Morning all here is a snip of a couple of posts I put in the Suse group
but have not any answers, I have not had any time yet to delve into
this deeply and cannot seem to find any ref anywhere on the net.

Can anyone shed any light on this problem for me or any pointers on
where to start please?

<<--snip from post No.1-->>

I did an apt over the weekend and have been out since then, came back
today and found that gkrellm and artsd had spawned more procs than soft
mick completely eating all mem and swap.

If I start KDE all is ok but if I enable the " sound system " ( arts )
every system sound spawns a new process and arts does not kill itself
after the time out, if I start gkrellm it keeps spawning a new process
approx every 5 secs until all memory is filled.

All was working fine before.

I know for a fact I did not update either of the 2 apps.

Also have noticed that my Kweather applet is opening loads of http refs.

Thought it might be related to a couple of gnome upgrades so renamed
all .gnome* and related files in my home

Have created a new user with a complete blank sheet and it looks like
that is doing the same as well.

Even tried running Gnome but that was a complete mess, however gkrellm
stopped spawning as far as I could tell ??????

If it is down to a Gnome problem then why would Arts be affected? that
is what is puzzling me?


<<--snip from post No.2-->>

Just realised I still had arts running so decided to shut it down from
the Control centre, and got a crash warning did a backtrace and got
this warning:-


<<--snip-->>

This backtrace appears to be of no use.
This is probably because your packages are built in a way which prevents
creation of proper backtraces, or the stack frame was seriously
corrupted in the crash.

<<--pins-->>

Anyone?


As far as I can determine all rpm`s and deps are met on my machine and
no "experimental" stuff has��been�put�on�for�weeks.

Ps. it now seems that knode is spawning extra procs as well but not on a
"time" basis, but still up to 23 in the last 1/2 hour?????? looks like
it might be related to every time it does a group update/check so might
have to shut it down for a while.

<<--end snips-->>



This morning I have had knode open for approx 1 hour and it is upto 42
separate procs, the first 4 look normal with Vmsize and VmRss looking
fine but the rest have 0 VmSize and an VmRss of 12 ????

No other KDE apps seem to suffer however I have not had much time to run
them over the last few days.

Any ideas from the boffins ?

TIA

-- 
Mark
Iligitimi Non Carborundum!
Twixt hill and high water, N.Wales, UK
onfxvgpnvfr-ng-tzk-qbg-pb-hx

baskitcaise adjusted his/her tin foil beanie and asbestos underwear to
write:

Bad form answering myself but more info follows:-

It appears as far as I can tell ( still not had any time to delve
deeper ) that kdeinit is spawning "certain" procs above and beyond what
is needed, now if it was purely gnome apps then I could understand it
might have something to do with the updates, but it happens with only
certain KDE apps as well, knode starts spawning every time it checks
for new messages ( I think, as I have had no time to study it
properly )

Other apps like Kmail which do periodic checks seem OK, gkrellm spawns
from the time it is started approx every 5 secs until the memory is
full.

Artsd will start OK but if a sound is played, as in a system sound then
the arts is not killed on timeout and the next time a sound is played
another proc is started.

Anyone give me any ideas please.

I would love to try and chase this down myself but at the moment I have
not got much time, at the moment it means that this machine has to be
shut down as I cannot trust it yet it still seems to run fine with seti
and mailcheck 24/7 just certain apps.

Could it be in one of the kdeinit scripts?


TIA
-- 
Mark
Iligitimi Non Carborundum!
Twixt hill and high water, N.Wales, UK
onfxvgpnvfr-ng-tzk-qbg-pb-hx

baskitcaise adjusted his/her tin foil beanie and asbestos underwear to
write:

> 


For anyone that might have the same problem:-

Well after having a couple of hours to have a look today it seems that
libxslt was to fault, I have un-installed and made a couple of symlinks
to various libs which had been removed and all now seems to be working.

If anyone wants more info or suffers from the same then please get to me
before my cells dissapear :)

-- 
Mark
Iligitimi Non Carborundum!
Twixt hill and high water, N.Wales, UK
onfxvgpnvfr-ng-tzk-qbg-pb-hx

baskitcaise adjusted his/her tin foil beanie and asbestos underwear to
write:

> baskitcaise adjusted his/her tin foil beanie and asbestos underwear to
> write:
> 
>> 
> 
> 
> For anyone that might have the same problem:-
> 
> Well after having a couple of hours to have a look today it seems that
> libxslt was to fault, I have un-installed and made a couple of
> symlinks to various libs which had been removed and all now seems to
> be working.
> 
> If anyone wants more info or suffers from the same then please get to
> me before my cells dissapear :)
> 

For all that might have watched this thread, it appears that I had not
solved this, I came back last night and gkrellm and artd had started
spawning procs again eating all the mem and eventually chewing the mem
til there was none left.

The machine has been fine alll day ( today ) going through the cron
jobs, so it does not seem t come from them.

I have noticed though that ksetispy gives up after approx 10 mins but
the seti client still runs.

Have been running ksetispy for over 4 years here and cannot see the
relationship.

Anyone ????



- 
Mark
Iligitimi Non Carborundum!
Twixt hill and high water, N.Wales, UK
onfxvgpnvfr-ng-tzk-qbg-pb-hx

baskitcaise adjusted his/her tin foil beanie and asbestos underwear to
write:

Latest news,

It seems the same happens after 12 or maybe 18 hours of use, gkrellm and
artsd start to spawn procs all over the place until memory is all used
up ( not related to any cron job though )

If I stop gkrellm all the related procs are stopped,  I have to stop all
artsd procs manually but I lose the sounds.

Everything else seems ok ( at the moment :)

Thanks for listening.

Might get an hour or two to look into it in the morning if I can get out
of looking for wall paper :)


Anyone ??????

-- 
Mark
Iligitimi Non Carborundum!
Twixt hill and high water, N.Wales, UK
onfxvgpnvfr-ng-tzk-qbg-pb-hx

baskitcaise wrote:

> I have noticed though that ksetispy gives up after approx 10 mins but
> the seti client still runs.
> 
> Have been running ksetispy for over 4 years here and cannot see the
> relationship.
> 
> Anyone ????

Maybe the aliens have taken over your machine.

Dave

-- 
Faculty of Divinity, University of Cambridge
West Road, Cambridge, CB3 9BS
http://www.cus.cam.ac.uk/~djg39/

Dave Goode adjusted his/her tin foil beanie and asbestos underwear to
write:


> Maybe the aliens have taken over your machine.
> 
> Dave
> 

<Do Dee Do Dee, Do Dee Do Dee>

It certainly feels that way at the moment :)

It looks like it is something to do with the latest "bleeding edge"
issue of Suse`s KDE or QT, funny how the multi spawning seems to have
stopped now after another update but I still cannot get ksetispy to
keep monitoring, it works for approx 10 mins then just drops the client
even though the client is still running, ksetiwatch monitors it fine.

Anyway no great shakes, I was only using ksetispy for the cache, which I
suppose I don`t really need.

And now that BOINC seems a bit more stable I should move over to that.
 
-- 
Mark
Iligitimi Non Carborundum!
Twixt hill and high water, N.Wales, UK
onfxvgpnvfr-ng-tzk-qbg-pb-hx