Terrace

Checked my year old OBS NDI Windows 10 VM to see if it could ping itself, so settings were just corrupt on the new VM that was unable to ping itself, built the whole thing again without the guide I used to set up OpenVPN and went straight for LogMeIn Himachi and it worked immediately, checking the VM was the right solution!

Having a problem with pinging and connecting to a game hosted on a Unraid VM, it's a LAN game so we tried to set up tunneling with hosting OpenVM on the VM and then when he couldn't connect we tried with (LogMeIn) "Himachi" tunneling in as well. There isn't a docker available for the game, but I don't understand why my VM won't allow me to host and have incoming connections or where to begin allowing incoming connections which seems to be the problem. So after not being able to get OpenVM or Himachi to connect to the Unraid VM we just decided to try connecting to each others gaming PC's and leave the Unraid VM that is suppose to be hosting the game out of it and immediately we could ping the assigned IP addresses through the application, and our own assigned IP addresses through windows command prompt, something I can't do on the VM for some reason. Internet browsing and such works fine on the VM, I can pull down files, watch youtube, access the Unraid shares, the VM "works" normally, I just don't get why I can't allow outside connections in. Friends Gaming PC - Can ping his own Himachi IP Address, and can ping My Gaming PC Himachi IP Address but not the VMW10 server Himachi IP Address My Gaming PC - Can Ping my Himachi IP Address, and can ping His PC Himachi IP Adress but not the server Himachi IP Address My Unraid PC with Windows 10 VM - Cannot ping it's own Himachi IP Address, cannot ping either of our gaming PCs Himachi IP Addresses

I'm following the first step, which is turning on the syslog mirror on the flash drive, but now that seems like I have to wait for it to randomly restart out of nowhere again to get anywhere with that, so this current restart event will just be a mystery? there's no way to solve it? It's a bit of a shame if it's hardware because it was 70 days stable at one point, and then i've manually turned the whole system off when someone came to change something with the gas supply and smart meter in my home. So it's been going really steady. I did spot one thing in red in the system log Sep 23 21:48:46 Tower kernel: 9pnet: Installing 9P2000 support Sep 23 21:48:46 Tower kernel: mce: [Hardware Error]: Machine check events logged Sep 23 21:48:46 Tower kernel: microcode: sig=0x906ea, pf=0x2, revision=0xf0 Sep 23 21:48:46 Tower kernel: microcode: Microcode Update Driver: v2.2. Microcode update into immediate hardware error?

Unraid restarted (unclean shutdown) whilst it was connected to a working UPS, losing me a lot of today's work, a large download that was almost complete and the ability to do any work tomorrow (20 hour parity check!) The power in the house didn't even go out, i've been on the second computer without a UPS the entire time, the light are on in the house, everything is fine except i just lost my Unraid to a restart that I didn't ask for. Rather than getting too angry about it and considering it some free time off, I do want to understand what happened though, I'm not very technically minded, so i might need it dumbed down a little for me. I'm uploading the whole diagnostic, but right at the start it seemed that it downloaded microcode and then crashed installing it? "Sep 23 21:48:46 Tower kernel: microcode: microcode updated early to revision 0xf0" If that's what did it, then I'd definitely like to know how to disable it doing that! I can't really afford this work computer to just go down and need a 20 hour parity check frequently. but of course it could be something else entirely. tower-diagnostics-20230923-2205.zip

So it's possible that as the bios was still showing 4.8ghz through multiple restarts that it really was running at 4.8 on the VM's and such? I wonder what the best way to test real cpu speed is. I'll take a look around on the internet. Seems like CPU-Z might show real speed rather than just reading out loud what's displayed in the windows panel. I'll try to find the time to turn the overclock settings back on and run CPU-Z. Thanks for pointing me in the right direction with unraid not turning off anything. I hope CPU-Z can still read a Virtual Machine setup

Oh sure I get that i'm on my own if i try it, I just wonderered if Unraid turns anything back off after it starts to load. I really did change the settings and double checked and made sure they were loaded, I can even enter the bios and see they are all set, but once i get into unraid everything seems to be back at stock. I've actually ended up breaking my entire network today with the introduction of my network cards, i can't even access unraid properly from the other PC anymore and managed to wreck everything. Trying to get some live help via the unraid discord networking section because I think i'll need to be walked through a lot in real time, but i don't think anyone is about at the moment.

After creating a lot of videos with Premiere Pro and using OBS with NDI on my VM It seems like my 8700k was the slowdown for premiere and loss of frames during streaming/recording on OBS. I'd decided today to do the research into overclocking, got the values entered, they looked good on the bios stating 4.8ghz instead of the 3.7ghz stock and was quite surprised when i turned everything on and it was set to stock, i noticed a unraid message saying: Aug 2 21:23:21 Tower kernel: tsc: Detected 3700.000 MHz processor Aug 2 21:23:21 Tower kernel: tsc: Detected 3699.850 MHz TSC went back into the bios, 4.8ghz, went back into unraid/VM 3.7ghz, then I did a lot of reading of people recommending not overclocking whilst using unraid, not even XMP (without knowing i'd had this on for 33 days so far) If I did want to try out overclocking before shelling out for a new MB/Processor, is there a safety I need to disable before unraid will allow me to overclock, or did perhaps something else go wrong that caused me to be at stock Ghz? Besides the money for a new MB/Processor I just brought the maximum amount of DDR4 ram THIS system could take it would be a shame to move onto another platform.

My cable just arrived today from China and the network was picked up immediately upon plugging it in, I needed a 12 core cable instead of a 24 core cable, and a type B cable instead of a type A! Thank you for the help here with narrowing down what could go wrong and the IP range validation and such!

I think standard could mean either of them, they're just listed on ebay for example 'ect ect type a cable' £60 and 'ect ect type b cable' £60 no colour change on the cable, no big description change, it's just as likely you'd hit an A or a B if a person went with top of page result

Trying to find the right cable is a bit of a pain, the ones i'm finding are '8 core' instead of '24 core' and i'm not sure if i need a 'patch cable' or a 'trunk cable' for a direct connection or one of the other types. And then there's 'type B Chanel 1 to Chanel 12' and I guess something else out there. Ugh, i wish the cable I have worked. https://www.fiber-optic-transceiver-module.com/difference-between-fiber-trunk-harness-and-patch-cables.html seems like Patch Cable is fine. Seems like the difference between and A and a B type is important and i have no idea which i need https://community.fs.com/news/type-b-polarity-mtp-cable-user-guide.html I think I need a Type-B Trunk Cable from reading this, I might be completely incorrect though

If 10.10.20.198 /24 is valid on the unraid and 10.10.20.196 and 255.255.255.0 is valid on the downstairs computer (because i've never ran two computers before, I wouldn't know) and everything else is ok, then I guess i'll have to bite the bullet and buy a second cable. If it's possible it got crushed in transit or if it was never working (purchased used on ebay a year ago, no way of testing it at the time of arrival) I guess if i start with a cable first then atleast I only need to buy one thing to test. I'll wait a day before i buy just in case any other ideas come up.

Are the diagnostics not visible from the last time i posted them? trurl asked me to post them as the first post, and then i immediately posted them. Is my attachment not visible to others?

Yep, Ethernet mode, both are as above: LINK_TYPE_P1 ETH(2) LINK_TYPE_P2 ETH(2) default was Auto (VPI?)which I read doesnt work, and the third option is infiniband(INF) which I also read wasn't right. both have been rebooted and checked that ETH is set

Here you go!tower-diagnostics-20230703-0017.zip Edit: I tried to follow along mostly with this Spaceinvader One video about using this brand of network cards with unraid and others And I have a plugin called "Mellanox Firmware Tools" from "ich777" but I wasn't overly sure if i needed it, or if i did need it if i needed to click anything.

So it's my first network that wasn't plugged in via a telephone cable or a Cat Cable and my first server (Unraid is really cool) I'm really enjoying having a file server and VM's to do my video production remotely from downstairs without making the Downstairs Computer basically unusable. Except all my Traffic, both file and internet is running through my poor Router because my ConnectX-3 cards show as cable unplugged. Downstairs is a https://www.asus.com/us/motherboards-components/motherboards/prime/prime-z690-a/techspec/ and it's in the lowest slot (Windows 10 / PopOS machine) Upstairs is a https://www.msi.com/Motherboard/Z370-SLI-PLUS/Specification and it's in the lowest slot (Unraid) 2x Mellanox CX354A ConnectX-3 FDR Infiniband 40GbE PCIex8 Prolabs OM3-24C-MPO/F-13.72M Fibre Optic Cable Finisar FTL414QL2C-E5 QSFP 4X14Gpbs FDR Transceiver pic of port 2 https://imgur.com/J7H3hnf Unraid pic of port 1 https://imgur.com/5QIHNec Windows10 / PopOS (since moved it to port 2 just to try) Transceivers plugged in with a good 'click' and won't pull backwards out without the tab pull Cable plugged in with a small 'click' Will the 'network display cable disconnected' error show up even if everything is plugged in correctly, but I haven't made a connection to the second computer as a settings error? or is cable disconnected more of an immediate problem for each computer like the hardware is faulty like the cable? I made downstairs IP 10.10.20.196 and subnet 255.255.255.0 rest blank, and upstairs 10.10.20.198/24 in unraid Both cards were flashed and everything on the same windows 10 same day running mstconfig -d 03:00.0 q on Unraids terminal gives me: Device type: ConnectX3 Device: 03:00.0 Configurations: Next Boot SRIOV_EN False(0) NUM_OF_VFS 8 LINK_TYPE_P1 ETH(2) LINK_TYPE_P2 ETH(2) LOG_BAR_SIZE 3 BOOT_PKEY_P1 0 BOOT_PKEY_P2 0 BOOT_OPTION_ROM_EN_P1 True(1) BOOT_VLAN_EN_P1 False(0) BOOT_RETRY_CNT_P1 0 LEGACY_BOOT_PROTOCOL_P1 PXE(1) BOOT_VLAN_P1 1 BOOT_OPTION_ROM_EN_P2 True(1) BOOT_VLAN_EN_P2 False(0) BOOT_RETRY_CNT_P2 0 LEGACY_BOOT_PROTOCOL_P2 PXE(1) BOOT_VLAN_P2 1 IP_VER_P1 IPv4(0) IP_VER_P2 IPv4(0) CQ_TIMESTAMP True(1) Win10 Downstairs running mlxconfig -d /dev/mst/mt4099_pci_cr0 query gives the exact same results as above. running flint -d mt4099_pci_cr0 query full on downstairs gives: Image Type: FS2 FW Version 2.42.5000 FW Release date 5.9.2017 MiC Version 2.0.0 Config Sectors 1 PRS Name cx3-1_MCX354A_fdr_09v.prs Product Version 02.42.50.00 Rom Info: type=PXE Version=3.4.752 Device ID:4099 GUIDS: lots of numbers Macs Lots of numbers VSD: -blank here- PSID MT_1090120019 Added a couple of other pictures here, but honestly i'm ready to try any advice, the website which suggested ConnectX-3 as something i should get don't really seem to have any help they wanted to give me. I was really hoping these would be plug'n'play and simple, and this has been making me miserable for a month and i don't know how to help myself other than basically buying a second set of everything and swapping it all out, which still might even do anything if it was settings. some device manager pic https://imgur.com/2YqFtRq unraids network page pic https://imgur.com/KeGAadP network settings window10 https://imgur.com/G2weRqL