On 2005 June 20 Monday 16:49, Ian Eure wrote:
> The 3.4.0 and 3.4.1 packages on alioth seem to have a pretty substantial
> memory leak in kded. Over the course of a week or so of uptime, kded grew
> to eat up over 500mb of memory, and bogged the system down severely. I
> closed /all/ open apps, and the size didn't decrease substantially.
>
> Right now, kded is eating 241mb of ram, and I've only been logged in since
> yesterday.
>
> Anyone else see this?

Yep. From top, sorting by "%MEM":

PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
9379 zack 15 0 2202m 409m 8640 S 0.0 40.4 9:31.38 kded
9070 root 6 -10 100m 64m 2300 S 1.2 6.4 12:53.23 XFree86
[snip]

I've got 1024MB of RAM, so it looks like kded is using over 400MB. The next
greediest process is XFree86, using about 60MB. Not cool.

Zack


--
To UNSUBSCRIBE, email to debian-kde-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org

On Tuesday, 21 de June de 2005 00:21, Zack Cerza shaped the electrons to
shout:
> Yep. From top, sorting by "%MEM":
>
> * PID USER * * *PR *NI *VIRT *RES *SHR S %CPU %MEM * *TIME+ *COMMAND
> *9379 zack * * *15 * 0 2202m 409m 8640 S *0.0 40.4 * 9:31.38 kded
> *9070 root * * * 6 -10 *100m *64m 2300 S *1.2 *6.4 *12:53.23 XFree86
> [snip]
>
> I've got 1024MB of RAM, so it looks like kded is using over 400MB. The
> next greediest process is XFree86, using about 60MB. Not cool.

top is not the right tool to measure memory consumption. OTH the xfree
memoryu consuption issue has been clarified several times. One is the top
misleading figures, the others is that it counts the graphic card memory.


--
ricardo galli GPG id C8114D34
http://mnm.uib.es/gallir/


--
To UNSUBSCRIBE, email to debian-kde-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org

* Zack Cerza [Mon, 20 Jun 2005 18:21:31 -0400]:

> > Anyone else see this?

> Yep. From top, sorting by "%MEM":

I can't confirm (session running for two days, will keep an eye during the week):

PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
5219 adeodato 16 0 74744 14m 10m S 0.0 2.8 1:10.90 kded


--
Adeodato Simó
EM: asp16 [ykwim] alu.ua.es | PK: DA6AE621

The problem I have with making an intelligent statement is that some
people then think that it's not an isolated occurrance.
-- Simon Travaglia


--
To UNSUBSCRIBE, email to debian-kde-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org

On 2005 June 20 Monday 18:29, Ricardo Galli wrote:
> On Tuesday, 21 de June de 2005 00:21, Zack Cerza shaped the electrons to
>
> shout:
> > Yep. From top, sorting by "%MEM":
> >
> > * PID USER * * *PR *NI *VIRT *RES *SHR S %CPU %MEM * *TIME+ *COMMAND
> > *9379 zack * * *15 * 0 2202m 409m 8640 S *0.0 40.4 * 9:31.38 kded
> > *9070 root * * * 6 -10 *100m *64m 2300 S *1.2 *6.4 *12:53.23 XFree86
> > [snip]
> >
> > I've got 1024MB of RAM, so it looks like kded is using over 400MB. The
> > next greediest process is XFree86, using about 60MB. Not cool.
>
> top is not the right tool to measure memory consumption. OTH the xfree
> memoryu consuption issue has been clarified several times. One is the top
> misleading figures, the others is that it counts the graphic card memory.

Er, I know that - but it's still ridiculous, no?

What I don't know (and you didn't say) is what the right tool is.

Zack

On Monday 20 June 2005 07:31 pm, Zack Cerza wrote:
> On 2005 June 20 Monday 18:29, Ricardo Galli wrote:
> > On Tuesday, 21 de June de 2005 00:21, Zack Cerza shaped the electrons to
> >
> > shout:
> > > Yep. From top, sorting by "%MEM":
> > >
> > > * PID USER * * *PR *NI *VIRT *RES *SHR S %CPU %MEM * *TIME+ *COMMAND
> > > *9379 zack * * *15 * 0 2202m 409m 8640 S *0.0 40.4 * 9:31.38 kded
> > > *9070 root * * * 6 -10 *100m *64m 2300 S *1.2 *6.4 *12:53.23 XFree86
> > > [snip]
> > >
> > > I've got 1024MB of RAM, so it looks like kded is using over 400MB. The
> > > next greediest process is XFree86, using about 60MB. Not cool.
> >
> > top is not the right tool to measure memory consumption. OTH the xfree
> > memoryu consuption issue has been clarified several times. One is the top
> > misleading figures, the others is that it counts the graphic card memory.
>
> Er, I know that - but it's still ridiculous, no?
>
> What I don't know (and you didn't say) is what the right tool is.
>
FWIW, I was using `ps axu | grep kded'. But it seems to give the same number
as top, so... I don't know.

Anyone?

On 6/21/05, Zack Cerza wrote:
> On 2005 June 20 Monday 18:29, Ricardo Galli wrote:
> > top is not the right tool to measure memory consumption. OTH the xfree
> > memoryu consuption issue has been clarified several times. One is the top
> > misleading figures, the others is that it counts the graphic card memory.
> Er, I know that - but it's still ridiculous, no?

It's not. In easy terms: You count your swap space as "memory" as
well, it's simply mapped into the addressable space of your system.
It's similar for the graphic card's memory.

> What I don't know (and you didn't say) is what the right tool is.

You could simply subtract the size of your card's memory from the
value you get with top or ps.
For more detailed information on memory usage:
# cat /proc/`pidof X`/status | grep ^Vm


Björn

On Tuesday 21 June 2005 01:54, Adeodato Simó wrote:
> * Zack Cerza [Mon, 20 Jun 2005 18:21:31 -0400]:
> > > Anyone else see this?
> >
> > Yep. From top, sorting by "%MEM":
>
> I can't confirm (session running for two days, will keep an eye during
> the week):
>
> PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
> 5219 adeodato 16 0 74744 14m 10m S 0.0 2.8 1:10.90 kded
>
>
> --
> Adeodato Simó
> EM: asp16 [ykwim] alu.ua.es | PK: DA6AE621
>
> The problem I have with making an intelligent statement is that some
> people then think that it's not an isolated occurrance.
> -- Simon Travaglia

I can't confirm a memory leak either (kded running for 5 days, kde-3.4.1):

# ps -o pid,cmd,rss,sz,vsz,pmem,lstart -C kded
PID CMD RSS SZ VSZ %MEM STARTED
3610 kded [kdeinit] 18308 8821 35284 1.7 Thu Jun 16 13:44:14 2005


Mihai


--
This message was scanned for spam and viruses by BitDefender.
For more information please visit http://www.bitdefender.com/