Just a quick update, after some experimenting with different values, the best configuration I found on my box is:
echo “64 32” > /proc/sys/vm/lowmem_reserve_ratio
I still get faults, but the least with that setup it seems (but that will likely depend on a lot of factors).
I am very interested in this and will try it, thanks Agerstia.
Believe it or not, I come to this thread almost every day just to check if there is any interesting findings. My disconnect problem still exists after updated to the latest firmware.
This 2-TB MyCloud becomes an experiment material, rather than a reliable NAS. What I don’t understand is some people seem never such a problem ? For me it is pretty consistent, every 36 to 48 hours after re-plug, the connection drops.
- it has now been 72 hours since I turned off “Media Streaming” and “Itunes Streaming” and the mycloud has not disconnected itself.
Again, since I don;t need those services, it’s not a big deal TO ME. However, it is a poor design if one has to disable major features of a device to get it to stay connected to the network.
- it has now been 72 hours since I turned off “Media Streaming” and “Itunes Streaming” and the mycloud has not disconnected itself.
Again, since I don;t need those services, it’s not a big deal TO ME. However, it is a poor design if one has to disable major features of a device to get it to stay connected to the network.
Lyle
We are still looking into this. Please respond if you get private messaged by support for further information.
Everyone, we are coming out with a firmware update for the WD My Cloud within the next two weeks to fix many of these issues. However, you may not have had the same root cause that created the issue in your circumstance. If that turns out to be the case, please make sure that you have correctly set up your WD My Cloud on your network. If you need help with this, there is a link below to a knowledgebase article for setting up the WD My Cloud.
@Aerodoq, @WD-Fan: I looked at my kernel logs (/var/log/kern.log), and with the “64 32” configuration I saw the least “page allocation failures” that I had talked about in my earlier/first post in this thread. Fewer than with “16 16”, but still not completely gone (only a kernel upgrade can fix that).
Ever since I got a replacment WD MyCloud I haven’t had any disconnect issue for about 1 month now. My previous drive I had to power cycle the WD Mycloud each day when I get home just to be able to get access to it. Now no issues with the dashboard and the Iphone app.
I really hope a solution of firmware update will finally available. I believe this problem is due to some interaction between the WD MyCloud box and the router. In the past month, I bought two with 3 TB and 4 TB respectively. The 3 TB one is installed in a network with an Apple Airport Extreme router and never had the disconnection problem even with the “auto sleep” on. The 4 TB one (bought two weeks later) is installed in a network with a Cisco Linksys E4200 router. The NAS (MyCloud box) disconnects from the network after two to three hours, even after “auto sleep” was turned off and updated the firmware to v03.03.01-156. I also manually set the router to preserve a fixed IP number to MyCloud. Disconnection constantly happened every night. Not until I adopted the “cron job” solution summarized by Steve511 (message #146) which was proposed by Pavel_Kuzub (message #71), my cloud has been up for two weeks. I have been happy with this solution, even though it sounds silly for a consumer-oriented product to require the user to write a crontab script.
As soon as the firmware update is available, I will test it. Would the community manager please give us some technical clue about the solution when the anticipated firmware update releases? In case a problem remains in certain network environment, we WD MyCloud user community can seek a solution educationally? Thank you in advance.
I think I found the solution. (but in general: This is the most unstable device i have ever owned. What good is a cloud that is not dependable)
I am on my 4 th WD Cloud Replacement.
1st drive disconnected from the network the 1st day
2nd drive worked for 2 months then started disconnecting every ~12 hours
3rd drive was a defective advance replacement it just initialized for hours and hours.
4th one disconnected from day 1 after ~12 hours
Problem: After reboot the WD My cloud attaches/registers with the router and works perfect. Then overnight it sits idle, in the morning it is no longer attached to the network. "In the morning the WD My cloud IP and Mac address are no longer listed on the Router IP lookup table. "
Solution to fix:
Configure the WD Mycloud to use a static IP: I used 192.168.0.3
Configure your Router DHCP server to not assign addresses that include the address used by the WD Mycloud: My DHCP Starts at 192.168.0.10 and ends at 192.168.0.100.
----- originally i assigned the WDMYCRAP to static 192.168.0.3 and the DHCP Start: 192.168.0.1 - 192.168.0.100. This did not work… The DHCP active range must not contain the WDMYCRAP address.
Its been up for about 72hours which is about 60hours longer than before… Im keeping my fingers crossed.
FAIL - this **bleep** disconnected after 90 hours… I hope WD releases a new firmware soon.
I just looked at agerstla 's post the memory scan thing at night makes allot of sense…
The idea of using a static IP over a DHCP’d IP is interesting, but certainly wasn’t the thing causing my box to drop the network. I was using a static IP from the initial setup, and the longest it stayed attached was 8 hours.
After making the vm.lowmem_reserve_ratio change, however, it’s been up now for 8.5 days.
It’s strange the different things that seem to work for some, but clearly don’t work for others.
A new firmware update has been released. Below are the fixes. We will continue to work on other fixes and push them out as soon as we can. Note: If you are still experiencing problems, please post them in the new firmware board.
Issues Resolved:
Resolved issue of dropped mobile cloud access connections after updating to firmware 03.03.01-156.
Resolved issue of a scheduled safepoint update that does not run automatically.
Resolved issue of device performance degradation caused by certain MPEG-4 (MOV/MP4) video formats used to record slow motion videos.
Just wanted to weigh in to say I have the same issue, and it’s pretty annoying. The one thing I hoped my Netwok-Attached Storage would do is stay Attched to the Network. I just updated the firmware, but the release notes do not seem to indicate that anything relevant has been fixed.
Mine was up and stable for a week when it was plugged into an old 10/100 4 port hub. I replaced the hub with a Trendnet GB switch two days ago and now it disappears after a few hours.
I am curious if anyone updated the firmware and the unit is staying connected. As mentioned earlier, the update doesn’t appear to address the dropped connectivity within the network.
So I updated to the -165 firmware yesterday. Since rebooting the box has been up for 1 day 11 minutes now, with no unexpected disconnections. The firmware update reverted my change to the vm.lowmem_reserve_ratio and I left it alone (“1 1”) instead of changing it again to “16 16”.
To recap, with the previous f/w it wasn’t able to stay connected for any longer than 8 hrs, often much shorter. After changing the vm.lowmem_reserve_ratio, I didn’t see any disconnections over 8+ days. And with the new f/w with the old vw.lowmem_reserve_ratio, I haven’t seen any disconnections yet in over 24 hours.
I don’t know what to make of all this, but as long as the box is stable I’m not too worried about it.
Upgraded to the latest version yesterday. Problem still exists.
I will now try the crontab solution.
Please note that you should use “date > /DataVolume/shares/Public/date.txt 2>&1” in stead of “date >> /DataVolume/shares/Public/date.txt 2>&1”
means append, so although a small amount of data is written to that file, it is written every minute and will use up space and will eventually fill up your disk.
will overwrite the file every time, so it will not grow
Edit: Problem occurred again after 1 hour and a half. Will now try with >> , see if that makes a difference.
Edit2: I think the issue was related to a temperature problem (I had it in a closed compartment). NAS has ben up & running for more than a day now.
