My DL4100 was running great until I loaded OS5. Now I am showing 3 faulted drives. I rebooted and came back with one faulted. Then a few hours later 2, the 3.
Help please.
What make and model of drives do you have installed in your DL4100? On the dashboard select the support option.
Create and save a report.
Access the file ZIP file archive that gets download and extract the file:
\MyCloudDL4100\log\var\log\Smartdata.log
Edit the extracted file to remove all the lines that start with: Serial Number:
Post a copy of the file without the drive serial numbers as a reply here.
Thanks for your assistance
(Attachment Smartdata.log is missing)
Paste the content (after removing any serial numbers/codes) into the reply.
For anyone who is wondering how that issue turned out.
The DL4100 in question has the standard 2Gb RAM. Something operating within the NAS was taking up ALL of the available working RAM and also used-up the entire swap-file. The result of this that a process or processes are selected for termination to release working memory.
In this case what was happening is that the netatalk process can not operate and the following is invoked.
netatalk invoked oom-killer: gfp_mask=0x14201ca(GFP_HIGHUSER_MOVABLE|__GFP_COLD), nodemask=(null), order=0, oom_score_adj=0
After the kernel does what it needs to do performed, repetitively, the following action. It can be any process what can’t start that can trigger the kernel reaper process.
Killed process 31766 (restsdk-server) total-vm:5815476kB, anon-rss:1408820kB, file-rss:2528kB, shmem-rss:0kB
restsdk-server gets re-launched and the process repeats.
So, look like of you have a DL4100 and then have lots stored on it, it will malfunction with all sorts of random faults showing up and it appears like it needs an extra 4Gb RAM installing.
Is this coincident with the Indexing process?
I have this suspicion that indexing hogs too many resources . . … which is resulting in the problems; and why the MyCloud mirrors struggle and why the single-bay MyClouds end up bricked.
I am also under the impression that maintaining “cloud” access also consumes an excessive amount of RAM for unknowable reasons. . . .which causes some problems even if the unit survives indexing.
Quite possibly.
