jbuszkie

Members
  • Posts

    693
  • Joined

  • Last visited

Posts posted by jbuszkie

  1.  

    But this is true regardless of which version of UnRAID you're using, so I'm a bit surprised the slowdown seem notably worse than it was with v5.    Did you have your "disk tunables" optimized with v5?  ... it could simply be that these are different with v6, and this is impacting the buffering for read/write operations and thus causing different performance during times when the disks are VERY busy.

    Since I don't know what "disk tunables" was for V5..  My guess is I didn't have it.  I did have the:

    for i in /dev/md*
    do
    echo Setting $i
      blockdev --setra 2048 $i
      blockdev --getra $i
    done

     

    In the go file for V5..  But I don't have it for V6.  Has anyone shown that it makes a difference?

     

    And I guess I should look up disk tunables?

     

    Thanks,

     

    Jim

     

  2. Ok..  I had the bios specifically slow down the CPU.  I did that because I didn't need the HP and to save power.  But now with V6 it seems I DO need the HP! So I bumped it up to the default.

     

    So my question still stands!  How does The kernel decide what the fastest and slowest speeds are for CPU?  Right now there are two speeds, 2.5G and 1.6G

    I'd like to slow down even further.  How do I do this?

     

    Thanks,

     

    Jim

     

  3. Ok..  so I solved/improved my parity check speed by changing to a 2 core processor...  see here. but now my array isn't responsive while doing a parity check.  It's reallly sssssllllloooooowwww when accessing files.  My automatic acronis backup

    fails when before it would happen just fine during a parity check with V5.

    Is there some way to lower the priority of the parity check or something else I can do?  I can't really have the array non responsive while the parity check is going on.

    I mean I can tolerate some slowness..  but if programs fail then that's just not good.

  4. I just put in a core2duo to replace my older celeron. (until I can upgrade the whole system)

    It has two cores with speedstep.  I'm trying to figure out how it determines to change the speed?

    The dashboard has both cores at 1200MHz at idle and rams them up to 1700MHz under load (parity test)  I think my cpu is capable of 2300MHz?

     

    Why didn't it go to 2300?  can I set the min to be lower than 1200?  like maybe 800MHz?

    Where are these things set?

     

    Thanks,

     

    Jim

  5. disk got a bunch of read errors!  Will this disk now be red balled?
    Theoretically the drive accepted the writes to replace the data that could not be read. When a drive has read errors, unraid reads all the other drives and calculates what should have been in the location that had read errors, and issues a write command to put the data that couldn't be read back onto the drive. If the write fails, the drive is red balled (X'ed?) otherwise if the write returns success, the drive stays in service and the error count for that drive is incremented.

     

    I'd disable spindown for that drive and run a long smart test, then get a smart report and post it.

     

    If you want help deciding what drive needs to be replaced next, get smart reports on all your drives and post them.

    Thanks!  So if the drive's not redballed now it should be there on a restart, right?

    Now does the long smart test have to be done with the array offline?

     

    And how can I just *read* everything on the drive to see if I get the errors again? something like dd to /dev/null.

     

    Jim

  6. Aww crud..  I start a parity check because I just upgraded to V6.1 and I was testing speeds and I had a couple drives with a current sector pending.  one went away..  but another disk got a bunch of read errors!  Will this disk now be red balled?  The parity check didn't get any errors.  but I got 113 read errors.

     

    Sep 24 03:05:01 Tower kernel: sd 7:0:0:0: [sdf] tag#3 CDB: opcode=0x28 28 00 a5 a6 78 5f 00 05 40 00 (Drive related)
    Sep 24 03:05:01 Tower kernel: blk_update_request: I/O error, dev sdf, sector 2779150872 (Errors)
    Sep 24 03:05:01 Tower kernel: ata7: EH complete (Drive related)
    Sep 24 03:05:01 Tower kernel: md: disk6 read error, sector=2779150808 (Errors)
    Sep 24 03:05:01 Tower kernel: md: disk6 read error, sector=2779150816 (Errors)
    Sep 24 03:05:01 Tower kernel: md: disk6 read error, sector=2779150824 (Errors)
    Sep 24 03:05:01 Tower kernel: md: disk6 read error, sector=2779150832 (Errors)
    Sep 24 03:05:01 Tower kernel: md: disk6 read error, sector=2779150840 (Errors)
    Sep 24 03:05:01 Tower kernel: md: disk6 read error, sector=2779150848 (Errors)
    Sep 24 03:05:01 Tower kernel: md: disk6 read error, sector=2779150856 (Errors)
    Sep 24 03:05:01 Tower kernel: md: disk6 read error, sector=2779150864 (Errors)
    Sep 24 03:05:01 Tower kernel: md: disk6 read error, sector=2779150872 (Errors)
    Sep 24 03:05:01 Tower kernel: md: disk6 read error, sector=2779150880 (Errors)
    Sep 24 03:05:01 Tower kernel: md: disk6 read error, sector=2779150888 (Errors)
    Sep 24 03:05:01 Tower kernel: md: disk6 read error, sector=2779150896 (Errors)
    .
    .
    .

     

    What should I do?  Since the parity check didn't give any errors??

    The disk is 5 years old.. but I have a couple that are 6 that are next in line to be replaced..

     

    Jim

  7. The V6 version isn't installed from unmenu..  from my experience...

     

    I got mine from here:

     

    http://lime-technology.com/forum/index.php?topic=31735.0

     

    For V6:

    Paste this line into the 'Install Plugin' line in the 'Plugins' webgui and click on install.  This will install the plugin.  Don't install the plugin manually.  Let the plugin manager handle it.

     

  8. I just installed the latest V6.1 and tried a parity check...  I'm getting about ~22-24MB/sec.

    It says it will take 2 days to  complete!!

    With V5 it's never taken 2 days!

    Is V6 expected to be slower?  My CPU is pegged.

     

    What could be going on here?

     

    I had this in my old go file... Does it need to be in the new one as well?

    for i in /dev/md*
    do
    echo Setting $i
      blockdev --setra 2048 $i
      blockdev --getra $i
    done
    

     

    Here are the top 3 processes...

     

     

    top - 16:00:12 up  1:57,  1 user,  load average: 3.08, 2.77, 2.61
    Tasks: 122 total,   3 running, 119 sleeping,   0 stopped,   0 zombie
    Cpu(s): 10.0%us, 50.8%sy,  0.0%ni,  0.0%id,  0.0%wa,  0.0%hi, 39.2%si,  0.0%st
    Mem:   2052560k total,  1050636k used,  1001924k free,   180024k buffers
    Swap:        0k total,        0k used,        0k free,   534564k cached
    
      PID USER      PR  NI  VIRT  RES  SHR S %CPU %MEM    TIME+  COMMAND
    1030 root       0 -20     0    0    0 S 38.1  0.0  15:30.87 kworker/0:1H
    1543 root      20   0     0    0    0 R 27.2  0.0  11:01.18 unraidd
    1433 root      20   0     0    0    0 D 18.9  0.0   7:21.20 mdrecoveryd
    

  9. Basically..  MyMain is the only thing I'm looking for.  I really like the fact that I can change the view..  like with the inventory I can set and view all my purchase dates and info and set usage notes..  I also use the network view sometimes.

     

    I agree the V6 GUI is a much improved interface..  but nothing still beats "MyMain"

     

    Oh..  I also like the color coded syslog too of unmenu!! :-)

     

     

  10. More info...

     

    I found the CPU info it's

    Intel Celeron 430 Conroe-L Single-Core 1.8 GHz LGA 775 35W BX80557430 Processor

     

    So it is only Single core.  I think I bought it because it was low power...  But now maybe it's too old?  Does it not have enough juice now?

    I bought it back in July of 2009!!!  I thought it was only 2 years old...  but I guess it's much older..

     

    New Question.....

     

    If I get a new MB/CPU..  Will my power consumption be less (taking the drives out of the equation)??  With all the throttling?

     

    Thanks,

     

    Jim

  11. What should my CPU utilization be?  I'm copying some files and my CPU (according to the dash board is 90-100%)

    SHould it be that high

     

    top - 10:33:09 up  1:29,  1 user,  load average: 1.87, 1.56, 0.79
    Tasks: 119 total,   2 running, 117 sleeping,   0 stopped,   0 zombie
    Cpu(s): 48.8%us, 34.0%sy,  0.0%ni, 11.1%id,  2.4%wa,  0.0%hi,  3.7%si,  0.0%st
    Mem:   2052560k total,  1965952k used,    86608k free,   101620k buffers
    Swap:        0k total,        0k used,        0k free,  1565172k cached
    
      PID USER      PR  NI  VIRT  RES  SHR S %CPU %MEM    TIME+  COMMAND
    2684 jbuszkie  20   0  294m  27m  17m R 25.9  1.4   1:28.24 smbd
    11455 nobody    20   0  289m  19m  17m S 18.9  1.0   1:16.68 smbd
    1660 root      20   0  290m  11m  668 S  9.3  0.6   0:46.48 shfs
    1432 root      20   0 89100 3552 3008 S  1.0  0.2   0:12.93 emhttp

     

    The dash board says the core is running at 1800MHz

     

    After the copy finished, The CPU is still about 60%?

     

    I just upgraded to V6 from V5.05 and I'm wondering if this is normal?  I can't remember what my old CPU utilization was...

     

    How can I tell  how many cores I have and what model of CPU?  I honestly can remember! :-)

     

    Thanks,

     

    Jim