nukeman

Thank you for all the help, I got it sorted out. I stopped the docker and vm services. Then I was able to restore the /system directory to its default location on /mnt/cache. That put the previous versions of system\docker\docker.img and system\ibvirt\libvirt.img back in their original locations. Once I started the docker and vm services back up, everything was back to normal.

I've been looking at my director structure. There's still folders for all my dockers in /mnt/cache/appdata There's still folders for all my vm's in /mnt/cache/domains Are you saying if I reinstall the dockers it will use the previous settings and not start over from scratch? That's good news.

First, I think I did this to myself but I'll try my best to recreate the sequence of events. Last night I was preparing to copy my share data to an external (unassigned) drive for backup. I was reviewing the filesystem and noticed that I had a /system folder on one of my array disks. I thought this was odd because my system share is set for Cache as primary storage and Array as secondary. There is plenty of available space on my Cache so I concluded that the /system folder on the array drive was left over from some previous evolution and wasn't needed. I confirmed that the dates on the files in the array system folder were several months old. Furthermore, there was a system folder on my cache drive with with files that were modified today. Assuming I didn't need it, I moved the array system folder to another location on the array. The destination doesn't use the cache and is excluded from my backups. I then proceed with backing up my share data to the external drive. I let it run all night and while I was at work I got a notification that a parity check had started. That was odd so when I got home I checked the GUI and saw that unassigned drive had an icon next to it saying "reboot" (I use the unassigned devices plugin). I've never seen that before so I paused the parity check, stopped the array and rebooted. After reboot I'm missing all my docker apps and VMs. I copied the /system back where I moved if from and stopped/started the array. Still no luck. I stopped there, and thought I'd reach out for help. I do use Appdata backup and those backups seem to be current. Happy to provide any information that will further help troubleshooting nas-diagnostics-20240220-1813.zip

I sent one of the cache pool drives back to Samsung. They sent a new(?) refurbished replacement. I added the replacement drive to the pool and Unraid did a parity check and found no errors. Is that all I need to do prior to RMA'ing the other, original, cache drive. I've never done this procedure before and want to make sure I'm good to remove the other faulty drive. Is "no balance found" relevant?

I've finally ran out of onboard SATA ports and I'd like to add a controller card to my system. I'm using a SuperMicro X10SLH-F motherboard and have open "PCI-E 3.0 x8" and "PCI-E 3.0 x8 (in x16)" ports. I've read the recommendation thread and I'm looking for an 8 port controller card. Based on that post, I'm looking for "9207-8i, 9300-8i and newer" for PCI 3.0, since my ports are PCI 3.0, right? Does this seem like a reasonable option? IBM branded, us seller with good feedback, breakout cables included, $60. Or, this one is LSI branded for $20 more. This is an x10 board from 2015. I want to make sure I utilize the full ability of the hardware without buying something so new it's not compatible. Any other recommendations?

ok, 45TBs sounded like a lot, but I guess I am moving large media files around frequently. Currently I have my downloads folder using the cache, perhaps it would make sense to not do so. I don't really care about fast writes from downloads. Regardless, I'm going to RMA the drives, one at a time, to keep the server up as much as possible. Plan is to: Remove one of the drives from the pool and return it When I get the replacement drive I'll put it into the pool and let BTRFS rebuild it Then I'll take the other old drive out of the pool and return it finally, I'll put the second replacement drive into the pool Or, should I do this procedure instead? This post says there's trouble rebuilding the pool in 6.9.2? But then, there's a workaround?

Are there any reports of excessive writing to cache drives or cache pools? My cache hosts 4-5 VMs and "normal" dockers (Radarr, Sonarr, etc). Both of these drives were introduced into my system when I initially created the cache pool. Prior to that I was just using one (smaller) ssd for cache and docker/vm hosting.

Well, I'm going to need some help translating the Wear_Leveling_Count. You're saying I've written 45TB to the drive?!?!? I don't understand how that could happen. Also, I'm not excited to read that report of other 870 EXOs failing...

The drives have 671GB free so they're 33% full. Here's the smart reports. 2947T SMART Report.txt 2900M SMART Report.txt

Recently I received a warning about some Reallocated Sectors on one of the SSDs in my cache pool. The other day I was doing some heavy copying with the cache and received a similar warning on the other drive in the pool. I've searched the forums and this seems to be either "watch it to see if it gets worse" or "critical fix it now" problem. I went back and found some old diagnostics and two months ago both drives had Reallocated_Sector_Ct=0 in their SMART reports. Now one drive has Reallocated_Sector_Ct=3 and Reallocated_Sector_Ct=7. Both of these drives are 1TB Samsung Evo 870's that were purchased in February of 2021. I'm not excited about the prospect of swapping out the drives as they contain several critical VMs for my home business as well as Unraid's Cache. That being said, I'm also not excited about both drives in my cache pool failing at the same time. Assuming these warnings are something I should take care of I started the RMA process with Samsung. I did get an RMA issued but they won't send out a replacement drive until I send the old one in for evaluation. Before I start down that road though, I wanted to get some opinions on what my next steps should be. Is this a warning that warrants replacing both drives? If so, how should I go about doing it? I thought Samsung SSDs were generally well regarded, maybe I just got unlucky? BTW - I'm happy to post the SMART reports but I'm wondering if there's any sensitive information I should remove? Like should I remove the serial numbers in the report prior to posting?

First I exported my video library to separate files by following these instructions. Then I setup MySQL like normal and imported the library from the exported nfo files. Finally, once I validated the import was successful, I deleted all the nfo files. Pay attention to the importwatchedstate and importresumepoint options when importing.

No, not with MariaDB. I installed a MySQL docker and everything is back to working like it should.

I'm experiencing a weird issue with kodi and a shared mariadb . Right now my setup contains a windows laptop, android tablet, shieldtv, and chromebook that all share the same video and music database. I think I've isolated the behavior to the following: if I open kodi on the windows laptop, I can no longer open kodi on the android devices. If I reset the mariadb docker container, android starts working again. I've been running a shared database for years without seeing this issue. It seems to be similar to what's mentioned here but there's no solution. I don't know if this has to do with the issues from that started popping up here around in August. These seemed to be related with the update from ubuntu to alpine. The fix was to downgrade mariadb then upgrade to latest. If this is the issue referenced above, would someone summarize the steps required to fix it? The steps are mentioned in this thread but it's kind of scattered and hard to follow. The relevant section of the chromebook's kodi log after the windows laptop launched kodi: 2021-11-15 19:07:18.664 T:5323 INFO <general>: MYSQL: Connected to version 10.5.13-MariaDB-log 2021-11-15 19:07:18.761 T:5323 ERROR <general>: SQL: [music-adults82] An unknown error occurred Query: SELECT idVersion FROM version 2021-11-15 19:07:18.763 T:5323 ERROR <general>: Process error processing job I realize "An unknown error occurred" isn't very helpful and I'd be happy to post more relevant logs from kodi and mariadb once we confirm what the problem is. Anybody else experience this?

There's been lots of responses since I started this post and I thought I'd provide the group an update. After changing my CPU Scaling Governor to "On Demand" the stability of my system improved dramatically. I could install the Ubuntu Desktop VM and its performance is acceptable. Also, the CPU use in the dashboard isn't spiking like it was previously. Sounds like the CPU Scaling Governor setting hasn't fixed this for everyone, but it seems to be working for me.

Um, so this is embarrassing. I have "Tips and Tweaks" installed but I don't remember ever changing any settings listed there. I changed "CPU Scaling Governor" to "On Demand" as you suggested. I also changed "Enable Intel Turbo/AMD Performance Boost" to "Yes". I rebooted the server for good measure. Performance inside the VMs seems much better and CPU usage on the dashboard has gone down as well. I'll keep an eye on this for and see if it remains stable. Thanks to @stor44 for the suggestion, hopefully it helps others too.

Just upgraded to 6.9 yesterday and noticed that the CPUs pinned to my Windows 10 VMs are pegged at 100%. Also, performance inside the Windows VM is horrible (slow spinning logon animation, 30 seconds to open Windows explorer, generally sluggish). Performance on these VMs was excellent in version 6.8.3, it was difficult to tell I was even using a VM through RDP. The two VMs in question are called "hcp" and "media". I created "media2" after the 6.9 upgrade to see if I could find any differences in the xml. Installation of Windows 10 was slow on "media2" and its performance is bad too. I've been using this instance of unraid since 2015 and it's been rock solid. I rely on these VMs for my home business and I can't figure out how to get their performance back to 6.8 .3 levels. Other items of note: The "nvr" VM is powered off I do not pass though a GPU but I do pass through a USB card to "media" I had to perform the steps outlined here after the upgrade to 6.9 I upgraded my cache drive a week before the 6.9 upgrade I created a cache pool for the first time during that upgrade htop shows /usr/bin/qemu-system-x86_64 using greater than 100% of CPU. That's weird, right? My CPU temps seem to stay stable. Perhaps these are false CPU percentage readings? Any ideas what could be going on? nas-diagnostics-20210304-1940.zip media.xml hcp.xml

I'm trying to set up an Ubuntu Desktop 20.04 VM. Eventually, the VM locks up and I notice its assigned CPU in UNRAID is pegged at 100%. I've tried several times and sometimes I can get the OS installed, and sometimes it locks up during installation. Wondering if it's the same issue as this? I've attached my XML and diagnostic info. Any ideas? Quite a lot of information for Windows guests here, but I'm having trouble finding equivalent information for Ubuntu guests. I have 3 Windows VM's running perfectly. BTW, I saw @jonp's advice to set UEFI boot mode, i440fx, and setting the CPU mode to emulated. However, I can't get the CPU setting to stick in the GUI. Every time I save it but when I go back into the form view it's set to Host Passthrough. ubuntu.xml nas-diagnostics-20200425-2003.zip

Wow Jon, thanks for the in depth reply. I plan on leaving this machine in the basement and headless. What I think I read in your reply is that unless I plan on physically plugging a monitor into the unraid machine, it really won't matter if I add a GPU, right? One thing I didn't make clear in my first post is that there is no graphics card in the machine at all, just the simple ASPEED on board adapter on the motherboard. When you use RDP to a windows client, you select VNC when making the virtual machine, right? You have to VNC into it long enough to enable remote access inside the host, then you're free to RDP. Correct? I'm anxious to see what others might recommend for remote controlling a Ubuntu GUI too. As I mentioned earlier, I have several Ubuntu sever clients, ssh works great for those. Again, thanks for the detailed reply.

I have a headless Unraid 6.1.3 instance running on a SuperMicro X10SLH-F-O with a E3-1231 v3 processor. I have successfully set up several Ubuntu Server VMs and now would like to move on to something with a GUI. I'd like to try either Ubuntu Desktop or Windows. My first attempt at Ubuntu Desktop (15.04) was less than successful. I got it to install but performance was pretty poor using the built in noVNC: the mouse response was sluggish, unity was slow etc. When I initially configured the VM I chose the VNC option for Graphics Card. The motherboard does have onboard graphics (an ASPEED controller), but when I selected that I lost the ability to use noVNC. Should I have picked VNC during the creation of the VM? If I don't pick VNC, how do I access the VM? Some external VNC Viewer? I made sure that both my motherboard and my processor support VT-x but should I have used something like the E3-1246 v3 that has onboard graphics? (Intel HD Graphics P4600) Would performance be better if I added a separate graphics card? Is this what GPU passthrough is all about? We use windows and rdp at work. Usually it's impossible to tell if I'm using a remote session, performance is that good. Is it reasonable to expect that level of performance from KVM? What's the best technology for remotely accessing a Ubuntu VM? VNC, RDP, Teamviewer, Guacamole, something else? I have no intention of gaming on the VM, just looking for a usable GUI environment. BTW, I did get an Xbuntu instance up and running with better performance, but eventually I'd like to have a Windows VM, so I figured I better try to get this worked out. Thanks for the help!

Ah, I didn't realize that help was context sensitive. Thanks!

Your question about size of the disk prompted me to reconsider that. I had been creating 2 GB disks. Just out of curiosity I tried 10 GB. It worked with no issues! With 2 GB the installer defaulted to creating a separate /home partition, and with 10 GB it didn't, don't know if that's related minimum requirement for disk space on Ubuntu server is 1 GB, wonder why 2 GB failed? Anyway I included the screenshot for reference. The only templates I see are "Custom" and "XML Expert". It does change the icon to Ubuntu's when I pick "Ubuntu" for the OS though. I thought it was strange that those were the only two options. I admit I'm late to the party with visualization. This is the first version of unraid I've used and have never had to configure vm's manually. Is there any guide that explains all the different choices for the configuration? Like I said before, I don't mind doing the research, but I've looked and haven't found anything specific. I'm spoiled by the folks in these forums, it's easy just to ask my questions here.

Just got my first unraid (6.0-rc3) box up and running and I'm now ready to try out vitalization. I thought I would start out simple and install an instance of Ubuntu server 14.04.2 but I'm having some trouble. I'm happy to do some reading and figure this out for myself, but I can't find anything that addresses my specific issues. Generally I can make it though most of the setup but it finally errors with a red screen that says "An installation step failed" at the step "Install the GRUB bootloader on a hard disk". I know the VM manager has been evolving through the betas and is a work in progress but is there anywhere that describes all the parameters so I know what to pick? I know I should pick "Ubuntu" for the OS, but should I choose Host Passthrough or Emulated (QEMU64)? What machine should I pick, Q35 or i440fx? What BIOS should I pick SeaBIOS or OVMF. I have found some posts that say OVMF is only necessary when I want to do GPU passthrough. I know that won't be the case when I'm installing the server version of Ubuntu. I get some errors at the beginning of the install that say "mount: mounting /dev/vda on /media failed: Invalid Argument" and "umount: can't umount /media: Invalid argument". Could they be related to the trouble I'm having with the install? Again, I hope I'm not missing something simple. I've tried all the combinations of configuration options I can think of but still get the same result. Any help is greatly appreciated. My configuration: SuperMicro X10SLH-F 16GB Memory SSD Cache

Just a note. I determined that the DNS-323 was the issue. Since it is RAID 1 I just took out one of the drives and connected it directly to the unraid box. I then mounted it (outside the array) and copied the data over.

Interesting, when I did do a copy from my windows laptop and as I mentioned, the speeds were about 2.5 MB/s. What I didn't mention was that was via wifi. I knew wifi would be slower, but I thought it would be better than 2.5 MB/s. Your post prompted me to plug my laptop into the gigabit network directly. When I started the same transfer, the speed was right around 100 MB/s! So, you may be on to something, it may be the DNS-323. Two questions: Is 2.5 MB/s on a 54 Mbps wifi connection reasonable? Is there anyting I can do to make the DNS-323 read faster? It's RAID 1. Can I just take a drive out and try to use it in the unraid box directly?

I'm hoping someone has some advise for my issues with my new unraid server. I just got the server stood up and my first task is to copy all my media off my previous nas (an ancient Dlink DNS-323). I was able to successfully copy some files but my speeds were horrible, less than 1MB/s write speeds. First my setup: SuperMicro X10SLH-F, Bios 3.0, 2x Intel i210AT LAN ports Xeon E3-1231 v3 2x8GB Hynix HMT41GU7MFR8C-PB (supported) 2x Toshiba 5TB 7,200 RPM PH3500U-1I72 (1 for data, 1 for partiy). Both drives passed preclear successfully, no SMART errors 1x Samsung 250GB 850 EVO MZ-75E250B/AM (cache) Consumer gigabit switch Dlink (DGS-2208) Fresh install of unRAIDServer-6.0-rc3-x86_64, NerdPack is the only plugin, Trial key at this point Both the unraid box and the old nas are connected with CAT6 cables to the gigabit switch. Gigabit is recognized on both devices My plan: I didn't enable the cache drive because I knew I would be copying TB's of data, and didn't want to fill it up I did enable both the parity and data drives initially. I allowed the parity to be built before I started. I used an rsync command line this: rsync -azPh root@nas_old.home:/mnt/HD_a2/media/test/ /mnt/disk1/test/ The results: Write speeds hover around 600 kB/s I'm gathering that with parity enabled I should be seeing 40-50 MB/s I removed the parity drive from the array leaving only the data drive and the results were the same I've been doing a lot of reading. I know problems like this have so many variables, everything from chipsets, LAN adapters, drivers, jumbo frames, cables, the list goes on and on. I also know it's difficult to do apples to apples comparisons with so many different variables. I'm no rsync expert, but that command seems pretty simple. I guess I could connect the unraid box to the dns-323 with a crossover cable and run the command from the IPMI interface. That would eliminate the LAN. I did try copying an iso from my windows laptop via the SMB share and the speeds were a little better at 2.5 MB/s (measured by windows). Eventually I'll have the cache (SSD) drive enabled so I might not feel this pain, but I want to make sure there's not something significantly wrong first. This thread seems to have a similar issue, but that's unraid 5 and X9 rather than X10. I'm hoping someone with a simliar setup has had simliar issues. Anyone have any ideas? Are there better ways for me to copy files? Are there better tools to measure transfer rates? How should I continue troubleshooting? syslog.zip