CUDA Out of Memory

Hi everybody, I have 1 rig of 6 cards P106-100 6gb (5x MSI, 1x ZOTAC). I’ve been running it for i think around 2-3 weeks and suddenly it stop running…
There’s a error stated CUDA out of memory, what does this mean ? I’m still very new to GPU mining this is my first rig…

Below are the error message:
2021.03.31:03:28:37.894: GPU1 GPU1: Allocating DAG (4.17) GB; good for epoch up to #406
2021.03.31:03:28:37.894: GPU1 CUDA error in CudaProgram.cu:388 : out of memory (2)
2021.03.31:03:28:37.895: GPU1 GPU1: CUDA memory: 5.94 GB total, 1.65 GB free
2021.03.31:03:28:37.895: GPU1 GPU1 initMiner error: out of memory
2021.03.31:03:28:37.895: wdog Fatal error detected. Restarting.
2021.03.31:03:28:40.352: main Eth speed: 0.000 MH/s, shares: 0/0/0, time: 0:00

Thanks in advanced !

Which miner are you using?
Did you intended to run it directly from the console?

I’m using PhoenixMiner, tried a few other miners also same problem… I’m planing of using T-Rex Miner.

“Did you intended to run it directly from the console?” meaning ?
Sorry my english not so good…

1 Like

ok, you should format and and reimage your USB, that kind error happens for two reasons, either the GPU doen’t have enough RAM or another process is already using the available memory. I’m pretty sure it was the second, just recently a lot people have been advising ton uninstall PhoenixMiner as it posible that is being used to attack miners. And is very possible, it just happened to me recently, the miner was running but pointing to an address that wasn’t mine, and to a pool I have never connected too, the Rig was offline in the web, and the SSH password was changed.

If you still have access to you physical machine do the following command:

ps -ef | grep -i miner

That instruction would give the list of miner running.

Yeah, and for ‘console’ I meant the user interface which appears when HiveOS starts, sorry for the delay

I’ve run the command you sent me, and below is the msg…

I don’t see the message :stuck_out_tongue:

root 719 1464 0 09:49 pts/2 00:00:00 bash -c ???log_name="/tmp/exec_“748283420”.log" ???(ps -ef | grep -i miner) > $log_name 2>&1 ???exitcode=$? ???[[ $exitcode -eq 0 ]] && ???cat $log_name | message info “ps -ef | grep -i miner” payload --id=748283420 || ???cat $log_name | message error “ps -ef | grep -i miner (failed, exitcode=$exitcode)” payload --id=748283420 ???rm $log_name ???
root 723 719 0 09:49 pts/2 00:00:00 bash -c ???log_name="/tmp/exec_“748283420”.log" ???(ps -ef | grep -i miner) > $log_name 2>&1 ???exitcode=$? ???[[ $exitcode -eq 0 ]] && ???cat $log_name | message info “ps -ef | grep -i miner” payload --id=748283420 || ???cat $log_name | message error “ps -ef | grep -i miner (failed, exitcode=$exitcode)” payload --id=748283420 ???rm $log_name ???
root 726 723 0 09:49 pts/2 00:00:00 grep -i miner
root 1971 1 0 09:37 ? 00:00:00 SCREEN -dm -c /hive/etc/screenrc.miner bash
root 1991 1971 0 09:37 pts/3 00:00:00 bash /hive/bin/miner-run t-rex 1
root 2030 1991 0 09:37 pts/3 00:00:00 bash /hive/bin/miner-run t-rex 1

For what I can see there are two instances of t-rex, I don’t know if there are duplicates of the miners on your fligh sheets. If you haven’t resolved your problem, next you could post a screen shot of your worker’s dash board, just remember to hide personal information, farm hash, username, wallet addresses, etc.