I currently have an 8-card Nvidia rig running the latest version of hiveos. I have a single ethernet cable going from my nighthawk router/modem to an 8-port switch. My 8-card rig and 3 other gaming PCs are plugged into that switch. It all ran perfectly for about a month. Then I started randomly losing internet connection for all the PCs while all the wifi worked perfectly. After much troubleshooting including replacing all the cables and even getting a new 8-port unmanaged switch, I was able to determine the hiveos mining rig was somehow causing the issue.
When I unplug the hiveos rig’s ethernet, everything else regains connection instantly.
I have setting static IPs for just the hiveos rig, and then for all the PCs connected via ethernet. I reinstalled drivers, and followed instructions from a post here to install the latest network driver on the hiveos machine.
I turned off DHCP on my router, and then couldn’t access my router’s home page XD so I factory reset it.
** The factory reset fixed the problem for three full days, but now it’s back. **
This problem is really hard to troubleshoot because I can lose connection in 20 seconds or up to 6 hours after plugging in the hiveos machine. It can happen while I’m doing things or randomly in the middle of the night with no activity. When it goes out, it never comes back on its own, and the wifi is just fine. All 4 machines are mining btw. 3 PCs on windows, all 4 machines using T-rex if that matters.
Any ideas? Halp pls T_T Sorry for the long post, trying to be thorough.
Since posting, I have not resolved the issue. I have completely rebuilt the offending rig, and now it is 100% new components (minus the GPUs). CPU cooler, motherboard, everything. I have 3 rigs I’m trying to run on HiveOS. One of them behaves, but the other two do the thing where they disconnect and take everything else with them (everything on the same branch of ethernet. Wifi is always unaffected.) When an offending rig is unplugged (hiveos is still running with the miner going as well), the others pop right back online, and 9 times out of 10 I can plug the offending PC back in seconds later and keep on hashing Ethereum.
The main change is that this happens after just a couple minutes now, not hours.
I gave EVERYTHING on my network a static IP, and ran the offending rigs one at a time with Wifi off while nothing else was connected via ethernet or wifi. Same result. Ran for a couple minutes, and then offline. I did this to rule out ARP spoofing. Any thoughts??
It’s also almost impossible to do any testing from the actual mining rig as it loses connection so fast, I can’t reliable ping things, etc.
Do you check the network cables? The is something with connectivity through cables…
Like you tell … the wifi plays without problems …So the problem is on cables or in the router (ethernet ports) …
I myself am experiencing this exact same issue as you basically word for word. I’ve replaced cables, modem, router and still no luck. This just started for me a couple weeks ago and I’m unsure where to go next. I’m thinking the network card on my H110 is busted so unsure if I should just buy a new motherboard or what
I had pretty much come to the same conclusion about my MPG Z390 Pro Carbon mobo, especially because I bought it used. However I even replaced the mobo (so it’s an entirely new system) and it’s still the same. My network seems to be perfectly fine. I got ahold of some of my software and hardware engineering buddies from college finally and they said “well if it’s not hardware, it’s software. Try another linux distro.” I put one of the rigs on Minerstat’s msOS last night. I have the only well-behaved hive rig runnig as well. Everything looks good right now. I am going to put the rest of them on msOS this weekend and report back. I’m convinced it’s hive. It seems to only really happen when I have multiple PCs running hive.
I had the same problems as you, checked and changed everything, nothing seems to help. However I found that the issue was indeed HiveOs and it was caused by a specific GPU in my rig. Whenever that GPU was not perfectly OC, the whole rig freezed and somehow disabled all the wired network. In my case it is an Asus Strix 3080 OC. So, check your OCs, and try to see the stats of each card just before the network crashed, one of them will have a temp or fan discrepancy, look for it and adjust the OC to an stable one. If you can’t find it then just try an stable OC with all the rig. Hope it helps.
I have the same issue with a EVGA 3080 non LHR. If I push the OC past 1900 after a couple hours it will freeze the rig and the network switch. Causing my whole farm to go offline until I reset the network switch.
Thank you everyone for the replies! Sorry you had similar issues, but I was glad to see I was not alone. I swaped over to minerstat’s msOS and have had zero network issues… or any issues really. I have been running 6 rigs stably for a month now, pushing 1.5GH on ETH over 8 workers. It seems it was a hive thing, or like same of you said, one GPU not being perfectly overclocked/underclocked. I used identical settings for all cards when I moved and had no issues, so one way or another it’s a hive thing.
I found this problem only happens in ASUS mother board with 2.5G ethernet port(but I run with 1G LAN).
When problem happen looks mass packet send from problematic platform. And when I unplug internet cable from it. My LAN will back to normal instantly.
I tried to update BIOS but useless.
So my current workaround is using WiFi instead ethernet cable.
My network router used Mikrotik RB750Gr3 and some unmanaged switch/poe switch. When problem happens, I can’t even ping any devices on LAN, neither enter router manage portal via Winbox(Mikrotik manage software)
And I guess probably cause by 2.5G Realtek ethernet chip set got some driver compatibility for Hive OS.
It looks like I have same issue with MSI 3080 supremeX. It was a mystery since I had one 6 card rig connected to a network with this card with no issues, I moved it to a rig with new MB Asus z590 2.5tb Ethernet and added 2 GPU (total 8 GPUs) and was no issues with network until: added another 8 cards rig on MSIZ590-A pro ( btw I have 4 rigs with this MSI Z590-A pro MB running at different location with no issues to network even when cards have error rig would restart and keep mining). With this second rig added to the same network now every day network will freeze wireless network works fine. It was nightmare since only one of the rig could work a one time. Until I noticed one card error. So with that MSI3080 GPU fan error the rig would restart and work fine by itself but when 2 rigs were mining on the same network I think that card would freeze the network with second rig running at the same time. I home that helps someone.
Hello, I currently have an 8-card Nvidia rig running the latest version of hiveos. I have a single ethernet cable going from my nighthawk … Reset your modem / router and restart your mining device or 192.168.l.254 rig . Use wired Ethernet instead of Wi-Fi, if possible. If not, at least move the rig … My rigs seem to be working. I tried: Downloading the same/ different image again and create new USB. Try Wifi. Windows is unable to manage multiple rigs from one interface. Where in dedicated operating systems like Hive OS, you can easily manage different …
