On Sat, Feb 9, 2013 at 2:37 PM, Josh Krupka wrote:
> Johnny,
>
> Sure thing, here's the system tap script:
>
>
Thank you for this!
> - I think you already started looking at this, but the linux dirty memory
> settings may have to be tuned as well (see Greg's post
> http://notemagnet.blogspot.c
On Thu, Feb 7, 2013 at 11:16 PM, Tony Chan wrote:
> Hi,
>
> May I know what is your setting for OS cache?
>
>
Tony:
Wasn't sure if you were asking me, but here's the output from "free":
# free
total used free sharedbuffers cached
Mem: 198333224 187151280
Josh:
Are you able to share your systemtap script? Our problem will be to try and
regenerate the same amount of traffic/load that we see in production. We
could replay our queries, but we don't even capture a full set because it'd
be roughly 150GB per day.
johnny
On Thu, Feb 7, 2013 at 12:49 PM
I appreciate all the responses on this thread, even though some are
conflicting :). We are going to try these one at a time, but we'll likely
need a day or so inbetween each to see what impact (if any), so it will
take time. But I will post back here our findings.
We'll start with dirty_background
On Tue, Feb 5, 2013 at 11:23 PM, Josh Krupka wrote:
> I've been looking into something on our system that sounds similar to what
> you're seeing. I'm still researching it, but I'm suspecting the memory
> compaction that runs as part of transparent huge pages when memory is
> allocated... yet to
On Wed, Feb 6, 2013 at 2:13 PM, David Whittaker wrote:
> We disabled THP all together, with the thought that we might re-enable
> without defrag if we got positive results. At this point I don't think THP
> is the root cause though, so I'm curious to see if anyone else gets
> positive results fr
On Wed, Feb 6, 2013 at 7:49 AM, Kevin Grittner wrote:
> "a...@hsk.hk" wrote:
> > Johnny Tan wrote:
>
> >>shared_buffers = 48GB# min 128kB
>
> > From the postgresql.conf, I can see that the shared_buffers is
> > set to 48GB which is not small,
# cat /sys/kernel/mm/redhat_transparent_hugepage/defrag
[always] never
On Tue, Feb 5, 2013 at 5:37 PM, Josh Krupka wrote:
> Just out of curiosity, are you using transparent huge pages?
> On Feb 5, 2013 5:03 PM, "Johnny Tan" wrote:
>
>> Server specs:
>> Dell R61
Server specs:
Dell R610
dual E5645 hex-core 2.4GHz
192GB RAM
RAID 1: 2x400GB SSD (OS + WAL logs)
RAID 10: 4x400GB SSD (/var/lib/pgsql)
/etc/sysctl.conf:
kernel.msgmnb = 65536
kernel.msgmax = 65536
kernel.shmmax = 68719476736
kernel.shmall = 4294967296
vm.overcommit_memory = 0
vm.swappiness = 0
vm
