Bitcoin Forum
November 21, 2017, 09:24:53 PM *
News: Latest stable version of Bitcoin Core: 0.15.1  [Torrent].
 
   Home   Help Search Donate Login Register  
Pages: « 1 ... 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 [53] 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 ... 287 »
  Print  
Author Topic: [OS] nvOC easy-to-use Linux Nvidia Mining v0019-1.4  (Read 327320 times)
fullzero
Legendary
*
Offline Offline

Activity: 1092



View Profile
June 22, 2017, 01:02:32 AM
 #1041

Anyone knows if EWBF will be updated to 0.3.4b ? Any reasons to do so ? I see 2% dev fee was added though but hashrate suppose to also increase by 2%.

Thanks
You can do it yourself if you want, see some pages back. It's about 1-2% faster but dev fee is a fixed 2%. Currently the only reason I see is that you'd want the new monitoring page. Personally I keep 3.3b running.

I agree with Maxximus007; but added the ability to choose between the two with v0016 as some members want to use 3.4.

I will run some rigs on each version and will be able to give solid data on relative performance later.

1511299493
Hero Member
*
Offline Offline

Posts: 1511299493

View Profile Personal Message (Offline)

Ignore
1511299493
Reply with quote  #2

1511299493
Report to moderator
Join ICO Now Coinlancer is Disrupting the Freelance marketplace!
Advertised sites are not endorsed by the Bitcoin Forum. They may be unsafe, untrustworthy, or illegal in your jurisdiction. Advertise here.
1511299493
Hero Member
*
Offline Offline

Posts: 1511299493

View Profile Personal Message (Offline)

Ignore
1511299493
Reply with quote  #2

1511299493
Report to moderator
1511299493
Hero Member
*
Offline Offline

Posts: 1511299493

View Profile Personal Message (Offline)

Ignore
1511299493
Reply with quote  #2

1511299493
Report to moderator
fullzero
Legendary
*
Offline Offline

Activity: 1092



View Profile
June 22, 2017, 01:04:19 AM
 #1042

using:

ETH_WORKER="miner7"
ETH_ADDRESS="ijduncan"
ETH_POOL="eth.suprnova.cc:5000"

anyone have a thought on that?

cheers
Wrong ETH_ADDRESS




ah ok figured it out

ETHADDR="$ETH_ADDRESS/$ETH_WORKER"

needed to be

ETHADDR="$ETH_ADDRESS.$ETH_WORKER"

for suprnova


You can solve this problem by setting
Code:
ETHERMINEdotORG="YES"

in oneBash

it is annoying that all the pools can't use the same syntax

fullzero
Legendary
*
Offline Offline

Activity: 1092



View Profile
June 22, 2017, 01:16:05 AM
 #1043

I tried mining monacoin last night but I got a cuda library error. How do I add the ability to mine lyra2rev2 algorithms?

You can copy or modify the VTC coin sections to use mona instead.

I will add a MONA coin selection to another oneBash; I will add this to the request list.

fullzero
Legendary
*
Offline Offline

Activity: 1092



View Profile
June 22, 2017, 01:25:12 AM
 #1044

Change Log:
v0016
  auto repair xorg: fixes single GPU not OCing bug (will auto detect / repair / reboot if using a single GPU or the xorg file has otherwise been damaged)
  implemented Maxximus007 suggestion to enable expect-less oneBash
  Thanks Maxximus007  Smiley
  powerlimit not setting bug is fixed (even for very slow USB keys)
  added Teamviewer with oneBash YES / NO switch to enable / disable it
  added openssh-server switch in oneBash YES / NO switch to enable / disable it
  added Desktop Sharing RDP program (enable / setup with GUI on the taskbar)
  added ifconfig output to the top of mining process (easy to get rig IP)
  added EWBF v3.4
  added Claymore v9.5
  added EWBF and Claymore version selectors to oneBash
  added EXTENTION_ARGUMENTS for each Ethash COIN (to add additional Claymore parameters)
  increased Claymore stability
  reduced boot time
  will now work with slow USB keys
    VTC COIN selection with VTC or BTC payout
    ZEN COIN selection
    NICE COIN selection




Fantastic work, fullzero. Tons of new features here, but I'm personally grateful for the continued under the hood optimization. Hopefully, the improvements to deal with slow USB sticks will help a lot of posters that may not have realized their USB write speeds were hindering them.

The slowest usb key I have is an old usb 2.0 kingston datatraveler.  I know there are slower usb keys out there; so if anyone has one that cannot OC / powerlimit with nvOC please let me know.

Tasbi
Newbie
*
Offline Offline

Activity: 9


View Profile
June 22, 2017, 01:27:25 AM
 #1045

Hi All,

I have decided to give nvOC a go, I am using an older Gigabyte GA-EX58-UD4P (Socket LGA1366) board and an i7 920 CPU.  I have two EVGA GTX 1080ti on risers in both PCI-E x16 slots.
I am able to boot fine but things just are not stable, I can run for about 15 hours but later GPU0 drops to about 150-170w and SOLs are suffering, I can reboot and get it working but it seems to happen again.

I have no idea if any BIOS changes need to be made to get this running stable or not, does anyone have any idea what I can try?

Also if I kill the mining process and make changes to the onebash config it doesn't seem to accept the changes, more specially the fan speed.  I attempt to launch the miner again and get the below error but the process does start;


Failed to connect to Mir: Failed to connect to server socket: No such file or directory
Unable to init server: Could not connect: Connection refused

ERROR: The control display is undefined; please run `nvidia-settings --help`
       for usage information.


When the mining process starts do you see OC messages like this?

https://ip.bitcointalk.org/?u=https%3A%2F%2Fs13.postimg.org%2Fq08huqnyv%2FIMG_0270.jpg%26t%3D577%26c%3Dyq0szP4ICxv47w&t=577&c=q_tbQ_FaeljjYQ

If you don't:

At any point did you boot with the monitor connected to the motherboard?

Did you at anytime boot with only one GPU attached?

If either of these is the case: ensure the monitor is attached to the primary GPU ( the one connected to the 16x slot closest to the CPU )

then follow this process:

https://bitcointalk.org/index.php?topic=1854250.msg19449945#msg19449945

I only booted with a monitor connected the first time, now it is completely headless.
Yes I might have booted with only one GPU attached at one point.
Thank you I will follow that process.  If I re-image the USB key and boot completely headless should I be seeing the OC messages via SSH as well?

With v0015 you would have to enable openssh server before you could SSH in.  You would also have to be able to know the rigs IP; which can be done in several ways.

With v0016 you can enable openssh server in oneBash, I would recommending trying it and seeing if it solves the problem.

I have now also tested with v0016, the mining process starts fine with all OC settings when viewing directly on the rig.  However I am still getting the error via SSH when trying to start.

Failed to connect to Mir: Failed to connect to server socket: No such file or directory
Unable to init server: Could not connect: Connection refused

ERROR: The control display is undefined; please run `nvidia-settings --help`
       for usage information.

What type of OS is the client computer; and what are the IPs of the client and rig?

Windows 10 running putty, Rig: 192.168.1.19 Client: 192.168.1.6

So when you enter

m1@192.168.1.19   

using port 22

with SSH selected

and click open it gives you the error above?


Sorry let me try and clear some things up.  I am able to SSH into the rig just fine but when executing the miner via ssh, this is when i am getting the error, so clock settings, fan speed and power limits are not being set but the mining process still starts.

I also really appreciate your support and rapid response.

Are you killing the existing mining process before launching another via SSH?

Also; when you are launching the mining process via SSH are you using the cmd:

Code:
bash '/media/m1/1263-A96E/oneBash'

I am SSHing and executing the following commands.

ps aux | grep gnome-terminal - finding the gnome-server PID
kill PID
screen -S rig1
bash '/media/m1/1263-A96E/oneBash'

Just before the miner starts I see the above error in place of the "attribute" commands but the mining process still starts.


after you have SSHed in enter the cmd:

Code:
echo $DISPLAY

and tell me what it outputs

That does not output anything at all.

It seems, im having the same troubles. So i will do the same here and post my results.
fullzero
Legendary
*
Offline Offline

Activity: 1092



View Profile
June 22, 2017, 01:34:04 AM
 #1046

Thanks for v0016, OP!  I'm running it on 1 of 2 rigs for now.  Definitely a big fan of the faster start time!

Let me know if you find any bugs.

fullzero
Legendary
*
Offline Offline

Activity: 1092



View Profile
June 22, 2017, 01:39:15 AM
 #1047

Trying to set up my first rig with the Asus z270 series board, Ubuntu 16.04, and 7 1060's.  Updated the bios, did all of the gen 2 switching ect yet still can't get more than 5 cards operating at once (It's random each time I reboot whether 4 or 5 seem to work). At first it was the cards with which I had individual sata riser power cables daisy chained (2 cards on each cable) but after unhooking everything and setting it back up that does not seem to be the issue.  I have also tried swapping out risers in order to see if one had gone bad also with no luck, ditto with the graphics cards.  I am kind of at a loss as what to try at this point in order to get all 7 operational at once.  

Which Asus z270 board?

Sorry, Prime Asus z270-A.

We have the same board, if you see above that is my exact setup including which PCI lanes I am using along with a video with step by step instructions.

Checked out that video earlier and made all of the bios changes (on 2 separate boards just in case one was failing) still the same issues.  Kind of at a loss.

Have you changed your PCI lanes you are using? And have your monitor output to the closest x16 to your CPU?

I switched the PCI-Lanes to 2.0, and I actually forgot I had taken my monitor out of GPU1 and hooked it directly the Motherboard earlier .... but when I put it back on GPU1 and rebooted that Card stopped working and I still only have 5 GPU's running.  GPU 5 which was not working on my previous boot started working this time in it's place.

Maybe v0016 will work.

fullzero
Legendary
*
Offline Offline

Activity: 1092



View Profile
June 22, 2017, 01:40:54 AM
 #1048

Hello,

I have an issue on version 0015:
I have severals rigs on Asus Z270 Prime-P but I do not manage to boot nvOC on a Asus Z270 Prime-A:
I loaded default bios, used the setting on the first page, plugged 8 cards using a m2 adaptater --> The bios POST, so everything seems to be OK. Then, I have a black screen as is should start loading the OS, the USB key seems to be accessed as it gets hot, but the black screen never ends. I waited 5 mins, nothing: only a black screen.
I tried to redo an other USB key without luck.
Do you have any idea what is blocking the boot?

Thanks

I have same issue with 0.14, 0.15 and 0.16. I am waiting 15-30 min with black sxreen, before system boot on. I have no idea what to do. I have different motherbords, usb flashes but time for boot on is so long
With V0016 boot time should be (very) much shorter. Please do a BIOS reset (jumper on MB). After that set the recommended BIOS settings, and boot with more than one card (on risers), monitor on the first GPU. Hope it helps! It's one of the motherboards I have on my shortlist for my next rigs.

I did BIOS reset with my ASUS H270, set up just 4G enable, but nothing changed, i can;t load mvOS. but with old MB i have this USB works very fast. I have no idea why its slow with new ASUS H270 and MSI Z270 MB

Did you update the bios?

with this motherboard it can be done by connecting an ethernet cable and entering the EZ Flash 3 Utility.  

Select DHCP and download then install the update


fullzero
Legendary
*
Offline Offline

Activity: 1092



View Profile
June 22, 2017, 01:43:28 AM
 #1049

Very nice updates in v0016 fullzero.  Smiley

Other than that, manual GPUFANSPEED doesn't work on my z270-a prime/evga 1080 Ti sc black.

What does the top of the mining process display where it should be showing fan control state and fan speed messages?


fullzero
Legendary
*
Offline Offline

Activity: 1092



View Profile
June 22, 2017, 01:47:47 AM
 #1050

I will respond to all posts later today: sorry if I have taken a longtime to respond.  I will make sure I respond to everyone at some point tonight.

v0016 is out see the OP for a link.

I didn't implement all requested changes, but I think I got most of them.  If I missed yours with v0016 I will get to it soon.

I have only tested this build on 8 rigs; so let me know if you find any bugs.  Grin

Change Log:
v0016
  auto repair xorg: fixes single GPU not OCing bug (will auto detect / repair / reboot if using a single GPU or the xorg file has otherwise been damaged)
  implemented Maxximus007 suggestion to enable expect-less oneBash
  Thanks Maxximus007  Smiley
  powerlimit not setting bug is fixed (even for very slow USB keys)
  added Teamviewer with oneBash YES / NO switch to enable / disable it
  added openssh-server switch in oneBash YES / NO switch to enable / disable it
  added Desktop Sharing RDP program (enable / setup with GUI on the taskbar)
  added ifconfig output to the top of mining process (easy to get rig IP)
  added EWBF v3.4
  added Claymore v9.5
  added EWBF and Claymore version selectors to oneBash
  added EXTENTION_ARGUMENTS for each Ethash COIN (to add additional Claymore parameters)
  increased Claymore stability
  reduced boot time
  will now work with slow USB keys
    VTC COIN selection with VTC or BTC payout
    ZEN COIN selection
    NICE COIN selection




keep getting low disk space
is something logging now?

There should be no logging other than the standard Ubuntu syslog; which should take up almost no space.

In v0016; to make the image smaller: I reduced the size of the primary partition by almost 1gb, which also reduced the free space available to the system. 

It is possible I removed too much, and should enlarge the partition for the next version.  Let me know if anything other than low disk messages happen; or if it appears performance is degraded in anyway as a result.


fullzero
Legendary
*
Offline Offline

Activity: 1092



View Profile
June 22, 2017, 01:53:53 AM
 #1051

Hi All,

I have decided to give nvOC a go, I am using an older Gigabyte GA-EX58-UD4P (Socket LGA1366) board and an i7 920 CPU.  I have two EVGA GTX 1080ti on risers in both PCI-E x16 slots.
I am able to boot fine but things just are not stable, I can run for about 15 hours but later GPU0 drops to about 150-170w and SOLs are suffering, I can reboot and get it working but it seems to happen again.

I have no idea if any BIOS changes need to be made to get this running stable or not, does anyone have any idea what I can try?

Also if I kill the mining process and make changes to the onebash config it doesn't seem to accept the changes, more specially the fan speed.  I attempt to launch the miner again and get the below error but the process does start;


Failed to connect to Mir: Failed to connect to server socket: No such file or directory
Unable to init server: Could not connect: Connection refused

ERROR: The control display is undefined; please run `nvidia-settings --help`
       for usage information.


When the mining process starts do you see OC messages like this?

https://ip.bitcointalk.org/?u=https%3A%2F%2Fs13.postimg.org%2Fq08huqnyv%2FIMG_0270.jpg%26t%3D577%26c%3Dyq0szP4ICxv47w&t=577&c=q_tbQ_FaeljjYQ

If you don't:

At any point did you boot with the monitor connected to the motherboard?

Did you at anytime boot with only one GPU attached?

If either of these is the case: ensure the monitor is attached to the primary GPU ( the one connected to the 16x slot closest to the CPU )

then follow this process:

https://bitcointalk.org/index.php?topic=1854250.msg19449945#msg19449945

I only booted with a monitor connected the first time, now it is completely headless.
Yes I might have booted with only one GPU attached at one point.
Thank you I will follow that process.  If I re-image the USB key and boot completely headless should I be seeing the OC messages via SSH as well?

With v0015 you would have to enable openssh server before you could SSH in.  You would also have to be able to know the rigs IP; which can be done in several ways.

With v0016 you can enable openssh server in oneBash, I would recommending trying it and seeing if it solves the problem.

I have now also tested with v0016, the mining process starts fine with all OC settings when viewing directly on the rig.  However I am still getting the error via SSH when trying to start.

Failed to connect to Mir: Failed to connect to server socket: No such file or directory
Unable to init server: Could not connect: Connection refused

ERROR: The control display is undefined; please run `nvidia-settings --help`
       for usage information.

What type of OS is the client computer; and what are the IPs of the client and rig?

Windows 10 running putty, Rig: 192.168.1.19 Client: 192.168.1.6

So when you enter

m1@192.168.1.19   

using port 22

with SSH selected

and click open it gives you the error above?


Sorry let me try and clear some things up.  I am able to SSH into the rig just fine but when executing the miner via ssh, this is when i am getting the error, so clock settings, fan speed and power limits are not being set but the mining process still starts.

I also really appreciate your support and rapid response.

Are you killing the existing mining process before launching another via SSH?

Also; when you are launching the mining process via SSH are you using the cmd:

Code:
bash '/media/m1/1263-A96E/oneBash'

I am SSHing and executing the following commands.

ps aux | grep gnome-terminal - finding the gnome-server PID
kill PID
screen -S rig1
bash '/media/m1/1263-A96E/oneBash'

Just before the miner starts I see the above error in place of the "attribute" commands but the mining process still starts.


after you have SSHed in enter the cmd:

Code:
echo $DISPLAY

and tell me what it outputs

That does not output anything at all.

Ok, tomorrow I will try to replicate this error; and see if I can figure out what is happening.  My guess is X11 is having a problem trying to output graphically.

If you have a linux computer other than your rig and you SSH into the rig with it; does it have the same error?

lbrasi
Newbie
*
Offline Offline

Activity: 25


View Profile
June 22, 2017, 02:08:09 AM
 #1052

Hi All,

I have decided to give nvOC a go, I am using an older Gigabyte GA-EX58-UD4P (Socket LGA1366) board and an i7 920 CPU.  I have two EVGA GTX 1080ti on risers in both PCI-E x16 slots.
I am able to boot fine but things just are not stable, I can run for about 15 hours but later GPU0 drops to about 150-170w and SOLs are suffering, I can reboot and get it working but it seems to happen again.

I have no idea if any BIOS changes need to be made to get this running stable or not, does anyone have any idea what I can try?

Also if I kill the mining process and make changes to the onebash config it doesn't seem to accept the changes, more specially the fan speed.  I attempt to launch the miner again and get the below error but the process does start;


Failed to connect to Mir: Failed to connect to server socket: No such file or directory
Unable to init server: Could not connect: Connection refused

ERROR: The control display is undefined; please run `nvidia-settings --help`
       for usage information.


When the mining process starts do you see OC messages like this?

https://ip.bitcointalk.org/?u=https%3A%2F%2Fs13.postimg.org%2Fq08huqnyv%2FIMG_0270.jpg%26t%3D577%26c%3Dyq0szP4ICxv47w&t=577&c=q_tbQ_FaeljjYQ

If you don't:

At any point did you boot with the monitor connected to the motherboard?

Did you at anytime boot with only one GPU attached?

If either of these is the case: ensure the monitor is attached to the primary GPU ( the one connected to the 16x slot closest to the CPU )

then follow this process:

https://bitcointalk.org/index.php?topic=1854250.msg19449945#msg19449945

I only booted with a monitor connected the first time, now it is completely headless.
Yes I might have booted with only one GPU attached at one point.
Thank you I will follow that process.  If I re-image the USB key and boot completely headless should I be seeing the OC messages via SSH as well?

With v0015 you would have to enable openssh server before you could SSH in.  You would also have to be able to know the rigs IP; which can be done in several ways.

With v0016 you can enable openssh server in oneBash, I would recommending trying it and seeing if it solves the problem.

I have now also tested with v0016, the mining process starts fine with all OC settings when viewing directly on the rig.  However I am still getting the error via SSH when trying to start.

Failed to connect to Mir: Failed to connect to server socket: No such file or directory
Unable to init server: Could not connect: Connection refused

ERROR: The control display is undefined; please run `nvidia-settings --help`
       for usage information.

What type of OS is the client computer; and what are the IPs of the client and rig?

Windows 10 running putty, Rig: 192.168.1.19 Client: 192.168.1.6

So when you enter

m1@192.168.1.19   

using port 22

with SSH selected

and click open it gives you the error above?


Sorry let me try and clear some things up.  I am able to SSH into the rig just fine but when executing the miner via ssh, this is when i am getting the error, so clock settings, fan speed and power limits are not being set but the mining process still starts.

I also really appreciate your support and rapid response.

Are you killing the existing mining process before launching another via SSH?

Also; when you are launching the mining process via SSH are you using the cmd:

Code:
bash '/media/m1/1263-A96E/oneBash'

I am SSHing and executing the following commands.

ps aux | grep gnome-terminal - finding the gnome-server PID
kill PID
screen -S rig1
bash '/media/m1/1263-A96E/oneBash'

Just before the miner starts I see the above error in place of the "attribute" commands but the mining process still starts.


after you have SSHed in enter the cmd:

Code:
echo $DISPLAY

and tell me what it outputs

That does not output anything at all.

Ok, tomorrow I will try to replicate this error; and see if I can figure out what is happening.  My guess is X11 is having a problem trying to output graphically.

If you have a linux computer other than your rig and you SSH into the rig with it; does it have the same error?

I tested from a linux VM I have and the same result Sad
Plumbus
Jr. Member
*
Offline Offline

Activity: 42


View Profile
June 22, 2017, 03:08:13 AM
 #1053

When I try to run lbry I get an error. Something to the effect of "980ti: invalid device function"

Works fine with ZEC and ZEN
ijduncan
Newbie
*
Offline Offline

Activity: 21


View Profile
June 22, 2017, 04:09:36 AM
 #1054

Amazing work ont eh update.

Got it working with no issues but then received the following error.  Dont know what to make of it.


 DCR: Stratum - connecting to 'dcr.suprnova.cc' <217.182.138.167> port 3252
ETH: Stratum - Connected (eth.suprnova.cc:5000)
 DCR: Stratum - Connected (dcr.suprnova.cc:3252)
 DCR: Authorized
 DCR: 06/22/17-00:04:37 - New job from dcr.suprnova.cc:3252

Received unknown response: {"id":null,"method":"mining.notify","params":["3247","129656acf5804305816f3dc0b32b47077c8cf4b8724f181efc529a93e4c1385b","999803e4fef6e5c5edd9968b3e00398713ca6f8d7f24812ea49ebf8373f11e8d",true]}
GPU0 t=54C fan=0%, GPU1 t=52C fan=0%, GPU2 t=49C fan=0%
ETH: Job timeout, disconnect, retry in 20 sec...


anybody have any thoughts?
Tasbi
Newbie
*
Offline Offline

Activity: 9


View Profile
June 22, 2017, 05:08:51 AM
 #1055

Version 0016 installed, I just tried teamviewer, PL is working good.
I tried with 9x gtx 1060 3gb mining ETHEREUM at 22.5 each with 880 watts on the wall. TOTAL 202 Mh/s

POWERLIMIT="YES"
POWERLIMIT_WATTS=84
__CORE_OVERCLOCK=100
MEMORY_OVERCLOCK=1400

Working temp 58C - 64C / fan at 75%

I had the issue of SSHing and executing bash '/media/m1/1263-A96E/oneBash',
it gave me this error,:

Failed to connect to Mir: Failed to connect to server socket: No such file or directory
Unable to init server: Could not connect: Connection refused

ERROR: The control display is undefined; please run `nvidia-settings --help` for usage information.

but now that I have teamviewer then I don´t need that anymore...

thanks fullzero




Maxximus007
Full Member
***
Offline Offline

Activity: 136


View Profile
June 22, 2017, 07:44:54 AM
 #1056

I will respond to all posts later today: sorry if I have taken a longtime to respond.  I will make sure I respond to everyone at some point tonight.

v0016 is out see the OP for a link.

I didn't implement all requested changes, but I think I got most of them.  If I missed yours with v0016 I will get to it soon.

I have only tested this build on 8 rigs; so let me know if you find any bugs.  Grin

Change Log:
v0016
  auto repair xorg: fixes single GPU not OCing bug (will auto detect / repair / reboot if using a single GPU or the xorg file has otherwise been damaged)
  implemented Maxximus007 suggestion to enable expect-less oneBash
  Thanks Maxximus007  Smiley
  powerlimit not setting bug is fixed (even for very slow USB keys)
  added Teamviewer with oneBash YES / NO switch to enable / disable it
  added openssh-server switch in oneBash YES / NO switch to enable / disable it
  added Desktop Sharing RDP program (enable / setup with GUI on the taskbar)
  added ifconfig output to the top of mining process (easy to get rig IP)
  added EWBF v3.4
  added Claymore v9.5
  added EWBF and Claymore version selectors to oneBash
  added EXTENTION_ARGUMENTS for each Ethash COIN (to add additional Claymore parameters)
  increased Claymore stability
  reduced boot time
  will now work with slow USB keys
    VTC COIN selection with VTC or BTC payout
    ZEN COIN selection
    NICE COIN selection




keep getting low disk space
is something logging now?

There should be no logging other than the standard Ubuntu syslog; which should take up almost no space.

In v0016; to make the image smaller: I reduced the size of the primary partition by almost 1gb, which also reduced the free space available to the system. 

It is possible I removed too much, and should enlarge the partition for the next version.  Let me know if anything other than low disk messages happen; or if it appears performance is degraded in anyway as a result.



You might want to run sudo apt update, sudo apt upgrade, sudo apt autoremove. This will remove around 1.5 Gb obsolete files. Apt was complaining about a file in /lib/modules/4.4.0-65.generic/kernel/drivers/media/usb/hdpvr/hdpvr.ko. I removed that PID file by hand, and ran sudo apt autoremove again. Currently 3.3Gb free space, that should be plenty.
** disclaimer ** it worked for me, YMMV
Maxximus007
Full Member
***
Offline Offline

Activity: 136


View Profile
June 22, 2017, 07:58:30 AM
 #1057

Hi All,

I have decided to give nvOC a go, I am using an older Gigabyte GA-EX58-UD4P (Socket LGA1366) board and an i7 920 CPU.  I have two EVGA GTX 1080ti on risers in both PCI-E x16 slots.
I am able to boot fine but things just are not stable, I can run for about 15 hours but later GPU0 drops to about 150-170w and SOLs are suffering, I can reboot and get it working but it seems to happen again.

I have no idea if any BIOS changes need to be made to get this running stable or not, does anyone have any idea what I can try?

Also if I kill the mining process and make changes to the onebash config it doesn't seem to accept the changes, more specially the fan speed.  I attempt to launch the miner again and get the below error but the process does start;


Failed to connect to Mir: Failed to connect to server socket: No such file or directory
Unable to init server: Could not connect: Connection refused

ERROR: The control display is undefined; please run `nvidia-settings --help`
       for usage information.


When the mining process starts do you see OC messages like this?

https://ip.bitcointalk.org/?u=https%3A%2F%2Fs13.postimg.org%2Fq08huqnyv%2FIMG_0270.jpg%26t%3D577%26c%3Dyq0szP4ICxv47w&t=577&c=q_tbQ_FaeljjYQ

If you don't:

At any point did you boot with the monitor connected to the motherboard?

Did you at anytime boot with only one GPU attached?

If either of these is the case: ensure the monitor is attached to the primary GPU ( the one connected to the 16x slot closest to the CPU )

then follow this process:

https://bitcointalk.org/index.php?topic=1854250.msg19449945#msg19449945

I only booted with a monitor connected the first time, now it is completely headless.
Yes I might have booted with only one GPU attached at one point.
Thank you I will follow that process.  If I re-image the USB key and boot completely headless should I be seeing the OC messages via SSH as well?

With v0015 you would have to enable openssh server before you could SSH in.  You would also have to be able to know the rigs IP; which can be done in several ways.

With v0016 you can enable openssh server in oneBash, I would recommending trying it and seeing if it solves the problem.

I have now also tested with v0016, the mining process starts fine with all OC settings when viewing directly on the rig.  However I am still getting the error via SSH when trying to start.

Failed to connect to Mir: Failed to connect to server socket: No such file or directory
Unable to init server: Could not connect: Connection refused

ERROR: The control display is undefined; please run `nvidia-settings --help`
       for usage information.

What type of OS is the client computer; and what are the IPs of the client and rig?

Windows 10 running putty, Rig: 192.168.1.19 Client: 192.168.1.6

So when you enter

m1@192.168.1.19   

using port 22

with SSH selected

and click open it gives you the error above?


Sorry let me try and clear some things up.  I am able to SSH into the rig just fine but when executing the miner via ssh, this is when i am getting the error, so clock settings, fan speed and power limits are not being set but the mining process still starts.

I also really appreciate your support and rapid response.

Are you killing the existing mining process before launching another via SSH?

Also; when you are launching the mining process via SSH are you using the cmd:

Code:
bash '/media/m1/1263-A96E/oneBash'

I am SSHing and executing the following commands.

ps aux | grep gnome-terminal - finding the gnome-server PID
kill PID
screen -S rig1
bash '/media/m1/1263-A96E/oneBash'

Just before the miner starts I see the above error in place of the "attribute" commands but the mining process still starts.


after you have SSHed in enter the cmd:

Code:
echo $DISPLAY

and tell me what it outputs

That does not output anything at all.

Ok, tomorrow I will try to replicate this error; and see if I can figure out what is happening.  My guess is X11 is having a problem trying to output graphically.

If you have a linux computer other than your rig and you SSH into the rig with it; does it have the same error?

I tested from a linux VM I have and the same result Sad

Confirming the issue, SSH into rig, kill gnome-terminal, bash '/media/m1/1263-A96E/oneBash'. Power limits are set, after that:
Code:
Failed to connect to Mir: Failed to connect to server socket: No such file or directory
Unable to init server: Could not connect: Connection refused

** EDIT** Have this with V0015 as well, but that image was updated with apt, unsure that has anything to do with it.
rampage999
Jr. Member
*
Offline Offline

Activity: 38


View Profile
June 22, 2017, 08:03:14 AM
 #1058

Hi,

Good work.

However it takes several minutes to boot from my USB 3.0 stick?
The booting seems have some problem.

V16 changed name to amdOC?
f00ch0w
Jr. Member
*
Offline Offline

Activity: 33


View Profile
June 22, 2017, 08:52:20 AM
 #1059

Any guide on how to update from v0015 to v0016? Or it's required to reimage again?
Maxximus007
Full Member
***
Offline Offline

Activity: 136


View Profile
June 22, 2017, 09:12:22 AM
 #1060

Any guide on how to update from v0015 to v0016? Or it's required to reimage again?
ATM the upgrade path is reimage indeed. -> Save your current oneBash: you have to mix and match with the V0016 oneBash to mine on your own accounts again, with desired OC, power and fan speeds.
Pages: « 1 ... 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 [53] 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 ... 287 »
  Print  
 
Jump to:  

Sponsored by , a Bitcoin-accepting VPN.
Powered by MySQL Powered by PHP Powered by SMF 1.1.19 | SMF © 2006-2009, Simple Machines Valid XHTML 1.0! Valid CSS!