Leaderboard

My apologies for long delay since the last release, for those that care about why, I'll leave it in the postscript. In this version there are some major changes in addition to the usual bug fixes. A big bug fix, is that the plugin should now be able to handle parenthesis in your VM names (thanks to squid for suggesting a solution to that). New features include the ability to have multiple configs so that you can run backups on different schedules, as well as the ability to run pre and post scripts with those configs. Compression now uses pigz for multi-threaded compression (though a faster inline compression using zstandard is still coming). Full change-log below. v0.2.0 - 2020/01/21 The Resistance - fixed new config options not always getting added to user config. - fixed parenthesis in vm names. - added ability to run pre and post scripts. - added option to run backup without array started. - added confirmation dialogs to destructive or dangerous buttons. - added support for multiple configs/schedules. - added option to allow multiple configs to run simultaneously. - added option to set compression level. - added option to set the number of threads used during compression. - changed compression program from gzip to pigz to support multi-threading. - changed alert dialogs to use sweet alerts. - updated help for custom cron. https://github.com/JTok/unraid.vmbackup/tree/v0.2.0 -JTok The delay in this release was caused by me overhauling some of the code to prevent the pages from needed refreshed to update content, and I wound up needing to implement my own version of unRAID's markdown and inline-help to make that work the way I wanted.

⚠️ NOTICE: I currently do not have the time to do more than maintain this project for myself at the moment. I will eventually do more work on it as my life permits, but for now development will be sporadic at best. I am more happy to add someone else to the project to maintain it, or even accept some PRs as that is much easier for me to find time for Mod note: this plugin has multiple known issues, use with care. This is a beta version of a plugin I am working on to replace the unraid-vmbackup script I have been maintaining here and here. It is based off the latest version of the script (v1.3.1), and I will be maintaining both for now as there are some use-cases for the script that the plugin cannot currently replace. This is provided as-is with no guarantees. That said, and without further ado, the current version of the plugin can be found here: https://github.com/JTok/unraid.vmbackup/tree/v0.2.8 -JTok Feel free to buy me a beer (or two 😋)

A plugin to create and modify vfio-pci.cfg from the unRAID webGui. Allows you to bind PCI devices by address to the vfio-pci driver. Also displays IOMMU groupings and which devices support reset. Unbound USB controllers display USB devices connected to them. Bound devices will be available to pass through to your VMs under "Other PCI Devices" in the Edit VM Form View. Please see the release notes for 6.7.0-rc1 for more information on this binding method. Search for "VFIO-PCI Config" in Community Applications Github repository: https://github.com/Skitals/unraid-vfio-pci/ URL for manual plugin install: https://raw.githubusercontent.com/Skitals/unraid-vfio-pci/master/plugins/vfio.pci.plg

Here is a guide about how to setup a Bitwarden server using the Bitwardenrs container. Also last part of video shows having setup the container how to use the chrome extension and importing passwords from chrome into Bitwarden. I hope you guys find it useful.

build 2020.01.21: 6.8.0-rc5 kernel with navi reset v2 vega reset patch* (used in conjunction with this script) k10temp Ryzen patches from 1/18/20 (new) Fix for onboard audio passthrough on x570 *modified to add 0x6863 in addition to 0x687f 6.8.0-rc5-Skitals-2020.01.21.zip

@bonienl The VM GUI editor is hard coded to set the thread count to 1 if it detects an AMD processor in libvirt.php // detect if the processor is AMD, and if so, force single threaded $strCPUInfo = file_get_contents('/proc/cpuinfo'); if (strpos($strCPUInfo, 'AuthenticAMD') !== false) { $intCPUThreadsPerCore = 1; } This was due to AMD reporting no support for hyperthreadding in a VM. With UNRAID 6.8.1 RC1, hyperthreadding is supported with CPU passthrough as is (and CPU cache) if the CPU feature topoext is enabled. Previously, the CPU had to be forced to report as an EPYC to get it to support hyperthreadding. <cpu mode='host-passthrough' check='none'> <topology sockets='1' cores='6' threads='2'/> <cache mode='passthrough'/> <feature policy='require' name='topoext'/> </cpu> Microsoft's CoreInfo returns Coreinfo v3.31 - Dump information on system CPU and memory topology Copyright (C) 2008-2014 Mark Russinovich Sysinternals - www.sysinternals.com Logical to Physical Processor Map: **---------- Physical Processor 0 (Hyperthreaded) --**-------- Physical Processor 1 (Hyperthreaded) ----**------ Physical Processor 2 (Hyperthreaded) ------**---- Physical Processor 3 (Hyperthreaded) --------**-- Physical Processor 4 (Hyperthreaded) ----------** Physical Processor 5 (Hyperthreaded) Note that changes to the CPU layout may not be detected in the VM until the VM is rebooted from inside the VM itself (for example: Start > Power > Restart) Prior to 6.8.1 RC1, I could not get CPU-Z to run, it would always hang at the 10%/Processors on load. It still takes a bit but does return now.

You don't have an E-mail recipient listed in the SMTP Settings.

Disabled disk rebuilt and now part of the array with a sweet green dot next to it. Just finished a parity check and have zero sync errors. Success! Thanks to everyone in this thread that helped me through this.

It might just be an update to the docker container and nothing to do with an actual Plex server version update. If your Plex server version changed, you can go to the Plex website for details. The Plex docker containers are just a Docker friendly packaging of the Linux version of PMS. None of the container authors does any development work on Plex itself and they do not know what may have changed in PMS.

If you can't boot and it stops before unraid is loaded, the issue is not unraid. If your bios is from 2016 it is about time you update it.

Yes! I received the Intel E-2278G a week ago, and 4 no. 32Gb ECC UDIMMS. I just booted for the first time into ESXi showing 128Gb of memory. the DIMMS are: 4 no. "32GB (1x32GB) Samsung M391A4G43MB1-CTD DDR4 ECC UDIMM, PC4-21300 (2666)" I'll carry on with the build now.

Open the container settings (edit) and change to advanced view. Check the setting for "WebUI" --> http://[IP]:[PORT:8888]

If the RAM is good (I'd reseat both DIMMs by swapping them over for good measure) I'd check the power connections to the motherboard, both the main ATX power and the CPU power. Try a different PSU, if you have one.

Hi there, Looks like you need to update your forum signature as your hardware has changed from AMD to Intel ;-). That said, I'm not seeing any events in the logs themselves that show a problem. And I'm not sure what this is supposed to mean: Also hoping that you removed this container while troubleshooting this: At this point, I would suggest hooking up a monitor and keyboard to the system and tailing the log (command is tail /var/log/syslog -f). This will begin printing the log out to the screen. Then try to get the system to crash again and capture whatever was printed to the screen (use your phone to take a picture if necessary). This should give us at least some information to point towards a cause. I'd also check for a BIOS update on your motherboard.

Another big thanks to @sjerisman for adding this feature. Again, for anyone using the plugin, the features that get added here will make their way into the plugin before too long. v1.3.1 - 2020/01/21 So Say We All - added option to create a vm specific log in each vm's sub-folder. Script here: https://github.com/JTok/unraid-vmbackup/tree/v1.3.1 -JTok

If you type ls -l /dev/disk/by-id you will see what the symlinks actually point to. Instead of using /dev/sdX in your configuration, use /dev/disk/by-id/Y where Y is the symlink (shown in light blue/cyan) that points to your /dev/sdX. If what I wrote doesn't make sense then post a screenshot of the ls -l command, above and confirm which sdX you are currently using (from your screenshot above it looks like you're using /dev/sde).

This is only for 6.8.0-rc5. To manually upgrade/downgrade, delete all the bz* files from the root of your usb drive (or rename them adding .bak to the end of the file name). Download rc5 zip here. Extract and copy the bz* files to your unraid usb drive. Download the kernel in the first post. Extract to your usb drive (overwriting bzimage, bzmodules, etc). All your settings are in /config (and maying /syslinux/syslinux.cfg) so back it up before mucking around.

Depends on where the bulk of the traffic is headed. If it's headed out the WAN port, then the router is the obvious choice to connect that specific switch. If it's staying local, then keeping it off the router is better. With the equipment you originally specced out, it really wouldn't make a whole lot of difference. This is the correct answer at this point in time for your layout IMHO. A single unit feeding all data lines, with uplink ports connecting to the POE branch. Now all you need is a real router and some Unifi access points hanging off that POE switch along with your cameras and you have a small office capable network.

Any audio transcoding is still handled by the CPU, but that would be significantly lower. All video is GPU. I don't have a P2000, but there are a few videos where people have got about 12/13 4K transcodes simultaneously (although 4K transcoding doesn't look right yet). If it's 1080p stuff, more. I have the 1231 processor, and I find it's now only taxed if one of the other dockers has a moment. Plex just works, and if I get any holdups, it's usually either a system pause whilst a drive spins up or network congestion.

It's not crazy. You are assuming perfect scaling (i.e. more core = better performance) but in real life, it's rarely ever like that. 44 (logical) cores sounds like a lot but they are hyperthreaded. An extra hyperthreaded core adds only about 30% performance (at best!) so 44 logical cores is roughly equivalent to 33 non-HT cores at best and 22 non-HT cores at worst. So now you only have to really "explain" the diff between my 24 non-hyperthreaded "odd-bank" config vs 33 non-hyperthreaded (at best) config. Firstly, there's diminishing return (i.e. performance) as you add cores and it can get to a point in which more cores makes things slower e.g. the scheduler of tasks can't keep up with the number of tasks being run. And then there's the idiosyncratic design of AMD Ryzen family of CPU (with CCX, CCD and how L3 cache is shared) e.g. how I have found uneven load on the CCX causes performance to drop. And then there's qemu bug. That's why Steve and Linus and others still run benchmarks. Otherwise, you would just say "hey more core, more performance, done deal, we are all fired". I would suggest you find a benchmark that matches your intended workload or even better, bench using your actual workload. Synthetic stuff sometimes doesn't quite match real life.

You need to move the data to another disk, with Krusader, mc (midnight commander), etc, then shrink the array, there are two options, both described below, 1st one is probably the best for this case, at least it would be the fastest. https://wiki.unraid.net/Shrink_array

You probably need to google some reviews I guess. I generally prefer PSU reviewed by jonnyguru. I would not be too concerned about the single-rail vs multi-rail argument, especially considering your load. It was a rather archaic argument (for tech stuff) back in the early 2000. It's more important that the PSU comes from a reputable brand and has at least decent review. PS 1: the Unraid wiki has to take into account people running 20+ HDD array. Those guys probably would prefer single rail. PS 2: the Unraid wiki is also quite old (for tech stuff) so back when it was written, multi-rail PSU had a certain notoriety of being cheap brandless knock-offs that randomly blow up.

Multiple arrays are not possible although I believe it is a roadmap item, but with no ETA. Having the 8TB as parity is over-kill at the moment, but it does mean you are in a position to add 8TB data drives going forward.

The fun never stops, I uploaded a new 6.8.0-rc5 kernel today that incorporates a bunch of goodies for this board. Brand new k10temp patches for Ryzen, and a fix for passing through onboard audio!

No worries. Thats why we are all here buddy. Because we all freak out sometimes.

So...I have a feeling the kernel panic was due to the upgrade to 6.8.1. For my part only 6.8 was stable. I say this because when I remade the USB stick with the old backup (6.7.2) and used the config files from the original stick, it booted up with the correct configuration and everything. I don't know if the guys at Unraid are aware of this issue but I won't be migrating to 6.8 until this gets fixed. All in all, thanks @PeteAsking. The config migration solution worked. Not sure why I didn't think of that. Much appreciated. I was freaking out a bit, I'll be honest.

Okay, thanks. Does this mean the redirect will only work on the non-SSL port? I would've figured the cert issue wouldn't have been an issue if it redirected to where I told it to go. Yes, the way the default page is configured in nginx makes it apply to non-SSL port only.

Have you tried making a backup of this working usb stick (that is an old config) and then just copying over the Config files from the non bootable usb stick that has the correct newer config? Copy across the config folder as this has the config in it.

bingo, i had to edit the xml instead of using the form view. The port was set to -1, I assume that means auto but I changed the websocket to 5703 and then it started. Thanks.

Some folks, when setting up an unRAID server, will load the data to than array without assigning a Parity drive. Of course, the data on the array will have no protection in the event of a disk failure but often the data is still on the original source drives so there is a backup. After the data is moved, the Parity drive is added and parity will have to be built. Moving massive amounts of data to an array is always a time consuming process. The actual transfer rate has many factors involved. Some of the most significant have to do with the characteristics of the hard drives themselves-- rotational speed, track-to-track movement times and data density are three prime ones. The transfer rate of a single 25GB BluRay ISO file will be much faster than moving 10GB of 10KB data files due the overhead of the file management routines and latency caused by the head movement from the files tables to the data area. Once, you have a initial loading done, you may want to consider using a SSD cache drive (or SSD cache pool) to speed file transfers to the server. The Mover can be programed to move the files from cache at convenient time of day. (Default is 3:00AM as I recall.)

I generally recommend Gigabyte mobo for new users because its BIOS allows you to pick which of the PCIe x16 slots as Initial Display Output (i.e. what Unraid boots with). Given you want to pass through a RTX 2080, chances are you will need a 2nd GPU (a cheapo single-slot like GT710 would work) to dump vbios and avoid error code 43. Having the ability to pick any PCIe x16 for Unraid to boot with allows you put the cheapo on the slowest slot (e.g. the one running at PCIe x4 speed) and reserve the fast slot (1st) slot for the RTX 2080.

got the firmware updated. Didn't get greeted w/a ton of udma crc errors (fingers crossed). Currently rebuilding array due to the disabled disk. Hope this works.

For the Metadata part please do the following (taken from Plex Forums that worked for me) 1. Stop PMS 2. Delete the HTTPCookies file under the Plug-in Support (the agents) 3. Start PMS 4. Refresh All Metadata 5. Give it time to walk through. For me the HTTPCookies files are in Plug-in Support/Data folder. Delete only the HTTPCookies files that are located in each folder. After doing that the Movies and TV shows Posters started populating. Sent from my iPhone using Tapatalk

Hmmm, Going to be honest, I didn't even realise realtek oot drivers were needed on this release, I must have missed that in the release notes. I'll build a new release, not a big problem. Sent from my Mi A1 using Tapatalk

Is it possible there was a regression for the RTL8125 driver in the Unraid Nvidia 6.8.1 build? I am running into the identical issue I had on the 8.8.0-rc5 build. Reverting to Unraid Nvidia 6.8.0 has my NIC working correctly again. Thanks again for all the work put into this project.

I have blue iris running in a windows 10 vm (igpu passthrough to that vm) and can transcode several 1080p plex streams without much of a sweat. You've got the k version so an extra 400mhz(?) to your base clock speed will do better. There's a also a 30 day trial, so you could just try it lol.

Possibly because of and your solution was to which didn't actually address the issue

Awesome! Bring on 6.9 with that new 5.4 kernel

I like the virtualisation feature and I would like to see overclocking features added in the future. This was a fun way to set up a competition!

I will update for 6.9 rc1 when it's out. 6.8 final reverted to a 4.x kernel. 6.9rc1 will be identical to 6.8 final with the addition of the 5.x kernel.

Ok so what exactly is your top score for space Invaders? [emoji16] Sent from my CLT-L09 using Tapatalk

What his real job is, Because his video's are top quality and great knowledge passed onto this community. How does he find the time. Much appreciated

If you're on latest release it's much easier, just click on the flash drive and change it here:

Basically, Unraid uses a even-parity calculation across all the data drives to be able to reconstruct the failure of a single data drive. See here for how it works: https://wiki.unraid.net/index.php/UnRAID_Manual_6#Parity-Protected_Array Raid is faster in data writing and retrieval operations. However if two drives fail in a RAID configuration, you lose all of the data in the array. Also all drives must be the same size. With Unraid, writing is much slower and the read process is limited to the read speed of the data drive which contains the actual file data. Also if two drives fail, you only lose the actual data stored on the drives which fail. The data on the other drives can be read on any Linux system. The capacity of the data drives can be different as long as the parity is as large or larger than the largest capacity data drive. It is also quite easy to add capacity to any Unraid server by simply adding another data drive. In some cases, it is possible to recover some of the data from the failed drives depending on the nature of the failure. But it is best not to count on it. The second parity drive uses a Reed-Solomon encoding scheme (which is basically a very complex matrix operation) to allow reconstruction (in use with the first parity drive) to be able to rebuilt two failed drives at the same time. If you wish to know more about Reed-Solomon, you can google it! (You will need the equivalent of a college degree in Mathematics...) One import thing to realize is that parity protection is NOT a substitute for a reliable multiple-copy backup scheme with one of those copies being in a remote location. Just think of having your server in a house fire, flood, hurricane or being stolen. It is also not a replacement for a due-diligent administrator who monitors his servers and sets up all of the notification services so that he receives the earliest possible notice when any problem arises!

Hello Unraid community. I am curious if I am able to limit the amount of resources a docker has access too. I recently installed the Handbrake docker and start my first job. I have a fairly decent servers for unraid with '2 x E5-2670's'. When I fired up my first handbrake job I saw all CPU cycles get consumed. As sexy as this is I have a VM that is my main base of operation also running on the server and its is very unresponsive with handbrake eating all cycles. Is there a way to restrict which cores Handbrake docker can use? Or at least limit how many cores it has access too? Thank you in advance! -Kopuz

https://forums.lime-technology.com/topic/57181-real-docker-faq/?page=2#comment-566088 https://forums.lime-technology.com/topic/57181-real-docker-faq/?page=2#comment-566087

The sas2hax creator has now created new tool for doing this process from within Linux itself. I haven't tested this but maybe an easier solution. https://github.com/marcan/lsirec Original Post Below I've add some script files to the method speed up the process. Updated Files - https://mega.nz/#!bp1lwAwZ!qrQics2qslPxCRFSI7pCaL-JECCauM_xqp00GQPloSg MIRROR - https://drive.google.com/file/d/18aULGYQEyMJcW7rUAk8sddUDX_Rozq68/view?usp=sharing Steps Create DOS USB using RUFUS and copy these files to the USB. GUIDE link Each task has 2 commands you can do. The first one saves you typing the commands by using script/bat files. Recommendations Unplug all drives but the USB, helps with the EFI shell bit Plug D2607 in to the first x16 slot, even if it means unplugging another card, had systems where you could only flash it if it was in the first slot. Make sure you have SAS to SATA breakout cable not SATA to SAS reverse breakout cable. (Thanks johnnie.black for pointing this out.) Boot the USB in non UEFI mode. (DOS) Then do the following commands Get SAS Address of the card 1.BAT or megacli -adpallinfo -aall | find /i "sas address" > sas_addr.txt Backup SBR incase we need it. We probably wont though. might need this if you get stuck for support. (Only run this once per card otherise you'll lose the original, before flashing the patched one below) 2.BAT or megarec -readsbr 0 O-SBR.bin Clean Flash the card. If the command fails to flash, just keep trying it until it completes. Can take 10+ attempts 3.BAT or megarec -cleanflash 0 Flash SBR for A11 or A21 - Just change the number in the command for the card you doing. 4-21.BAT or megarec -writesbr 0 SBR-A21.bin Clean Flash the card 5.BAT or megarec -cleanflash 0 Reboot in to UEFI shell on the usb Some motherboards you can go into the boot menu and do the UEFI option for the USB stick, best to check out in the manual or online. Your mount point for your USB might not be FS0, could be FS1, FS2 and so on. Either unplug all drives so it's just the USB drive or, when the EFI shell starts it prints the drives it see's in yellow, you might be able to work it out from there. Once booted in to EFI Shell, do the following commands Change to USB fs0: Flash IT Mode Firmware 6.NSH or sas2hax -o -f 2118it.p20.bin It will error about it can't reset the adapter, don't worry its normal. Reboot and then boot the UEFI on the usb again Do the following commands fs0: Now we Flash IT Mode Firmware again 7.NSH or sas2hax -o -f 2118it.p20.bin -b mptsas2.p20.rom Get SAS Address we saved 8.NSH or type SAS_ADDR.TXT Set SAS Address of the card. Change the address to the one that got printed out. sas2hax -o -sasadd 5003005700dbf920 Reboot the USB in non UEFI mode. (DOS) Do the following commands Flash SBR for A11 or A21 - Just change the number in the command for the card you doing. 4-21.BAT or megarec -writesbr 0 SBR-A21.bin Notes: Making up a SAS address is fine, just make sure you don't flash two cards with the same address. You can find more help cross flashing the D2607 card here. He made the sas2hax application. https://marcan.st/2016/05/crossflashing-the-fujitsu-d2607/ I add no -noreset when I was flashing some card but seems to cause issues, so I left it without. Had to reboot twice when ever I did it, not sure why though. Orignal post: https://forums.lime-technology.com/topic/12114-lsi-controller-fw-updates-irit-modes/?do=findComment&comment=522038 Flash Fujitsu - D2607.zip

How do I limit the CPU resources of a docker application? Two methods: The first is the easiest and doesn't involve much thought about what's going on under the covers: Pin the application to one or more cores of the CPU. This way the application will only execute on that particular core(s), leaving the other cores open for other applications / vms / etc On 6.6+ on the Settings - CPU Pinning you can select which core(s) the docker application will run on. NOTE though that if you are also isolating CPU cores from the OS, then you do not want to pin a docker container to multiple isolated cores. If you do that the container will only run on the lowest isolated core. (This may seem counter-intuitive, but it is NOT a bug anywhere but is actually correct operation) Alternatively, you can also prioritize one docker app over another: eg: You're running the folding@home app and plex. When Plex starts transcoding you want to give it as much CPU as possible at the expense of folding@home To do something like this, you would add to the extra parameters section of folding@home the following: --cpu-shares=2 This gives folding@home the absolute lowest cpu priority, so that if another docker app requires / wants to use all the cpu available it will. Note that if the other docker apps are idle and doing nothing, then folding@home will use as much as it can (subject to its own internal settings) This is a rather simple example, only utilizing 2 apps. Here is a better example (and an explanation of what's actually happening when using 3+ apps) (from the docker run reference): Note that it is more complicated (and beyond the scope of a FAQ) to prioritize non-docker applications over a docker application. For those so-inclined, review the docker run reference and play around

Please see Why do I see csrf errors in my syslog?

Why do I see csrf errors in my syslog? Starting with 6.3.0-rc9 unRaid includes code to prevent CSRF vulnerabilities. (See here) Some plugins may have needed to be updated in order to properly work with this security measure. There are 3 different errors that you may see logged in your syslog: missing csrf_token - This error happens if you have plugins that have either not been updated to conform to the security system or the version of the plugin you are running is not up to date. Should you see this error, then check for and install updates for your plugins via the Plugins tab. To my knowledge, all available plugins within Community Applications have been either updated to handle csrf_tokens or they were not affected in the first place. If updating your plugins does not solve your issue, then post in the relevant support thread for the plugin. There will be hints on the log line as to which plugin generated the error. wrong csrf_token - CSRF tokens are randomly generated at every boot of unRaid. You will see this error if you have one browser tab pointed at a page in unRaid and on another tab you initiate a restart of unRaid. Note that the browser in question can also be on any device on your network. This includes other computers, tablets, phones, etc. IE: Close the other browser tabs. This error can also be caused by mobile apps such as ControlR checking the status of unRaid but the server has been rebooted after the app was started. Restart the application to fix. unitialized csrf_token - Thus far the community has never once seen any report of this being logged. Presumably it is an error generated by unRaid itself during Limetech's debugging period (ie: not plugin related), and should you see this you should post your diagnostics in the release thread for the version of unRaid you are running. EDIT: There is a possibility that if your rootfs is completely full due to misconfiguration of an application that you may see this particular token error.