yeah not happening 
on my to-do list
You donāt need to use unused credit cards. It does not damage the card if you keep the magnetic strip away from the my cloud.
let me borrow your credit cards then
Nice try. The cards are not used to pry open the device. They are only used to release the four catches that hold the case together. They should be placed between the back and the side and inserted about 1/4 inch. Once the four cards are inserted. Just pry the cover off by using a small screw driver on the top of the unit. This will cause the sides to move away from the top.
That is a quick and dirty way.
I gave up. This is absolutely crazy that a 200GB time machine backup would set off a 3 day media scan complete with fan blasting at full speed continuously.
Seriously this is unacceptable.
Good thing that we can kill the scanning process of which I finally did because I simply couldnāt stand it anymore.
Makes you wonder why WD is still wondering why their products donāt sell well?
More, they are wondering why their attempts at dictating reality arent working, given the way they blatantly ignore suggested fixes to their firmware with no feedback.
prosumers are going to be linux-heads, and putting big ugly mittens on them is only going to make them crabby. Being blatant in the attempts to shut down owning the device by the site admin, and threatening with voiding warranties for simple config changes (when a slightly bigger mtd0 device with an emergency restore payload, like basically all consumer routers have, that can accept a binary transmission over tftp when the system fails to find a bootable image, or gets sent a special token on a reboot, would negate nearly all of the need to void warranties since the firmware could be re-pushed effortlesslyā¦ But no. āLOCK THAT STUFF DOWN! MITTENS MODE ON!ā
Granted, they get kudos for not using digital signature magic on the kernel and ramdisk images. Makes repurposing these boards nearly painless, and that is awesome.
Really, I would suggest rebuilding the cramfs image for the device. I have considered it many times on my Gen2. There is no check performed for veracity of the image, and replacing it with a new one would enable much on these boxes. (For instance, putting fun_plug on /mnt/HD_a4, or FIXING THE BROKEN CSS on the web ui, and turning all the purposefully disabled stuff back on. Like āshutdown.ā)
I thought that I came up with something new. But it has already been done. The /usr/local/config partition does not get over written. So one could put the fun_plug script on this partition. Then as this post describes modify the config.xml to add a cron job.
Iāll seeā¦ but there isnāt any motivation but yeah perhaps just to get rid of that click click click click when wdmcserverd and wdphotodbmerger are turned off.
A few years ago after a firmware update, my device never came back online and was dead locked with my connected USB drive. At that time I didnāt know what I know now and I kept pulling the plug and rebooting with no success. Luckily I was able to recover by pulling the USB drive and doing the 40 second reset but because of the missing public folder the firmware update was a disaster because some of the firmware code depended on the existence of the public folder. File permissions had to be manually reset by SSHāing into the device.
The lessons that I carried forward from that day was I wanted to reboot to a virgin My Cloud something that the guys at WD wanted complete with the useless Public Private Folder.
I do like the idea of using the fun_plug feature to modify the boot and perhaps replace the boot with a personal init bypassing the internal boot or perhaps not if the fun_plug is called from within init and if that is the case, the least I can do is to deactivate the scans or add a cron job to deactivate the scans, yeah I think that isnāt happening . The difference is that by removing the USB, the My Cloud reboots normally.
If we rebuild the cramfs image, then we can do an if-then-else block, checking for existence of fun_plug. (If found, branch that way, else do normal init)
As things currently are though, the init process is passed to a script inside the cramfs volume which does all the horseshit, then at the bottom, looks for and executes (if it exists) fun_plug.
I would rather it branch and treat it as a system-init than treat fun_plug as a user-level init.
if you rebuild the cramps image you have to be consistentā¦ like leaving fun_plug alone but you can add if then system_init_fun_plug then treat system_init_fun_plug as a system_init. In this fashion fun_plug continues to be fun_plug.
I decided to become a normal user by moving the directories with humogous (half million files) entries which includes ebooks, midi sheet music, pdf, backups with sparse indexing which are basically thousands of 8GB blocks of data (these are the worse as even Microsoft windows couldnāt even copy them and just hangs on 100GB Sparse index file) into two 6TB My Books Duo; one formatted as NTFS and the other as a Mac journal.
This leaves about 5TB of movies and music on the Main My Cloud and when cloud access was turned on it only took about 3 hours to re-index instead of 4 frigging days. No cloud access to the USB My Books.
What Iām discovering these days, in fact for over a year now, I donāt really need cloud access to my data as Netflix, DramaFever, Amazon Video are providing all the movie streaming I need and Apple Music Match, Google Music and Amazon Music provides all the music streams that I need.
Moving the My Cloud that was full of data that was connected to the Airport extreme router over to the switch resolve the barrage of wake ups and has been sleeping soundly with the default My Cloud setting; even with the cron left running.
I have two My Clouds with one with no data and one with about 200GB of data that sleeps soundly despite their connection directly to the airport extreme router.
It is interesting that I havenāt notice any device waking up and this is without any modifications to the My Clouds; they boot up as-is. They are waking up, Iām sure, because the cron is left on but Iām not hearing or noticing them.
Even with the reduced data the Main My Cloud it is still ticking and grumbling whenever it is awake. It is annoying butā¦ it could be worse, and it was, but it is what it is, until somedayā¦ BAM I get a QNAPā¦
So that is thatā¦
addendum: I found that after rebuilding the media indexes for Cloud access that the My Cloud keeps waking every hour or every half hour or so. Even with turning off Cloud access, the My Cloud continued waking. So I hibernated the device and un-plug and re-plug the power as opposed to re-boot just because I had two usb drive plugged in and I had to disconnect them before turning the power back on. I learn the hard way back four years ago that you shouldnāt reboot with the USB drive plugged in. After a power cycle the drive is now quiet and sleeping.
My best guess is that openvpn once opened will continue to ping the device every 30 minutes. It is either that or someone is watching your movies stored on your Cloud. This is why in the past I would kill openvpn before the cloud settles down for the night.
So after copying 2GB into my sleeping, no cloud access, default boot up My Cloud, the device started to wake up several times an hour. No scans are running, no openvpn. no PC with windows, no cloud access. Just the annoying wake up with clicks every two seconds. Killing crond doesnāt do anything. I even killed restsdk-server without any effects.
Doing a Top every time the device wakes up shows
system_daemon
up_read_daemon
However these shows up even when the device is sleeping.
The only solution to this is a full hibernate and power up again. Reboot doesnāt do it.
So after copying files to the device, something is left running saying that data has been updated but a flag or a file that contains total size is never updated thereby causing the device to wake up every 10 to 20 minutes.
A hibernate and power up seems to reset that flag or file allowing the device to sleep pretty much soundly.
This isnāt a great solution at all as it means that every time I copy data to the device, I need to hibernate and power up. This is no longer the scans anymore as there are no scans that are running.
Please tell WD to fix @Bill_S, @Great_Scottt
Thanks
Havenāt seen ether of them for some time. [edit: Bill seems to have gone. Great_Scott is still around].
Anyway: you expect WD to fix sleep issuesā¦? Come onā¦
Is he?
Hello Great_Scottā¦ do something for goodness sakesā¦
yes
Iām an optimist and Iām hoping that there is still one programmer left that knows how to implement thisā¦
If you run the following script when the system is in the wakeup loop. It will tell you what files are being accessed that cause the wakeup.
TestGen2:/mnt/HD/HD_a2/Public# cat checkgen2.sh
#!/bin/bash
sda=(" " " " " " " " " " " " " " " " );
function test () {
if [ ā$3ā != ā$2ā ]; then
da=date +%k-%M-%S
A=$2
B=1
ior_sda=${A#-}
iow_sda=${A%-}
A=$3
ior_sdaold=${A#-}
iow_sdaold=${A%-}
let a=ior_sda-ior_sdaold
let b=iow_sda-iow_sdaold
if [ $4 -ne 1 ]; then
echo -n $da " "
fi
printf "%4s %4s %4s " āsdaā$1 $a $b
fi
}
function GetData {
for (( i=1; i<8;i++ ));
do
y=printf "%s%s" $3 $i
x=awk -v disk="$y" '{if ($3==disk) printf "%s-%s",$10,$6}' /proc/diskstats
sda[$i]=$x
done
}
while :; do
GetData ior_sda1 iow_sda1 āsdaā
B=0
for (( i=1; i<8;i++ ));
do
test $i ā${sda[$i]}ā ā${sdb[$i]}ā $B
done
sdb=(ā${sda[@]}ā)
if [ $B -eq 1 ]; then
echo
./find /mnt/HD/HD_a2 /mnt/HD_a4 /usr/local/config -cmin -1
fi
sleep 4
done
Iāll see if I want to expend the energy to investigateā¦ thanksā¦
Iām pretty certain the loop that is causing the wakeups is in init which use to be in that sleep loop.
Whenever you copy files over to the device, it triggers the device to start scanning (I think it is the total volume size), however since there is no scans the (total volume size) doesnāt get reset thus this little task that makes the device go click click click is probably comparing directories to see what has changed thus waking the device up. This gets rest when you hibernate the device and reboot.
After a hibernate and reboot, the device sleeps perfectly, until you add or delete files from the device.
So if I were to invest time to investigate, Iāll look through the init script for that loopā¦
It is so close in becoming the perfect NASā¦
You do know between you and WD the on,y one that will expend energy looking into this problem is you.
I knowā¦
Alternatively I have been looking at QNap too and the the cheapest I can get in on QNap is a dual bay mirror drive for only $300.
Throw my two 10TB drives into that, consolidate all my files into one deviceā¦ Sell all my WDs and Iām done. It is only one click awayā¦
Here you goā¦
| Forums | News and Announcements | Register | Help | Forum Guidelines |
Copyright Ā© 2021 Western Digital Technologies, Inc. All rights
reserved.
|
Trademarks
|
Privacy
|
Community Terms of Service
|
Site Terms of Use
|
Copyright
|
Contact WD
|