[lug] load average constant at 2

Atkinson, Chip CAtkinson at Circadence.com
Thu Sep 14 15:25:54 MDT 2000


It means that it checks out ok.  The man page says that all discrepancies
are displayed, implying that no discrepencies are not displayed.  

Ok, do you have network access?  can you ping someone like yahoo.com?
Sometimes network timeouts can cause a problem.  Also, what do you see when
you run top for a while?  You could go through and start killing processes
and watch your load.  That often works for me when I've tried everything
else.

Chip

-----Original Message-----
From: Jason Bowen [mailto:Jason.Bowen at Colorado.EDU]
Sent: Thursday, September 14, 2000 3:21 PM
To: 'lug at lug.boulder.co.us'
Subject: RE: [lug] load average constant at 2


On Thu, 14 Sep 2000, Atkinson, Chip wrote:

> How about rpm --verify procps-2.0.6-12mdk?

Well that is the filename, I had mistyped earlier but I get nothing
printed out.  Does that mean it checks out?  The man page says nothing
about having no output from the run.

> 
> Here's a script that goes through /proc to see if ps returns every process
> found in /proc:
> 
> for i in $(ls -d /proc/[0-9][0-9]*); do
> PROC=$(ps auxw | grep $(basename $i))
> if [ -z "$PROC" ]; then
>   echo $PROC is not shown by ps
> fi
> done

Nothing was printed.

> 
> -----Original Message-----
> >From: Jason Bowen [mailto:Jason.Bowen at Colorado.EDU]
> Sent: Thursday, September 14, 2000 3:07 PM
> To: 'lug at lug.boulder.co.us'
> Subject: RE: [lug] load average constant at 2
> 
> 
> On Thu, 14 Sep 2000, Atkinson, Chip wrote:
> 
> > That's odd.  What's rpm -qa | grep ps show?
> > 
> 
> getty_ps-2.0.7j-9mdk
> psmisc-18-2mdk
> tetex-dvips-1.0.6-6mdk
> psutils-p17-3mdk
> xlispstat-3.52.9-4mdk
> gnapster-1.3.12-0mdk_helix_1
> procps-X11-2.0.6-5mdk
> psacct-6.3-2mdk
> procps-2.0.6-12mdk
> 
> 
> > -----Original Message-----
> > >From: Jason Bowen [mailto:Jason.Bowen at Colorado.EDU]
> > Sent: Thursday, September 14, 2000 3:03 PM
> > To: lug at lug.boulder.co.us
> > Subject: RE: [lug] load average constant at 2
> > 
> > 
> > I had reinstalled procps but here in intersting thing to me.
> > 
> > [/bin]# rpm -qf ps
> > procps-2.0.6-5mdk
> > [/bin]# rpm --verify procps-2.0.6-5mdk
> > package procps-2.0.6-5mdk is not installed
> > 
> > 
> > On Thu, 14 Sep 2000, Atkinson, Chip wrote:
> > 
> > > The crude version of the ps patch that was installed on my machine was
> > "too
> > > small".  If you 
> > > ls -l $(which ps) and see something that's around 6K bytes, rather
than
> > 60K
> > > bytes, it might be a tipoff.  You can use RPM to help perhaps:
> > > rpm -q --whatprovides /bin/ps
> > > returns something like procps-2.0.6-5
> > > followed by 
> > > rpm --verify procps-2.0.6-5
> > > 
> > > Chip
> > > 
> > > -----Original Message-----
> > > >From: Jason Bowen [mailto:Jason.Bowen at Colorado.EDU]
> > > Sent: Thursday, September 14, 2000 2:47 PM
> > > To: 'lug at lug.boulder.co.us'
> > > Subject: RE: [lug] load average constant at 2
> > > 
> > > 
> > > ps doesn't show anything unusual either.  I thought about an intruder
> but
> > > I don't see anything to be honest.
> > > 
> > > On Thu, 14 Sep 2000, Atkinson, Chip wrote:
> > > 
> > > > What does ps show?  Any chance there is a secret process that's
> running
> > as
> > > > the result of an intrusion?  It was something I found on my machines
> > > anyway.
> > > > It didn't appear to affect the load too much though.  When I see
that
> > > > problem, it's often something like netscape.
> > > > 
> > > > Chip
> > > > 
> > > > -----Original Message-----
> > > > >From: Jason Bowen [mailto:Jason.Bowen at Colorado.EDU]
> > > > Sent: Thursday, September 14, 2000 2:38 PM
> > > > To: lug at lug.boulder.co.us
> > > > Subject: [lug] load average constant at 2
> > > > 
> > > > 
> > > > For about 2 weeks now, at least that is when I noticed it first, my
> load
> > > > average sits at 2.  This is with top reporting the idle processes at
> > close
> > > > to 100%.  When I reboot the machine the 5 and 15 minute load
averages
> > > > slowly creep towards 2.00 while the 1 minute average drops to 2.00
> > > > after letting the machine idle after I login at the console.  Has
> > anybody
> > > > seen something like this before?
> > > > Jason
> > > > 
> > > > 
> > > > _______________________________________________
> > > > Web Page:  http://lug.boulder.co.us
> > > > Mailing List: http://lists.lug.boulder.co.us/mailman/listinfo/lug
> > > > 
> > > > _______________________________________________
> > > > Web Page:  http://lug.boulder.co.us
> > > > Mailing List: http://lists.lug.boulder.co.us/mailman/listinfo/lug
> > > > 
> > > 
> > > 
> > > _______________________________________________
> > > Web Page:  http://lug.boulder.co.us
> > > Mailing List: http://lists.lug.boulder.co.us/mailman/listinfo/lug
> > > 
> > > _______________________________________________
> > > Web Page:  http://lug.boulder.co.us
> > > Mailing List: http://lists.lug.boulder.co.us/mailman/listinfo/lug
> > > 
> > 
> > 
> > 
> > _______________________________________________
> > Web Page:  http://lug.boulder.co.us
> > Mailing List: http://lists.lug.boulder.co.us/mailman/listinfo/lug
> > 
> > _______________________________________________
> > Web Page:  http://lug.boulder.co.us
> > Mailing List: http://lists.lug.boulder.co.us/mailman/listinfo/lug
> > 
> 
> 
> _______________________________________________
> Web Page:  http://lug.boulder.co.us
> Mailing List: http://lists.lug.boulder.co.us/mailman/listinfo/lug
> 
> _______________________________________________
> Web Page:  http://lug.boulder.co.us
> Mailing List: http://lists.lug.boulder.co.us/mailman/listinfo/lug
> 


_______________________________________________
Web Page:  http://lug.boulder.co.us
Mailing List: http://lists.lug.boulder.co.us/mailman/listinfo/lug




More information about the LUG mailing list