Re: postgresql recommendation memory

On Thu, Nov 7, 2013 at 2:13 AM, Евгений Селявка <evg(dot)selyavka(at)gmail(dot)com>wrote:

> All my sar statistics
>
...

> sar -u ALL
> 11:40:02 AM CPU %usr %nice %sys %iowait %steal
> %irq %soft %guest %idle
> 01:15:01 PM all 8.57 0.00 1.52 1.46 0.00
> 0.00 0.05 0.00 88.40
> 01:20:01 PM all 8.50 0.00 1.53 1.61 0.00
> 0.00 0.05 0.00 88.31
> 01:25:01 PM all 9.00 0.00 1.78 1.27 0.00
> 0.00 0.06 0.00 87.89
> 01:30:01 PM all 9.58 0.00 1.63 1.71 0.00
> 0.00 0.06 0.00 87.01
> 01:35:01 PM all 8.75 0.00 1.47 1.57 0.00
> 0.00 0.06 0.00 88.15
>

Did a freeze-up occur in there someplace? Otherwise, that is not not so
useful.

You should try to decrease the sar interval to 1 min if you can. The extra
overhead is negligible and the extra information can be very valuable. I'd
also have something like "vmstat 5" running and capture that. Although
perhaps one of the options to sar other than -u capture that same
information, I know little of the other sar options.

Cheers,

Jeff