Re: [CentOS] Weird output of system load

2014-09-12 Thread Kai Schaetzl
Mingfei Hua wrote on Thu, 11 Sep 2014 15:05:33 +:

> This issue happened again, who can help on this? system load up to
> 72, but /proc/stat only show that 2 processes were running and no
> process blocked

As others have already said, you are not giving and you are not looking at 
relevant information. You want to look in top at the running processes, 
for instance. It doesn't matter at all that you have "only 2" processes 
running. As it seems you don't even know which were running. Find out!

Btw, quoting old and new information adds confusion.

The number of sleeping *2534* processes seems a bit high for me. What the 
hell have you running there in the background on a "tiny" 4 GB machine 
with "only" two CPUs/cores? You can either have a few (or just one) 
processes needing a lot of ressources or many processes waking up, each 
one just needing a little bit, but in total they count ...


Kai


___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos


Re: [CentOS] Weird output of system load

2014-09-12 Thread Leon Fauster
Am 11.09.2014 um 21:05 schrieb m.r...@5-cent.us:
> Leon Fauster wrote:
>> Am 11.09.2014 um 17:05 schrieb Mingfei Hua :
>>> This issue happened again, who can help on this? system load up to 72,
>>> but /proc/stat only show that 2 processes were running and no process
>>> blocked. could it be a bug?
>> 
>> show us
>> 
>> $ ps r -A
>> 
> The load depends on what the running jobs are *doing*. I *frequently* see
> loads of 70 and some over 100, on servers with 48 or 64 cores... but we do
> *really* serious number crunching around here.



for such systems an effective load between 1,56 - 2,1 

:-)

--
LF


___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos


Re: [CentOS] Weird output of system load

2014-09-11 Thread m . roth
Leon Fauster wrote:
> Am 11.09.2014 um 17:05 schrieb Mingfei Hua :
>> This issue happened again, who can help on this? system load up to 72,
>> but /proc/stat only show that 2 processes were running and no process
>> blocked. could it be a bug?
>
> show us
>
> $ ps r -A
>
The load depends on what the running jobs are *doing*. I *frequently* see
loads of 70 and some over 100, on servers with 48 or 64 cores... but we do
*really* serious number crunching around here.

  mark

___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos


Re: [CentOS] Weird output of system load

2014-09-11 Thread Leon Fauster
Am 11.09.2014 um 17:05 schrieb Mingfei Hua :
> This issue happened again, who can help on this? system load up to 72, but 
> /proc/stat only show that 2 processes were running and no process blocked. 
> could it be a bug?


show us 

$ ps r -A

--
LF
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos


Re: [CentOS] Weird output of system load

2014-09-11 Thread Mingfei Hua
This issue happened again, who can help on this? system load up to 72, but 
/proc/stat only show that 2 processes were running and no process blocked. 
could it be a bug?

/proc/stat
cpu  5584507 61716 7213361 166019596 244884 146923 396136 0 0
cpu0 4186208 9686 2596560 81685397 112653 146699 365382 0 0
cpu1 1398298 52029 4616801 84334199 132230 224 30753 0 0
intr 1121292737 139 8 0 0 0 0 0 0 0 0 0 0 106 0 0 73 0 0 550904491 0 0 0 0 0 0 
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 4022511 0 0 0 0 0 
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
0 0
ctxt 1114886293
btime 1409530629
processes 10263085
procs_running 2
procs_blocked 0
softirq 1469123445 0 380686538 182314 632630211 3763285 0 2 56074856 1852476 
393933763
/proc/loadavg
72.27 33.47 12.45 3/1906 13601


On Sep 11, 2014, at 2:54 PM, Mingfei Hua 
mailto:mf...@aerohive.com>> wrote:

All,

One server of mine running CentOS 6.3. The load is very high, but few process 
in running or blocked.
Followed is the output of /proc/stat, /proc/loadavg and top
/proc/stat
cpu  5351723 61716 6974590 161365578 240734 141769 380525 0 0
cpu0 4016881 9686 2510787 79408769 110721 141549 351075 0 0
cpu1 1334842 52029 4463802 81956808 130013 220 29449 0 0
intr 1081538370 139 8 0 0 0 0 0 0 0 0 0 0 106 0 0 73 0 0 531522246 0 0 0 0 0 0 
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 3929190 0 0 0 0 0 
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
0 0
ctxt 1074666354
btime 1409530629
processes 9684115
procs_running 1
procs_blocked 0
softirq 1418158114 0 367862840 177073 609942882 3679591 0 2 53962455 1790447 
380742824
/proc/loadavg
11.21 22.50 10.74 1/2536 16895
Top –H
top - 05:08:02 up 10 days,  4:50,  1 user,  load average: 13.24, 23.26, 10.86
Tasks: 2535 total,   1 running, 2534 sleeping,   0 stopped,   0 zombie
Cpu(s):  3.1%us,  4.0%sy,  0.0%ni, 92.5%id,  0.1%wa,  0.1%hi,  0.2%si,  0.0%st
Mem:   3925040k total,  3780944k used,   144096k free,31464k buffers
Swap:  2064376k total, 1196k used,  2063180k free,  3201568k cached


Sincerely,
Mingfei Hua

___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos


Re: [CentOS] Weird output of system load

2014-09-11 Thread Eliezer Croitoru

On 09/11/2014 09:54 AM, Mingfei Hua wrote:

ll,

One server of mine running CentOS 6.3. The load is very high, but few process 
in running or blocked.
Followed is the output of /proc/stat, /proc/loadavg and top


And what??
You are not giving any relevatn information about what machine what are 
the wanted results nor what this server purpose is..

It will be considered pretty OK for some services to be like that.

Eliezer
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos


[CentOS] Weird output of system load

2014-09-10 Thread Mingfei Hua
All,

One server of mine running CentOS 6.3. The load is very high, but few process 
in running or blocked.
Followed is the output of /proc/stat, /proc/loadavg and top
/proc/stat
cpu  5351723 61716 6974590 161365578 240734 141769 380525 0 0
cpu0 4016881 9686 2510787 79408769 110721 141549 351075 0 0
cpu1 1334842 52029 4463802 81956808 130013 220 29449 0 0
intr 1081538370 139 8 0 0 0 0 0 0 0 0 0 0 106 0 0 73 0 0 531522246 0 0 0 0 0 0 
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 3929190 0 0 0 0 0 
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
0 0
ctxt 1074666354
btime 1409530629
processes 9684115
procs_running 1
procs_blocked 0
softirq 1418158114 0 367862840 177073 609942882 3679591 0 2 53962455 1790447 
380742824
/proc/loadavg
11.21 22.50 10.74 1/2536 16895
Top -H
top - 05:08:02 up 10 days,  4:50,  1 user,  load average: 13.24, 23.26, 10.86
Tasks: 2535 total,   1 running, 2534 sleeping,   0 stopped,   0 zombie
Cpu(s):  3.1%us,  4.0%sy,  0.0%ni, 92.5%id,  0.1%wa,  0.1%hi,  0.2%si,  0.0%st
Mem:   3925040k total,  3780944k used,   144096k free,31464k buffers
Swap:  2064376k total, 1196k used,  2063180k free,  3201568k cached


Sincerely,
Mingfei Hua

___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos