You are here: Home » Topic » Disk not spinning down

Disk not spinning down

This topic contains 6 replies, has 2 voices, and was last updated by  mwalsh 10 years, 7 months ago.

Viewing 7 posts - 1 through 7 (of 7 total)
  • Author
    Posts
  • #1229

    mwalsh
    Participant

    Hi,

    I’ve successfully unslung my NSLU2 and loaded Firefly and pointed it at my iTunes folder moved from My Music to the NSLU2 using the TweakUI modification to the My Music folder location as detailed here:

    http://forums.fireflymediaserver.org/viewtopic.php?t=5470&postdays=0&postorder=asc&start=15

    This seems to be working fine from iTunes and also accessing Firefly through iTunes as well but I don’t yet have a SoundBridge to test it with.

    I’ve unslung to disk 1 which is a Seagate 320GB drive and the only drive attached to the NSLU2, which from the reviews below I was hoping would spin down no problem:

    “what i like about this is that its virtually silent when running and the drive spins down automatically after a period of inactivity. its solid construction with legendary seagate reliability. Connect direct to you pc via the supplied usb cable or, as i have done, attach to a network storage device ( nslu2 ). whatever you do with it, 320gigs is going to take some filling !”

    “Originally intended to buy just one of these drives to go with a 120GB usb drive attached to a Linksys NSLU2. After testing found to be much faster quieter than existing drive. I was so impressed with the quality and performance of this drive that I purchased another about a month later and now have both connected to the NSLU2 as a network storage system. For this type of use they are great as the power save feature spins down the drives after about five or so minutes.”

    Source: http://www.ebuyer.com/customer/products/index.html?rb=26638297344&action=c2hvd19wcm9kdWN0X3Jldmlld3M=&product_uid=116092

    It does spin down after about 5 minutes when the NSLU2 is switched off so I get the impression the spin down is working it’s just that there must be some sort of activity that is keeping the drive spinning all the time? Perhaps this is a side effect of not using a USB memory stick for the system drive?

    My slug started with the R23 Linksys firmware and I’m pretty sure the spin down was working then. During the unslinging process I upgraded the firmware to R63 as I was getting the “no enough memory” error, I’ve seen reports that the R63 firmware has caused spin down problems? If that’s the case can I roll back my unslung to R29?

    I’m a complete newbie to linux and I’ve only managed to get this far thanks to the excellent resources on the web, this being one of them.

    mt-daapd/Firefly Version: version svn-1518
    Server Hardware: NSLU2
    Server Software: V2.3R63-uNSLUng-6.8-beta
    iTunes 6 for PC

    #9794

    richdunlop
    Participant

    @mwalsh wrote:

    It does spin down after about 5 minutes when the NSLU2 is switched off so I get the impression the spin down is working it’s just that there must be some sort of activity that is keeping the drive spinning all the time? Perhaps this is a side effect of not using a USB memory stick for the system drive?

    You’re right, something is accessing the disk from time to time. There are various things you can do about this in a single disk set up such as mounting certain areas of the filesystem (e.g. /var) in a ramdisk or mounting things with -noatime. You’ll need to pick through the Unslung wiki but all the info is on there. Here’s a page that might help:

    http://www.nslu2-linux.org/wiki/HowTo/SetSpinDownTimeOnMaxtorOneTouch

    The starting point would be to figure out what’s doing the disk access though…

    #9795

    mwalsh
    Participant

    Thanks for the reply. I really appreciate it.

    OK, this is where my linux knowledge is hampering me as I have no idea how I can create the rc.local and rc.bootbin files using telnet!

    #9796

    richdunlop
    Participant

    @mwalsh wrote:

    Thanks for the reply. I really appreciate it.

    OK, this is where my linux knowledge is hampering me as I have no idea how I can create the rc.local and rc.bootbin files using telnet!


    cd /unslung
    touch rc.bootbin
    touch rc.local

    However before you get too sucked into this it might be worth figuring out why your disk isn’t spinning down first. I used to have a single disk setup and the disk was spundown quite a lot of the time. For example, what rescan interval have you set for Firefly? Are there frequently scheduled jobs in crontab such as ntpclient stuff? If there isn’t much running on the slug it shouldn’t be doing much disk access.

    #9797

    mwalsh
    Participant

    It’s working!

    I followed the details you provided here:

    http://www.nslu2-linux.org/wiki/HowTo/SetSpinDownTimeOnMaxtorOneTouch

    I originally avoided this page as I’m not using a Maxtor drive (I’m using a Seagate) so maybe that should be renamed to be inclusive of all drives just so people don’t cast it to one side.

    I added the rc.local edit:

    # Kill the onetouch_detect processes so that the drive can spin down
    /bin/killall onetouch_detect

    I also un-commented the “return 1” line at the bottom.

    Lastly I renamed sda1 to sdb1 in rc.local and then typed “mount” in telnet to check that the disk had been mounted with “noatime”.

    I did this by unmounting the drive using a Knoppix boot CD (just a matter booting Knoppix with the drive attached and then shutting down) and then using Ext3 IFS to access the disk in Windows. I couldn’t edit the files using Knoppix but I could in Windows. Then I used a nifty little utility called dos2unix to convert the line breaks to unix. Probably the long way around I know but it worked in the end!

    I haven’t made any edits to the crontab but perhaps I should as I have a feeling the drive might wake up due to the clock being reset every hour? We’ll see.

    Thanks so much for your help and I hope this thread will help others who hit the same problem.

    #9798

    richdunlop
    Participant

    @mwalsh wrote:

    It’s working!

    😀

    @mwalsh wrote:

    Lastly I renamed sda1 to sdb1 in rc.local and then typed “mount” in telnet to check that the disk had been mounted with “noatime”.

    Keep the noatime behaviour in mind. If in the future you install a piece of software that makes use of file timestamps then you might see some odd behaviour.

    @mwalsh wrote:

    I haven’t made any edits to the crontab but perhaps I should as I have a feeling the drive might wake up due to the clock being reset every hour? We’ll see.

    The slug clock drifts quite a lot. It’s probably worth leaving the hourly synch in there. In fact, if you haven’t done so already there’s an unslung wiki page that explains how to setup ‘ntpclient’ on the slug.

    #9799

    mwalsh
    Participant

    OK, will do.

    Thanks again.

Viewing 7 posts - 1 through 7 (of 7 total)

You must be logged in to reply to this topic.