• kolik CPU/RAM zere jaky proces se zjisti snadno, napriklad vypisem top
• monitorovani IO je zlozitejsi a dost casto zpomaluje system nejvic

• existuji programy:
  • iotop
  • atop
  • pidstat
  • dstat (–topio, –topbio)
  • vmstat (-d)
  • iostat
  • iozone
  • sar

Here’s how you use it. First, get it:

wget http://aspersa.googlecode.com/svn/trunk/iodump

Then turn on kernel messages about I/O:

echo 1 > /proc/sys/vm/block_dump

This makes the kernel start writing messages about every I/O operation that takes place. Now all you have to do is get those messages and feed them into my script:

while true; do sleep 1; dmesg -c; done | perl iodump

Wait a little while, then cancel the script. The results should look something like the following:

root@kanga:~# while true; do sleep 1; dmesg -c; done | perl iodump
^C# Caught SIGINT.
TASK                   PID      TOTAL       READ      WRITE      DIRTY DEVICES
firefox               4450       4538        251       4287          0 sda4, sda3
kjournald             2100        551          0        551          0 sda4
firefox              28452        185        185          0          0 sda4
kjournald              782         59          0         59          0 sda3
pdflush                 31         30          0         30          0 sda4, sda3
syslogd               2485          2          0          2          0 sda3
firefox              28414          2          2          0          0 sda4, sda3
firefox              28413          1          1          0          0 sda4
firefox              28410          1          1          0          0 sda4
firefox              28307          1          1          0          0 sda4
firefox              28451          1          1          0          0 sda4

I deliberately generated a bunch of I/O by deleting my Firefox history and cache.

Do dmseq se zapisuje neco jako….

firefox(4450): WRITE block 25058328 on sda4
firefox(4450): WRITE block 25058336 on sda4
firefox(4450): WRITE block 25059104 on sda4

A quick way is to look at the /proc/[PID]/io file.

Example:

$ cat /proc/1944/io
read_bytes: 17961091072
write_bytes: 8192000
cancelled_write_bytes: 32768

#!/bin/bash

while [ 1 ]
do

  #CPIDS=`pgrep cache`
  CPIDS=`ls /proc/ | grep "[0-9]"`

  i=0
  for P in $CPIDS
  do
    STATA[i]=`cat /proc/$P/io 2>/dev/null | grep "^write_bytes:" | cut -d " " -f 2`
    i=$((i+1))
  done
 
  sleep 1
 
  i=0
  for P in $CPIDS
  do
    STATB[i]=`cat /proc/$P/io 2>/dev/null | grep "^write_bytes:" | cut -d " " -f 2`
    i=$((i+1))
  done
 
  i=0
  for P in $CPIDS
  do
    [[ ${STATB[$i]} -gt 0 ]] && [[ ${STATA[$i]} -gt 0 ]] && DIFF=$(( ${STATB[$i]} - ${STATA[$i]}  )) || DIFF=0
    [[ $DIFF -gt 0 ]] && {
      DIFF=$((DIFF/1024))
      echo "PID $P: $DIFF kB/s (${STATB[$i]} - ${STATA[$i]})"
    }
    i=$((i+1))
  done
 
  echo "---" 

done

   wa -- iowait (Amount of time the CPU has been waiting for I/O to complete)

• procesy ve stavu D

PROCESS STATE CODES
 D uninterruptible sleep (usually IO)
 R running or runnable (on run queue)
 S interruptible sleep (waiting for an event to complete)
 T stopped, either by a job control signal or because it is being traced.
 W paging (not valid since the 2.6.xx kernel)
 X dead (should never be seen)
 Z defunct ("zombie") process, terminated but not reaped by its parent.

• vypis procesu ve stavu D

 for x in `seq 1 1 10`; do ps -eo state,pid,cmd | grep "^D"; echo "----"; sleep 5; done

• zdroj (http://bencane.com/2012/08/06/troubleshooting-high-io-wait-in-linux)