Used Memory Not Clearing Down On Redhat Machines.

[Smith, Albert]

> -----Original Message-----
> From: redhat-list-bounces at redhat.com 
> [mailto:redhat-list-bounces at redhat.com] On Behalf Of Tolga Evren
> Sent: Monday, September 19, 2005 3:22 AM
> To: General Red Hat Linux discussion list
> Subject: RE: Used Memory Not Clearing Down On Redhat Machines.
> 
> Hi ,
> 
> Look at my memifo pls:
> 
>         total:    used:    free:  shared: buffers:  cached:
> Mem:  8490827776 8118534144 372293632        0 252133376 4320608256
> Swap: 25165725696 113704960 25052020736
> MemTotal:      8291824 kB
> MemFree:        363568 kB
> MemShared:           0 kB
> Buffers:        246224 kB
> Cached:        4215008 kB
> SwapCached:       4336 kB
> Active:        2952512 kB
> Inact_dirty:   2277600 kB
> Inact_clean:    211712 kB
> Inact_target:  1088352 kB
> HighTotal:           0 kB
> HighFree:            0 kB
> LowTotal:      8291824 kB
> LowFree:        363568 kB
> SwapTotal:    24575904 kB
> SwapFree:     24464864 kB
> Committed_AS: 621197056 kB
> HugePages_Total:     0
> HugePages_Free:      0
> Hugepagesize:    262144 kB
> 
> 
> The inact_dirty is 2277600 kB .  
> 
> > http://www.redhat.com/magazine/001nov04/features/vm/
> > Pages in the inactive dirty state ought to be getting 
> flushed out to 
> > disk, but it looks like this isn't happening.
> > 
> > I'm no kernel expert, but it seems to me that something is 
> causing the 
> > kernel not to properly launder the inactive dirty memory.  
> This could 
> > be a kernel bug, so if you are sure you don't have any hung 
> processes 
> > that could be holding the memory open, you should check to see if 
> > there is a newer kernel available for your system.
> 
> 
> How can i be sure that this is a bug?
> 
> Kind Regards,
> Tolga Evren
> 
> 
> 
> 
> -----Original Message-----
> From: redhat-list-bounces at redhat.com
> [mailto:redhat-list-bounces at redhat.com] On Behalf Of Smith, Albert
> Sent: Thursday, September 15, 2005 8:35 PM
> To: General Red Hat Linux discussion list
> Subject: RE: Used Memory Not Clearing Down On Redhat Machines.
> 
> > -----Original Message-----
> > From: redhat-list-bounces at redhat.com
> > [mailto:redhat-list-bounces at redhat.com] On Behalf Of Eris Caffee
> > Sent: Friday, August 12, 2005 11:17 AM
> > To: General Red Hat Linux discussion list
> > Subject: Re: Used Memory Not Clearing Down On Redhat Machines.
> > 
> > > 15:20:33  up 14 days,  2:16,  1 user,  load average: 0.00,
> > 0.00, 0.00
> > > 49 processes: 48 sleeping, 1 running, 0 zombie, 0 stopped
> > > CPU states:  cpu    user    nice  system    irq  softirq  
> > iowait    idle
> > >            total    0.8%    0.0%    0.2%   0.0%     0.6%    
> > 0.0%  198.4%
> > >            cpu00    0.0%    0.0%    0.0%   0.0%     0.0%    
> > 0.0%  100.0%
> > >            cpu01    0.8%    0.0%    0.2%   0.0%     0.6%    
> > 0.0%   98.4%
> > > Mem:  3599032k av, 2926964k used,  672068k free,       0k 
> > shrd,   82052k
> > > buff
> > >                     469844k actv, 1731916k in_d,   54940k in_c
> > > Swap: 2044072k av,       0k used, 2044072k free             
> >     2614264k
> > > cached
> > 
> > You have an awful lot of memory in the in_d state.  That is 
> "Inactive 
> > Dirty".  There's a description of it and the other memory 
> states (for 
> > 2.4 kernels at least) here:
> > http://www.redhat.com/magazine/001nov04/features/vm/
> > Pages in the inactive dirty state ought to be getting 
> flushed out to 
> > disk, but it looks like this isn't happening.
> > 
> > I'm no kernel expert, but it seems to me that something is 
> causing the 
> > kernel not to properly launder the inactive dirty memory.  
> This could 
> > be a kernel bug, so if you are sure you don't have any hung 
> processes 
> > that could be holding the memory open, you should check to see if 
> > there is a newer kernel available for your system.
> > 
> > Of course, you should also double check to make sure there really 
> > isn't some process from the analysis job still running and 
> holding all 
> > of this memory open.
> > 
> > Eris Caffee
> > 
> > 
> > --
> > redhat-list mailing list
> > unsubscribe 
> mailto:redhat-list-request at redhat.com?subject=unsubscribe
> > https://www.redhat.com/mailman/listinfo/redhat-list
> > 
> > 
> 
> I did some investigation with RH EL4 and it looks like this 
> memory problem is addressed with 2.6.x line of kernel. I am 
> currently compiling
> 2.6.13 manually on my EL3 to see the results.
> 
> I will keep you posted.
> 
> Albert Smith
> Sr. Unix Systems Administrator
> HPCSA, RHCT
> Genex Services
> 440 E. Swedesford Rd.
> Wayne, PA 19087
> albert.smith at genexservices.com
> (610) 964-5154
> 
> 
> 
> --
> redhat-list mailing list
> unsubscribe mailto:redhat-list-request at redhat.com?subject=unsubscribe
> https://www.redhat.com/mailman/listinfo/redhat-list
> 
> --
> redhat-list mailing list
> unsubscribe mailto:redhat-list-request at redhat.com?subject=unsubscribe
> https://www.redhat.com/mailman/listinfo/redhat-list
> 
> 

I have seen this on my systems running oracle and/or dell openmange
software. So far I removed the dell openmange system and I am currently
doing some testing under 2.6 kernel. If you want to see if there is an
official bug listed check bugzilla.redhat.com but I don't think I saw
one listed as of this time, but acording to dell/redhat the problem is
with kernel the 2.4 kernel.

Albert Smith
Sr. Unix Systems Administrator
HPCSA, RHCT  
Genex Services
440 E. Swedesford Rd.
Wayne, PA 19087
albert.smith at genexservices.com
(610) 964-5154