| From: | Daniel Farina <daniel(at)heroku(dot)com> |
|---|---|
| To: | Jeff Davis <pgsql(at)j-davis(dot)com> |
| Cc: | Simon Riggs <simon(at)2ndquadrant(dot)com>, Heikki Linnakangas <heikki(dot)linnakangas(at)enterprisedb(dot)com>, pgsql-hackers <pgsql-hackers(at)postgresql(dot)org> |
| Subject: | Re: Sync Rep v17 |
| Date: | 2011-02-26 01:32:03 |
| Message-ID: | AANLkTimnYtVHuzw48FXV=rDDd1bPdNcJ-2-ScxXj3GwT@mail.gmail.com |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
On Fri, Feb 25, 2011 at 4:52 PM, Jeff Davis <pgsql(at)j-davis(dot)com> wrote:
> On Wed, 2011-02-23 at 22:42 -0800, Daniel Farina wrote:
>> Oh, yes, this reproduces past shutdowns/startups, and there's quite a
>> few txids before I catch up. I'm also comfortable poking around with
>> gdb (I have already recompiled with debugging symbols and
>> optimizations off and was poking around, especially at
>> MemoryContextStats(TopMemoryContext), but was not rewarded.
>
> Where is all of that memory going during recovery? Recovery shouldn't
> use much memory at all, as far as I can tell.
>
> What's even allocating memory at all?
I noticed this is RSS fooling with me. As pages get touched in shared
memory, for some reason RSS was constantly getting increased, along
with SHR at the same time.
Still, the long recovery time was mystifying to me, considering the
lack of unclean shutdowns.
--
fdr
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Daniel Farina | 2011-02-26 01:35:17 | Re: disposition of remaining patches |
| Previous Message | Josh Berkus | 2011-02-26 01:21:50 | Re: disposition of remaining patches |