Gentoo Forums
Gentoo Forums
Gentoo Forums
Quick Search: in
Intel XEON & Top [SOLVED]
View unanswered posts
View posts from last 24 hours

 
Reply to topic    Gentoo Forums Forum Index Duplicate Threads
View previous topic :: View next topic  
Author Message
bgregorcy
n00b
n00b


Joined: 19 May 2006
Posts: 61
Location: SLC

PostPosted: Wed Feb 03, 2010 6:28 pm    Post subject: Intel XEON & Top [SOLVED] Reply with quote

Hi I am seeing an odd issue with our Gentoo servers with the intel XEON processor. When I run `top` is shows that the machine is using almost 100% of the memory. Here is what the line looks like:
Code:

Mem:  16432312k total, 15968200k used,   464112k free,    29784k buffers


Free reports this:
Code:

humboldt ~ # free
             total       used       free     shared    buffers     cached
Mem:      16432312   15974760     457552          0      30440   15560136
-/+ buffers/cache:     384184   16048128
Swap:      2008116         32    2008084



But everything seems fine, the server is responsive. The only reason I noticed it is that a sys admin from a different department brought it up that the same thing is happening on there servers. The only difference being they are running Red Hat (and I am sure a different kernel), and we are running Gentoo (kernel-2.6.24-gentoo-r8)

If anybody know has any idea why we would be interested in the answer.

Thanks for the help


Last edited by bgregorcy on Tue Feb 09, 2010 6:56 pm; edited 2 times in total
Back to top
View user's profile Send private message
Veldrin
Veteran
Veteran


Joined: 27 Jul 2004
Posts: 1945
Location: Zurich, Switzerland

PostPosted: Wed Feb 03, 2010 7:21 pm    Post subject: Reply with quote

Code:
total = used + free
used = eff.used + cached + buffered


You got confused with what top regards as used. basically any non used memory is occupied (in the view of top).
But used may be really used (eff.used), cached or buffered.

To get back to your example:
total = 16432312k (16G)
free = 457552k (~450M)
used = 15974760k (~15.5G)
eff.used = 384184k (~380M)
cached = 15560136k (~15.1G)
buffered = 30440k (~30M)
and if you add those 3 numbers (eff.used + cached + buffered) you should end up with the same amount as used. (The numbers in brackets are not exact, but should give you some idea about the magnitude)

top itself is not that transparent. It misses the cached amount - which got you confused.


just my .02$
V.
Back to top
View user's profile Send private message
bgregorcy
n00b
n00b


Joined: 19 May 2006
Posts: 61
Location: SLC

PostPosted: Wed Feb 03, 2010 8:25 pm    Post subject: Xeon Reply with quote

But why are we not seeing the same issue on other CPU types, for example my desktop (Inter Core 2 Duo)


Code:

Mem:   8192732k total,  2299708k used,  5893024k free,    23208k buffers
Swap:  4008208k total,        0k used,  4008208k free,   881776k cached



Code:

harley # free
             total       used       free     shared    buffers     cached
Mem:       8192732    2324148    5868584          0      23544     885664
-/+ buffers/cache:    1414940    6777792
Swap:      4008208          0    4008208



Quote:

You got confused


No doubt about that
Back to top
View user's profile Send private message
Veldrin
Veteran
Veteran


Joined: 27 Jul 2004
Posts: 1945
Location: Zurich, Switzerland

PostPosted: Wed Feb 03, 2010 9:22 pm    Post subject: Reply with quote

hmm...
have you noted, that the cached entry is on the swap line - can you confirm that for the Xeon System? (Just checked 2 Core2, a Athlon64 and an PentiumIII System - They all seem fine)

cheers
V.
Back to top
View user's profile Send private message
bgregorcy
n00b
n00b


Joined: 19 May 2006
Posts: 61
Location: SLC

PostPosted: Wed Feb 03, 2010 9:34 pm    Post subject: Xeon Reply with quote

yep this is what top looks like for me:

Code:

Tasks: 242 total,   1 running, 241 sleeping,   0 stopped,   0 zombie
Cpu(s):  0.0%us,  0.1%sy,  0.0%ni, 99.6%id,  0.2%wa,  0.0%hi,  0.1%si,  0.0%st
Mem:  16432312k total, 15696120k used,   736192k free,    72788k buffers
Swap:  2008116k total,       32k used,  2008084k free, 14969756k cached


and free

Code:

humboldt gregorcy # free
             total       used       free     shared    buffers     cached
Mem:      16432312   15722504     709808          0      72896   14989576
-/+ buffers/cache:     660032   15772280
Swap:      2008116         32    2008084


Is that what you were looking for?
Back to top
View user's profile Send private message
bgregorcy
n00b
n00b


Joined: 19 May 2006
Posts: 61
Location: SLC

PostPosted: Wed Feb 03, 2010 9:39 pm    Post subject: Cpu Reply with quote

This is the cpuinfo:

Code:


humboldt gregorcy # cat /proc/cpuinfo
processor   : 0
vendor_id   : GenuineIntel
cpu family   : 6
model      : 23
model name   : Intel(R) Xeon(R) CPU           E5410  @ 2.33GHz
stepping   : 6
cpu MHz      : 2327.573
cache size   : 6144 KB
physical id   : 0
siblings   : 4
core id      : 0
cpu cores   : 4
fpu      : yes
fpu_exception   : yes
cpuid level   : 10
wp      : yes
flags      : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx lm constant_tsc arch_perfmon pebs bts rep_good pni monitor ds_cpl vmx est tm2 ssse3 cx16 xtpr dca sse4_1 lahf_lm
bogomips   : 4658.67
clflush size   : 64
cache_alignment   : 64
address sizes   : 38 bits physical, 48 bits virtual
power management:

processor   : 1
vendor_id   : GenuineIntel
cpu family   : 6
model      : 23
model name   : Intel(R) Xeon(R) CPU           E5410  @ 2.33GHz
stepping   : 6
cpu MHz      : 2327.573
cache size   : 6144 KB
physical id   : 1
siblings   : 4
core id      : 0
cpu cores   : 4
fpu      : yes
fpu_exception   : yes
cpuid level   : 10
wp      : yes
flags      : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx lm constant_tsc arch_perfmon pebs bts rep_good pni monitor ds_cpl vmx est tm2 ssse3 cx16 xtpr dca sse4_1 lahf_lm
bogomips   : 4655.22
clflush size   : 64
cache_alignment   : 64
address sizes   : 38 bits physical, 48 bits virtual
power management:

processor   : 2
vendor_id   : GenuineIntel
cpu family   : 6
model      : 23
model name   : Intel(R) Xeon(R) CPU           E5410  @ 2.33GHz
stepping   : 6
cpu MHz      : 2327.573
cache size   : 6144 KB
physical id   : 0
siblings   : 4
core id      : 1
cpu cores   : 4
fpu      : yes
fpu_exception   : yes
cpuid level   : 10
wp      : yes
flags      : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx lm constant_tsc arch_perfmon pebs bts rep_good pni monitor ds_cpl vmx est tm2 ssse3 cx16 xtpr dca sse4_1 lahf_lm
bogomips   : 4655.22
clflush size   : 64
cache_alignment   : 64
address sizes   : 38 bits physical, 48 bits virtual
power management:

Back to top
View user's profile Send private message
Veldrin
Veteran
Veteran


Joined: 27 Jul 2004
Posts: 1945
Location: Zurich, Switzerland

PostPosted: Wed Feb 03, 2010 9:42 pm    Post subject: Reply with quote

yep - and they look exactly how they should....

And (I should have explained better) it is rather normal that the System uses all its available RAM (at least after some time). Caches and Buffers (don't ask for the exact difference) are used to keep data from the HDD in RAM to have faster access to it. So the speak, RAM is used for 2 applications: first to run a program (as in eff.used, check 2nd line in free) and second to cache/buffer data to be accessed faster.
On you Xeon System, most RAM is used for buffer/cache (some 15GB!) while a rather small amount is really needed to run apps (~380MB). In short, your Xeon System is not using all RAM (for running apps).
But it is true, that most RAM is occupied by some data.

I hope that clears it up a little.
cheers
V.
Back to top
View user's profile Send private message
Mad Merlin
Veteran
Veteran


Joined: 09 May 2005
Posts: 1155

PostPosted: Fri Feb 05, 2010 2:37 am    Post subject: Reply with quote

You will find this helpful: http://www.linuxatemyram.com/
_________________
Game! - Where the stick is mightier than the sword!
Back to top
View user's profile Send private message
bgregorcy
n00b
n00b


Joined: 19 May 2006
Posts: 61
Location: SLC

PostPosted: Mon Feb 08, 2010 9:58 pm    Post subject: thx Reply with quote

Cool thanks for the help
Back to top
View user's profile Send private message
linuxtuxhellsinki
l33t
l33t


Joined: 15 Nov 2004
Posts: 700
Location: Hellsinki

PostPosted: Tue Feb 09, 2010 6:54 pm    Post subject: Re: thx Reply with quote

bgregorcy wrote:
Cool thanks for the help

You should edit your 1st post's subject line and add [Solved] to it :wink:


MadMerlin: http://www.linuxatemyram.com/ Great link :P
_________________
1st use 'Search' & lastly add [Solved] to
the subject of your first post in the thread.
Back to top
View user's profile Send private message
timeBandit
Bodhisattva
Bodhisattva


Joined: 31 Dec 2004
Posts: 2719
Location: here, there or in transit

PostPosted: Tue Feb 09, 2010 7:41 pm    Post subject: Reply with quote

Moved from Kernel & Hardware to Duplicate Threads.

See link above and Linux Memory Management or 'Why is there no free RAM?'
_________________
Plants are pithy, brooks tend to babble--I'm content to lie between them.
Super-short f.g.o checklist: Search first, strip comments, mark solved, help others.
Back to top
View user's profile Send private message
Display posts from previous:   
Reply to topic    Gentoo Forums Forum Index Duplicate Threads All times are GMT
Page 1 of 1

 
Jump to:  
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum