| 1 |
On Wednesday 25 May 2011 17:49:29 Alan McKinnon wrote: |
| 2 |
> Apparently, though unproven, at 16:20 on Wednesday 25 May 2011, Alex |
| 3 |
> Schuster |
| 4 |
> |
| 5 |
> did opine thusly: |
| 6 |
> > Oh, even weirder: The phone just rang, and five minutes later, swap has |
| 7 |
> > gone to 860M. I was running rdiff-backup --list-increment-sizes, maybe |
| 8 |
> > this uses much memory, and caches the stuff. Now the command has |
| 9 |
> > finished, and paging has stopped. The rdiff-backup process itself does |
| 10 |
> > not use much memory. |
| 11 |
> > |
| 12 |
> > BTW, does anyone else's kwin use 750M? That's pretty high, I think it |
| 13 |
> > used to be more like 300M. |
| 14 |
> |
| 15 |
> I've been noticing this kind of thing too for a while now. In my case it's |
| 16 |
> the nepomuk/akonadi/virtuoso stack doing it - it seems to trigger full |
| 17 |
> scans at weird times and does other special things after a resume from |
| 18 |
> suspend. |
| 19 |
> |
| 20 |
> Virtuoso can sometimes get as high as 800M RES memory in top. Which is all |
| 21 |
> quite bizarre, I suspect a dodgy config on my part. |
| 22 |
> |
| 23 |
> As for kwin - what column are you reading the value from? Here kwin uses |
| 24 |
> more like 60M |
| 25 |
|
| 26 |
This is kwin usage from a 32bit Pentium 4 box with 3G of RAM: |
| 27 |
|
| 28 |
VIRT SHR RES |
| 29 |
143m 22m 29m |
| 30 |
|
| 31 |
At this moment I am compiling chromium (which will take close to 2 hours) and |
| 32 |
it's eaten up 1560K swap. Under normal usage the 3G or RAM is more than |
| 33 |
adequate. Small amounts of swapping happen only when I emerge something large |
| 34 |
(e.g. OOo) or when I fire up VirtualBox and have umpteen apps open, browsers |
| 35 |
with dozens of tabs, etc. |
| 36 |
|
| 37 |
PS. I have switched off desktop search and although I can see akonadi/nepomuk |
| 38 |
there's no virtuoso, strigi or other such stuff showing up. |
| 39 |
-- |
| 40 |
Regards, |
| 41 |
Mick |
Archives