Pstark

UPDATE; I moved the NIC that wasn't populating in UNRAID for a 3rd or 4th and then it showed up. I'm going to move pfSense off the server and use an old hardware I have laying around to make a dedicated pfSense box. I'm sure my family will appreciate it since the network won't go down if I reboot my server anymore.

This is my eth0. This is good and not a problem. These are the onboard NIC's from the Supermicro board that I don't want to use. There should be another X520 2 port sfp 10G NIC but I cant get it to show up after moving it. I don't know if the vfio-pcie is stored somewhere else and it's still being stubbed. I cant boot the VM without this NIC available. tower-diagnostics-20230408-1515.zip

I also want to know if using my router as an access point effects the network? Because it's a 1G device plugged in to a 10G network does that limit all devices connections on the network to 1G?

If in the future I need more 10G rj45, would it cause any issues to buy an additional switch to expand the 10G ports? The QSW-M1208-8C looks pretty nice but will I be able to plug in additional switches into this? ie. plug my 2.5G switch in to a 10G rj45 and still retain 10G speeds on the rest of the ports on the QNAP?

I want to improve my home networks speed. I have a 10Gb intel NIC (X520-DA1 with an ioPlex asf-10G-T transceiver) in my server and I want to be able to move large amounts of data quickly. I have questions on planning everything out so I don't waste time and money. This is my current network setup: Internet (1Gb) comes into my house to my home to my Arris SB8300 modem. --cat6a--> into pfSense WAN (VM inside UNRAID via X520-DA2 ioPlex asf-10g-t) --cat6a--> LAN port of pfSense (DA2 second port fs sfp-1-g-t) ---cat6a--> TrendNet s350 (2.5G 5 port switch): 1 cat6a ---> Unraid 2 cat6a ----> 1Gb TP-Link 8 port switch (below) 3 cat6a ---> Game PC 4 cat6a <----- LAN connection from pfSense 5 empty TP-Link 8 Port Gbe switch 1 cat6a ---> Amplifi Alien mesh network (pass through mode for wifi access to network) --cat5e--> ethernet backbone to second router/meshpoint 2 cat6a --->Gaming PC 3 cat6a ---> iot device 4 cat6a ---> iot device 5 cat6a <---- LAN connection from TrenNet switch 6 cat6a ----> IPMI port 7 empty 8 empty I'm looking for a 10 port (or more) switch with at least 5 10GbE ports (can be sfp+). Things to note: have 2 fiber SFP+ transceivers not being used vlan and switch management support is wanted but not needed. I want to learn about this and use if these options can be beneficial and this is a good time to make this kind of decision. I make my own cat6a cables. I have started buying hardware for this upgrade over the past few months (NIC's) but before I get anything further, I would like the communities opinion on what options I have for finishing my network and what I have already as well as any issues with my current setup?

Yes there should be an Intel DA2 SFP+ NIC

tower-diagnostics-20230407-0056.zip

After reboot my second NIC it not being seen.

Thanks! This was what I was thinking I had to do but I didn't know what the file was called. I ended up moving my LSI card to another spot and it booted without issue. It's booting again after deleting the vfio-pci.cfg | my pfSense NIC wasn't found in system devices so I went ahead and deleted the file you referenced. Will update after the reboot.

I moved PCIE devices around so that my LSI card would receive better cooling. I also repasted it (not sure if this could be the cause) before putting it back in. First boot after moving these cards was fairly normal with the exception of my pfSense VM not booting. I plugged in a second router and navigated to the webUI and the VM was stopped. I hit start and error PCIE device missing. I physically checked the NIC assigned to the VM and the lights were flashing as if it were being used. (I have a second NIC that's assigned to eth0) I powered down the server and checked the NIC I passed through and reseated it. Then rebooted the server. But UNRAID will no longer boot.... fully. Everything is normal during boot until, during the boot sequence while watching the log, the screen clears and then nothing. No drive activity or anything on the console. LSI card is initialized during boot but since I repasted it, I took it out and rebooted the server. It boots enough to ssh into it but no webUI. I'm thinking since the NIC being passed through was moved, it's causing issues with other devices but unsure on how to fix it. syslog

I'm completely new to pfSense and I followed this guide to create a policy for qBittorrent WAN traffic to be routed through a VPN. Now radarr can't connect to qBittorrent.

pfSense uses cloudflare DNS servers. So Unraid wants the gateway IP for both primary and secondary DNS servers?

I attached a screenshot of the settings I used when attempting to set a static IP. Unraid doesn't like my DNS settings but only with the static IP. pfSense gives the same DNS settings. I loose br0 after every reboot as well.

Current build: Supermicro H12SSL-I EPYC 7302P 96GB ECC 3200 Mhz LSI 9300 16i HBA Array; Parity drive: 18 TB EXOS SAS Disk 1: 18 TB Ultrastar DC HC550 SATA Disk 2: 18 TB EXOS SATA Disk 3: 18 TB EXOS SATA Disk 4: 12 TB EXOS SATA Disk 5: 12 TB Iron Wolf (not pro) SATA Disk 6: 10 TB HGST He10 SAS Disk 7: 8 TB Red Plus SATA Bought the SAS EXOS drive with the intention of removing the 10 and 8 TB drives from the array and I expected to get better IO performance due to the SAS drive being the parity drive. I'm not using turbo write so drives not being used can spin down but ~150MB/s is not what I was expecting. Am I expecting to much? All drives are healthy. Enabling turbo write doesn't affect IO performance either. Disks 0 (parity) through 4 are in a NORCO SS-500 Disks 5 though 7 are in another NORCO SS-500 I use these cables- Internal Mini SAS to SATA Cable, SFF-8643 to SATA Breakout https://a.co/d/2hSam69

I did this and then got DNS errors causing not access to the internet.

I was running it as a VM as well. It was painful getting it running the first time but today I received a new drive and had to stop the array (loosing pfSense) then I couldn't connect to Unraid anymore. I connected directly to Unraid but the connection very inconsistent. I gave up on that and used my old router, finished the new config but was never able to connect to the pfSense network. My adapter couldn't get an IP from the DHCP server. I think my pfSense vdisk may have been changed but I will investigate tomorrow.

So in pfSense it gives Unraid a static IP but in Unraid it gets it IP from DHCP server. So should I set Unraid (in Unraid network settings) to the same static IP? I thought this could cause problems. I will look into this. Is it possible to do the same with Unraid networking?

One of the other problems I was having (pending if fixed) was some of my docker containers would error on starting the container saying the container doesn't exist. I found out that all the containers that were using "br0" network were now getting this error when attempting to start. The containers were still on the docker tab and I just went into the container and the network would be set to "None", I just changed it to any other network available, save and then go back in and change it to "br0". Problem fixed and run without issue. I just don't know if I restart the server again I will loose br0 again? Another thought I'm having: Since I'm running pfSense as a VM could it cause this issue since Unraid wont get an IP until after the pfSense VM boots?

Updated my isolation and now I'm getting an error from FCP.

Is it normal for Unraid to only use 1 thread? I isolated 2 cores for the host to use but I have never seen it use any more than 1 thread. I imagine it would make the UI snappier if it could utilize more.

Micron 32GB PC4-3200AA 3200Mhz DDR4 ECC Registered Ram Memory pn: 36ASF4G72PZ-3G2J3

Diagnostics tower-diagnostics-20230329-0827.zip

Built my new server (below) with the intention of running pfSense as a VM. I got everything up and running with minimal issue. The first issue I noticed was I didn't have a br0 network for my dockers. To fix: 'rm /var/lib/docker/network/files/local-kv.db" then "/etc/rc.d/rc.docker restart" then created my docker network and everything seemed fine. But since doing this, things seem to not work as before the new build. For example I cant connect to photoprism webui. I spent most of yesterday trying to solve it to no avail. I restarted the server in the process and once again br0 is gone. Could pfSense be the cause of these issues? H12SSL-I EPYC 7302P 96GB EEC 3200MHz X520-DA1 (was being used but currently waiting for transceiver to arrive) X520-DA2 (pfSense) GPU LSI 16 port

Any update on this? Mine is pausing after 1-2 hours.

While I already have my new system up and running, I can confirm you are correct. I had to cut the slot and my HBA works in a x1 slot surprisingly.