2015-10-29 12:54:22 +1100, Dave Chinner wrote:
> On Thu, Oct 29, 2015 at 12:23:12AM +0100, Andres Freund wrote:
> > The blocking/latency of the fsync doesn't actually matter at all *for
> > this callsite*. It's called from a dedicated background process - if
> > it's slowed down by a couple seco
On Thu, Oct 29, 2015 at 12:23:12AM +0100, Andres Freund wrote:
> Hi,
>
> On 2015-10-29 07:48:34 +1100, Dave Chinner wrote:
> > > The idea of using SYNC_FILE_RANGE_WRITE beforehand is that
> > > the fsync() will only have to do very little work. The language in
> > > sync_file_range(2) doesn't insp
On Thu, Oct 29, 2015 at 07:48:34AM +1100, Dave Chinner wrote:
> Hi Andres,
>
> On Wed, Oct 28, 2015 at 10:27:52AM +0100, Andres Freund wrote:
> > On 2015-10-25 08:39:12 +1100, Dave Chinner wrote:
>
> > > Data integrity operations require related file metadata (e.g. block
> > > allocation trna
Hi,
On 2015-10-29 07:48:34 +1100, Dave Chinner wrote:
> > The idea of using SYNC_FILE_RANGE_WRITE beforehand is that
> > the fsync() will only have to do very little work. The language in
> > sync_file_range(2) doesn't inspire enough confidence for using it as an
> > actual integrity operation :/
Hi Andres,
On Wed, Oct 28, 2015 at 10:27:52AM +0100, Andres Freund wrote:
> On 2015-10-25 08:39:12 +1100, Dave Chinner wrote:
> > Data integrity operations require related file metadata (e.g. block
> > allocation trnascations) to be forced to the journal/disk, and a
> > device cache flush iss
Hi,
Thanks for looking into this.
On 2015-10-25 08:39:12 +1100, Dave Chinner wrote:
> WB_SYNC_ALL is simply a method of saying "writeback all dirty pages
> and don't skip any". That's part of a data integrity operation, but
> it's not what results in data integrity being provided. It may cause
>
Hi,
On Thu 22-10-15 15:15:55, Andres Freund wrote:
> postgres regularly has to checkpoint data to disk to be able to free
> data from its journal. We currently use buffered IO and that's not
> going to change short term.
>
> In a busy database this checkpointing process can write out a lot of
>
On Thu, Oct 22, 2015 at 03:15:55PM +0200, Andres Freund wrote:
> Hi,
>
> postgres regularly has to checkpoint data to disk to be able to free
> data from its journal. We currently use buffered IO and that's not
> going to change short term.
>
> In a busy database this checkpointing process can wr
Hi,
postgres regularly has to checkpoint data to disk to be able to free
data from its journal. We currently use buffered IO and that's not
going to change short term.
In a busy database this checkpointing process can write out a lot of
data. Currently that frequently leads to massive latency spi
10 matches
Mail list logo