-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Thanks for you replay.
It seems lsof does not display which process is write or read from the block device. As refered in kernel document Documentation/filesystems/proc.txt, I read the document Documentation/laptop-mode.txt: If you want to find out which process caused the disk to spin up, you can gather information by setting the flag /proc/sys/vm/block_dump. When this flag is set, Linux reports all disk read and write operations that take place, and all block dirtyings done to files. This makes it possible to debug why a disk needs to spin up, and to increase battery life even more. The output of block_dump is written to the kernel output, and it can be retrieved using "dmesg". When you use block_dump and your kernel logging level also includes kernel debugging messages, you probably want to turn off klogd, otherwise the output of block_dump will be logged, causing disk activity that is not normally there. It works. and any tools provide maybe better. David Fox wrote: > On Mon, Apr 14, 2008 at 7:23 PM, Ding Honghui <[EMAIL PROTECTED]> wrote: >> 1. Is it means some processes are accessing my disk heavly? > >> 1000 ? I'd say yes. > > For comparison, my (rather) idle box is doing: > > [EMAIL PROTECTED]:~$ vmstat -a 4 > procs -----------memory---------- ---swap-- -----io---- -system-- ----cpu---- > r b swpd free inact active si so bi bo in cs us sy id wa > 2 0 61812 7992 210820 507152 0 0 13 42 47 44 83 7 10 0 > 0 0 61812 7960 210832 507248 0 0 0 59 546 2127 39 11 50 0 > 1 0 61812 7836 210916 507280 0 0 0 36 539 2093 43 8 44 4 > 0 0 61812 7804 210960 507236 0 0 0 3 547 1943 52 10 37 0 > 1 0 61812 7556 211056 507256 0 0 0 21 543 2102 41 10 49 0 > 0 0 61812 8300 210628 507072 0 0 0 40 536 2084 44 10 46 0 > 1 0 61812 8300 210636 507120 0 0 0 44 541 2065 35 10 55 0 > 0 0 61812 8300 210680 507088 0 0 0 36 544 2066 41 11 48 0 > 0 0 61812 8300 210680 507056 0 0 0 0 701 2198 42 12 46 0 > 1 0 61812 8052 210792 507184 0 0 0 26 691 2382 45 15 41 0 > 0 0 61812 7672 211052 507264 0 0 0 42 553 2110 48 12 40 0 > 0 0 61812 7640 211132 507280 0 0 0 104 602 2010 49 11 39 0 > 1 0 61812 7516 211188 507288 0 0 0 26 806 2470 47 13 40 0 > > The (bo) block out io is coming mostly from cdparanoia which is ATM > trying to read in a CD to wav, and the CD is kind of dirty :(. > > >> 2. If so, is there any IO diagnose tools exist to find out which process is? > > lsof probably: > > [EMAIL PROTECTED]:~$ lsof /dev/hdd > lsof: WARNING: can't stat() reiserfs file system /dev/.static/dev > Output information may be incomplete. > COMMAND PID USER FD TYPE DEVICE SIZE NODE NAME > cdparanoi 7526 dfox 3u BLK 22,64 3624 /dev/hdd > cdparanoi 7526 dfox 4u BLK 22,64 3624 /dev/hdd > >> - -- >> Best Regards, >> Ding Honghui >> -----BEGIN PGP SIGNATURE----- >> Version: GnuPG v1.4.6 (GNU/Linux) >> Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org >> >> iD8DBQFIBBGdJo9Njjkvy34RAv1NAKCRIOnyL93LGbpPt+wEZszeXBDrEgCfa8EH >> XGXrnYwI4GZHGgKls/FEPpU= >> =D4Nm >> -----END PGP SIGNATURE----- >> >> >> -- >> To UNSUBSCRIBE, email to [EMAIL PROTECTED] >> with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED] >> >> > > - -- Best Regards, Ding Honghui -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.6 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFIBDGfJo9Njjkvy34RArLCAKCGSTqcspZLkR0Lxi1rEZWcz2kDgACdHudj JIcgOJBg4Zeo/6N+L0kJcVI= =1nbZ -----END PGP SIGNATURE----- -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
