| From: | Greg Smith <gsmith(at)gregsmith(dot)com> |
|---|---|
| To: | |
| Cc: | pgsql-hackers(at)postgresql(dot)org |
| Subject: | Re: PANIC caused by open_sync on Linux |
| Date: | 2007-10-27 02:39:12 |
| Message-ID: | Pine.GSO.4.64.0710262218020.28838@westnet.com |
| Views: | Whole Thread | Raw Message | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
On Fri, 26 Oct 2007, Tom Lane wrote:
>> The 8.3 bgwriter keeps running even during checkpoints, so it's feasible
>> to add such a feature now.
> I wonder though whether the walwriter wouldn't be a better place for it.
I do, too, but that wasn't available until too late in the 8.3 cycle to
consider adding this feature to there either.
There's a couple of potential to-do list ideas that build on the changes
in this area in 8.3:
-Aggressively pre-allocate WAL segments
-Space out checkpoint fsync requests in addition to disk writes
-Consider re-inserting a smarter bgwriter all-scan that writes sorted by
usage count during idle periods
--
* Greg Smith gsmith(at)gregsmith(dot)com http://www.gregsmith.com Baltimore, MD
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Camilo Porto | 2007-10-27 02:46:42 | URGENT HELP about 'duration' stats |
| Previous Message | Tom Lane | 2007-10-27 02:06:19 | Avoiding planning redundant backwards merges |