Pages:
Author

Topic: BAMT - Easy persistent USB key based linux for dedicated miners/mining farms - page 8. (Read 167467 times)

sr. member
Activity: 369
Merit: 250
My rig has a hard time rebooting from the ssh command.  Most of the time it sits there until you manually reboot it yourself.  I'm getting some errors like "cant read os media" etc.  Is this a motherboard specific problem?

never type reboot

always type coldreboot

and above all.. leave that thing alone and let it mine

ok, I'll remember that.  Any specifics why 'reboot' doesn't work?
hero member
Activity: 616
Merit: 506
My rig has a hard time rebooting from the ssh command.  Most of the time it sits there until you manually reboot it yourself.  I'm getting some errors like "cant read os media" etc.  Is this a motherboard specific problem?

never type reboot

always type coldreboot

and above all.. leave that thing alone and let it mine
sr. member
Activity: 369
Merit: 250
My rig has a hard time rebooting from the ssh command.  Most of the time it sits there until you manually reboot it yourself.  I'm getting some errors like "cant read os media" etc.  Is this a motherboard specific problem?
hero member
Activity: 616
Merit: 506
I am almost there!! lol


why is this showing up??

http://108.50.132.175/mgpumon/





You need to set your miners to send mgpumon info to that address. In /etc/bamt/bamt.conf on both your miners, set "do_direct_status: " to the local (192.168.x.x) IP of the machine you set up port forwarding to.

this is usually not necessary.  only needed if the mgpumon is running on a different network than the miners, which i do not think is the case in this situation.
hero member
Activity: 616
Merit: 506
I am almost there!! lol


why is this showing up??

http://108.50.132.175/mgpumon/





There isn't anything I can tell you that isn't already contained within the message you are getting there.
hero member
Activity: 560
Merit: 500
Ad astra.
I am almost there!! lol


why is this showing up??

http://108.50.132.175/mgpumon/





You need to set your miners to send mgpumon info to that address. In /etc/bamt/bamt.conf on both your miners, set "do_direct_status: " to the local (192.168.x.x) IP of the machine you set up port forwarding to.
hero member
Activity: 910
Merit: 1000
Items flashing here available at btctrinkets.com
I have a problem I assume is caused by the latest update. Is there any way for me to do a rollback or do I need to do a complete "reinstall" ?

..and btw the problem is 70 or so missing mhs on one of my cards, 5870 @ core 1020Mhz, mem 300, and core voltage: 1.17 I used to get a solid 445Mhs out of this, now its pushig our 373Mhs.
To me it would seem like a problem applying the overclock.

My other 5870, that dosent clock asfar isnt showing any Mhs loss.


there is no "roll back".

it would help if you were specific about what is meant by the "latest" update.. do you mean  #31?  

no update has changed the over clocking code in some months, so this would be strange.

did you verify that the card is getting the correct settings using atitweak?

are you sure you didn't break something in the config file?


Aye, #31

I cant find atitweak (sorry, im clueless with Linux) .. however I found AMDOverdriveCtrl and it's telling me that Im running a core clock of 850mhz and stock voltage (1.125)

Im fairly sure nothing's broken in the config file (triple checking atm)

It sounds like BAMT has disabled overclocking on that GPU because it locked up.  You need to reduce your overclocking settings.
Then, remove the control file that is preventing overclocking.  It will be in /live/image/BAMT/CONTROL/ACTIVE

ps... to use atitweak just type:

atitweak





You were correct on BAMT disabling the Oc. I managed to delete the file and the card is running as usual.

I did type atitweak, didnt get a response, tried atitweak -help ... turns out -h is better Tongue

Will adding: "detect_defunct: 0" to the settings part of bamt.conf prevent this from heppening again ?
..also Im fairly sure me messing around on the running rig caused the crash, those clocks on that rig have had a record uptime of 8 days (and even then it was me that chose to reboot it)

Also thank you for your help.

turning off detect_defunct will prevent BAMT from recovering from a locked up GPU, yes.  It won't prevent the GPU from locking up, and without detection and recovery the machine will simply sit there locked up if it happens.  Your call.


8-day uptime says to "kill this feature with fire" imo, once again thank you for your time.
hero member
Activity: 616
Merit: 506
I have a problem I assume is caused by the latest update. Is there any way for me to do a rollback or do I need to do a complete "reinstall" ?

..and btw the problem is 70 or so missing mhs on one of my cards, 5870 @ core 1020Mhz, mem 300, and core voltage: 1.17 I used to get a solid 445Mhs out of this, now its pushig our 373Mhs.
To me it would seem like a problem applying the overclock.

My other 5870, that dosent clock asfar isnt showing any Mhs loss.


there is no "roll back".

it would help if you were specific about what is meant by the "latest" update.. do you mean  #31?  

no update has changed the over clocking code in some months, so this would be strange.

did you verify that the card is getting the correct settings using atitweak?

are you sure you didn't break something in the config file?


Aye, #31

I cant find atitweak (sorry, im clueless with Linux) .. however I found AMDOverdriveCtrl and it's telling me that Im running a core clock of 850mhz and stock voltage (1.125)

Im fairly sure nothing's broken in the config file (triple checking atm)

It sounds like BAMT has disabled overclocking on that GPU because it locked up.  You need to reduce your overclocking settings.
Then, remove the control file that is preventing overclocking.  It will be in /live/image/BAMT/CONTROL/ACTIVE

ps... to use atitweak just type:

atitweak





You were correct on BAMT disabling the Oc. I managed to delete the file and the card is running as usual.

I did type atitweak, didnt get a response, tried atitweak -help ... turns out -h is better Tongue

Will adding: "detect_defunct: 0" to the settings part of bamt.conf prevent this from heppening again ?
..also Im fairly sure me messing around on the running rig caused the crash, those clocks on that rig have had a record uptime of 8 days (and even then it was me that chose to reboot it)

Also thank you for your help.

turning off detect_defunct will prevent BAMT from recovering from a locked up GPU, yes.  It won't prevent the GPU from locking up, and without detection and recovery the machine will simply sit there locked up if it happens.  Your call.
hero member
Activity: 910
Merit: 1000
Items flashing here available at btctrinkets.com
I have a problem I assume is caused by the latest update. Is there any way for me to do a rollback or do I need to do a complete "reinstall" ?

..and btw the problem is 70 or so missing mhs on one of my cards, 5870 @ core 1020Mhz, mem 300, and core voltage: 1.17 I used to get a solid 445Mhs out of this, now its pushig our 373Mhs.
To me it would seem like a problem applying the overclock.

My other 5870, that dosent clock asfar isnt showing any Mhs loss.


there is no "roll back".

it would help if you were specific about what is meant by the "latest" update.. do you mean  #31?  

no update has changed the over clocking code in some months, so this would be strange.

did you verify that the card is getting the correct settings using atitweak?

are you sure you didn't break something in the config file?


Aye, #31

I cant find atitweak (sorry, im clueless with Linux) .. however I found AMDOverdriveCtrl and it's telling me that Im running a core clock of 850mhz and stock voltage (1.125)

Im fairly sure nothing's broken in the config file (triple checking atm)

It sounds like BAMT has disabled overclocking on that GPU because it locked up.  You need to reduce your overclocking settings.
Then, remove the control file that is preventing overclocking.  It will be in /live/image/BAMT/CONTROL/ACTIVE

ps... to use atitweak just type:

atitweak





You were correct on BAMT disabling the Oc. I managed to delete the file and the card is running as usual.

I did type atitweak, didnt get a response, tried atitweak -help ... turns out -h is better Tongue

Will adding: "detect_defunct: 0" to the settings part of bamt.conf prevent this from happening again ?
..also Im fairly sure me messing around on the running rig caused the crash, those clocks on that rig have had a record uptime of 8 days (and even then it was me that chose to reboot it)

Also thank you for your help.
hero member
Activity: 616
Merit: 506
I have a problem I assume is caused by the latest update. Is there any way for me to do a rollback or do I need to do a complete "reinstall" ?

..and btw the problem is 70 or so missing mhs on one of my cards, 5870 @ core 1020Mhz, mem 300, and core voltage: 1.17 I used to get a solid 445Mhs out of this, now its pushig our 373Mhs.
To me it would seem like a problem applying the overclock.

My other 5870, that dosent clock asfar isnt showing any Mhs loss.


there is no "roll back".

it would help if you were specific about what is meant by the "latest" update.. do you mean  #31?  

no update has changed the over clocking code in some months, so this would be strange.

did you verify that the card is getting the correct settings using atitweak?

are you sure you didn't break something in the config file?


Aye, #31

I cant find atitweak (sorry, im clueless with Linux) .. however I found AMDOverdriveCtrl and it's telling me that Im running a core clock of 850mhz and stock voltage (1.125)

Im fairly sure nothing's broken in the config file (triple checking atm)

It sounds like BAMT has disabled overclocking on that GPU because it locked up.  You need to reduce your overclocking settings.
Then, remove the control file that is preventing overclocking.  It will be in /live/image/BAMT/CONTROL/ACTIVE

ps... to use atitweak just type:

atitweak



full member
Activity: 164
Merit: 100
OK, I got it working 99%......

when I ssh to one box, I can see both rig on mgpumon.

However, through the web, when i put http://myip/mgpumon/  i get

Not Found

The requested URL /mgpumon/ was not found on this server.


when i use this,

http://myip/cgi-bin/status.pl

I get only 1 rig

hero member
Activity: 910
Merit: 1000
Items flashing here available at btctrinkets.com
I have a problem I assume is caused by the latest update. Is there any way for me to do a rollback or do I need to do a complete "reinstall" ?

..and btw the problem is 70 or so missing mhs on one of my cards, 5870 @ core 1020Mhz, mem 300, and core voltage: 1.17 I used to get a solid 445Mhs out of this, now its pushig our 373Mhs.
To me it would seem like a problem applying the overclock.

My other 5870, that dosent clock asfar isnt showing any Mhs loss.


there is no "roll back".

it would help if you were specific about what is meant by the "latest" update.. do you mean  #31?  

no update has changed the over clocking code in some months, so this would be strange.

did you verify that the card is getting the correct settings using atitweak?

are you sure you didn't break something in the config file?


Aye, #31

I cant find atitweak (sorry, im clueless with Linux) .. however I found AMDOverdriveCtrl and it's telling me that Im running a core clock of 850mhz and stock voltage (1.125)

Im fairly sure nothing's broken in the config file (triple checking atm)
hero member
Activity: 616
Merit: 506




Ok, I got it working on 1 of them

My farm has 2 rigs. Do i have to setup port forwarding to both? or just 1?

thanks


make sure both are set to broadcast status, then you only need to run mgpumon on either one and you'll see status of both.

note that all this business with port forwarding and networking might be better served in another thread.  it's nothing specific to BAMT, this is networking 101 stuff.


Sorry lodcrappo.....lol!!! I am not a networking guy.


it's no problem, just trying to keep things at least somewhat BAMT focused here.  the requirements of BAMT for remote access are exactly the same as any remote access to any machine running any OS or any software, this is why the questions are not BAMT specific.
hero member
Activity: 616
Merit: 506
I have a problem I assume is caused by the latest update. Is there any way for me to do a rollback or do I need to do a complete "reinstall" ?

..and btw the problem is 70 or so missing mhs on one of my cards, 5870 @ core 1020Mhz, mem 300, and core voltage: 1.17 I used to get a solid 445Mhs out of this, now its pushig our 373Mhs.
To me it would seem like a problem applying the overclock.

My other 5870, that dosent clock asfar isnt showing any Mhs loss.


there is no "roll back".

it would help if you were specific about what is meant by the "latest" update.. do you mean  #31? 

no update has changed the over clocking code in some months, so this would be strange.

did you verify that the card is getting the correct settings using atitweak?

are you sure you didn't break something in the config file?
full member
Activity: 164
Merit: 100




Ok, I got it working on 1 of them

My farm has 2 rigs. Do i have to setup port forwarding to both? or just 1?

thanks


make sure both are set to broadcast status, then you only need to run mgpumon on either one and you'll see status of both.

note that all this business with port forwarding and networking might be better served in another thread.  it's nothing specific to BAMT, this is networking 101 stuff.


Sorry lodcrappo.....lol!!! I am not a networking guy.
hero member
Activity: 910
Merit: 1000
Items flashing here available at btctrinkets.com
I have a problem I assume is caused by the latest update. Is there any way for me to do a rollback or do I need to do a complete "reinstall" ?

..and btw the problem is 70 or so missing mhs on one of my cards, 5870 @ core 1020Mhz, mem 300, and core voltage: 1.17 I used to get a solid 445Mhs out of this, now its pushig our 373Mhs.
To me it would seem like a problem applying the overclock.

My other 5870, that dosent clock asfar isnt showing any Mhs loss.
hero member
Activity: 616
Merit: 506




Ok, I got it working on 1 of them

My farm has 2 rigs. Do i have to setup port forwarding to both? or just 1?

thanks


make sure both are set to broadcast status, then you only need to run mgpumon on either one and you'll see status of both.

note that all this business with port forwarding and networking might be better served in another thread.  it's nothing specific to BAMT, this is networking 101 stuff.
hero member
Activity: 504
Merit: 500
you have to setup your router with 2 of them...port x goes to one machine, port y goes to thee other, like with the phone numbers you need to add one for room 410 and another for room 411...
hero member
Activity: 560
Merit: 500
Ad astra.
Splirow, it works like a hotel room...you dial 410 on the phone and get room 410. but if you dial 410 from home it will not work. so you need the hotels telephone number...when you call the telephone number you ask the guy at the front desk to connect you to room 410.

so there are 2 kinds of ip addresses, the ones in the building, that pc's use to talk to each other, and outside the building, just like the phone numbers.

so when you call the hotel, you must ask the guy to transfer you to the room, same wih the ip address...you must get the outside ip address to transfer you to the right pc.

in order to do this you need to forward a port to that pc, or to your main pc you use at work.

the pc at work should have a static ip address.

if you look at the site:

http://portforward.com/english/routers/port_forwarding/routerindex.htm find your router and follow the instructions...

if this is too complicated go to logmein.com register and choose the free option. not the trial, the free one...then add yur pc at work, and you can log into it from anywhere as long as you have an internet connection.


Ok, I got it working on 1 of them

My farm has 2 rigs. Do i have to setup port forwarding to both? or just 1?

thanks

Well, you can have your router forward another port and change the SSH port on the other rig. That requires editing some configuration files. Alternatively, SSH into your first rig and then SSH from that to the second one using the local (192.168.x.x) address.
full member
Activity: 164
Merit: 100
Splirow, it works like a hotel room...you dial 410 on the phone and get room 410. but if you dial 410 from home it will not work. so you need the hotels telephone number...when you call the telephone number you ask the guy at the front desk to connect you to room 410.

so there are 2 kinds of ip addresses, the ones in the building, that pc's use to talk to each other, and outside the building, just like the phone numbers.

so when you call the hotel, you must ask the guy to transfer you to the room, same wih the ip address...you must get the outside ip address to transfer you to the right pc.

in order to do this you need to forward a port to that pc, or to your main pc you use at work.

the pc at work should have a static ip address.

if you look at the site:

http://portforward.com/english/routers/port_forwarding/routerindex.htm find your router and follow the instructions...

if this is too complicated go to logmein.com register and choose the free option. not the trial, the free one...then add yur pc at work, and you can log into it from anywhere as long as you have an internet connection.


Ok, I got it working on 1 of them

My farm has 2 rigs. Do i have to setup port forwarding to both? or just 1?

thanks
Pages:
Jump to: