how to check which process consuming a lot of HDD I/O ? Do You know any good command which can show me which process saving something big on the storage system ? "iostat" or maybe "ps" ? Would be great if somebody could past me here nice command.
View 3 RepliesIs there a command to check specific processes that's using the most IO/disk usage? I know sar and ps but I want more specific details on IO on individual processes
View 5 Replies View RelatedHow to check server usage (process and memory) for last 4months in RHEL5
View 3 Replies View RelatedWe have an internal application which takes below 10% cpu usage normally. However sometimes it hit above 85% or more (and stays there) causing the server to become very slow. Currently we have to monitor this manually, kill the process and manually start it.
command / script that will kill the process (using pid will be the best option as i think) when it reaches 85% or higher
(I saw another question that suggested to use monit but I have to get this done using the standard commands available in the current installation - RHEL5)
I use linux and Unix and I want to monitor the memory usage for process. To prevent memory leakage and out of memory of the system. Any command or sytnax , have more better and presentable data than below command about memory usage of one process ?
View 3 Replies View RelatedI am using qmail in my office. I am noticing this for more than a month. It is my daily routine to check the cpu utilization using top command and by using top different processes can be viewed. the apache process after 1 or 2 days shows cpu utilization of 100%. There is no affect on the server whatsoever of 100% because mail server has 2 processors of 4 core, i think so that may be the reason. Can any one explain why is this happening. What i do after that is i restart the httpd process and cpu utilization comes to normal. What are the reasons behind this and how to solve this permanently.
View 1 Replies View RelatedI'm trying to check my server's bandwidth usage in real time, installed the following programs but none worked so far.
Iptraf - No results even when using iptraf -u
Tcptrack - Error : pcap_loop: cooked-mode frame doesn't have room for sll header
Iftop - No results, everything 0b
Are there any programs that displays bandwidth usage in real time and actually works on VPSes? Or getting real time bandwidth usage on a VPS is simply impossible?
Is there any way to monitor one process' CPU usage and RAM usage over time on Linux? I am trying to change to a cheaper VPS and need to work out what level of CPU and RAM I need!
View 2 Replies View RelatedCan anyone explain this output? I'm not sure what to make of it. 36% usage and no files??
root@host:/backups# ls -la
total 8
drwxr-xr-x 2 root root 4096 Mar 8 20:12 .
[code]....
I was trying to get the status of memory usage and disk usage using sigar in windows and ubuntu. done this in windows by just copying the sigar library into jdk library. But i was unable to do so in ubuntu. I've copied the library to java-6-sun library but still can't run the program.
View 14 Replies View RelatedI'm running into a problem where my system is running out of disk space on the root partition, but I can't figure out where the runaway usage is. I've had a stable system for a couple of years now, and it just ran out of space. I cleaned some files up to get the system workable again, but can't find the big usage area, and I'm getting conflicting results.For example, when I do a df it says I'm using 44GB out of 58 GB:
Code:
[root@Zion ~]# df -h
Filesystem Size Used Avail Use% Mounted on
[code]....
Code:
df -h
Filesystem Size Used Avail Use% Mounted on
/dev/sda3 50G 47G 602M 99% /
How to find why Avail space shows only 602M?
how can I check if disk devices are running at the proper speed (read/write access)?What are, for example, the correct values for ATA o SCSI device, if I test them with hdparm?
View 2 Replies View Related[root@linux root]# fdisk -l
Disk /dev/hda: 40.0 GB, 40020664320 bytes
255 heads, 63 sectors/track, 4865 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes
Device Boot Start End Blocks Id System
/dev/hda1 * 1 13 104391 83 Linux
/dev/hda2 1276 4864 28828642+ f Win95 Ext'd (LBA)
/dev/hda3 14 395 3068415 83 Linux
/dev/hda4 396 526 1052257+ 82 Linux swap
/dev/hda5 1276 3187 15358108+ 7 HPFS/NTFS
/dev/hda6 3188 3249 497983+ 8e Linux LVM
/dev/hda7 3250 3311 497983+ 8e Linux LVM
Here /dev/hda5 taken of How much capacity for NTFS (need space in MB).
When level 3, we can use only command. What can I do to check for Hard Disk usage such as used space, free space, total space.
View 4 Replies View RelatedHow to schedule check disk at boot time in CentOS 5.3 and OpenSUSE 11.1?
View 5 Replies View RelatedSome thing is using up a huge amount of my disk space about 10G and I can not determine what it is. When I look at my disk usage in system monitor it say I have used about 25G and when I scan the directory in disk usage analyzer the entire file system used is 15G.
View 1 Replies View RelatedI wonder to know the command or the procedure to get the overall CPU utilisation in linux. I have used top, iostat, mpstat but the outputs are not the way i needed. Is it possible to get the output like...
CPU usage: .....% or something like that.
I am running Slackware 13.0. I am aware of free -m, vmstat, top, etc. However, none of these programs display how much ram each program is using. Is there a program that displays how much ram each program is using? I run a headless so I'd need a program that runs in CLI.
I know that top command shows %MEM (only two programs were using 0.1%MEM) but after running free -m I only have a total of 400 MB ram left out of my 1.5 GB of ram. Where is all that lost ram?
I have one dedicated server in godaddy. Now I got mail regarding overage bandwidth. I don't know how to check this and I must give report how its happen.
View 1 Replies View RelatedLinux OS: Ubuntu 10.10..In Windows, we can use Device Manager to check the power usage of my USB devices (like my WiFi adapter use 500mA). How to check this under linux?The purpose of asking this question is just to learn.
View 1 Replies View RelatedI am currently writing a JAVA script to monitor certain unix processes through JConsole. Upon having lots of trouble with runtime.exec, i decided to bypass the top/ps command call and just get the information straight from /proc/*pid*/whatever.Now i can pull back any information from any of the files I want, and the current way i determine the CPU usage of a process is as follows:Add the UTime and STime of that process from /proc/pid/stat then divide my pidCpu by UTime + STime + NTime from /proc/stat, then multiply that by 100, should give me the % cpu usage a process is using, right?Theory being if I get the jiffies assigned to my process, I can divide that by the total jiffies the cpu assigns.
However, my results seem to vary from the ones gathered from top and ps. What am I doing wrong?
If I, using sar or vmstat, found that there is a CPU load on my Linux box. How can I know the top processes consuming the CPU?
View 9 Replies View RelatedI"m unable to find the memory usage however top shows me the following results.
Tasks: 90 total, 2 running, 88 sleeping, 0 stopped, 0 zombie
Cpu(s): 21.1%us, 1.4%sy, 0.0%ni, 74.0%id, 2.4%wa, 0.0%hi, 1.0%si, 0.0%st
Mem: 8313376k total, 7969976k used, 343400k free, 185496k buffers
[code]....
and free shows the following.
total used free shared buffers cached
Mem: 8118 7764 354 0 181 7335
-/+ buffers/cache: 247 7870
Swap: 4000 0 4000
Total: 12119 7764 4354
how can i find the usage of memory which is being showed by top and free.
How to check if some folder is on a shared disk or on a local disk?
View 1 Replies View RelatedPossible Duplicate:
Linux: Is there something similar to "top" for I/O?
I am trying to do some CPU intensive operation on a Linux machine and it is showing very high load.
I want to figure out if this load is due to I/O or CPU usage. How could I get the percentage of I/O usage by process?
How can I maximum the CPU usage for a process? And also use both cores if possible.
View 5 Replies View RelatedI'm writing a shell script which aims to create a safe gtared (xxx.sql.gz) copy of MySQL databases.This script is planned to be Cron-Jobed.
Well, what I need to add to this shell, is something that limit CPU usage for the whole process (just in case if the database being generating is a huge one.)So, after few time of googling I found couple of solutions:
- Using cpulimit.
I tried to place the code in Position(1) and Position(2) but it didn't seem to be working fine.. Any idea about the right use?
And the other Solution is:- Using nice.
Well, assuming I named my shell script (sqlbacker)..
Finally, this is my first time I ever write a shell, so correct me if somewhere I made a mistake :-)
(The script itself works perfectly)
I am currently writing a JAVA script to monitor certain unix processes through JConsole. Upon having lots of trouble with runtime.exec, i decided to bypass the top/ps command call and just get the information straight from /proc/*pid*/whatever.Now i can pull back any information from any of the files I want, and the current way i determine the CPU usage of a process is as follows:
Add the UTime and STime of that process from /proc/pid/stat then divide my pidCpu by UTime + STime + NTime from /proc/stat, then multiply that by 100, should give me the % cpu usage a process is using, right?Theory being if I get the jiffies assigned to my process, I can divide that by the total jiffies the cpu assigns. However, my results seem to vary from the ones gathered from top and ps. What am I doing wrong?
There's a disfunctional process eating copious cpu time.Is there a way to effectively assign it a high nice value? I need to do this whenever it runs, for whatever reason, and I can't be bothered to track down all the scripts and scenarios that cause it to run and change the script to use nice. And I can't be bothered to manually run renice whenever I notice that it's running.
I want the OS to automatically assign a high nice value to this process, perhaps based on the processes name. Is this possible?Presumably, a cron job could run every 5 minutes and run "renice" on every process matching a given name, but I'm hoping for a solution with more finess.