Ubuntu Servers :: High Load Average But Idle CPU?
Nov 26, 2010
I am having a problem with the server that I use to host my personal site. The load average quite often spikes to exceed 1.00 for the 1 and 5 minute intervals, and the 15 minute interval gets above .5. This occurs while the server is idle, serving very few or no requests and with the CPU 99% idle with <1% IOWAIT usage. I have checked top and vmstat, but neither one provides any useful info. Top continues to say the CPU is 99% idle, and vmstat says that there are 0 runnable and 0 blocking tasks. Occasionally, vmstat will say that there is 1 runnable task, but this doesn't even coincide with the load average spikes. I have already searched for other solutions to this problem, but everything I have seen says to use top and/or vmstat, but those aren't showing anything out of the ordinary. Can anyone recommend anything else I might do?
My server has a Pentium 4 HT 3gHz processor, 2GB RAM, and runs Kubuntu 10.10. (The reason it runs Kubuntu instead of Ubuntu Server is that it needs an X environment so that the Nvidia driver can initialize and put its graphics card into a power saving mode.)
View 8 Replies
ADVERTISEMENT
May 5, 2010
Since upgrading from 9.10 to 10.04, my system is noticeably less responsive and exhibits halting behavior for at times 10s of seconds. There is nothing obvious in the various logs that I can find. The only objective indication I can find that demonstrates that something is seriously wrong is that my load average never drops below 0.5 and is often over 1.5, even at idle. In this situation, top/ps/whatever shows very few processes running (usually just top). This suggests to me that either the new kernel scheduler is horrible or that something new is resulting in blocking I/O or other uninterruptable sleeps.Typical top output:
Quote:
top - 12:18:01 up 1 day, 49 min, 5 users, load average: 1.42, 0.86, 0.71
Tasks: 262 total, 1 running, 261 sleeping, 0 stopped, 0 zombie
Cpu(s): 0.9%us, 0.8%sy, 0.0%ni, 97.6%id, 0.0%wa, 0.0%hi, 0.7%si, 0.0%st
[code]....
View 9 Replies
View Related
Dec 7, 2010
I recently upgraded from Fedora 11 to Fedora 14 on my Dell Latitude E6400 notebook PC. It's a Dual core processor, as seen in the uname provided below:
uname -a:
Linux COMPUTERNAME 2.6.35.6-48.fc14.i686 #1 SMP Fri Oct 22 15:34:36 UTC 2010 i686 i686 i386 GNU/Linux
redhat-release:
Fedora release 14 (Laughlin)
Since the upgrade, I've noticed some unexplainable behavior: The load average is consistently 0.5 - 1.75 when my PC is completely idle (99% according to top). No processes are in the D (I/O) state, and I have no Zombie processes. This happens on a fresh boot, every time. It gets worse and less worse, and bounces from unusable (cursor stops moving as I type in the terminal, to actually losing keystrokes at times). I'm wondering if anybody else is experiencing similar issues, or knows of a (the) kernel bug which is causing this behavior? Actively researching the issue. I've come across the following: Tens of wakes per second in "[kernel scheduler] Load balancing tick" on Core 2 Duo even with only 1 core enabled
[Code]....
View 3 Replies
View Related
Aug 10, 2011
I wrote a script to extract and get the the name of *.gz in a foler . Since running that script every 10 minutes, load average on my server increases more than 10.I checked with 'top' and it showed many D process.
View 2 Replies
View Related
Dec 14, 2009
The load average is almost 1.06, but cpu is not 100% utilized... I am just wondering why load average is 1.06 in that case still ? (well i monitored cpu for long time, it never exceeded above 40%)can anyone explain the reason behind it ? Also is the system over utilized in this case ?
View 3 Replies
View Related
Aug 18, 2011
I am having Red Hat LINUX 5 Enterprise Server and facing problem regarding very very high server load (load average is going high up to 60-70)due to which server is getting hang.
View 7 Replies
View Related
Oct 15, 2010
It's the fourth time now since Maverick that I had to cold reboot my system because it was totally unresponsive. The system monitor in my taskbar shows 100% on the background blue graph (I guess that's just one core then) and almost nothing on the other core. On the last freeze I managed to open 'sudo top' before it went totally unresponsive and saw that there was no high CPU usage at all, but a load average spiking above 24. Also the swap seemed to be full although my machine usually never uses swap. I was watching a movie with VLC this time, but I'm not sure if VLC ran the other times my OS froze. I made a snapshot with my cell phone: [URL]. How I can prevent a process from causing so much load?
View 6 Replies
View Related
Feb 10, 2010
I have High load on my server and my investigation shows nothing (so i believe that my investigations is wrong ), the load average in this moment is 10.13, 9.47, 8.24. , mentioning the below.
- The disk utilization (all the disks) is near 0, as the result of the IOSTAT
- There is no blocked processes (as a result of VMSTAT).
- I have two processors (dual core) , the maximum load average should be something around 4.
- The server always have above 8 load average in all times interval.
btw , my OS is RHEL AS release 4 (Nahant Update 7)Kernel :Linux 2.6.9-78.ELhugemem #1 SMP i686 i686 i386 GNU/Linux
View 8 Replies
View Related
Jan 13, 2010
I have high load average on the servers with no identified reason (From my point of view)
View 2 Replies
View Related
Jul 6, 2010
I have a Nagios server with a lot of hosts and services; around 400 services (in all) and 150 hosts. Most of these services are programmed in bash language. The problem is the server has a high load average; between 5 and 11. The server has the next features:
- Intel Xeon 2.66GHz Dual Core
- 4MB cache memory
- 1GB RAM memory
- 50GB hard disk
Is this load average normal? Should I program the plugins with C?
View 3 Replies
View Related
Nov 24, 2010
I have several CentOS 32-bit VMs running on ESX 4. Those that were updated to the most current patch level ("yum update", accepting all updates available last week) started showing load average of ~0.4 when completely idle. After comparing the problematic VMs with those that show zero load average at idle, and then modifying them in all kinds of ways, I narrowed it down to the combination of a recent kernel patch (2.6.18-164 is fine, 2.6.18-194 is not) and E1000 network adapter. Replacing the network adapter on a problematic VM with VMXNET3 fixes the load statistics.
View 3 Replies
View Related
May 10, 2010
I upgraded webserver to new ubuntu server 10.04 (x86-64). After upgrade the increased load from 0,3 to 1,4. On webserver running phpbb, which generating slow quieres, which not before upgrade to lucid. HW conf: Intel Core i7, 8GB ram, WD Raptor 10k rpm. Week17 upgrade to new version.
View 9 Replies
View Related
Oct 29, 2010
My laptop has a very high consumption when running on idle (I mean nothing that write on the disk or video etc..). I read on lenovo forums that people could get 10-12W and I can hardly get lower than 20W. My cpufreq is set to powersave and my cpus are always at the lowest scale.
configure my laptop to save some power. It used to last almost 7h on battery (sadly I don't remember the consumption) on Ubuntu and now it can barely last more than 3h with slackware.
My config is:
T400 intel duo core 3,06ghz
14.1" TFT display with 1440x900 (WXGA+) resolution with LED backlight
Switchable Graphics with Intel GMA 4500MHD and ATI Mobility Radeon HD 3470 (256 MB)
320GB 7200rpm 2.5" SATA HDD
Intel Wifi Link 5300 (AGN)
9 cells battery
View 1 Replies
View Related
Mar 11, 2011
how to setup an Active/Active Load Balanced and High Available (If one of the nodes is down the system still runs) MySQL cluster. I have found quite a few howto's but I have some things unclear in my mind. I found a few solutions like this one: [URL] or this: [URK] Those are using two or four MySQL nodes, two Load Balancers to avoid a single point of failure but only one MySQL cluster management server. What happens if the MySQL cluster management fails?
I have also found a "MySQL Master-Master Circular Replication" technique but from what I read, with this option there is a chance that conflicts will arise if node A and node B both insert an auto-incrementing key on the same table.
View 1 Replies
View Related
Jun 15, 2009
I read the man pages of batch and it says that commands or scripts scheduled using the batch command will only execute when the load average goes below 0.8. As a newbie I hardy understand what it means. Does it mean that the system resources are busy?Ans since I want to see the o/p of the commands I have set using batch, is there any way i can bring the sys load average below 0.8?
View 3 Replies
View Related
Jul 12, 2011
I was struggling to find the information so I thought it would be easier to ask here. What does the load average as reported by top mean ? To me those are 3 mysterious values, which most likely refer to some average CPU usage and this is all I know about it (not sure if it's true).What is the maximum value for this parameter (I guess the minimum is 0.00) and what does it mean ?
View 5 Replies
View Related
May 19, 2010
How to get the load average for each CPU core in multi core(eg:duel core machine) processor environment. I tried using,
Code:
1. cat /proc/loadavg
2. uptime
3. top
But all of those commands gives the load average for whole system but not particular CPU core. Are there a way to take the load average for CPU core(Or any mechanism that can be done programmatic manner).
View 6 Replies
View Related
Feb 11, 2010
How to reduce the load average? I have two linux machines.. now my problem is that both the machines having same configuration and same processes are running in those two machines... But in one machine the load average is more than 3 and in other it is 0.5(when i am using top command).For this load average one machine(where load average is more) is slow when i am running application on it..
View 5 Replies
View Related
Jan 18, 2011
I have two servers. One in production (lets call it the OLD ONE) and the other (lets call it the NEW ONE) in tests to replace the OLD ONE.This is the basic hardware of each one: (I can post more detailed info if you need, but beside the erros on dmesg, look at the the L2 cache of the NEW ONE )
Old one:
2 quad core processors that linux recognize as 8 x Intel(R) Xeon(R) CPU E5440 @ 2.83GHz - 32Kb L1 - 6Mb L2
48Gb RAM
New one:
8 quad core processor that linux recognize as 32 x AMD Opteron(tm) Processor 6136 @ 2.400Mhz (- 64Kb L1 - 512Kb L2
128Gb RAM
The scenario:We run a dataflex system on the old one, with average of 3000 users, with tops at 3300 users and sometimes less then 2000. In the old one, we have a load average from 2 to 6 with the 3000 users, depending on the type of application (sometimes we run reports to txt files that take more then 6 hours to complete and the load average can raise to 12). 85% of this conections are from remote links.
Dataflex is a language that derives from C that have their own sgbd (if we can call it that way), and have a limit of 2gb per table. This size we almost have on 5 tables and we use a dataflex feature to compress the data.
The problem:We are migrating (or trying to) the tables to oracle, so we bought new machines for the DB and the new one to replace the old one, becouse we think that could not handle the job with 2 oracles (load balance).In some tests we could see (or suposed) that the oracle database was not so fast with more then 1000 users (opening same table and doing the same task) and we decided to test the new one with the system that is in production right now, with dataflex tables, to ensure that the problem could be oracle.We change the HDs and IP. Started the system on the new one, and started to monitorate as the real users start their jobs. At 800 users the Load Average raised to 26 and with 1300 users we had more then 115 on Load Average. More users login in and the TOP become slow, pointing 400 of LA. From here we started to get some "Lock time out" erros and we had to change to the old one again, to prevent corruption on the tables. I'm analizing all report tools I know about performance and hardware and I cant see nothing. I saw some errors on dmesg, but I can say that is related to that problem.
View 6 Replies
View Related
May 7, 2010
A server of mine previously running ubuntu 9.10 used to be between 0.01 and 0.10 load average during the normal load of users using various server programs on it (mostly apache with php scripts).
Now, after upgrading to 10.04 (which went smoothly for the most part), the load average is much greater under the same user workload, hovering between 0.1 and 0.3 under very light work and up to and over 1 regularly when more users are accessing the same scripts.
Are there any known issues that would cause greater usage of the same resources in lucid, or are there any ways I can trace what's causing the higher load? Downgrading or starting with a fresh install are last resorts, as there are a lot of customized options specifically set up for this server and I'd rather not go through backing them all up and restoring them after a complete wipe.
View 5 Replies
View Related
Jun 10, 2011
installed squeeze, but top keeps reporting 0.00 most of the time, and def it's wrong cause we have another server same hardware but lenny, and its over 2.00 same load. what can i check or do to fix this? thx
Linux 2.6.32-5-amd64 #1 SMP Wed May 18 23:13:22 UTC 2011 x86_64 GNU/Linux
[Code]...
View 12 Replies
View Related
Jan 20, 2011
Anyone know what means the values of Load Average?
load average: 0.00, 0.00, 0.04
load average: 17.50, 13.44, 12.70
Sometime is low and many times is high.
View 1 Replies
View Related
Apr 20, 2010
I don't know what is going on with one of my servers.
It is sitting with a load average of 10.09 10.08 10.02
However If I look at top I see the server between 95% and 99% idle!
Is it possible that my server is running some hidden processes or something?
How I can investigate the cause of this.
View 9 Replies
View Related
Apr 7, 2009
Is it possible with vixie cron to limit cron's activities by the current load average? Basically I'm looking for the same functionality that's in fcron, but I'm restricted to use the vixie cron with CentOS 5.3.If not, does anyone have any recommendations on limiting the impact that cron has when it runs 20-30 user-specified crontabs at once?
View 4 Replies
View Related
Nov 9, 2010
I just wonder if anyone can give some hint regarding my problem. I have a server based on RHEL 4.4.What I have noticed is that server load average is raising in approx. 7-10 days by approx. 7, even if has very low CPU utilization, memory utilization on approx 80% and low on I/O stats.Linux apollo 2.6.9-67.ELsmp #1 SMP Wed Nov 7 13:58:04 EST 2007 i686 i686 i386 GNU/Linux
[Code]...
Server is used as compilator / build machine to build aplication from source code stored in IBM Rational clearcase envinronment, and as a test server to test those builds with network equipment. Does anyone can tell what could be causing this behaviour?
View 3 Replies
View Related
Oct 27, 2009
We have a server that is running RHEL4 that occasionally spikes in load average above 10 and we have no idea what is causing it. We would like to know if there are any free tools or a script that when the load average hits a certain point it will trigger the system to start logging the processes to see what is happening. Usually by the time we get logged into the system the load average is on its way down. If someone has a better idea please let me know.
View 4 Replies
View Related
Feb 10, 2011
I'm new to the linux world and its been quit sometime since i've done any programming. However, I'm writing a program which simply calculates the load average of a process. In doing this I need to use the uptime command for linux in a java program. I've done a little bit of searching on the net and it mentions this is possiable by using java runtime command. Unfortunatly though I have yet to find a working example of this. I've tried just simply reading the /proc/uptime file but I have no clue how to format the 2 numbers in seconds to make it the same as if you just typed in uptime in the linux command prompt.
View 2 Replies
View Related
May 26, 2010
For reasons I won't get into, I need to copy directories so long as the average system load is low. Can someone help me write a BASH script that will copy the contents of a directory, but check to make sure the average system load is below X before copying each file, and if not, wait Y seconds and try again?
View 3 Replies
View Related
Nov 2, 2015
I had noticed that my workstation (running debian testing) is having a constant load of about 0.9 (+/-0.1), even while the system is completly idle.The system starts with a normal load and then goes up to about 0.9 within a few minutes and then will stay there forever (well it can get higher, but never lower). This is the load graph after the system runs some time: URL....
I have tried everything I know to find the cultprint but there is no cpu usage and no io usage. I even have disabled the desktop and still have this problem.Also I have tried booting an older Kernel (4.1 and 3.16) with no difference in the load.Here is some information about my system:
Code: Select all» uname -a
Linux localhorst 4.2.0-1-amd64 #1 SMP Debian 4.2.3-2 (2015-10-14) x86_64 GNU/Linux.
View 7 Replies
View Related
Oct 27, 2010
Can anyone explain the difference between IOWAIT%/IDLE%/LOAD AVG? We have a 3 servers (oracle rac) attached to a SAN. When we added another CPU to our 3rd server, the IOWAIT across the oracle rac dropped a lot. We used to get pretty high IOWAIT % before. So I don't really understand how the extra CPU could do this. I thought the IOWAIT % is high because the server is waiting for the SAN, so I thought the SAN was the bottleneck. So I don't really understand the difference between the 3 things (iowait, idle and load avg).
View 1 Replies
View Related
