Herdo

I've read the FAQ, but I'm not sure I fully understand any of it. I think I have everything configured correctly, but I keep getting messages telling me the docker image is getting full. http://i.imgur.com/JUkOpDo.png I'm thinking it is maybe built up log files. I found this thread through the FAQ: http://lime-technology.com/forum/index.php?topic=45249.0 Unfortunately, the command found there, isn't working for me: du -ah /var/lib/docker/containers/ | grep -v "/$" | sort -rh | head -60 I'm on unRAID version 6.2.4 and Docker version 1.10.3. Can someone help me figure this out?

It's possible your routers DHCP server reset or something else changed. I know for me, my physical server is x.x.x.11, and if I type that into my browser I get a SM login screen. x.x.x.10 however pulls up the unRAID login.

Your physical server and the unRAID "server" should have separate IP addresses. If you are getting the SM login, you are using the wrong one. Check your router for "attached devices" or something similar and see if you can't identify the unRAID IP address, then try that.

Great, I'll try this out. Quick question though, what exactly does disabling the USB 2.0 controller do? Will this limit the speed of the port?

I used the web ui to update my unRAID install. As for the new/test flash drive, I downloaded the 6.2 zip from the website, and yes, I copied everything over including the syslinux folder. You have quite a mysterious case Have you tried the USB stick in a different USB port (there are known issues with USB3). Oh yea, I should have mentioned that. I tried 2 USB 2.0 ports and a USB 3.0 port and it still wouldn't work. I also tried unplugging my other USB devices (UPS and keyboard) and just having the boot flash drive plugged in, but got the same results.

I used the web ui to update my unRAID install. As for the new/test flash drive, I downloaded the 6.2 zip from the website, and yes, I copied everything over including the syslinux folder.

Thanks garycase. OK, so I ran memtest for about 42 hours and 38 full passes with 0 errors. I then installed unRAID 6.2 on a new flash drive, and I had the same issue. Then I tried 6.19 on that new flash drive, and it booted just fine. So apparently it isn't my flash drive. Does anyone have any other ideas? I'm not sure if it's getting hung up on "loading bzroot....", because the screen goes black but it flashes back for a split second and I think I can see it says "loading bzroot...... ok". It may be directly after that, which I think (at least in 6.19) says: "early console in decompress_kernel Decompressing Linux..." The 6.2 GUI version still works completely fine.

The ACPI error is an expected thing and can be ignored, not related to your issue. Have you checked your flash device for corruption? My experience with these kind of loading problems is the flash has some file system corruption and fixing it using Windows check disk usually works, you may want to try that. Thanks for the suggestion bonienl. Chkdsk was one of the first things I did. I also tried reloading 6.2 and replacing the bzimage, bzroot, and bzroot-gui files. Same issue though.

Hey thanks garycase, that is a good idea and it's something I should have thought of.

I'm not sure how much this will help now, but I generated a diagnostics zip. It's odd because the GUI version still has to run bzroot before it runs bzroot-gui, yet it works. I noticed I am getting an ACPI error: Sep 15 19:42:13 unRAID kernel: ACPI: Core revision 20150930 Sep 15 19:42:13 unRAID kernel: ACPI Error: [\_SB_.PCI0.XHC_.RHUB.HS11] Namespace lookup failure, AE_NOT_FOUND (20150930/dswload-210) Sep 15 19:42:13 unRAID kernel: ACPI Exception: AE_NOT_FOUND, During name lookup/catalog (20150930/psobject-227) Sep 15 19:42:13 unRAID kernel: ACPI Exception: AE_NOT_FOUND, (SSDT:xh_Zumba) while loading table (20150930/tbxfload-193) Sep 15 19:42:13 unRAID kernel: ACPI Error: 1 table load failures, 7 successful (20150930/tbxfload-214) unraid-diagnostics-20160915-1943.zip

With 6.2 final, I'll be removing that test shortly as its no longer applicable. Good to know thank you. As for the possible memory issues. I actually did have memory issues initially. Apparently the Dell H310 in a PCIe slot and memory sticks in slots A1 and B1 don't play well together. I actually had to place a small piece of tape over some of the pins. You can see more about this issue here: http://yannickdekoeijer.blogspot.com/2012/04/modding-dell-perc-6-sas-raidcontroller.html Not sure if it's related at all, but thought I'd mention it. I've rebooted the server several times now in both the GUI and non-GUI versions. Non-GUI version still does not work. I get: loading bzimage........... ok loading bzroot............. and then the screen goes black, but for a split second it flashes back to the boot menu and I believe I can see it say loading bzimage........... ok loading bzroot............. ok Maybe it's getting hung up AFTER bzroot? The GUI version continues to boot fine and I've watched closely and can't see any errors while it's loading.

That's what I was hoping for too, but I just tried it and it's still not working. The GUI mode still works fine though.

Yes, I can access everything fine from the web UI, all disks and shares are present, etc. Nothing seems to be wrong with it now. The only thing odd was that I got a Fix Common Problems warning telling me I shouldn't be placing my Dockers "appdata" in a user share. I set the default to /mnt/cache/appdata/ (which is where is was before, and where all my Dockers appdata is currently stored) and the warning went away. I must have read the 6.2 announcement post three times, but I did forget to disable the Dockers before updating. I just checked them all and they all seem to be functioning correctly now though. Maybe that was the problem? I will run memtest overnight as well. Thanks for the replies everyone!

Thanks Squid. I replaced those two files and I still have the same issue. However there was one thing I didn't try, the "GUI" version. Just tried selecting it from the boot menu and it works fine. I can easily change the boot loader settings to automatically launch the GUI version, but I still am wondering why the non GUI version is failing.

I just updated to 6.2 stable and upon restarting I noticed my servers web GUI didn't reload. Went to the server and found it constantly rebooting itself. It starts up fine, gets to the boot loader, counts down and auto selects the main unRAID option, then prints out: "loading bzimage" then "loading bzroot" but then it immediately reboots and starts the whole process over again. I left it a while and it just keeps doing it. I tried the safe mode option and the same thing happened. I'm going to try and repair the USB device using chkdsk, but I'm wondering if it would be safe to try and replace bzroot. Maybe it got corrupted somehow? There is unfortunately no way for me to get a diagnostics file. Any other ideas?

If you lose the parity drive, it's no different then losing any other drive. With unRAID, you never lose the entire array. So for instance, if you lost one drive (parity or data) all of your data would still be intact. Let's say you lose two drives; you only lose the data on those two drives. The others in the array would still be accessible and the data would be intact. This makes it much better than a typical RAID 5 array in that regard. Actually, if you were going to lose two drives, one of them being the parity would be the best case scenario, because that's means you are only losing one drive worth of actual data. unRAID 6.2 also allows two parity drives, which means you could lose up to two drives without any data loss.

Perfect, thank you. It's showing another computer on my local network (10.0.0.3). That's the HTPC that Plex is currently installed on, but Plex isn't showing any activity. I've checked that PCs network activity though and it's definitely busy. Showing about 900Mbps receive and 20Mbps send. EDIT: It's definitely PMS up to something. When I shut it down, all streams stop instantly. EDIT 2: After looking around on the HTPC it appears to be updating some "Deep Analysis" files which I now know are updated automatically (I believe every few days). I guess this is just another reason I should be moving PMS over to my unRAID server; it's eating up all of my network bandwidth, haha.

I'm just wondering if it's normal to see shares being accessed when they are not actively being used? For instance: I have nothing currently accessing the "Movies" share. Also, just prior to taking this screenshot it was saying there were 3 streams accessing "TV", which isn't accurate. I checked all my docker containers and none of them were doing anything (CouchPotato, Sonarr, Syncthing), so I'm not sure what this could be. I also noticed that when this happens, I am suddenly using a lot more CPU. Is this something I should be concerned about? How can I narrow down what/who is accessing these user shares? EDIT: If I refresh the "Main" tab in the web GUI, I can see the "read" is going up on my disks. Something is definitely accessing my data. I checked the logs and no unRAID processes like parity check/sync, mover, etc are currently running.

What case are you using? What backplane? How are you connecting the backplane?

I can't say for certain so don't quote me on this, but I believe SAS, SAS2, SATA, etc should all be backwards compatible. All you should need is two forward breakout cables.

This is a new one for me. However, I have a Yubikey (a second-factor authentication device) that I have plugged into one of the USBs port on the front of my main computer. Whenever I plug in a USB drive into the other one, the Yubikey goes offline, and I have to remove it and replug after I take the drive out. (The key has a LED on it so I know when it happens.) It could be something similar... I was thinking the same thing, but I tried the APC daemon again using the USB 3.0 port and it screwed up again. That may have been my fault though because I forgot to plug the UPS back in and it immediately got the "shutdown" command when I booted it back up, haha. I'll try it again later tonight. Either way, the NUT plugin is working great. Thanks for the help.

Both the USB flashdrive and the USB UPS cord are connected to USB 2.0 ports. It's like it's getting stuck mounting the disks. I can usually SSH in and see /mnt/disk1, but there is no /mnt/disk2 and disk1 is empty. EDIT: I thought I'd try a USB 3.0 port because I'm out of ideas. So far, it's working perfectly. I first tried the NUT plugin instead of the default APC plugin, but it had the same issue. After switching to the USB 3.0 port, the NUT plugin is working beautifully. I haven't tested the APC plugin yet, but I think I'm going to keep using the NUT plugin anyways as it can actually shut down the UPS immediately after the server is shutdown. I still have no idea why this would be happening.

Thanks Frank1940, I've made a backup of the flash drive. I've also found the problem, and unfortunately it IS the UPS. Parity-sync completed fine. I shutdown and restarted the server 3 times without any issue before I decided to try the UPS again. It actually booted and allowed me into the web GUI this time, but all the disks were listed as "missing". I did a clean shutdown and tried again. Same problem; all disks missing. So I unplugged the USB and the power cords from the UPS and started up the server. This time there were no problems whatsoever. All disks were there where they should be and the array started on its own. Very weird. Maybe it has something to do with my settings? Aside from the "Start APC UPC Daemon" which I normally have set to "Yes", these are the settings I am using. Because the server didn't fully lock up like it normally does, I was able to grab a diagnostics report. The UPS I'm using is the CyberPower CP1000PFCLCD. unraid-diagnostics-20160830-1438.zip

Well the parity-sync completed without issue so I went back to test out my new UPS. Same issue again. The web gui is completely timed out and inaccessible so I tried powering down the server using: /sbin/powerdown It initiated a shutdown, but then got stuck on "Starting Diagnostics Collecting..." It was stuck like that for a good 30 minutes before I pressed ctrl+C, which then caused the terminal to display a warning saying "SERVER WILL HALT IMMEDIATELY!" I left it like that for another 20 minutes or so before I gave up waiting and I held the power button down. I found a thread where someone else seemed to be having a similar issue and they deleted their super.dat file which seemed to fix the problem. https://lime-technology.com/forum/index.php?topic=47659.0 I'm currently rebuilding parity again without the UPS hooked up. I had this issue, or at least a similar issue before the UPS so I'm fairly certain that isn't the problem, but I'm going to test it anyways. I'm also going to remove the flash drive tomorrow morning and run chkdisk. Is there anything else I should be looking at? Is it possible the server is just busy doing something? I think I read a thread where someone mentioned the server could take a couple hours doing something (can't remember what) when initially started and the web gui might be unresponsive. EDIT: OK, the parity-sync was going extremely slow. It usually runs at about 150MB/sec but was running at about 15MB/sec. I noticed there were several "streams" running on my shares, which was weird because I wasn't accessing them in any way. I decided to stop the parity-sync and check the flash drive right away. I plugged it into a Windows machine and immediately I got a prompt saying "There are issues with this device". I clicked on the "Scan and Fix" button and Windows did its thing. It finished, said it fixed the device, and there were no errors. I ran chkdsk on it anyways which also showed no problems. I then plugged it back in to the unRAID server and everything booted up perfectly. It's rebuilding parity right now at 165MB/sec so I'm going to leave it overnight to finish. I'm not sure what this means exactly. Maybe I need a new flash drive, or maybe it just got shutdown wrong or something?

Sorry, I should have been more clear. I assigned the drives as data drives, then SSHed into the unRAID server and checked their contents. disk1(sdd) had data, and disk2(sde) had data. sdc was the parity drive. EDIT: Also, from the web gui while it's doing the parity sync I can navigate to "Main" and click the little folder icon labeled "View" for disk1 and disk2 and see the shares. I have "Movies" assigned to only disk1 and "TV" assigned to only disk2, and they are both showing their expected contents. Thanks jonathanm, I'm going to catalog their serial numbers for when this happens again.