mossiemosforth

I love the flexibility of the system, meaning I can run everything I want to on 1 piece of hardware. What I would like to see is the ability to snapshot VMs brought into the GUI.

Evening all, I'm after advice/opinions for a replacement cache drive in my server. My current Server specs are in my signature, but it boils down to 3x 4tb array, SSHD cache drive and a 240GB SSD I have mounted with UD that my VMs and Docker run on. As My array is around 90% full I want to buy another 4TB drive, so I don't want to spend too much more on a cache. That and given the Proliant only has SATA II capability I don't get (nor am I bothered by) read write speeds. Currently I have a Seagate SSHD (Model ST1000DX001) that is dying a horrible death with over 51K 'reported uncorrects', so I do need to replace it. Now looking I can get another SSHD, either 500GB or 1TB for a reasonable price, but I wondered if these types of drives are really that suitable as a cache drive? I do however need the drive to have a decent size capacity, as I do host a fair amount of data permanently on it (such as my music share) so it's always readily available. Anyway, I've waffled on long enough. Any advice or opinion would be greatly appreciated. TIA

Evening all, I'm hoping someone here more knowledgeable than I with S.M.A.R.T. could have a look at a disk I'm experiencing issues with. I have a SanDisk 240GB SSD I mount with UD, on which I run my VMs and Docker config directories. When upgrading to 6.3.5 I rebooted my server and it didn't come back up. Checked the console and and the bios complained that my SSD has SMART issues. I continued to boot so I could pull off the data (which I have managed to btw) and unraid also complained it failed SMART check. So i checked the drives warranty, 3yrs and I've had it 18months. Easy I though, I'll raise an RMA. Did that and SanDisk asked me to install their tool and run an extended test, sending them a screenshot of the results. Did that on my windows box, and it passed!! I haven't sent this to SanDisk as yet, as I'm sure they'll just tell me to do one. So place it back in my server and now it states it passed SMART checking, but I'm still receiving lots of warnings regarding 'retired block count' and 'reported uncorrect' (screenshot attached). Even though it now 'passes', every time I try to interact with the drive it's either slow, or ends up unresponsive. I'm kind of hoping I can get another failed reading that I can send to SanDisk. I'm stuck, can anyone have a look at the attached SMART report, give me advice. Thanks in advance unraid-smart-20170530-2202.zip

+1 for XFS, mines been working flawlessly for about a year now, mounted with Unassigned Devices

I can't even see my signature block in this new forum format, completely forgot about it. Updated now I've been keeping up with the releases as and when they come, currently running v6.3.2 with no issues

I run my VMs on an SSD mounted with unassigned devices at /mnt/disks/SSD and have done since before v6.1, with no issues. I also run my docker config folders on the same drive, so a subfolder for each - /mnt/disks/SSD/vms and /mnt/disks/SSD/docker My default path for VMs is set to /mnt/disks/SSD/vms, not had any issues using this, just wanted to re-assure you

Sounds like you have done everything correctly. Only piping up as I too have my VMs disks sat on an SSD outside the array using the unassigned devices plugin with no issues. When you have modified the VMs xml it should look similar to this; <disk type='file' device='disk'> <driver name='qemu' type='qcow2' cache='writeback'/> <source file='/mnt/disks/SSD/VMs/Win10VM/vdisk1.qcow2'/> <backingStore/> <target dev='hdc' bus='virtio'/> <boot order='1'/> <alias name='virtio-disk2'/> <address type='pci' domain='0x0000' bus='0x00' slot='0x05' function='0x0'/> </disk> So the source file starts with /mnt/disks (this is the path that UA plugin uses, the bit in bold) Also ensure the boot order is 1, or if you still have an ISO attached then remove it. Lastly you could also try deleting the address type line (in italic above), this will be automatically re-entered the next time you start the VM as KVM allocates it. I only mention this as I've had occasions where I've edited properties of the VM and I've needed to refresh this, when realistically if you're not changing the VMs hardware details you shouldn't have to. Hope you get it running ok.

Morning guys. I'm trying to set up an unraid vm for testing 6.2 before moving to it and I'm a little stuck. I have the USB created, but when trying to convert it to an img i get "Invalid parameter 'raw' " Any ideas? EDIT - never mind, n00b error. capital -O for options *facepalm*

Is it only the ssd share that's asking for credentials? Have you set the SMB security settings for Unassigned devices itself? Go to; Settings Tab >> Unassigned Devices (under user utilities, at the bottom of the page) Set the SMB settings as per what you wish, it's the same manner as any other share in unraid I only know as I was in the same position as you a little while ago, performing everything on my UD ssd through putty. I believe this is a global security setting for all UD. I only have the 1 drive setup outside the array, so can't prove it

If you're running v6 you can create a vm, pass the printer through to it and share that way. I've done it with mine so anyone in the house can use it Sent from my Nexus 5 using Tapatalk

Thanks for your input guys. I have just placed an RMA with ebuyer to return the card i have and will be purchasing an MSI GT720 instead - http://www.ebuyer.com/661551-msi-gt-720-1gb-ddr3-vga-dvi-hdmi-pci-e-graphics-card-n720-1gd3hlp Sadly i had no luck with the GT610 out of my daughters pc, iirc that's also an asus branded card. good job i'm not in a massive rush for this, i'm actually running my OE instances on Raspberry Pi's atm, i just want to beef up the living room one so i can add retroarch and whoop my kids at some classic games

My hardware is a HP proliant server board, which has its own igpu for the unraid console. The card is independent of that at least. I did manage everything successfully before using a GT610 borrowed from my daughters PC. Problem with that was the card didn't reset when you rebooted the vm, only after rebooting the whole server would it work again (hence me getting a new card).

Apologies if anyone has covered this, but has anyone managed to managed to get this working with an AMD HD5450? I've managed to pass the card through ok, but it will not boot if i pass the audio through also (leaving no audio for kodi), if i pass through the audio also it hangs on boot with a 'kernal panic' error. I managed to pass both to a windows 7 VM i have, but even after installing the drivers it still complains about not having any audio devices. I also changed between ACS override on and off with no change to symptoms. Therefore i believe it's the card at fault, not sure what is going on. The card is new, so i could return it for another if that's best option. Any advice appreciate.

you sir, are a true gent. That previous version has worked a treat for me. Now the hideous task of migrating my VMs and Docker to the new drive

I've come looking as i am also having this very same issue. Installed an SSD to host my VMs and Docker, all seems to be good until i tried to mount the disk and the icon never goes green saying mounted. That being said it does actually mount the disk and shares it. Another weird one is for some reason the share name is in lowercase (ssd), whereas in all config files i have defined it in uppercase (SSD), anyone have any idea to fix that? Checked my logs and i also get the same php errors as you all (after reporting it is mounted correctly); Fri Jan 15 18:29:56 GMT 2016: Mounting '/dev/sdf1' Fri Jan 15 18:29:56 GMT 2016: Drive found with the following attributes: name='sdf1', device='/dev/sdf1', part='1', disk='/dev/sdf', label='SanDisk_SDSSDA240G_154608406660-part1', fstype='btrfs', mounted='', mountpoint='/mnt/dis$ Fri Jan 15 18:29:56 GMT 2016: Adding disk ... Fri Jan 15 18:29:56 GMT 2016: PHP Warning: set_error_handler() expects the argument () to be a valid callback in /usr/local/emhttp/plugins/unassigned.devices/UnassignedDevices.php on line 428 Fri Jan 15 18:29:56 GMT 2016: Mounting drive with command: mount -t btrfs -o auto,async,noatime,nodiratime '/dev/sdf1' '/mnt/disks/SSD' Fri Jan 15 18:29:56 GMT 2016: Successfully mounted '/dev/sdf1' on '/mnt/disks/SSD' Fri Jan 15 18:29:56 GMT 2016: Partition '/dev/sdf1' successfully mounted. Fri Jan 15 18:29:56 GMT 2016: Defining share 'SSD' on file '/SSD.conf'. Fri Jan 15 18:29:56 GMT 2016: Adding share 'SSD' to /boot/config/smb-extra.conf. Fri Jan 15 18:29:56 GMT 2016: Reloading Samba configuration. Fri Jan 15 18:29:56 GMT 2016: Directory '/mnt/disks/SSD' shared successfully. Fri Jan 15 18:29:56 GMT 2016: Command not available, skipping. Fri Jan 15 18:29:58 GMT 2016: PHP Warning: set_error_handler() expects the argument () to be a valid callback in /usr/local/emhttp/plugins/unassigned.devices/UnassignedDevices.php on line 428 Fri Jan 15 18:29:58 GMT 2016: PHP Warning: set_error_handler() expects the argument () to be a valid callback in /usr/local/emhttp/plugins/unassigned.devices/UnassignedDevices.php on line 428 Fri Jan 15 18:29:58 GMT 2016: PHP Warning: set_error_handler() expects the argument () to be a valid callback in /usr/local/emhttp/plugins/unassigned.devices/UnassignedDevices.php on line 428 Fri Jan 15 18:29:58 GMT 2016: PHP Warning: set_error_handler() expects the argument () to be a valid callback in /usr/local/emhttp/plugins/unassigned.devices/UnassignedDevices.php on line 428 Fri Jan 15 18:30:00 GMT 2016: PHP Warning: set_error_handler() expects the argument () to be a valid callback in /usr/local/emhttp/plugins/unassigned.devices/UnassignedDevices.php on line 428 Fri Jan 15 18:30:00 GMT 2016: PHP Warning: set_error_handler() expects the argument () to be a valid callback in /usr/local/emhttp/plugins/unassigned.devices/UnassignedDevices.php on line 428 Fri Jan 15 18:30:02 GMT 2016: PHP Warning: set_error_handler() expects the argument () to be a valid callback in /usr/local/emhttp/plugins/unassigned.devices/UnassignedDevices.php on line 428 Fri Jan 15 18:30:02 GMT 2016: PHP Warning: set_error_handler() expects the argument () to be a valid callback in /usr/local/emhttp/plugins/unassigned.devices/UnassignedDevices.php on line 428 Fri Jan 15 18:30:04 GMT 2016: PHP Warning: set_error_handler() expects the argument () to be a valid callback in /usr/local/emhttp/plugins/unassigned.devices/UnassignedDevices.php on line 428 Fri Jan 15 18:30:04 GMT 2016: PHP Warning: set_error_handler() expects the argument () to be a valid callback in /usr/local/emhttp/plugins/unassigned.devices/UnassignedDevices.php on line 428 Fri Jan 15 18:30:06 GMT 2016: PHP Warning: set_error_handler() expects the argument () to be a valid callback in /usr/local/emhttp/plugins/unassigned.devices/UnassignedDevices.php on line 428 Fri Jan 15 18:30:06 GMT 2016: PHP Warning: set_error_handler() expects the argument () to be a valid callback in /usr/local/emhttp/plugins/unassigned.devices/UnassignedDevices.php on line 428 Fri Jan 15 18:30:08 GMT 2016: PHP Warning: set_error_handler() expects the argument () to be a valid callback in /usr/local/emhttp/plugins/unassigned.devices/UnassignedDevices.php on line 428 Fri Jan 15 18:30:08 GMT 2016: PHP Warning: set_error_handler() expects the argument () to be a valid callback in /usr/local/emhttp/plugins/unassigned.devices/UnassignedDevices.php on line 428 Fri Jan 15 18:30:10 GMT 2016: PHP Warning: set_error_handler() expects the argument () to be a valid callback in /usr/local/emhttp/plugins/unassigned.devices/UnassignedDevices.php on line 428 Fri Jan 15 18:30:10 GMT 2016: PHP Warning: set_error_handler() expects the argument () to be a valid callback in /usr/local/emhttp/plugins/unassigned.devices/UnassignedDevices.php on line 428 Fri Jan 15 18:30:12 GMT 2016: PHP Warning: set_error_handler() expects the argument () to be a valid callback in /usr/local/emhttp/plugins/unassigned.devices/UnassignedDevices.php on line 428 Would moving back to SNAP be an option do you think?

Short, sharp, to the point! I like it Can't shut the server down now as the kids are watching something. Job for the morning me thinks, just hope the copying/remapping it all etc doesn't take too long!

Just read about this - http://lime-technology.com/forum/index.php?topic=38635.0 is this the way to go now then?

Is the image qcow2? A dynamically expanding disk? This is the only thing I can think of

I'm surprised no-one has answered this for you. The short answer is yes you can. When you first create a VM you'll specify where the virtual disk will reside (this can be anywhere you like). Should you wish to move it (after say fitting a new cache drive etc) you'll only need to; - Shutdown the VM - Copy/move the virtual disk to the new location/drive - Edit the xml of the VM, changing the file path section for the virtual disk - Start the VM Simples

Hi guys. I've just received a new SSD which I intend on putting in my server to store my docker image, containers & VMs, I have a question about mounting it outside of the array. I did this once before with an old vertex 2 drive using SNAP, the downside I now see is that I never had a SMART status on this drive. If I had then it may have prevented me loosing all my VMs and dockers when it died I had to rebuild everything from scratch onto my cache drive, but as this is a spinning disk it does put a fair bit of wear on the drive along with the larger power consumption (electricity in the UK isn't cheap). Is SNAP still the best method? Has it improved and now you can see the status of the drive on the dashboard? or is there a better method? Advice welcome, TIA

Thanks for the reply Jon, sounds like promising news. Sadly the VM dropped out on me while watching a TV Show last night, kind of lost my faith in it a little. Where can I find the logs, as to better find out what is going on with it? I've had to plug my raspberry pi back in for now, for stability, so waiting out for a new release should be too bad. On that note do you have a time frame for release? or an idea when Openelec v6.0.0 willl be an available template?

I figured that was the case, as the old AMD 3450 i had ran like a charm and was able to be reset (if it was HDMI i'd keep it) I have both bios types with no avail, shame as the card itself seems to perform well for a low end card. As for trying another rom, i have followed the wiki, but I have no idea what rom 'could' even work with this card. Anything in the region of this model i have tried failed miserably. Seems like buying another card is my best solution, current one that's caught my eye is a Radeon HD5450, I can currently get one for circa £20, should be good enough to run kodi and a few emulators, anyone have have one/know if they work?

I've just re-read all this and though 'wow, i was waffling on'. So i've edited it to bold up my main question, for those like me who hate to read blocks of text. next time i'm not going to write a post so late after staring at a screen for hours.

I got the pulse-eight adaptor working by adding the address manually to the xml for the VM. <hostdev mode='subsystem' type='usb' managed='yes'> <source> <vendor id='0x2548'/> <product id='0x1002'/> <address bus='7' device='5'/> </source> <alias name='hostdev1'/> </hostdev> The above is the code specific to my installation but it might be of some use. Smashing Phil, thanks for the confirmation. Thanks for the code, did you have to manually add it? Just it seems to be the same entered when using the tick boxes on the 'edit VM' screen. Either way, useful non-the-less I think i'll get one ordered once I've solved my graphics card issue.

Hi guys, a small issue I have and I'm after clarification/best solution. I have been running VMs for quite a while now, including a linux box with 2 passthrough NICs (firewall/DMZ into my home network) and have never had a issue. Also passed through singular USB and even one of my motherboards USB controllers to different VMs. My issue now lies with graphics pass through. I've achieved it before but never really had a use for it, until my wife saw me playing Super Mario and asked if it was possible on the living room TV so we can play together. I then came across the retroarch/advancedlauncher guide johnoden wrote and thought i'll give it a go. So i've ripped out the old AMD HD3450 card I had previously used (with no issue passing through) and put in a GT610 I 'borrowed' from my daughter machine as it has HDMI so I can send video and audio to the TV. Already a big fan of KODI and i run multiple instances around the house i figured i could use limetechs OpenELEC VM to make this all happen and replace the raspberry pi I currently use in the living room. I have indeed managed all the above (and may write this up at some point, UCD post maybe), including pass through of a USB keyboard and xbox360 wireless receiver, but i have an issue with the graphics card. Specifically if I reboot the OpenELEC VM I get no video once booted again. I can gain graphics again, but have to reboot the entire server to get it, which takes ages, along with this the VMs logs are full of this libusb: error [submit_bulk_transfer] submiturb failed error -1 errno=28 (i think this may be to do with the xbox receiver). From what I can gather the issue lies with the GT610 itself, it doesn't reset properly when the VM restarts, am I right? If so is there a way of forcing it to? I have also read about pointing VMs at a different ROM that would allow this (but that's outside my knowledge and if anyone could point me in the right direction i'd be much obliged) Failing all else I can also go out and get another card, but wanted to ask before i spend money unnecessarily. If this is my only option does anyone have a cheap recommendation (i'm in the UK and parts can be pricey). Last question (maybe have my fingers slapped for wrong forum), but any idea when limetech will release an OpenELEC v6.0.0 template. I'm a growing fan of it and theres a few addtions to v6 that i would make use of.