Smartcoin Linux mining administration. [MULTI-MACHINE SUPPORT NOW IN!]

[pdawg]

yep, started yesterday with BTCGuild and the pool was shutdown completely.  Running my alternate profile with multiple pools.  Can't wait for failover to be implemented.

[1714021469]

1714021469

[1714021469]

1714021469

[1714021469]

1714021469

[mono]

Is it possible to mine BTCGuild in first priority (with full power) and when they go down (like the last days because of DDOS) smartcoin switches completly to a backup like deepbit? If this is possible, where can i find this?

[Fletch]

I have two feature requests. Not sure if they have been requested already.

1. Built-in overclocking using aticonfig. Could just be two added parameters for the device, one for memory clock and one for core.
2. Automated fan control. In other words, if temp is over 50 degrees, set fan level 30. If temp goes over 60 degrees, set fan level 40 etc.

Is it possible to mine BTCGuild in first priority (with full power) and when they go down (like the last days because of DDOS) smartcoin switches completly to a backup like deepbit? If this is possible, where can i find this?

This is an upcoming feature.

[jondecker76]

Kennel

I do plan on adding some simple aticonfig stuff like you are suggesting down the line.

I roughed out plans for the failover system last night, and think I have an idea of a reasonable way to do it.
Failovers will work with manual profiles only.  There will be a new "Configure Failover?" menu option. Under this menu option, you will be able to set a failover order of your manual profiles. Lastly, to run in failover mode, there will be a new dynamic profile that you can select from the Choose Profile menu named "Failover".
I should have the failover system ready in the next few days

[Fletch]

Kennel

I do plan on adding some simple aticonfig stuff like you are suggesting down the line.

I roughed out plans for the failover system last night, and think I have an idea of a reasonable way to do it.
Failovers will work with manual profiles only.  There will be a new "Configure Failover?" menu option. Under this menu option, you will be able to set a failover order of your manual profiles. Lastly, to run in failover mode, there will be a new dynamic profile that you can select from the Choose Profile menu named "Failover".
I should have the failover system ready in the next few days

I'm loving everything you just said. As soon as I have this up and running, I'm donating.

[elrock]

Just wanted to say that this is a great tool for remote administration.  Like others, I eagerly await implementation of the failover option, especially now that many of the large pools are experiencing connection problems.

[plantucha]

I have two feature requests. Not sure if they have been requested already.

1. Built-in overclocking using aticonfig. Could just be two added parameters for the device, one for memory clock and one for core.
2. Automated fan control. In other words, if temp is over 50 degrees, set fan level 30. If temp goes over 60 degrees, set fan level 40 etc.

i don't think so...
100% fan will kick in 150 °C ?
by then is your PC toast........
...........
Fans have to be set manually   @ desired noise level ( in my case 100%, other people can handle 50%only)
and overclock will be based on temperature.

30°C overclock up until temperature is 70°C then step down if 68°C then step up,
so working temperature is always around 70°C (or any other max)
=> if is outside PC too hot and cooling is not enough, GPUs auto underclock and temp will be still 70°C
if is outside PC colder card will be overclocking.
if is card more then 85°C = fan is broken = PC auto shut down

[ptshamrock]

following problem !


while installing it..

autodetecting local installed miners ..

./smartcoin_install.sh:line 167: locate: command not found

which miner listed above do u want to be the default miner?


?what to type??

invalid seleciotn...


all under newest linuxcoin

[Fletch]

i don't think so...
100% fan will kick in 150 °C ?
by then is your PC toast........

Huh? What I said was just an example. It doesn't have to be linear. I could continue it with: If temp goes over 70 degrees, set fan level 80 etc.

My fans are currently at 55% with GPU temps slightly under 70 degrees. However, I have no control over the ambient temperature in the room where the mining rig is placed, so I would love to have the security of an application that can boost the fan speeds should the temps get out of control.

PS. Fans are much more likely to fail while running at 100% than GPUs running at, say, 90 degrees.

[jondecker76]

ollowing problem !


while installing it..

autodetecting local installed miners ..

./smartcoin_install.sh:line 167: locate: command not found

which miner listed above do u want to be the default miner?

What distro are you running?  I've personally never seen one that didn't include the locate command...
You should try installing 'findutils' and re-running the installer.
debian/ubuntu:

Code:

sudo apt-get install findutils

fedora:

Code:

sudo yum findutils

What I may end up doing in an upcoming update, is asking first thing if you want to use autodetection. If you choose not to use it, then a minimal install is done instead and then you can set things up manually on the control screen.

[ptshamrock]

hmmm..

have run  it and it said newest verstion installed..then tried again..but same error

linuxcoin newest version with persistence

[Jen4538]

i will have to do alot of reading before i understand how to use this application ,  for now i have to do as i always have.   it looks like a very nice application however understanding how to set up a miner is beyond me .   


Jen

[CydeWeys]

hmmm..

have run  it and it said newest verstion installed..then tried again..but same error

linuxcoin newest version with persistence

You want to install the "locate" package.  Worked for me in Linuxcoin.

[jondecker76]

ptshamrock:

For now, you can comment-put line 167.

Code:

#phoenixMiner=`locate phoenix.py | grep -vi svn` 

Then after install, you will have to:
1) In the controll screen, go to configure miners->Add and add an instance of  phoenix.  Make sure your path is good, and use a launch string like:

Code:

python <#path#>phoenix.py -v -u http://<#user#>:<#pass#>@<#server#>:<#port#>/ -k phatk device=<#device#> worksize=128 vectors aggression=11 bfi_int fastloop=false

2) In the control screen, go to Edit Settings, and change the 'Phoenix installation location' to the path of your phoenix location. I,e,

Code:

/home/your_user/phoenix/

That should get you up and running until I make some changes to the installer.



Jen4538:
Just give it a try, its actually not too hard!



Everyone else:

Just thought I would make you all jealous by telling you that the new Failover system is soooo fun! I've been testing it by manually setting a new flag in the database, and its working great!
To finish it off, I need to:
A) Make a new function which sets the flag  when a profile fails/returns to normal
B) Make a  new control screen option which allows you to set failover order of profiles. (it will order by 1) Priority and then 2) order in which the profile was created.

[jondecker76]

cydeWeys:

You want to install the "locate" package.  Worked for me in Linuxcoin.

Thank you very much for this bit  of information. I will pass this on to the LinuxCoin author so that installs go smoother!

[Jen4538]

basicly i tried to create a working but list of pools i am not in then i didnt understand how to exit this part of and go back to main menu to try something else.  that then created a miner which i couldnt get out of either so more frustration to me than i can stand right now.  i know very little linux and this only made it harder to me to understand.   so i go back to being dumb and paste and copy miner login info to have them working

Jen

[plantucha]

basicly i tried to create a working but list of pools i am not in then i didnt understand how to exit this part of and go back to main menu to try something else.  that then created a miner which i couldnt get out of either so more frustration to me than i can stand right now.  i know very little linux and this only made it harder to me to understand.   so i go back to being dumb and paste and copy miner login info to have them working

Jen

with fresh installation

you do need create workers only.

most used pools are already pre-filed, no need to touch it unless you are using some tiny sketchy one

[Jen4538]

basicly i tried to create a working but list of pools i am not in then i didnt understand how to exit this part of and go back to main menu to try something else.  that then created a miner which i couldnt get out of either so more frustration to me than i can stand right now.  i know very little linux and this only made it harder to me to understand.   so i go back to being dumb and paste and copy miner login info to have them working

Jen

with fresh installation

you do need create workers only.

most used pools are already pre-filed, no need to touch it unless you are using some tiny sketchy one

so then your saying any pool that isnt on the list is sketchy?  i not sure i understand why this is


Jen

[plantucha]

basicly i tried to create a working but list of pools i am not in then i didnt understand how to exit this part of and go back to main menu to try something else.  that then created a miner which i couldnt get out of either so more frustration to me than i can stand right now.  i know very little linux and this only made it harder to me to understand.   so i go back to being dumb and paste and copy miner login info to have them working

Jen

with fresh installation
you do need create workers only.
most used pools are already pre-filed, no need to touch it unless you are using some tiny sketchy one

so then your saying any pool that isnt on the list is sketchy?  i not sure i understand why this is

Jen

well...yes
until pool is not established and trusted by many people......it is sketchy
Pool owner can claim any time block for himself/herself.

Most known and trusted are already in list.

[jondecker76]

Code:

smartcoin Management System r344 Thu Jul 7 00:36:39 EDT 2011
--------------------------------------------------------------------------------
Host: localhost
GPU[0]: Temp: 71.50 load: 99%
GPU[1]: Temp: 64.00 load: 99%
GPU[2]: Temp: 48.00 load: 99%
CPU Load : 7.40%

Profile: Failover
--------BTCGuild--------
GPU[0]: <<<DOWN>>>
GPU[1]: <<<DOWN>>>
GPU[2]: <<<DOWN>>>
Total : [0 hash/sec] [0 Accepted] [0 Rejected]
--------DeepBit--------
GPU[0]: [358.51 Mhash/sec] [1 Accepted] [0 Rejected] [RPC (+LP)]
GPU[1]: [359.99 Mhash/sec] [3 Accepted] [0 Rejected] [RPC (+LP)]
GPU[2]: [359.30 Mhash/sec] [1 Accepted] [0 Rejected] [RPC (+LP)]
Total : [1077.80 MHash/sec] [5 Accepted] [0 Rejected]

Grand Total: [1077.80 Mhash/sec] [5 Accepted] [0 Rejected] [0% Rejected]

The new Failover system is testing tonight.  Should be good timing, as I expect BTCGuild to come online tonight.
If you look above, BTCGuild is <<<DOWN>>>, so it failed over to DeepBit.  When BTCGuild comes back online, the DeepBit workers will fall back off of the list.  All tests so far show it working very good!  Failure detection threshold is currently hard-coded for about 1 minute (a profile has to be showing as DOWN for 1 minute), after which the next available profile in the failover list is tacked on to the end.  

As a preview, setting up the FailOver system goes like this:
1) Create some profiles.  
2) Chose the "Failover" profile from the control screen. Failover will happen by default in the order the profiles were created.
3) If you wish to change the failover order, use the new control tab option 12) Set Failover Order.  Changes take effect immediately with no restarts required.