Kevinb123213

Oh, thank you for this additional info. I’m glad you were able to find a way for it to persist after reboot.

If you have a backup of your USB. Try and restore from there. Looks like unclean shutdowns are the cause of corrupting files.

TL;DR Files ending in a dash seem to indicate some sort of corruption. Long version: Found the solution. I think I should share it as it may help others. Apparently my setup cuts power and resets when I passthrough one of my GPUs - specifically the Nvidida Quadro K600. Knowing that I had an unclean shutdown, I went down the thought processes of data corruption. Which made me remember, a while back I noticed on a corrupted copy unraid usb, the dash ending files (syslinux-) appeared there too. So, I ran the following and pressed option 2 to restore from backup. fsck /boot At this point I restarted and was able to start libvirt. Now time to figure why my K600 is causing reset. -Kevin

Hello, I had an unclean shutdown (power cut and powered back up itself) for some reason when passing a gpu to a vm. Since then, I have not been able to start libvirt. Diag attached. My /etc/libvirt/ dir seems to have been automatically renamed to /etc/libvirt-/. What does a dash at the end of a directory signify? Can I just do the following? mv /etc/libvirt-/ /etc/libvirt/ Noteworthy errors: syslog root: /usr/sbin/libvirtd: error: Unable to initialize network sockets. Check /var/log/messages or run without --daemon for more info. libvirt 17277: error : virNetSocketNewListenUNIX:511 : Failed to bind socket to '/var/run/libvirt/libvirt-sock': Address already in use BTW, deleting libvirt img did not fix the issue. Thanks in advance, Kevin everest-diagnostics-20220926-1422.zip

TL;DR: You can enable blanking (black screen output) as a screen saver by doing the following: 1. Backup slim.conf cp /etc/slim.conf /etc/slim.conf.orginal 2. Edit slim.conf nano /etc/slim.conf 3. look for this line. xserver_arguments -novtswitch -allowMouseOpenFail -quiet -xinerama -v -s 0 -dpms -fp /usr/share/fonts/TTF vt07 4. Change to the following xserver_arguments -novtswitch -allowMouseOpenFail -quiet -xinerama -v -s 1 dpms -fp /usr/share/fonts/TTF vt07 You can disable screen output (no signal output) as a screen saver by doing the following: 1. Repeat steps 1 - 3. 2. Change to the following xserver_arguments -novtswitch -allowMouseOpenFail -quiet -xinerama v -s 1 dpms -fp /usr/share/fonts/TTF vt07 DISCLAIMER: Do this at your own risk. You are responsible if you break your unraid setup. Edit: Is there really no way to blank the output of the screen?

Hello, I have a monitor that has Auto Input select. I would like to have the UnRaid GUI be able to sleep after x minutes (blank the video output) when booted into GUI mode. Is there a file I can edit to enable this or command I can execute? I do not see a way to do it the the GUI itself. Ideally, i want the monitor to automatically switch to the secondary input when Unraid's GUI blanks (sleeps) and still be able to wake via usb keyboard/mouse to resume video output. Thank you in advance, Kevin Sent from my iPhone using Tapatalk

I tried that too. However, it doesn't give the error just clicking apply does nothing. Issue resolved, Thank you. I feel like an idiot LOL

Hello community, I believe I've found a minor bug in the Dynamix WireGuard. The Local endpoint field doesn't allow certain domain names that are valid. For example the following FQDN will not work although it is valid: 0xFE80FFFF.ddns.net or any domain with a hexadecimal address in it. Please see screenshot attached. Thank you, Kevin

Looks like my motherboard ASUS P8Z77-M are all connected via the Z77 chipset (figure 1). Is it possible altering USB BIOS settings could help? If so, which ones? If I were to go the path of getting a USB PCIe card which would you recommend for best compatibility with UnRAID? I've tested all 4 back USB 2.0 ports and they all seem to give crashes shown in the initial post. I've never had this issue in the past. I probably should note the only change I made before this issue occurred is that I re-flashed same BIOS via ASUS EZ-Flash tool in BIOS. Figure1: Thanks again, Kevin

Thanks, I'll give this a shot. May take some time to get back to you because I have a few USB 2.0 ports.

Hello, I'm experiencing a strange issue where the web GUI seem corrupted with a banner across the top saying: "System notifications are disabled..." So, I took a look at my syslog and I seem to be having a disconnect issue with the USB drive I presume. The only fix i've found is rebooting the server. However, it will happen again 2-3 days later. Thanks in advance, Kevin everestsrvr-diagnostics-20191030-2227.zipeverestsrvr-diagnostics-20191030-2227.zip

UPDATE: still crashes in safe mode.. Please help. Suggestions greatly appreciated. Same behavior -- numlock is lit up (can't toggle it as keyboard is frozen) and no access to server via ssh, telnet, ping, nor http/https while in this frozen state. -Kevin

Unfortunately, I'm not using VMs. I was still getting these soft-locks with virtualization disabled in BIOS. However, there was a point in time which I was using VMs and passing hardware through. Just today, I noticed that my Syslinux configuration had hardware ids for passing through old Network cards and graphics cards of which are no longer in the server. I have removed those pcie ids boot flags. If I understand you correctly, you are saying that this could cause these soft-locks?

Bump. I am 23 days stable *knock on wood* but only in safe mode with drives mounted and docker running. What else does safe mode disable besides plugins? Can I assume at this point it is a plugin issue? I've attached my stable logs here. Thanks, Kevin everestsrvr-diagnostics-20190525-1108.zip

I thought it might be a SMART/HDD problem because I was receiving offline uncorrectable errors on two drives from unassigned devices. I have pulled those drives from sata and power and removed from chassis. However, freezing continues. I've also done a memtest from unraid boot. It passed 2 passes (not sure if thats enough). My next steps. Boot UnRAID in safe mode and test stability. Thanks, Kevin

Thank you PeteB for the prompt reply. I've taken a look at your post and I am running the latest BIOS 2203 for P8Z77-M but there hasn't been an update since 2016 for the BIOS. Looking at some of the replies on your post i'm going to try the following and get back to you: 1. tail -f /var/log/syslog Thanks to glennv on your post. 2. Post a picture of console screen with tail command at hung state. I have same lockup behavior where it will freeze randomly between just a few hours and 11 days. I have the following plugins installed and up-to-date: CA Auto Update Applications -- 2019.02.22 Community Applications -- 2019.04.06a Dynamix Active Streams -- 2019.01.03 Dynamix SSD TRIM -- 2017.04.23a Fix Common Problems -- 2019.03.26 Nerd Tools -- 2019.01.25 Speedtest Command Line Tool -- 2018.02.11 Tips and Tweaks -- 2019.03.29 Unassigned Devices -- 2019.03.31 I have the following Docker Containers installed and up-to-date:

UPDATE: the cursor remains to flash but keyboard is frozen. I toggle caps lock and num-lock and the lights (caps and num) just stay as it was before freezing. Hello, I've attached my FCPsyslog_tail.txt & everestsrvr-diagnostics-20190408-0031.zip. Please note that I am unable to capture a syslog before rebooting as no network connectivity and unable to type in console during hung state. I would take a picture when it's frozen but it's just a CLI login prompt with a frozen solid cursor (not flashing). Any help is greatly appreciated! Thank you, Kevin

Hello lime-tech community, I am unable to autostart a Windows 10 VM after turning the server on. Please note the following is 100% reproducible on my system. (5/5 trials) 1. Shut down the VM and host machine running unRAID 6.3.5. 2. unplug the power cable to all power supplies for 5 minutes 3. power button for a good 30 seconds. 4. press the power button to turn on unRAID and let it boot 5. Observe that the Windows 10 VM will freeze shown in attached picture "stuck-bootup.JPG" 6. Go to the webGUI of your unRAID Server and log in. 7. You are able to navigate through all tabs and settings in web GUI EXCEPT VM tab at this point without any lag or issue. I even downloaded the Diagnostics at this point (will be attached). 8. Click the VM tab in the web GUI. Note: this may take up to 6 and a half minutes to load, as it took over 6 minutes for my trial. 9. Observe the VM tab of Web GUI will finally load showing the VM as running as seen in attached picture "VMtabBeforeRefresh.PNG" 10. refresh the VM tab in the browser. 11. Observe that the VM will be stopped now as shown in the attached picture "VMtabAfterRefresh.PNG" 12. Start the same VM from Web GUI. 13. Observe that Windows 10 will now boot up and log in. Any help in the right direction in fixing this issue is greatly appreciated. tower-diagnostics-20170601-1459.zip tower-diagnostics-20170601-1527-After-Successful-Boot.zip

Thank you! That helps.

Hello, Firstly, I have 4 cache drives, three of them are 120GB, and the fourth is 512GB. I was wondering why unRAID only shows exactly half of the storage? (120*3)+512 = 872GB is the actual storage. However, unRAID only shows 436GB for the cache pool (exactly half of 872GB). Is this a RAID 1 or mirror? Secondly, My windows 10 VM was pausing because unRAID was reporting that the cache pool was full when I had 150GB free. What would cause this? Is my unRAID cache pool setup correctly? I found a temporary fix by freeing up an old 60GB "vdisk.img". Why can I not even use half of the storage on the drives? It just doesn't make sense. Any help towards me understanding what is going on is greatly appreciated. Thank you, Kevin syslog.1.txt syslog.txt btrfs-df.txt btrfs-show.txt Samsung_SSD_840_EVO_120GB_S1D5NSAF541803M-20170502-1325.txt Samsung_SSD_850_EVO_120GB_S21TNXAG719705J-20170502-1325.txt Samsung_SSD_850_EVO_120GB_S21TNXAG719706W-20170502-1325.txt Samsung_SSD_950_PRO_512GB_S2GMNX0H500113K-20170502-1325.txt

I have an issue after using the windows 10 VM for a while I do a clean shutdown through windows 10, and unraid becomes unresponsive either in GUI or CLI mode. won't accept keyboard or mouse input anymore. Also when i press the physical power button on my case it doesn't do the normal two beeps. So I am forced to press and hold the power button.. I have the troubleshoot plugin installed and have attached the FCPsyslog and tower-diagnostics. Could someone please point me in the correct direction to resolve this issue? Thanks in advance, Kevin FCPsyslog_tail.txt tower-diagnostics-20170420-2022.zip

https://i.gyazo.com/ef203a011ecb733bb88940a43d545dc0.png Thank you trurl. Please mark as solved. I resolved this issue with chmod 777

Hello, I have a strange issue where two folders (Documents, and Downloads) are set to Read & execute permissions under Windows 10. However the Remainder folders are just fine (Videos, Music, etc). All these folders belong under a SMB Shared called 'Media'. To illustrate it is \\tower\Media\Downloads & \\tower\Media\Documents are the folders giving me issues. This share is set to public under web GUI for UnRAID. Things I have tried: 1. Changing Permissions in windows 10: Right click either downloads or documents folder and select properties. Navigate to 'Security' tab. Click 'Advanced' button. Change user 'Everyone' from 'Read & Execute' to 'Full Control'. Click Apply. The following error message apears: https://i.gyazo.com/49e2b8732a68e44f76324c5b57ca8aa1.png Click ok and this error apears: https://i.gyazo.com/5d0a95e56ea645d0a557209221ef61a6.png Here are the current permissions for Downloads and Documents: https://i.gyazo.com/bacce9437a2344cf8598ae6e033480bc.png Here are the current permissions for Videos (a working directory): https://i.gyazo.com/47b5589693b516aa744a8c68a61a330b.png 2.Right click the Downloads or Documents directory and 'Taking Ownership' option: i get the following output: "ERROR: This security ID may not be assigned as the owner of this object." 3.Ensure the SMB share is public. Which it is: https://i.gyazo.com/b6f93f935fe2c1fb75cc63ebe4338e98.png Please help me, I've been struggling for hours. Thanks, Kevin