Roma4593

Hello friends, I am looking for some help- I am very new to networking so apologies for dumb questions. I had my private domain up and running flawlessly with a reverse proxy on my unraid server via swag and cloudflareddns. All of my intended subdomain/hosts were accessible over WAN which was incredible. I unfortunately had a disk failure and had to shutdown my array to rebuild. I have the following script to run whenever I restart the array in order to rebuild my docker networks -> otherwise I lose custom br0 for all my containers. rm /var/lib/docker/network/files/local-kv.db /etc/rc.d/rc.docker restart I subsequently recreated my proxy network via cmd as I had previously done. docker network create ****** I started up cloudflareddns and swag without any noticeable log errors but unfortunately now I am getting ERROR 521 when trying to connect over WAN in the browser. Ive turned off all the proxys in cloudflare and purged the chache in cloudflare. When I make A names directly to my WAN IP I get "This site can't be reached". I've done everything short of just shutting down the array/rebooting the server. Not sure where along restarting my array things broke. There's unfortunatley not a lot of information regarding error 521 and the one or two threads addressing this error seem to thing its on cloudflare's end? I am going to hard reset the server once my drive is rebuilt. I sincerely appreciate anyones input/help on the matter. James halo-diagnostics-20210320-1246.zip

Thank you!!! James

Hello all, I had a quick question regarding network ports. I am in the process of setting up pihole and received an error message that the port 80 was already in use (error attached). None of my current dockers -all of which were OFF anyways contain that port assignment. I did "lsof -Pni | grep LISTEN" in the terminal and it shows a "NGINX" "LISTEN" on port 80 (also attached). Is this a standard UNRAID process? is there a way to remove "NGINX"? Is this Malicious process??? Side note - I set up pihole in the past briefly without any problems using the standard template ports and completely removed it for a time. When I change the template ports to finish the install it does not work properly/block ads. I appreciate anyone's help - networking is very complex to me unfortunately! VR James halo-diagnostics-20210225-1344.zip

I am running it in unraid. I was just wondering where the file was that the template modifies that includes the server name, password, ports, etc. Thanks

My friends and I are absolutely loving the Valheim server docker. Just quick question, where is the start_headless_server.bat’-file (Linux: ‘start_server.sh’) equivalent so that I can manually change server name, instance name and password etc besides going through the UNRAID docker GUI. Thanks for a fantastic piece of software James

Will probably stick with windows at this point. There are sync features in krusader that I wish I could take advantage of. I appreciate all of the help.

I'm beginning to wonder if it is something with my network settings ie something wrong with my routing table. Unfortunately I know very little about networking and I am worried I would brick access to my unraid web UI. I've attached the noted speeds under interface as well as a screen of my network settings. Seems that bond0 is "fault-tolerance" eth0 is GB eth1 down? Please let me know if anyone has thoughts on this.

Thank you for your responses so far. I had not looked into MC. After learning about it, I managed to create a connection between my UNRAID server and my PR4100 and had the same slow speeds as with the krusader GUI! Photo attached. I would also receive an error when writing UNRAID -> PR4100 " Error ERRDOS - ERRbadpath creating directory \Public\Test\Books" for example - however transfer speed is still my biggest issue at the moment. best

I get the full GB + speed when I transfer within a windows VM between the UNRAID machines when mapped as network drives. UNRAID server 1 -> windows vm -> UNRAID server 1 mapped network drive <----> UNRAID server 2 mapped network drive = 110+ MB/s using krusader SMB - UNRAID server 1 <-------> UNRAID server 2 = 2-8 Mib/s Very confusing why this would be the case. Same hardware paths.

Hello all I am a newer UNRAID user and new to the forum. I was hoping for some help - I now have krusader to facilitate LAN file transfers between my other backup UNRAID server as well as a WD PR4100 NAS. I have gigabit connection when transferring files within my UNRAID server via windows VM to and from the other devices as "network drives" however krusader SMB connection transfers over the exact same path and hardware are getting between 2-8Mib/s. Completely unusable. I hope someone can help. For networking I have asus RT AX92U over aimesh between my UNRAID server and other devices. Diagnostics are attached. halo-diagnostics-20210206-2115.zip

Wow. As I was on the verge of scrapping the whole project, I did what Linus did and put a an old potato graphics card in PCIe1 and moved my GTX1080 to the second PCIex16 slot...... checked the GTX in the VM gui and it works.... This seems like a ghetto solution... but I guess I will take it unless there is a better fix. Thanks JR

Hello friends, I was looking for a storage NAS solution as well as the possibility of hosting gaming VMs with the same machine. have followed spaceinvader one's series on setting up UNRAID and creating daily driver/gaming VM. I have been able to launch a windows 10 vm and setup NAS which works beautifully --- however--- I have spent hours on hours of troubleshooting GPU passthrough and I am unfortunately still getting this dreaded error code #43. FX8320 with 990FXA UD3 mobo FB bios Asus gtx1080 turbo HBA pcie card with 4x WD 2tb drives 1TB m.2 on PCIE for cache 16gb ddr3 ram I've essentially done all of the aforementioned troubleshooting in this thread. 1. IOMMU groups are natively separated 2. disabled hyper-v through both unraid gui (does not seem to change xml) and via xml script 3. assigned multifunction to GPU and assigned sound to appropriate slot - ive also tried just starting the VM with just the GPU and no sound 4. have trailed alternative machine types 5. I've attempted to perform Siwat's "advanced" workaround however even with cuda 9.1.85 there is an error with attempting to sign the file in powershell 6. I've attempted to apply drivers directly with nvidia dch installer 7. I've attempted to run my GPU concurrently with vnc to update drivers/install nvidia control panel for which VNC interface becomes dead (probably because of splashtop in the background) 8. I've attempted to uninstall/reinstall GPU drivers via splashtop 9. hardware was rechecked -gpu works flawlessly on primary win10 computer 10. software updated - windows updates up to date, bios is latest installment (FB) 11. legacy only boot settings in bios 12. I have attempted to install drivers via geforce experience however i am only able to do so while running vnc and GPU concurrently as windows basic drivers limit my resolution to 680 and does not support many native system viewer windows much less geforce experience ui. When I have attempted to place graphics ram BIOS with a modified techpowerup .dump file - the VM will start but does not boot and curiously pegs an assigned CPU core to 100%. Ive attempted varying the VM cpu assignments with the same thing happening regardless. As a final note - I have had some of this hardware for quite some time - my motherboard and CPU are are 8 years old but still capable- it would seem that this shouldn't necessarily be the issue but if it is then I can settle with no GPU passthrough for this particular UNRAID machine (although gpu passthrough is specifically why I chose UNRAID). My ultimate goal was to run this as a test machine and ultimately run a gaming vm/backup storage NAS UNRAID system with a ryzen r9 chipset. Would newer hardware pose just as much difficulty with GPU passthrough??? It seems like generally people dont seem to have much problem except those of us in here. Any help would be sincerely appreciated. Please let me know of any further information I can provide that could help.