TheFreemancer

Thank God this exists. When nerd tools stoped working this is the solution.

Thank you! This is a very useful information. Just found out what happened. I have a second lan port for docker containers (since those are exposed to the WAN). This cable were left unplug when I restarted the server and there was no Br1 network interface so the docker container failed. I'm going to restore those, thank you. One more question, should I update to 7.2.2 from 7.0.1?

This is really odd but I think unraid is removing the wrong orphan image from the docker container. I have updated binhex-sabnzbd container and it was gone. Updated ich777/radarr and it was gone too. I'm on 7.0.1. Should I update? Is there a way to get the containers back?

Just wanted to say thank you very much for this. After using your guide, plus a bit of help from chatgpt and also https://www.reddit.com/r/Ubiquiti/comments/v42avk/fix_for_adopting_offline_loop_on_docker_unifi/ I had this very annoying problem that adopting access pointes would never finish. All i had to do was point to my docker container IP. This was a few hours battle but I was finally able to migrate from my personal computer to my unraid server as my unifi controller. Thank you!

The commented line is from the network-rules.cfg example from the post I linked earlier. I only made adjustements for the MAC ADDRESS which I replaced with ***** to hide it here. I actually got it working as I wanted. I had another spare 2.5NIC with 4 ports. Intel chipset. Put those on unraid and interface rules showed up. Still no option to switch eth0, only eth1 2 3 4. So I took it to chat gpt and got a way to block the onboard nic via syslinux.cfg. Now I can switch anyway I want all 4 and eth1 became eth0.

Spoke too soon. After rebooting and getting the right NIC as ETH0.. unraid created a ETH123?? wtf.. And this would not show up in settings and it seems this is not persistent when rebooting. EDIT: Gave up, I'm swaping the cables and have to give up 2500 network for now. I'll order another PCIE NIC and see if unraid enforces so badly a NIC to be it's default one.

FINALLY GOT IT WORKING: This is sadly the only solution: Basically, deleting network.cfg won't solve my problem because when rebooting unraid will create a new network.cfg with active backup and bonding between both interfaces. What worked for me was: create the network-rules.cfg like this: # PCI device 0x15b3:0x1003 (mlx4_core) SUBSYSTEM=="net", ACTION=="add", DRIVERS=="?*", ATTR{address}=="*******", ATTR{dev_id}=="0x0", ATTR{type}=="1", KERNEL=="eth*", NAME="eth1" # PCI device 0x8086:0x1533 (igb) SUBSYSTEM=="net", ACTION=="add", DRIVERS=="?*", ATTR{address}=="******", ATTR{dev_id}=="0x0", ATTR{type}=="1", KERNEL=="eth*", NAME="eth0" set each mac address to the interface you want. And I had to have a network.cfg ready for this to work, no bonding, and everything set up as I wanted. and you can't reboot from ssh, reboot from unraid gui. Not sure why but it seems to be a different kind of reboot to apply those settings. This is such an old bug from unraid, not having the interface rules section sucks. I gave 4+ hours trying different solutions to such a simple problem. This should be fixed ASAP.

So, after years using my unraid server with a i7 6700k I finally upgraded to a xeon e5 2696v4. Upgrade went smooth and all the drives are in the right place but I have a problem here: The onboard NIC is gigabit only and I have a second NIC which is 2.5gbit. The onboard NIC is set as ETH0 and the PCIE is set as ETH1. All I want to do is to invert it but sadly I'm one of those that interface rules section doesn't show up at all. What I've tried so far: deleting /boot/config/network.cfg create /boot/config/network-rules.cfg moving 2.5gbit NIC from PCI-E slots upgrading unraid from 6.12.10 to 7.0.1 Tried disabling onboard NIC but there's no option in bios to do so. Is there any way to replace ETH0 with ETH1 and vice versa?

Got it working wget https://slackware.uk/slackware/slackware64-15.0/slackware64/n/openvpn-2.5.5-x86_64-1.txz wget https://slackware.uk/slackware/slackware64-current/slackware64/n/openssl-3.4.0-x86_64-1.txz installpkg openvpn-2.5.5-x86_64-1.txz installpkg openssl-3.4.0-x86_64-1.txz get https://github.com/alfredopalhares/openvpn-update-resolv-conf cp update-resolv-conf.sh /etc/openvpn/update-resolv-conf enjoy

I'm going to necrobump this since there was no answer since 2021 and it was the only result that asked exactly what I'm trying to do here. Connect to a openvpn server within unraid, not using docker containers.

Yeah, I was going to try to contact over github but there's no way I suppose. Too bad. I'll try to install it without a template and see how it goes.

@SmartPhoneLover any chance you still have the repository backed up? Gerbera is an exceptional UPnP media server and it's not available in community applications anymore.

I have a gigabit internet and while using SABNZBD in a docker container I'm not maxing out my internet connection and getting around 70 MB/s. So after a few days googling I found this post on reddit: https://www.reddit.com/r/SABnzbd/comments/1ap888p/download_speed_limited_by_disk_speed_with_an_ssd/ That points out that the user was doing the same as I: using 2 mount points for incomplete and complete downloads. In my case incomplete: /mnt/cache/incomplete (a NVME PCI-E SSD) and complete in /mnt/disks/XXXXX/usenet/complete (a different NVME PCI-E SSD). So I tried putting incomplete and complete in the NVME and got almost 78MB/s. So this is getting better. Now the problem is, the only way to have both showing under the same path would be mounting /mnt (which is a bad idea) and also it doesn't work because /mnt/cache isn't visible. Mounting /mnt/user/INCOMPLETE a share I made that points to /mnt/cache isn't a good idea because being part of the array I'm limited to read and write speeds compared to direct disk access. So is there any way to mount both /mnt/cache and /mnt/disks under the same volume being passed to the docker container? In any safe way?

It isn't? I solved it by disabling my onboard network adapter. It worked and I still don't have the interface rules. I've tried creating network-rules.cfg on /boot/config but this file is deleted at every boot. As I've seen in other threads this is a long lasting bug of unraid.

Hey, I have an onboard lan interface that I believe is Intel on my Asus motherboard. I recently got a Intel 2.5gbit NIC and as far as I can see it is blinking when the cable is plugged in. Besides that I can't see it on system devices and can't select it under network settings. There are several topics here saying you need to change it under interface rules but there's nothing of sort under network settings. I'm attaching the diagnostics here to find out what's going on. tower-diagnostics-20230709-1819.zip

Answering my own question XFS is faster compared to BTRFS: root@Tower:/mnt/cache# dd if=/dev/zero of=/mnt/cache/test2.img bs=3G count=1 oflag=dsync 0+1 records in 0+1 records out 2147479552 bytes (2.1 GB, 2.0 GiB) copied, 1.59658 s, 1.3 GB/s and doing a different test: root@Tower:/mnt/cache# dd if=/dev/zero of=/mnt/cache/test.img bs=1M count=5120 5120+0 records in 5120+0 records out 5368709120 bytes (5.4 GB, 5.0 GiB) copied, 2.34594 s, 2.3 GB/s Almost the same speed I was getting on windows.

Sorry to revive this topic but I need to ask one more question. I have replaced my kingston nv1 snvs/1000g for a adata XPG gammix s11 Pro that I was using in another PC. The XPG is much faster compared to the Kingston, I was getting 3GB/s with it on Windows. The m.2 connection on this motherboard is PCIE 3.0 X4 so it should max out at 32gbit/s and from previews tests with the kingston I was getting 1.3GB/s if writting directly to the cache. After replacing it: root@Tower:~# dd if=/dev/zero of=/mnt/cache/INCOMPLETE/test1.img bs=3G count=1 oflag=dsync 0+1 records in 0+1 records out 2147479552 bytes (2.1 GB, 2.0 GiB) copied, 2.32709 s, 923 MB/s Much slower. Before I try anything else, this cache drive is formated as BTRFS and I don't remember if the previews one was XFS. Should I go XFS for this drive to max out it's performance? Or file system matters little in this case?

That's fine, and thanks for the info. As I use it mostly for incomplete downloads from sabnzbd and for the rclone cache, which is already limited in size, this is fine.

Disk share would be the case I need better performance from another device on my LAN. That's a good info there. Mapping to /mnt/cache seems to be the answer. On Sabnzbd I've mapped to /mnt/cache/INCOMPLETE and the most performance difference I'm seeing is instead of using /mnt/user/rclone_cache is using /mnt/cache/rclone_cache/ Is there any complication that can come from using direct disk access instead of the user folder in the cache drive? I'm aware this is a problem when writting files to disks that are indeed part of the array.

It is set to only. Even tried prefer. As I understand this only changes how the mover will treat the folder. I want it to be outside of the array but still on the cache drive so I don't have to buy another NVME SSD just for this.

I replaced my 240GB SATA SSD with a 1TB M.2 NVME drive and I use it as cache drive. This cache drive is used for appdata, system, isos, VM img, docker img, all good part of the unraid system. But I also used it for partial downloads (incomplete folder) and for rclone cache. Now here's the problem: root@Tower:/tmp# dd if=/dev/zero of=/mnt/user/INCOMPLETE/test1.img bs=3G count=1 oflag=dsync 0+1 records in 0+1 records out 2147479552 bytes (2.1 GB, 2.0 GiB) copied, 4.36276 s, 492 MB/s and if it's outside of the array: root@Tower:/tmp# dd if=/dev/zero of=/mnt/cache/INCOMPLETE/test1.img bs=3G count=1 oflag=dsync 0+1 records in 0+1 records out 2147479552 bytes (2.1 GB, 2.0 GiB) copied, 1.68525 s, 1.3 GB/s How can I still have the INCOMPLETE folder but don't have it inside the array?

I've been reading the cache and cache pool FAQ but I still have some doubts and really would like some advice for my scenario. Right now I have a 240gb SSD sata drive as cache and a 2TB as unsigned device outside of the array. I don't use mover and only my appdata is on my cache drive. What I use my cache drive is for partial usenet downloads (incomplete folder) and I set on my unsigned device the complete folder. Usenet requires large unpaking and now that I'm getting a gigabit internet I want to improve my download speeds and specially unrar speed on completed downloads. I'm considering add a ssd nvme m.2 500gb drive and set it as cache. My question is, if I add it as cache pool the size missmatch will be a problem, right? If using raid0? Also this wouldn't be ideal because it would download on both and unpack on both. So, would be best to leave the 240gb as a complete download drive, and leave it as unsgined device and partial download on the new ssd nvme 500gb drive as a cache drive? And one final question, if appdata is not in any drive of the array, only on cache disk, if the cache disk dies, all data is gone, right? So, the best way to back it up, excluding the backup plugin and storing elsewhere, would be creating a script to copy it's content to the array every 24 hours, right? So if the drive dies, you just add another drive, set as cache and copy from the array to the cache drive. Does it sound as a good alternative?

So far so good. If the drive indeed fails, I can get another one and start rebuilding again, right?

Ok It's rebuilding now. Thanks for the help, hopefully everything goes well.

How do I do it? When rebooting the server I can assign the disk to the same slot but it is still disabled. Do I have to start it once with the slot disk set to none, reboot and assign the disk to that same slot again? Also when I set the disk4 slot to none the drive shows up as unsassigned device. Can the unassigned devices plugin do any harm to the files?