Find the answer to your Linux question:
Results 1 to 7 of 7
kernel 2.6.20 included new real-time process IO statistics. Here is a fragment from the announcement: So this IO accounting implements per-process statistics of "storage I/O" (i.e.: I/O that _really_ does ...
Enjoy an ad free experience by logging in. Not a member yet? Register.
  1. #1
    Just Joined!
    Join Date
    Mar 2007
    Location
    Germany
    Posts
    5

    where is /proc/<pid>/IO ??


    kernel 2.6.20 included new real-time process IO statistics.

    Here is a fragment from the announcement:
    So this IO accounting implements per-process statistics
    of "storage I/O" (i.e.: I/O that _really_ does I/O on the storage device -
    Linux already had I/O storage statistics but it's not per-task). The data
    is reported through taskstats and procfs (/proc/$PID/io) (commit 1, 2, 3,
    4, 5, 6, 7, 8, 10, 11)

    I see no such thing in the several recent Linux distros I have tried out, with kernels up to 2.6.22. There are no /proc/<pid>/IO directories (I did see one of these, but the disk statistics were dead (small values that did not change)).

    There must be something special I must do to enable this.
    Where can I find information?

    thanks
    Mike

  2. #2
    Just Joined!
    Join Date
    Nov 2007
    Posts
    3
    You have to build support into the kernel as it's not enabled by default and once you've done that you can display the stats in a format similar to 'ps' with collectl - see Process I/O Stats but that's not all collectl can do. See the home page to learn more.
    -mark

  3. #3
    Just Joined!
    Join Date
    Mar 2007
    Location
    Germany
    Posts
    5
    Thanks for the tip.

    Why is this neat feature disabled by default? Is there an overhead problem? I cannot imagine why that would be the case.

  4. $spacer_open
    $spacer_close
  5. #4
    Just Joined!
    Join Date
    Apr 2008
    Posts
    2

    Ubuntu supports

    The new ubuntu scheduled for about 4 days from now comes with this feature enabled by default... I am currently using it to design a system resource monitor....

    now the output for this io file is as follows
    rchar: 296684
    wchar: 237786
    syscr: 2110
    syscw: 1654
    read_bytes: 516096
    write_bytes: 348160
    cancelled_write_bytes: 0

    Does anyone know the meaning of all these... some of them seem self explanitory but a few make confused... and its not something that is easy to find a legend for because as you mentioned not many distros use it for who knows what reason.

  6. #5
    Just Joined!
    Join Date
    Mar 2007
    Location
    Germany
    Posts
    5

    numbers do not change

    I just tried Ubuntu 8.04 and although the file /proc/<pid>/io is present, the data does not change even after thousands of disk accesses.

    What gives? Are you using the server or desktop version? Mine is desktop.

  7. #6
    Just Joined!
    Join Date
    Apr 2008
    Posts
    2

    unfortunatly

    I have the same issue as do... I decided that watching the characters written and read (rchar and wchar which do increase) would help me identify a system hog as well... although not my ideal choice... its what im stuck doing at the moment.

    if I needed access to this information on a 7.10 machine... how would I recompile the kernel to include this module???

  8. #7
    Just Joined!
    Join Date
    Mar 2007
    Location
    Germany
    Posts
    5
    There are numerous documents on the web about how to compile a kernel, including one in the Ubuntu community web site. I have never tried it myself. I did read that to enable the process level I/O statistics all you have to do is set a #define switch and then compile. There is no other module that needs to be added.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •