Good afternoon, I set the miner to crontab to reset every day at the same time, the rig goes to retire, but freezes, and then I need to manually run the rig,Can anyone give me advice on this problem?
Topic: [OS] rxOC easy-to-use Linux AMD Mining v_stopgap - page 3. (Read 31370 times)
I guess so, I did with my rx580s and I dont have that loop. What cards are you using?
I use a Sapphire Rx580 Pulse 8GB, now running with rxOS in one rig...I am doing experiments on another USB disk on my main PC, using QEMU...now tried to upgrade but cannot, missing libs and dependencies...which is the correct procedure for upgrading the driver? Step by step?
like:
1) uninstall amdgpu-pro ->reboot
2) enable sources in software update
3) upgrade the system ->reboot
4) install amdgpu-pro 17.50
I guess so, I did with my rx580s and I dont have that loop. What cards are you using?
...when the driver is wrong for the present GPU (like when you are in an emulator) you are stuck at the ubuntu login. Even if you enter the correct user and pwd, the system cannot login...
BTW, I tried again to install the 17.50 driver...it does not install, many dependencies not satisfied...missing libs...have to update, upgrade the system before attempting to install the new driver...?
BTW, I tried again to install the 17.50 driver...it does not install, many dependencies not satisfied...missing libs...have to update, upgrade the system before attempting to install the new driver...?
sorry, not sure what do you mean by login loop?
@etherdave: you can update to latest amdgpupro drivers for yourself: https://support.amd.com/en-us/kb-articles/Pages/AMDGPU-PRO-Install.aspx
@rgr_rgr: thanks for your help, I am sorry my reply took so long but I'm very busy atm. I think utilization and temp of gpus would be minimum to be aware of. I had a look by myself and stumbled on the same thinking as you so if you can deliver some telegram code thats fine, if I get some time I will also
@rgr_rgr: thanks for your help, I am sorry my reply took so long but I'm very busy atm. I think utilization and temp of gpus would be minimum to be aware of. I had a look by myself and stumbled on the same thinking as you so if you can deliver some telegram code thats fine, if I get some time I will also
Thanks for your reply! I did attempt to update the driver with no success. This is what I did:
1) uninstall AMDGPU-PRO -> reboot
2) Install new AMDGPU-PRO ->reboot
3) I was at the Ubuntu login loop...but ok, I was running it in QEMU...so I put the USB key in the rig...but was still trapped in the login loop...:-(
...any ideas? Will it go better using new 17.50 driver?
@etherdave: you can update to latest amdgpupro drivers for yourself: https://support.amd.com/en-us/kb-articles/Pages/AMDGPU-PRO-Install.aspx
@rgr_rgr: thanks for your help, I am sorry my reply took so long but I'm very busy atm. I think utilization and temp of gpus would be minimum to be aware of. I had a look by myself and stumbled on the same thinking as you so if you can deliver some telegram code thats fine, if I get some time I will also
@rgr_rgr: thanks for your help, I am sorry my reply took so long but I'm very busy atm. I think utilization and temp of gpus would be minimum to be aware of. I had a look by myself and stumbled on the same thinking as you so if you can deliver some telegram code thats fine, if I get some time I will also
Hi guys, my best compliment to the creator of this very nice rxOS...but, no more developing on this? Really? It is working very well for me, but the beta driver feels like a bit old...any news? Thanks in advance for your time
3. will it be very hard to implement telegram, auto temp, wtm profit switcher from nvoc? I am looking forward very hard to build an image and if working, I am willing to distribute.
I am already able to send Telegram messages. Sendig is easy. But I see a lot of nvdia specific stuff in the telegram script used bei nvOC.
Parts could be done as well, but needs to get rewritten. What is the most important informations? How often to send?
If your autostart is still not working try to put "/bin/su -l m1 -c /home/m1/2unix' into /etc/rc.local
Update: When using my Vega FE (my intended GPUs to use) I receive an error that the dipay is running in low quality and I am not able to get past that point. I switched out to an RX580 and was able to boot into the system without any issues. The AMD Blockchain Beta drivers support Vega FE. Please advise how to get my FE's working correctly.
Thanks,
Hello,
I am using a Gigabyte z170x - Gaming 5 and it does not recognize the USB. I switched to an ASUS Prime z270-A and the USB is recognized and boots to a login screen, but when I enter the password miner1 the login screen simply reloads. Can you advise on what steps may be needed for my Gigabyte motherboard and/or how to get past the password screen on the ASUS?
I didn't manage to get the low graphics message fixed but you can ssh into the rig and control everything fine. Thanks,
Hello,
I am using a Gigabyte z170x - Gaming 5 and it does not recognize the USB. I switched to an ASUS Prime z270-A and the USB is recognized and boots to a login screen, but when I enter the password miner1 the login screen simply reloads. Can you advise on what steps may be needed for my Gigabyte motherboard and/or how to get past the password screen on the ASUS?
Update: When using my Vega FE (my intended GPUs to use) I receive an error that the dipay is running in low quality and I am not able to get past that point. I switched out to an RX580 and was able to boot into the system without any issues. The AMD Blockchain Beta drivers support Vega FE. Please advise how to get my FE's working correctly.
Thanks,
Hello,
I am using a Gigabyte z170x - Gaming 5 and it does not recognize the USB. I switched to an ASUS Prime z270-A and the USB is recognized and boots to a login screen, but when I enter the password miner1 the login screen simply reloads. Can you advise on what steps may be needed for my Gigabyte motherboard and/or how to get past the password screen on the ASUS?
Thanks,
Hello,
I am using a Gigabyte z170x - Gaming 5 and it does not recognize the USB. I switched to an ASUS Prime z270-A and the USB is recognized and boots to a login screen, but when I enter the password miner1 the login screen simply reloads. Can you advise on what steps may be needed for my Gigabyte motherboard and/or how to get past the password screen on the ASUS?
so for 2: it seems that 'sudo shutdown -r -f now' works
Code:
$ sudo systemctl reboot --force
https://www.freedesktop.org/software/systemd/man/systemctl.html#-f
so for 2: it seems that 'sudo shutdown -r -f now' works
ok, I have a couple of problems and I will list them all. Maybe anyone can give me a tip, as soon as I will find a solution I will report here.
1. I am still getting low graphics mode when using more than 6 cards. I can start mining with SSH and bash 2unix but unfortunately mining doesn't autostart anymore after reboot. anyone has a tip on how to fix low graphics mode, or disabling GUI or other way to autostart mining?
2. somehow rigs dont do sudo shutdown or sudo reboot. they loose SSH connection but nothing happens. need to power off.
3. will it be very hard to implement telegram, auto temp, wtm profit switcher from nvoc? I am looking forward very hard to build an image and if working, I am willing to distribute.
1. I am still getting low graphics mode when using more than 6 cards. I can start mining with SSH and bash 2unix but unfortunately mining doesn't autostart anymore after reboot. anyone has a tip on how to fix low graphics mode, or disabling GUI or other way to autostart mining?
2. somehow rigs dont do sudo shutdown or sudo reboot. they loose SSH connection but nothing happens. need to power off.
3. will it be very hard to implement telegram, auto temp, wtm profit switcher from nvoc? I am looking forward very hard to build an image and if working, I am willing to distribute.
i have 8 cards running by manually starting with bash 2unix. But how can I powerlimit these cards? what is the nvidia-smi equivalent? I installed amdgpu-pro drivers already
e: ok so it seems that i can see current power consumption per card with "./rocm-smi -P" just in case anyone is as lost as me
e: ok so it seems that i can see current power consumption per card with "./rocm-smi -P" just in case anyone is as lost as me
any fix for "low graphics mode"? after acrd 7 i get this error
I spend roughly 2 days now trying to get 8x rx580 running with rxoc. I have another rig with 8x 1060 and nvoc running. Hardware i use:
Asus prime z270a
Hp dps1200 with breakout board
160w pico psu
8gb ddr4 ram hynix
Celeron 3900
When i use 1-2 cards in the main 16x pcie slots they boot fine. But a 3rd sometimes doesnt come up (using the 3rd big pcie slot) and on 4th card sometimes screen freezes up on boot or i get strange pcie error output in terminal (using the 1x speed slot on the top / nearest to cpu)
Bios settings i did the same as the other rig i am using sith same asus board. On this video you can see wich settings: https://www.google.de/url?sa=t&source=web&rct=j&url=%23&ved=0ahUKEwi6oY6Er9XYAhVOZVAKHadWCukQwqsBCB8wAA&usg=AOvVaw0VTJzWxrN8ZFOD4xbU2nov
Double checked the settings. I switched mainboard, psu, riser, usb cables, cards, hdd. I tried every card on its own as single card setup, they all boot up fine.
With one power cable 6pin pcie i power riser 6pin pcie and psu 8pin pcie by y-cable. I get power for the ssd from the pico psu wich gets power from an own pcie port of the dps1200.
The cards have a bios switch. I tried both modes without success.
I really hope someone has an idea because i am lost.
BR
just want to add some information. first, my bios settings:Asus prime z270a
Hp dps1200 with breakout board
160w pico psu
8gb ddr4 ram hynix
Celeron 3900
When i use 1-2 cards in the main 16x pcie slots they boot fine. But a 3rd sometimes doesnt come up (using the 3rd big pcie slot) and on 4th card sometimes screen freezes up on boot or i get strange pcie error output in terminal (using the 1x speed slot on the top / nearest to cpu)
Bios settings i did the same as the other rig i am using sith same asus board. On this video you can see wich settings: https://www.google.de/url?sa=t&source=web&rct=j&url=%23&ved=0ahUKEwi6oY6Er9XYAhVOZVAKHadWCukQwqsBCB8wAA&usg=AOvVaw0VTJzWxrN8ZFOD4xbU2nov
Double checked the settings. I switched mainboard, psu, riser, usb cables, cards, hdd. I tried every card on its own as single card setup, they all boot up fine.
With one power cable 6pin pcie i power riser 6pin pcie and psu 8pin pcie by y-cable. I get power for the ssd from the pico psu wich gets power from an own pcie port of the dps1200.
The cards have a bios switch. I tried both modes without success.
I really hope someone has an idea because i am lost.
BR
https://cdn.discordapp.com/attachments/376037211717763072/402035746598289410/IMG-20180112-WA0004.jpg
when I insert 4 cards, sometimes the rig hangs. but sometimes on boot i see this:
https://giphy.com/gifs/3o7527iajvsJPiRk5y
saying something like "line dropped... pcieport failure... corrected" hard to read because blurry when appears.
found a solution online https://steemit.com/mining/@maebog/running-12x-gpu-on-the-biostar-tb250-btc-pro-with-simplemining-os
this should be implemented in the next release imho. keyword pcie=noaer (Advanced Error Reporting)
Stuck on the login screen every time i enter password miner1 simply appears to refresh and sends me back to enter it again anybody have a solution?
If this is the first time you have tried using the rxOC image on a newly built drive then I expect you may have not edited the oneBash file before booting the system.
I had this same issue the first time I used it.
There are some default settings in there for mining address and such that won't allow the system to boot all the way so it needs to be cleaned up.
If you had it running before then I'm sorry I have no helpful ideas.
Since fullzero has stopped the support help for rxOC (and nvOC as well) I don't think any of the other code writers have picked up for AMD.
nvOC is still getting quite a bit of activity and great support on that blog so I've decided to rid myself eventually of my AMD setup unless I find a substitute.
I have mostly Nvidia cards anyway.
I spend roughly 2 days now trying to get 8x rx580 running with rxoc. I have another rig with 8x 1060 and nvoc running. Hardware i use:
Asus prime z270a
Hp dps1200 with breakout board
160w pico psu
8gb ddr4 ram hynix
Celeron 3900
When i use 1-2 cards in the main 16x pcie slots they boot fine. But a 3rd sometimes doesnt come up (using the 3rd big pcie slot) and on 4th card sometimes screen freezes up on boot or i get strange pcie error output in terminal (using the 1x speed slot on the top / nearest to cpu)
Bios settings i did the same as the other rig i am using sith same asus board. On this video you can see wich settings: https://www.google.de/url?sa=t&source=web&rct=j&url=%23&ved=0ahUKEwi6oY6Er9XYAhVOZVAKHadWCukQwqsBCB8wAA&usg=AOvVaw0VTJzWxrN8ZFOD4xbU2nov
Double checked the settings. I switched mainboard, psu, riser, usb cables, cards, hdd. I tried every card on its own as single card setup, they all boot up fine.
With one power cable 6pin pcie i power riser 6pin pcie and psu 8pin pcie by y-cable. I get power for the ssd from the pico psu wich gets power from an own pcie port of the dps1200.
The cards have a bios switch. I tried both modes without success.
I really hope someone has an idea because i am lost.
BR
just want to add some information. first, my bios settings:Asus prime z270a
Hp dps1200 with breakout board
160w pico psu
8gb ddr4 ram hynix
Celeron 3900
When i use 1-2 cards in the main 16x pcie slots they boot fine. But a 3rd sometimes doesnt come up (using the 3rd big pcie slot) and on 4th card sometimes screen freezes up on boot or i get strange pcie error output in terminal (using the 1x speed slot on the top / nearest to cpu)
Bios settings i did the same as the other rig i am using sith same asus board. On this video you can see wich settings: https://www.google.de/url?sa=t&source=web&rct=j&url=%23&ved=0ahUKEwi6oY6Er9XYAhVOZVAKHadWCukQwqsBCB8wAA&usg=AOvVaw0VTJzWxrN8ZFOD4xbU2nov
Double checked the settings. I switched mainboard, psu, riser, usb cables, cards, hdd. I tried every card on its own as single card setup, they all boot up fine.
With one power cable 6pin pcie i power riser 6pin pcie and psu 8pin pcie by y-cable. I get power for the ssd from the pico psu wich gets power from an own pcie port of the dps1200.
The cards have a bios switch. I tried both modes without success.
I really hope someone has an idea because i am lost.
BR
https://cdn.discordapp.com/attachments/376037211717763072/402035746598289410/IMG-20180112-WA0004.jpg
when I insert 4 cards, sometimes the rig hangs. but sometimes on boot i see this:
https://giphy.com/gifs/3o7527iajvsJPiRk5y
saying something like "line dropped... pcieport failure... corrected" hard to read because blurry when appears.
Jump to: