NETGEAR is aware of a growing number of phone and online scams. To learn how to stay safe click here.
Forum Discussion
Disk spin-down problem, RN104, fw 6.2.0
Hello,
I'm using RN104, 2 x 4GB WD RED in JBOD/Flex mode (I need no raid). No apps installed, samba, dlna services stopped to debug this issue. Http/https/ssh - enabled. I've enabled disk spin down after 5 minutes from GUI menu. System logs shows that disks are stopped, but then immediately started after 4-8 seconds.
After enabling write log: echo 1 > /proc/sys/vm/block_dump I see this in the dmesg:
I do believe, these writes wakes up the disks, becaue they happen in every 4-8 seconds. I've tried to kill leafp2p process, it doesn't seem to be the problem source. I can suspect that it's these writes are made by kernel, but don't have any knowledge about raid stuff. Same behaviour is when I type "hdparm -y /dev/sda". It spinds down, and wakes up in a second. Please advise.
I'm using RN104, 2 x 4GB WD RED in JBOD/Flex mode (I need no raid). No apps installed, samba, dlna services stopped to debug this issue. Http/https/ssh - enabled. I've enabled disk spin down after 5 minutes from GUI menu. System logs shows that disks are stopped, but then immediately started after 4-8 seconds.
Disk [0] going to standby...
Disk [0] spinning up...
Disk [0] going to standby...
Disk [0] spinning up...
After enabling write log: echo 1 > /proc/sys/vm/block_dump I see this in the dmesg:
md0_raid1(694): WRITE block 8 on sda1 (1 sectors)
md0_raid1(694): WRITE block 8 on sdb1 (1 sectors)
jbd2/md0-8(719): WRITE block 3684640 on md0 (8 sectors)
jbd2/md0-8(719): WRITE block 3684648 on md0 (8 sectors)
jbd2/md0-8(719): WRITE block 3684656 on md0 (8 sectors)
jbd2/md0-8(719): WRITE block 3684664 on md0 (8 sectors)
jbd2/md0-8(719): WRITE block 3684672 on md0 (8 sectors)
jbd2/md0-8(719): WRITE block 3684680 on md0 (8 sectors)
jbd2/md0-8(719): WRITE block 3684688 on md0 (8 sectors)
jbd2/md0-8(719): WRITE block 3684696 on md0 (8 sectors)
leafp2p(1282): WRITE block 775816 on md0 (8 sectors)
md0_raid1(694): WRITE block 8 on sda1 (1 sectors)
md0_raid1(694): WRITE block 8 on sdb1 (1 sectors)
I do believe, these writes wakes up the disks, becaue they happen in every 4-8 seconds. I've tried to kill leafp2p process, it doesn't seem to be the problem source. I can suspect that it's these writes are made by kernel, but don't have any knowledge about raid stuff. Same behaviour is when I type "hdparm -y /dev/sda". It spinds down, and wakes up in a second. Please advise.
23 Replies
Replies have been turned off for this discussion
- Do you have any apps installed? If so, which ones?
What services are you running? - There are no installed apps at the moment. Only http, https and ssh services enabled.
By the way, lsof /dev/sda, lsof /dev/sdb, doesn't show any access to files too. I think I could try reseting to factory defaults, to make sure that it's not something i've installed back ago. But I guess there is no way save data on the disks? Even if I disconnect them during reset, they will be wiped out once I connect them to fresh system? - Doing a factory reset with no disks installed achieves nothing as the whole purpose of a factory reset is to wipe the disks installed and do a clean install of the OS.
Backup your data, do a factory default (wipes all data, settings, everything) and restore your data form backup if you wish.
Any clues in the logs zip file around the time the disks are spun up which might explain what is going on? - I have a RN104 too.
I had the same problem. I delete and uninstall any apps and still did not work.
Running the top command from ssh I saw instances of mysql (but if I have not installed). I install mysql and then delete now Spindown works well.
When I reboot the NAS, mysql instances be running again !! Spindown now does not work well.
In the end I renamed the "mysql" script in the init.d folder to NAS not run this. Now Spindown works well with everything stopped.
When operating Transmission does not stop any of the 4 HD that I have, is this normal?
Not if it will be the same problem, I hope you can help with anything explained.
regards, - We use systemctl to disable services in OS6.
Transmission makes lots of reads. The disks won't spin down with that running. - Is "Continuous Protection" bit-rot? After turning that off I could get to 15 minutes but I guess I have too many apps and I couldn't figure out what was triggering all the activity so I gave up.
Is spindown incompatble with "Continuous Protection"?
steve - Continuous protection is snapshots, these will be taken at the frequency you set. I would assume the disks need to spin up to create a snapshot. Continuous protection should be turned off on transmission shares.
Bitrot protection is different.
I'm not certain how bitrot protection works, but I'm pretty sure it wouldn't be the root cause of your problem.
I have both active on several shares and my disks spin down as expected.
What I don't have are lots of apps, just one-LMS. - Thank you all for the answers. I've reset system to factory defaults, spin down worked ok. Ater installing Qbittorrent client, it seems that it was preventing from standby, even if app was removed later. I've just reset everything for the second time, and will try to stay away from qbittorrent. Maybe it's all it takes.
- As per support team, they told me that bitrot protection automatically detects and correct corruption due to media degradation.
- Yes that's what it does, question is, how?
NETGEAR Academy
Boost your skills with the Netgear Academy - Get trained, certified and stay ahead with the latest Netgear technology!
Join Us!
