To use sar, sadc (system activity data collector) needs to be running. Check its status or start it with rcsysstat {start|status}.
sar can generate extensive reports on almost all important system activities, among them CPU, memory, IRQ usage, IO, or networking. With its many options, it is too complex to explain further here. Refer to the man page for extensive documentation with examples.
iostat monitors the system device loading. It generates reports that can be useful for better balancing the load between physical disks attached to your system.
The first iostat report shows statistics collected since the system was booted. Subsequent reports cover the time since the previous report.
tux@mercury:~> iostat
Linux 2.6.27.19-3.2-default (geeko@buildhost) 23.3.2009 _x86_64_
avg-cpu: %user %nice %system %iowait %steal %idle
0,49 0,01 0,10 0,31 0,00 99,09
Device: tps Blk_read/s Blk_wrtn/s Blk_read Blk_wrtn
sda 1,34 5,59 25,37 1459766 6629160
sda1 0,00 0,01 0,00 1519 0
sda2 0,87 5,11 17,83 1335365 4658152
sda3 0,47 0,47 7,54 122578 1971008
When invoked with the -n option,
iostat adds statistics of network filesystems (NFS)
load. The option -x shows extended statistics
information.
You can also specify which device should be monitored at what time
intervals. For example, iostat -p sda 3
5 will display five reports at three second intervals for
device sda.
The utility mpstat examines activities of each available processor. If your system has one processor only, the global average statistics will be reported.
With the -P option, you can specify the number of
processors to be reported (note that 0 is the first processor). The
timing arguments work the same way as in the case of
iostat command. Entering mpstat
-P 1 2 5 prints five reports for the second
processor (number 1) at 2 second intervals.
tux@mercury:~> mpstat -P 1 2 5 Linux 2.6.27.19-5-default (geeko@buildhost) 03/23/2009 _x86_64_ 08:57:10 AM CPU %usr %nice %sys %iowait %irq %soft %steal \ %guest %idle 08:57:12 AM 1 4.46 0.00 5.94 0.50 0.00 0.00 0.00 \ 0.00 89.11 08:57:14 AM 1 1.98 0.00 2.97 0.99 0.00 0.99 0.00 \ 0.00 93.07 08:57:16 AM 1 2.50 0.00 3.00 0.00 0.00 1.00 0.00 \ 0.00 93.50 08:57:18 AM 1 14.36 0.00 1.98 0.00 0.00 0.50 0.00 \ 0.00 83.17 08:57:20 AM 1 2.51 0.00 4.02 0.00 0.00 2.01 0.00 \ 0.00 91.46 Average: 1 5.17 0.00 3.58 0.30 0.00 0.90 0.00 \ 0.00 90.05
If you need to see what load a particular task applies to your system, use pidstat command. It prints activity of every selected task or all tasks managed by Linux kernel if no task is specified. You can also set the number of reports to be displayed and the time interval between them.
For example, pidstat -C top 2 3
prints the load statistic for tasks whose command name includes the
string “top”. There will be three reports printed at two
second intervals.
tux@mercury:~> pidstat -C top 2 3 Linux 2.6.27.19-5-default (geeko@buildhost) 03/23/2009 _x86_64_ 09:25:42 AM PID %usr %system %guest %CPU CPU Command 09:25:44 AM 23576 37.62 61.39 0.00 99.01 1 top 09:25:44 AM PID %usr %system %guest %CPU CPU Command 09:25:46 AM 23576 37.00 62.00 0.00 99.00 1 top 09:25:46 AM PID %usr %system %guest %CPU CPU Command 09:25:48 AM 23576 38.00 61.00 0.00 99.00 1 top Average: PID %usr %system %guest %CPU CPU Command Average: 23576 37.54 61.46 0.00 99.00 - top
The utility free examines RAM usage. Details of both free and used memory and swap areas are shown:
tux@mercury:~> free
total used free shared buffers cached
Mem: 2062844 2047444 15400 0 129580 921936
-/+ buffers/cache: 995928 1066916
Swap: 2104472 0 2104472
The options
-b,-k,-m,-g
show output in bytes, KB, MB, or GB, respectively. The parameter
-d delay ensures that the display is refreshed every
delay seconds. For example, free -d
1.5 produces an update every 1.5 seconds.
It can be useful to determine what processes or users are currently
accessing certain files. Suppose, for example, you want to unmount a
file system mounted at /mnt.
umount returns "device is busy." The command
fuser can then be used to determine what processes
are accessing the device:
tux@mercury:~> fuser -v /mnt/*
USER PID ACCESS COMMAND
/mnt/notes.txt tux 26597 f.... less
Following termination of the less process, which was
running on another terminal, the file system can successfully be
unmounted. When used with -k option,
fuser will kill processes accessing the file as well.
The Linux kernel keeps certain messages in a ring buffer. To view these messages, enter the command dmesg:
tux@mercury:~> dmesg [...] end_request: I/O error, dev fd0, sector 0 subfs: unsuccessful attempt to mount media (256) e100: eth0: e100_watchdog: link up, 100Mbps, half-duplex NET: Registered protocol family 17 IA-32 Microcode Update Driver: v1.14 <tigran@veritas.com> microcode: CPU0 updated from revision 0xe to 0x2e, date = 08112004 IA-32 Microcode Update Driver v1.14 unregistered bootsplash: status on console 0 changed to on NET: Registered protocol family 10 Disabled Privacy Extensions on device c0326ea0(lo) IPv6 over IPv4 tunneling driver powernow: This module only works with AMD K7 CPUs bootsplash: status on console 0 changed to on
Older events are logged in the files
/var/log/messages and
/var/log/warn.
To view a list of all the files open for the process with process ID
PID, use -p. For example, to
view all the files used by the current shell, enter:
tux@mercury:~> lsof -p $$ COMMAND PID USER FD TYPE DEVICE SIZE NODE NAME bash 5552 tux cwd DIR 3,3 1512 117619 /home/tux bash 5552 tux rtd DIR 3,3 584 2 / bash 5552 tux txt REG 3,3 498816 13047 /bin/bash bash 5552 tux mem REG 0,0 0 [heap] (stat: No such bash 5552 tux mem REG 3,3 217016 115687 /var/run/nscd/passwd bash 5552 tux mem REG 3,3 208464 11867 /usr/lib/locale/en_GB. [..] bash 5552 tux mem REG 3,3 366 9720 /usr/lib/locale/en_GB. bash 5552 tux mem REG 3,3 97165 8828 /lib/ld-2.3.6.so bash 5552 tux 0u CHR 136,5 7 /dev/pts/5 bash 5552 tux 1u CHR 136,5 7 /dev/pts/5 bash 5552 tux 2u CHR 136,5 7 /dev/pts/5 bash 5552 tux 255u CHR 136,5 7 /dev/pts/5
The special shell variable $$, whose value is the
process ID of the shell, has been used.
The command lsof lists all the files currently open when used without any parameters. Because there are often thousands of open files, listing all of them is rarely useful. However, the list of all files can be combined with search functions to generate useful lists. For example, list all used character devices:
tux@mercury:~> lsof | grep CHR bash 3838 tux 0u CHR 136,0 2 /dev/pts/0 bash 3838 tux 1u CHR 136,0 2 /dev/pts/0 bash 3838 tux 2u CHR 136,0 2 /dev/pts/0 bash 3838 tux 255u CHR 136,0 2 /dev/pts/0 bash 5552 tux 0u CHR 136,5 7 /dev/pts/5 bash 5552 tux 1u CHR 136,5 7 /dev/pts/5 bash 5552 tux 2u CHR 136,5 7 /dev/pts/5 bash 5552 tux 255u CHR 136,5 7 /dev/pts/5 X 5646 root mem CHR 1,1 1006 /dev/mem lsof 5673 tux 0u CHR 136,5 7 /dev/pts/5 lsof 5673 tux 2u CHR 136,5 7 /dev/pts/5 grep 5674 tux 1u CHR 136,5 7 /dev/pts/5 grep 5674 tux 2u CHR 136,5 7 /dev/pts/5
When used with -i, lsof lists
currently open Internet files as well:
tux@mercury:~> lsof -i pidgin 4349 tux 17r IPv4 15194 0t0 TCP \ jupiter.example.com:58542->www.example.net:https (ESTABLISHED) pidgin 4349 tux 21u IPv4 15583 0t0 TCP \ jupiter.example.com:37051->aol.example.org:aol (ESTABLISHED) evolution 4578 tux 38u IPv4 16102 0t0 TCP \ jupiter.example.com:57419->imap.example.com:imaps (ESTABLISHED) npviewer. 9425 tux 40u IPv4 24769 0t0 TCP \ jupiter.example.com:51416->www.example.com:http (CLOSE_WAIT) npviewer. 9425 tux 49u IPv4 24814 0t0 TCP \ jupiter.example.com:43964->www.example.org:http (CLOSE_WAIT) ssh 17394 tux 3u IPv4 40654 0t0 TCP \ jupiter.example.com:35454->saturn.example.com:ssh (ESTABLISHED)
udevadm monitor listens to the kernel uevents and events sent out by a udev rule and prints the device path (DEVPATH) of the event to the console. This is a sequence of events while connecting a USB memory stick:
![[Important]](admon/important.png) | Monitoring udev events. |
|---|---|
Only root user is allowed to monitor udev events by running udevadm command. | |
UEVENT[1138806687] add@/devices/pci0000:00/0000:00:1d.7/usb4/4-2/4-2.2 UEVENT[1138806687] add@/devices/pci0000:00/0000:00:1d.7/usb4/4-2/4-2.2/4-2.2 UEVENT[1138806687] add@/class/scsi_host/host4 UEVENT[1138806687] add@/class/usb_device/usbdev4.10 UDEV [1138806687] add@/devices/pci0000:00/0000:00:1d.7/usb4/4-2/4-2.2 UDEV [1138806687] add@/devices/pci0000:00/0000:00:1d.7/usb4/4-2/4-2.2/4-2.2 UDEV [1138806687] add@/class/scsi_host/host4 UDEV [1138806687] add@/class/usb_device/usbdev4.10 UEVENT[1138806692] add@/devices/pci0000:00/0000:00:1d.7/usb4/4-2/4-2.2/4-2.2 UEVENT[1138806692] add@/block/sdb UEVENT[1138806692] add@/class/scsi_generic/sg1 UEVENT[1138806692] add@/class/scsi_device/4:0:0:0 UDEV [1138806693] add@/devices/pci0000:00/0000:00:1d.7/usb4/4-2/4-2.2/4-2.2 UDEV [1138806693] add@/class/scsi_generic/sg1 UDEV [1138806693] add@/class/scsi_device/4:0:0:0 UDEV [1138806693] add@/block/sdb UEVENT[1138806694] add@/block/sdb/sdb1 UDEV [1138806694] add@/block/sdb/sdb1 UEVENT[1138806694] mount@/block/sdb/sdb1 UEVENT[1138806697] umount@/block/sdb/sdb1
xrestop provides statistics for each connected X11 client's server-side resource. The output is very similar to Section 13.6.4, “Processes: top”.
xrestop - Display: localhost:0
Monitoring 40 clients. XErrors: 0
Pixmaps: 42013K total, Other: 206K total, All: 42219K total
res-base Wins GCs Fnts Pxms Misc Pxm mem Other Total PID Identifier
3e00000 385 36 1 751 107 18161K 13K 18175K ? NOVELL: SU
4600000 391 122 1 1182 889 4566K 33K 4600K ? amaroK - S
1600000 35 11 0 76 142 3811K 4K 3816K ? KDE Deskto
3400000 52 31 1 69 74 2816K 4K 2820K ? Linux Shel
2c00000 50 25 1 43 50 2374K 3K 2378K ? Linux Shel
2e00000 50 10 1 36 42 2341K 3K 2344K ? Linux Shel
2600000 37 24 1 34 50 1772K 3K 1775K ? Root - Kon
4800000 37 24 1 34 49 1772K 3K 1775K ? Root - Kon
2a00000 209 33 1 323 238 1111K 12K 1123K ? Trekstor25
1800000 182 32 1 302 285 1039K 12K 1052K ? kicker
1400000 157 121 1 231 477 777K 18K 796K ? kwin
3c00000 175 36 1 248 168 510K 9K 520K ? de.comp.la
3a00000 326 42 1 579 444 486K 20K 506K ? [opensuse-
0a00000 85 38 1 317 224 102K 9K 111K ? Kopete
4e00000 25 17 1 60 66 63K 3K 66K ? YaST Contr
2400000 11 10 0 56 51 53K 1K 55K 22061 suseplugge
0e00000 20 12 1 50 92 50K 3K 54K 22016 kded
3200000 6 41 5 72 84 40K 8K 48K ? EMACS
2200000 54 9 1 30 31 42K 3K 45K ? SUSEWatche
4400000 2 11 1 30 34 34K 2K 36K 16489 kdesu
1a00000 255 7 0 42 11 19K 6K 26K ? KMix
3800000 2 14 1 34 37 21K 2K 24K 22242 knotify
1e00000 10 7 0 42 9 15K 624B 15K ? KPowersave
3600000 106 6 1 30 9 7K 3K 11K 22236 konqueror
2000000 10 5 0 21 34 9K 1K 10K ? klipper
3000000 21 7 0 11 9 7K 888B 8K ? KDE Wallet