Unwanted spin up, how to correctly use blkdevMonitor

Discussion about hard drive spin down (standby) feature of NAS.
manny2003
Getting the hang of things
Posts: 74
Joined: Sun Feb 15, 2015 8:57 am

Unwanted spin up, how to correctly use blkdevMonitor

Post by manny2003 » Sun Feb 15, 2015 9:47 am

Hi everybody,
I'm a new Qnap user that just received his brand new HS-251. I'm trying to instruct myself on how to configure my NAS in the best way possible and to do this I'm trying to read as much as possible of the threads in this forum, but I still need some answers if it is possible.

I would like the NAS to spin down, just because it is sometimes unused for long period (it is an home installation) and also because it is installed in my bedroom. I don't want to shutdown the NAS since someone in other parts of the house would like to use it also late at night, but I would like it to stay as quieter as possible when effectively nobody is using it, so I would like it to spin down.
Luckily I have no problem to have the disk to spin down after 20min, but during the day it spin up also when there is no devices powered on in the network. I understand there could be external packet that could wake it up, but I switched off myqnapcloud and services other than photo and music station. No DHCP client, no NTP, no download station.
I know sometime there will be scheduled maintenance, I'm ok with them, but it wake up from 5 to 10 times per day without apparently reason. So I would like to better understand the reason why this happens and I found this very useful script, blkdevMonitor. I just would like to understand if I know how to correctly use it and if it suites my need.

1) I installed the script and I run it through an SSH session. If I have well understood I need to let it run until it finishes the 100 tests without using the NAS in anyway. Is this correct?

2) As I said, the NAS goes in sleep mode after 20 min (in my case), but will it stay awake during the test due to the "echo /proc/sys/vm/block_dump" command?

3) What is in fact a "test"? The occurrence of a write/read activity?

4) I'm running the tests at the moment and after 40 minutes all I have are md1_raid1 WRITE and vgdisplay READ processes. What does this mean? They are in fact activities to the disk that logically will keep it awake, but are they just related to the test itself?

5) I understand that the script is a great tool to discover processes that keep the NAS awake, but to find out exactly which process is the one who wake up the NAS once it is sleeping, is there any other method that will log the first process who asked the NAS to spin up?

Sorry for my many questions, I try hard to understand on my own before wasting your times, but some help in order to make me a better investigator will be very appreciated!
Thank!

User avatar
pwilson
Guru
Posts: 22568
Joined: Fri Mar 06, 2009 11:20 am
Location: Victoria, BC, Canada (UTC-08:00)

Re: Unwanted spin up, how to correctly use blkdevMonitor

Post by pwilson » Mon Feb 16, 2015 1:55 pm

manny2003 wrote:Hi everybody,
I'm a new Qnap user that just received his brand new HS-251. I'm trying to instruct myself on how to configure my NAS in the best way possible and to do this I'm trying to read as much as possible of the threads in this forum, but I still need some answers if it is possible.

I would like the NAS to spin down, just because it is sometimes unused for long period (it is an home installation) and also because it is installed in my bedroom. I don't want to shutdown the NAS since someone in other parts of the house would like to use it also late at night, but I would like it to stay as quieter as possible when effectively nobody is using it, so I would like it to spin down.
Luckily I have no problem to have the disk to spin down after 20min, but during the day it spin up also when there is no devices powered on in the network. I understand there could be external packet that could wake it up, but I switched off myqnapcloud and services other than photo and music station. No DHCP client, no NTP, no download station.
I know sometime there will be scheduled maintenance, I'm ok with them, but it wake up from 5 to 10 times per day without apparently reason. So I would like to better understand the reason why this happens and I found this very useful script, blkdevMonitor. I just would like to understand if I know how to correctly use it and if it suites my need.

1) I installed the script and I run it through an SSH session. If I have well understood I need to let it run until it finishes the 100 tests without using the NAS in anyway. Is this correct?

2) As I said, the NAS goes in sleep mode after 20 min (in my case), but will it stay awake during the test due to the "echo /proc/sys/vm/block_dump" command?

3) What is in fact a "test"? The occurrence of a write/read activity?

4) I'm running the tests at the moment and after 40 minutes all I have are md1_raid1 WRITE and vgdisplay READ processes. What does this mean? They are in fact activities to the disk that logically will keep it awake, but are they just related to the test itself?

5) I understand that the script is a great tool to discover processes that keep the NAS awake, but to find out exactly which process is the one who wake up the NAS once it is sleeping, is there any other method that will log the first process who asked the NAS to spin up?

Sorry for my many questions, I try hard to understand on my own before wasting your times, but some help in order to make me a better investigator will be very appreciated!
Thank!


  1. It isn't very useful if you don't provide the output. Please post the output to pastebin.com, and then provide the URL to your pastebin.com output.
  2. Did you remember to run it from the RAMDisk?

    Code: Select all

    cd /tmp
    wget ftp://csdread:csdread@ftp.qnap.com/NAS/spindown/blkdevMonitor.sh
    chmod +x /tmp/blkdevMonitor.sh
    /tmp/blkdevMonitor.sh >/tmp/blkdevMonitor.log &
    tail -f /tmp/blkdevMonitor.log
    #done

  3. Have you reviewed QNAPedia article: Hard drives spindown issues yet?
  4. Have you reviewed QNAPedia article: How to know which process prevents the NAS HDD from spindown?

Those commands above with create /tmp/blkdevMonitor.log. You can "cat" it in order to display it a second time so that you can cut&paste it to pastebin.com.

Code: Select all

cat /tmp/blkdevMonitor.log
cp /tmp/blkdevMonitor.log /share/Public/
sed 's/$'"/`echo \\\r`/" /tmp/blkdevMonitor.log > /share/Public/blkdevMonitor.txt
#done



You can also download this output from your "Public" share under the filename: "blkdevMonitor.log" in Linux text format, or as "blkdevMonitor.txt" to get it in Windows/DOS text format.

Patrick M. Wilson
Victoria, BC Canada
QNAP TS-470 Pro w/ 4 * Western Digital WD30EFRX WD Reds (RAID5) - - Single 8.1TB Storage Pool FW: QTS 4.2.0 Build 20151023 - Kali Linux v1.06 (64bit)
Forums: View My Profile - Search My Posts - View My Photo - View My Location - Top Community Posters
QNAP: Turbo NAS User Manual - QNAP Wiki - QNAP Tutorials - QNAP FAQs

Please review: When you're asking a question, please include the following.

manny2003
Getting the hang of things
Posts: 74
Joined: Sun Feb 15, 2015 8:57 am

Re: Unwanted spin up, how to correctly use blkdevMonitor

Post by manny2003 » Mon Feb 16, 2015 2:09 pm

Thank you very much Patrick for your reply.

I will post the result as soon as I can get it, it took about 10 hrs to run the 100 tests last time.
I knew about the links you gave me, and the first one is the one I followed. Since it says to run "cp /share/Public/blkdevMonitor.sh /root" I did not run the command from the RAMDisk! I was missing this information and it was the one I was looking for in order to avoid logging the script itself.

I will experiment a little more and I will come back. Thank you again for your support, very appreciated.

User avatar
pwilson
Guru
Posts: 22568
Joined: Fri Mar 06, 2009 11:20 am
Location: Victoria, BC, Canada (UTC-08:00)

Re: Unwanted spin up, how to correctly use blkdevMonitor

Post by pwilson » Mon Feb 16, 2015 2:28 pm

manny2003 wrote:Thank you very much Patrick for your reply.

I will post the result as soon as I can get it, it took about 10 hrs to run the 100 tests last time.
I knew about the links you gave me, and the first one is the one I followed. Since it says to run "cp /share/Public/blkdevMonitor.sh /root" I did not run the command from the RAMDisk! I was missing this information and it was the one I was looking for in order to avoid logging the script itself.

I will experiment a little more and I will come back. Thank you again for your support, very appreciated.


/root is also part of the RAMDisk, so that will work too. Many people don't like having /root in a RAM Disk, so some NASAdmins do strange things with it. Using /tmp, is safer, especially if I don't know what modifications the NAS admin has done to their NAS devices.

Patrick M. Wilson
Victoria, BC Canada
QNAP TS-470 Pro w/ 4 * Western Digital WD30EFRX WD Reds (RAID5) - - Single 8.1TB Storage Pool FW: QTS 4.2.0 Build 20151023 - Kali Linux v1.06 (64bit)
Forums: View My Profile - Search My Posts - View My Photo - View My Location - Top Community Posters
QNAP: Turbo NAS User Manual - QNAP Wiki - QNAP Tutorials - QNAP FAQs

Please review: When you're asking a question, please include the following.

manny2003
Getting the hang of things
Posts: 74
Joined: Sun Feb 15, 2015 8:57 am

Re: Unwanted spin up, how to correctly use blkdevMonitor

Post by manny2003 » Mon Feb 16, 2015 2:34 pm

Ok! So is it normal that after running the script the disk never spin down or this should not happen?
In order to test I set the spin down timer to 5 min and it works well but during the running of the script the disks never sleep.
Thanks again.
Qnap NAS: HS-251 OS Ver. 4.2.2 20160901 | HD: 2x WD40EFRX | RAID: 1 | Used OS: Windows 10 x64
Qnap NAS: TS-251-4GB OS Ver. 4.2.2 20160901 | HD: 1x WD40EFRX | RAID: 0 | Used OS: Windows 10 x64

User avatar
pwilson
Guru
Posts: 22568
Joined: Fri Mar 06, 2009 11:20 am
Location: Victoria, BC, Canada (UTC-08:00)

Re: Unwanted spin up, how to correctly use blkdevMonitor

Post by pwilson » Mon Feb 16, 2015 2:54 pm

manny2003 wrote:Ok! So is it normal that after running the script the disk never spin down or this should not happen?
In order to test I set the spin down timer to 5 min and it works well but during the running of the script the disks never sleep.
Thanks again.


I was just trying to assist you in getting the output you said you wanted. I personally don't care if my drives spindown or not, so I don't worry about it here at all. My NAS operates 24/7/365. As such I can't help you very much with troubleshooting this.

If you are running your NAS as a Local Master Browser, or are running XBMC or any of the QPKGs available for your NAS it may never sleep. You can provide the output (via Pastebin.com) if you want others to look at it. No one can help you troubleshoot this if you don't actually provide the output.

Patrick M. Wilson
Victoria, BC Canada
QNAP TS-470 Pro w/ 4 * Western Digital WD30EFRX WD Reds (RAID5) - - Single 8.1TB Storage Pool FW: QTS 4.2.0 Build 20151023 - Kali Linux v1.06 (64bit)
Forums: View My Profile - Search My Posts - View My Photo - View My Location - Top Community Posters
QNAP: Turbo NAS User Manual - QNAP Wiki - QNAP Tutorials - QNAP FAQs

Please review: When you're asking a question, please include the following.

manny2003
Getting the hang of things
Posts: 74
Joined: Sun Feb 15, 2015 8:57 am

Re: Unwanted spin up, how to correctly use blkdevMonitor

Post by manny2003 » Mon Feb 16, 2015 3:09 pm

Yes I can understand your point of view about let the nas operate 24/7. I was just trying to understand how blkdevMonitor works and if it prevent itself the spin down, since I would like to have the disk spinning down during the test to understand what then spin them up.
I was thinking the problem was the fact that I was not using the ramdisk, but finally you told me root is in ram, so at the moment I don't understand why the disk spin down except when blkdevMonitor is testing. I will study some more... :)
Qnap NAS: HS-251 OS Ver. 4.2.2 20160901 | HD: 2x WD40EFRX | RAID: 1 | Used OS: Windows 10 x64
Qnap NAS: TS-251-4GB OS Ver. 4.2.2 20160901 | HD: 1x WD40EFRX | RAID: 0 | Used OS: Windows 10 x64

User avatar
schumaku
Guru
Posts: 43664
Joined: Mon Jan 21, 2008 4:41 pm
Location: Kloten (Zurich), Switzerland -- Skype: schumaku
Contact:

Re: Unwanted spin up, how to correctly use blkdevMonitor

Post by schumaku » Tue Feb 17, 2015 2:32 am

Suspect the legacy blkdevMonitor.sh does not cover the new volume manager structures, essential information might be lost resp. filtered and not shown on the screen. All thse will end in the Kernel output - just a ring buffer accessible using # dmesg

User avatar
pwilson
Guru
Posts: 22568
Joined: Fri Mar 06, 2009 11:20 am
Location: Victoria, BC, Canada (UTC-08:00)

Re: Unwanted spin up, how to correctly use blkdevMonitor

Post by pwilson » Tue Feb 17, 2015 3:24 am

schumaku wrote:Suspect the legacy blkdevMonitor.sh does not cover the new volume manager structures, essential information might be lost resp. filtered and not shown on the screen. All thse will end in the Kernel output - just a ring buffer accessible using # dmesg


I suspect you are right. Yes, "dmesg" output is very useful too, but the OP was explicitly asking about blkdevMonitor.sh, so I didn't provide this information to him.

Patrick M. Wilson
Victoria, BC Canada
QNAP TS-470 Pro w/ 4 * Western Digital WD30EFRX WD Reds (RAID5) - - Single 8.1TB Storage Pool FW: QTS 4.2.0 Build 20151023 - Kali Linux v1.06 (64bit)
Forums: View My Profile - Search My Posts - View My Photo - View My Location - Top Community Posters
QNAP: Turbo NAS User Manual - QNAP Wiki - QNAP Tutorials - QNAP FAQs

Please review: When you're asking a question, please include the following.

manny2003
Getting the hang of things
Posts: 74
Joined: Sun Feb 15, 2015 8:57 am

Re: Unwanted spin up, how to correctly use blkdevMonitor

Post by manny2003 » Tue Feb 17, 2015 7:36 am

schumaku wrote:Suspect the legacy blkdevMonitor.sh does not cover the new volume manager structures, essential information might be lost resp. filtered and not shown on the screen. All thse will end in the Kernel output - just a ring buffer accessible using # dmesg


I think I found an answer to my first concern: why the blkdevMonitor in my case prevent the normal spin down?
I blindly followed the instruction given here: http://wiki.qnap.com/wiki/Hard_drives_spindown_issues where it says to execute:

Code: Select all

cp /share/Public/blkdevMonitor.sh /root

I did not notice that doing that I was creating the root folder that does not exist originally in this firmware release as it is instead called new_root.
I then modified the script to use new_root instead of root to save logs (instead of tmp since I had some trouble with tmp path using sftp connection that found a different tmp under .qpkg :ashamed: ) and now the script seems to run without preventing the normal spin down.

Second thing I noticed is that I needed to disable the download station also if it was not in use. It seems that the DHT service run anyway generating network traffic and maybe triggering some unwanted spin up events.

I found a thread where you discuss with SnakeByte who modified the script in order to deal with the missing info in dmesg.

I understood I should replace this

Code: Select all

while [ 1 ]; do
      /bin/dd if=/proc/kmsg of=$_klog bs=1 count=10240 2>/dev/null 1>/dev/null
      for i in $MD_DEVIVES; do
         /bin/cat $_klog | /bin/grep $i | /bin/grep "dirtied inode"
         if [ $? = 0 ]; then
            /bin/cat $_klog | /bin/grep $i | /bin/grep "dirtied inode" >> $BLKDEV_LOG
            _standby=0
         fi
         /bin/cat $_klog | /bin/grep $i | /bin/grep "block"
         if [ $? = 0 ]; then
            /bin/cat $_klog | /bin/grep $i | /bin/grep "block" >> $BLKDEV_LOG
            _standby=0
         fi
      done
      for i in $SD_DEVIVES; do
         /bin/cat $_klog | /bin/grep $i | /bin/grep "dirtied inode"
         if [ $? = 0 ]; then
            /bin/cat $_klog | /bin/grep $i | /bin/grep "dirtied inode" >> $BLKDEV_LOG
            _standby=0
         fi
         /bin/cat $_klog | /bin/grep $i | /bin/grep "block"
         if [ $? = 0 ]; then
            /bin/cat $_klog | /bin/grep $i | /bin/grep "block" >> $BLKDEV_LOG
            _standby=0
         fi
      done
      if [ $_standby = 0 ]; then
         return 1
      fi
   done


with this

Code: Select all

while [ 1 ]; do
      /bin/dd if=/proc/kmsg of=$_klog bs=1 count=10240 2>/dev/null 1>/dev/null
      /bin/cat $_klog | /bin/grep -E 'dirtied inode|block' | /bin/grep -v -E 'tmpfs|proc'
                if [ $? = 0 ]; then
                        /bin/cat $_klog | /bin/grep -E 'dirtied inode|block' | /bin/grep -v -E 'tmpfs|proc' >> $BLKDEV_LOG
                        _standby=0
                fi      
                if [ $_standby = 0 ]; then
         return 1
      fi
   done


Hope to have well understood, I cannot hide I am not used to work a lot with Linux system and shell scripting.

At the moment the NAS seems to sleep quite well, but I really would like to be ready with a good tool to investigate future unwanted spin up, at least in this first period where I am trying to find the best way to organize the use of the NAS for my needs.

As usual, thank you very much.
Qnap NAS: HS-251 OS Ver. 4.2.2 20160901 | HD: 2x WD40EFRX | RAID: 1 | Used OS: Windows 10 x64
Qnap NAS: TS-251-4GB OS Ver. 4.2.2 20160901 | HD: 1x WD40EFRX | RAID: 0 | Used OS: Windows 10 x64

User avatar
schumaku
Guru
Posts: 43664
Joined: Mon Jan 21, 2008 4:41 pm
Location: Kloten (Zurich), Switzerland -- Skype: schumaku
Contact:

Re: Unwanted spin up, how to correctly use blkdevMonitor

Post by schumaku » Wed Feb 18, 2015 6:54 am

Looks OK to me - except of the /root thingie you talk about.

QNAP does still use modded sshd and telnetd ignoring the uer default folder from /etc/passwd - if login as admin, the default login folder will be /root ... and it readily exists in the RAM root (/) disk:

[~] # pwd
/root
[~] # cd ..
[/] # ls -ls | grep root
0 drwxr-xr-x 2 admin administ 40 Jan 27 16:59 new_root/
0 drwxr-xr-x 2 admin administ 260 Feb 17 14:39 root/

manny2003
Getting the hang of things
Posts: 74
Joined: Sun Feb 15, 2015 8:57 am

Re: Unwanted spin up, how to correctly use blkdevMonitor

Post by manny2003 » Wed Feb 18, 2015 8:52 am

schumaku wrote:Looks OK to me - except of the /root thingie you talk about.

QNAP does still use modded sshd and telnetd ignoring the uer default folder from /etc/passwd - if login as admin, the default login folder will be /root ... and it readily exists in the RAM root (/) disk:

You are right. This is due to my lacks in Linux system! :ashamed:
I was looking in mount -l and I did not find the / mount type, I finally founded with cat /proc/mounts where it is well a tmpfs. So it has been a coincidence that now the script run without prevent the spin down.

I still have some unwanted spin up, but I will dig into them in the next days using the modified script. I am also investigating at the moment in some constant spin up that always happens at a specific time. There must be a process that schedule something at 00:45 everyday because in crontab I cannot see any scheduled jobs at that time.

Code: Select all

# m h dom m dow cmd
0 5 * * * /sbin/qthin_check
0 2 * * * /sbin/qfstrim
0 3 * * 0 /etc/init.d/idmap.sh dump
0 4 * * * /sbin/hwclock -s
0 3 * * * /sbin/vs_refresh
0 3 * * * /sbin/clean_reset_pwd
0-59/15 * * * * /etc/init.d/nss2_dusg.sh
30 7 * * * /sbin/clean_upload_file
30 3 * * * /sbin/notice_log_tool -v -R
0 3 * * * /bin/rm -rf /mnt/HDA_ROOT/twonkymedia/twonkymedia.db/cache/*
10 15 * * * /usr/bin/power_clean -c 2>/dev/null
57 * * * * /sbin/qddns_check 2>/dev/null
4 3 * * 3 /etc/init.d/backup_conf.sh
0 1 * * * /usr/bin/qcloud_cli -c


I will update the post as soon as I will make some discoveries. :wink:
Qnap NAS: HS-251 OS Ver. 4.2.2 20160901 | HD: 2x WD40EFRX | RAID: 1 | Used OS: Windows 10 x64
Qnap NAS: TS-251-4GB OS Ver. 4.2.2 20160901 | HD: 1x WD40EFRX | RAID: 0 | Used OS: Windows 10 x64

manny2003
Getting the hang of things
Posts: 74
Joined: Sun Feb 15, 2015 8:57 am

Re: Unwanted spin up, how to correctly use blkdevMonitor

Post by manny2003 » Tue Mar 17, 2015 2:54 pm

I have not had enough time yet to dig into the unwanted spin up issue (it happens often at specific times also if crontab show no entries for that periods :' ) but I thought it would be helpful to see a human readable timestamp when using the blkdevMonitor during investigations.
I modified the script for this purpose, and I thought to share it just in case anyone else would find it useful:

Code: Select all

analyse_kmsg()
{
   _klog=/.klog
   /bin/touch $_klog
   _standby=1

   # read /proc/kmsg
   while [ 1 ]; do
      /bin/dd if=/proc/kmsg of=$_klog bs=1 count=10240 2>/dev/null 1>/dev/null
      /bin/cat $_klog | /bin/grep -E 'dirtied inode|block' | /bin/grep -v -E 'tmpfs|proc' | sed "s/^<.*>\[[ ]*\?\([0-9.]*\)\] \(.*\)/\\1 \\2/" |
     while read timestamp message; do
      printf "[%s] %s\n" "$(date -d "$(</proc/uptime awk '{print $1}') seconds ago + $timestamp seconds" +"%a %b %d %T %Y")" "$message"
     done
            if [ $? = 0 ]; then
            /bin/cat $_klog | /bin/grep -E 'dirtied inode|block' | /bin/grep -v -E 'tmpfs|proc' | sed "s/^<.*>\[[ ]*\?\([0-9.]*\)\] \(.*\)/\\1 \\2/" |
            while read timestamp message; do
               printf "[%s] %s\n" "$(date -d "$(</proc/uptime awk '{print $1}') seconds ago + $timestamp seconds" +"%a %b %d %T %Y")" "$message"
            done >> $BLKDEV_LOG
            _standby=0
         fi     
      if [ $_standby = 0 ]; then
         return 1
      fi
   done
}


Ps. I have a little problem with an offset of 6 minutes that it is not so serious, but I cannot understand from where it comes since the same script works well if executed on a different Linux machine.
Qnap NAS: HS-251 OS Ver. 4.2.2 20160901 | HD: 2x WD40EFRX | RAID: 1 | Used OS: Windows 10 x64
Qnap NAS: TS-251-4GB OS Ver. 4.2.2 20160901 | HD: 1x WD40EFRX | RAID: 0 | Used OS: Windows 10 x64

jmpage1911
First post
Posts: 1
Joined: Sun Dec 13, 2015 7:55 am

Re: Unwanted spin up, how to correctly use blkdevMonitor

Post by jmpage1911 » Sun Dec 13, 2015 8:06 am

Hi Manny,

I recently purchased an HS-251+ and had the same questions as you. I learned a lot from this thread and it seems like download station was also responsible for my random spin ups.

Have you learned anything else in the last 9 months that would be useful to a newbie? Are you still getting periodic spin ups and if so do you know what apps are causing it?

manny2003
Getting the hang of things
Posts: 74
Joined: Sun Feb 15, 2015 8:57 am

Re: Unwanted spin up, how to correctly use blkdevMonitor

Post by manny2003 » Mon Dec 14, 2015 2:46 am

jmpage1911 wrote:Hi Manny,

I recently purchased an HS-251+ and had the same questions as you. I learned a lot from this thread and it seems like download station was also responsible for my random spin ups.

Have you learned anything else in the last 9 months that would be useful to a newbie? Are you still getting periodic spin ups and if so do you know what apps are causing it?


Hi jmpage1911,
it is a bit hard, at least for my limited linux knowledge, to deeply understand all the reasons regarding the unwanted spin-ups. What I can tell you is that first of all the script that I have also modified to include a readable timestamp is your first step to debug your issue. A part from this, some general considerations that I have collected are that:

- It is useful to have the crontab entries configured as much as possible in the same time window (I don't have long time consuming jobs so I run all of them during a same hour)
- There are services on the NAS, like the Qnapmycloud, the internet time sync and many others that periodically make some checks. The script can help to identify them but it is not always so easy to link one process to one specific service or program, at least for me, and in any case most of them are part of the OS and you don't really know why and what they are precisely doing.
- Pay attention to network shares, sometime just booting up one PC on my network wakes up the NAS.
- Use a static IP address to avoid DHCP pooling
- Disable all the unused services as much as possible

The next 2 points are my direct experience for which I have never found any discussions about in the forum, I can be wrong but...

- You should have less problem than me since your NAS have 2GB of ram, but if you, for some reason, saturate at some point your ram to the level that your NAS will start to use the disk swap partition, it is possible that some internal memory handling will trigger some spin-up also when the NAS is idle and you have closed the ram consuming application. Consider that once the swap has been used you don't know exactly if and when this memory will be released, since freeing the ram does not necessarily means that the swap will be immediately released as well.
- I noticed that some sort of HDMI dialog made with the TV when connected can trigger spin-up. I say this because I noticed that spin-ups are less frequent if I disconnect it from the TV. Also when the TV is switched off it is still detectable and it is possible that the TV will transmit some low level connection state.

Hope this could help you a bit in your investigation. :wink:
Qnap NAS: HS-251 OS Ver. 4.2.2 20160901 | HD: 2x WD40EFRX | RAID: 1 | Used OS: Windows 10 x64
Qnap NAS: TS-251-4GB OS Ver. 4.2.2 20160901 | HD: 1x WD40EFRX | RAID: 0 | Used OS: Windows 10 x64

Post Reply

Return to “HDD Spin Down (HDD Standby)”