2.4.0-test5, raidtools2 and ReiserFS 3.6.11 hang

From: Russell Coker (russell@coker.com.au)
Date: Wed Aug 09 2000 - 17:02:28 EDT

  • Next message: Russell Coker: "2.4.0-test5 NFS kernel server with ReiserFS and 2.4.0-test4 NFS"

    I have been doing some more intensive testing on RAID with recent Linux.
    This machine has a RAID-1 array of two IBM 46G ATA-66 disks on the same
    channel (hda and hdb). Here is the df output:

    Filesystem 1k-blocks Used Available Use% Mounted on
    /dev/ide/host0/bus0/target0/lun0/part3
                           1228916 1180672 48244 96% /
    /dev/ide/host0/bus0/target0/lun0/part2
                             23333 7405 14724 33% /boot
    /dev/md/0 43510604 18926016 24584588 43% /mnt

    /mnt is the RAID array in question. It is a ReiserFS file system, it had
    previously been run as Ext2 and I had run all the same tests on it (and
    more). However I am not certain that it is ReiserFS, it could be that
    ReiserFS triggers some other bug in the kernel.
    Here is the top output showin the kswapd process being in D state
    permanently, and nothing much happening (the load average is high because of
    a few "ls" processes that are stuck on "ls -la /mnt").

     11:01pm up 5 days, 4:42, 2 users, load average: 3.53, 2.58, 2.21
    42 processes: 40 sleeping, 2 running, 0 zombie, 0 stopped
    CPU states: 0.0% user, 7.2% system, 2.3% nice, 4.3% idle
    Mem: 256120K av, 207644K used, 48476K free, 0K shrd, 38104K buff
    Swap: 265032K av, 13336K used, 251696K free 153024K cached
     
      PID USER PRI NI SIZE RSS SHARE STAT LIB %CPU %MEM TIME COMMAND
        2 root 20 0 0 0 0 SW 0 60.2 0.0 3046m kapmd
        1 root 9 0 96 48 40 S 0 0.0 0.0 0:15 init
        3 root 9 0 0 0 0 DW 0 0.0 0.0 74:43 kswapd
        4 root 9 0 0 0 0 SW 0 0.0 0.0 5:55 kflushd
        5 root 9 0 0 0 0 SW 0 0.0 0.0 0:07 kupdate
        6 root -1 -20 0 0 0 SW< 0 0.0 0.0 0:00 mdrecoveryd
        7 root 9 0 0 0 0 SW 0 0.0 0.0 0:04 kreiserfsd
       33 root 9 0 232 164 72 S 0 0.0 0.0 0:00 devfsd
       66 root -1 -20 0 0 0 SW< 0 0.0 0.0 0:10 raid1d
       67 root 19 19 0 0 0 SWN 0 0.0 0.0 0:19 raid1syncd
      195 root 9 0 276 216 172 S 0 0.0 0.0 0:00 syslogd

    Here is the PS output showing the bonnie++ process that is stuck. At this
    time the hard drive access light was always off (so the system is indeed
    stuck).

    root@test:~#ps ax | grep bonn
     3759 ? DN 34:31 /usr/sbin/bonnie++ -q -s 40960 -n 0 -f

    This test is using 40 files of size 1G for testing (40G of temporary storage
    in total). Previous tests of 4G and 8G went fine.

    Please let me know of any other data you would like. 

    -- 
My current location - X marks the spot.
X
X
X

    -
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/

    This archive was generated by hypermail 2b29 : Wed Aug 09 2000 - 17:46:44 EDT