[Minera v0.9.1] Your next mining dashboard - Antminer/CPUminer/CGminer/BFGminer

[stefna]

Just if anyone has wondered: Minera works with up to 32 5-Chip Gridseeds on a raspberry pie.

I think that is the rasberry limit. I know that other distributions for rpi have a limit of sometimes 18 GS Devices. The only distribution I have come across so far that could run up to 32 devices is the zoomhash image without webinterface.

Great great work!!

Michellem: it would be great if there was a client/host version of minera that made it possible to run minera on a dedicated server to control a couple of Raspberries in a LAN network. What do you think? Do you like the idea?

Cheers, Stefna

[michelem]

http://pastebin.com/dUZ1CXB9

real    0m0.326s
user    0m0.010s
sys     0m0.040s

Please can you pastebin me also the /var/log/minera/log-2014-05-18.php file?

http://pastebin.com/kUdZmeUn

Something else of note, I just had to bounce cpuminer, it stopped sending shares entirely.

Well, if you cannot see the dashboard stats within this config I think the problem could be related on how much time needs Minera to get the whole stats output. Probably it timeouts before getting everything and so you see {"error": true}.

I can say this because trying to inject directly your stats output into Minera works very fine, so the problem should be on getting it from minerd when there are so large output.

But now debugging this it's quite hard for me, you could do this:

* find the file: application/model/util_model.php
* at the beginning of the file you find the "callMinerd" function
* change it with this:

Code:

	// Call minerd to get the stats and retry before give up
	public function callMinerd($i = 0)
	{
		if(!($fp = @fsockopen("127.0.0.1", 4028, $errno, $errstr, 3)))
		{
			return array("error" => true, "msg" => $errstr);
		}
	
		stream_set_blocking($fp, false);
		
		$out = json_encode(array("get" => "stats"))."\n";
		
		fwrite($fp, $out);
		
		usleep(250000);
		
		$out = false;
		
		while(!feof($fp))
		{
		    if(!($str = fread($fp, 4096))) break;
		    $out .= $str;
		}

		fclose($fp);
			
		return json_decode($out);
	}

* save the file and retry the dashboard/stats page

Since version 0.1.10a this function is in application/model/cpuminer_model.php

Version 0.1.10a is out:

* bug fixes
* latest cpuminer
* added extra rc.local commands option
* added JSON conf

So for those of us running multiple G-Blades should we add that code or is it already included in the 0.1.10a update?
What should we do next to at least get the Minera stats working?

Jess yes those are in 0.1.10a yet, but I think this is not solving stats problem with lot of Blades.

Can you try to disconnect some Blades and retry? I mean try running Minera with only 5 Blades instead 10 (or less), are the stats showing up?

If with few Blades all work fine so the problem is definitely that. Minera has problems getting large amount of data from the get_stats command.

Please if someone can confirm this could be very useful.
Thanks

[michelem]

To clarify Pools main/failover issues:

Minera checks against Pool for Dead or Alive are made only checking the URL/TCP connectivity FULL-STOP.

So while url/port are reachable Pool is Alive for Minera, it becomes Dead only if url/port are not reachable.

And this is how Minera shows you Dead/Alive, but I don't know how cpuminer do the same job, so a dead pool for Minera web interface could be alive for minerd and viceversa.

If you know how to check a pool more in depth please let me know.

[quakefiend420]

To clarify Pools main/failover issues:

Minera checks against Pool for Dead or Alive are made only checking the URL/TCP connectivity FULL-STOP.

So while url/port are reachable Pool is Alive for Minera, it becomes Dead only if url/port are not reachable.

And this is how Minera shows you Dead/Alive, but I don't know how cpuminer do the same job, so a dead pool for Minera web interface could be alive for minerd and viceversa.

If you know how to check a pool more in depth please let me know.

Any thoughts on the issue that I'm having with the BetaRigs pool?  Or do we need to ask Sandor on that one?

[michelem]

To clarify Pools main/failover issues:

Minera checks against Pool for Dead or Alive are made only checking the URL/TCP connectivity FULL-STOP.

So while url/port are reachable Pool is Alive for Minera, it becomes Dead only if url/port are not reachable.

And this is how Minera shows you Dead/Alive, but I don't know how cpuminer do the same job, so a dead pool for Minera web interface could be alive for minerd and viceversa.

If you know how to check a pool more in depth please let me know.

Any thoughts on the issue that I'm having with the BetaRigs pool?  Or do we need to ask Sandor on that one?

Yes, please for everybody, any cpuminer strictly related questions must be asked to Sandor and not to me.
I mean you can ask them to me but I probably don't know what to answer you because that's not my project, my project is Minera and Minera acts only as "frontend" for cpuminer.
Thanks

[n00bminer]

Great Work!!

Is the fix in so /etc/rc.local does not get wiped and reset on save?

[LexMine]

Just if anyone has wondered: Minera works with up to 32 5-Chip Gridseeds on a raspberry pie.

I think that is the rasberry limit. I know that other distributions for rpi have a limit of sometimes 18 GS Devices. The only distribution I have come across so far that could run up to 32 devices is the zoomhash image without webinterface.

Great great work!!

Michellem: it would be great if there was a client/host version of minera that made it possible to run minera on a dedicated server to control a couple of Raspberries in a LAN network. What do you think? Do you like the idea?

Cheers, Stefna

Good to know. I´m getting 100 Grids this week and just ordered some more Raspberries.

Is this a limitation of the Raspberry or general under Unix? Does a Unix PC handle more Devices?

Does someone know the max amount of grids under windows?

[sandor111]

Just if anyone has wondered: Minera works with up to 32 5-Chip Gridseeds on a raspberry pie.

I think that is the rasberry limit. I know that other distributions for rpi have a limit of sometimes 18 GS Devices. The only distribution I have come across so far that could run up to 32 devices is the zoomhash image without webinterface.

Great great work!!

Michellem: it would be great if there was a client/host version of minera that made it possible to run minera on a dedicated server to control a couple of Raspberries in a LAN network. What do you think? Do you like the idea?

Cheers, Stefna

Good to know. I´m getting 100 Grids this week and just ordered some more Raspberries.

Is this a limitation of the Raspberry or general under Unix? Does a Unix PC handle more Devices?

Does someone know the max amount of grids under windows?

It's a software limitation, can easily be bypassed by increasing http://lxr.free-electrons.com/source/drivers/usb/class/cdc-acm.h#L22
Someone will need to compile Raspbian though.

[LexMine]

Just if anyone has wondered: Minera works with up to 32 5-Chip Gridseeds on a raspberry pie.

I think that is the rasberry limit. I know that other distributions for rpi have a limit of sometimes 18 GS Devices. The only distribution I have come across so far that could run up to 32 devices is the zoomhash image without webinterface.

Great great work!!

Michellem: it would be great if there was a client/host version of minera that made it possible to run minera on a dedicated server to control a couple of Raspberries in a LAN network. What do you think? Do you like the idea?

Cheers, Stefna

Good to know. I´m getting 100 Grids this week and just ordered some more Raspberries.

Is this a limitation of the Raspberry or general under Unix? Does a Unix PC handle more Devices?

Does someone know the max amount of grids under windows?

It's a software limitation, can easily be bypassed by increasing http://lxr.free-electrons.com/source/drivers/usb/class/cdc-acm.h#L22
Someone will need to compile Raspbian though.

Maybe michelem will do this for the image file?

[sharteel]

Hi,

any scheduler in minera ? (same cgwatcher) ?


great job +++

[Flep182]

Just if anyone has wondered: Minera works with up to 32 5-Chip Gridseeds on a raspberry pie.

I think that is the rasberry limit. I know that other distributions for rpi have a limit of sometimes 18 GS Devices. The only distribution I have come across so far that could run up to 32 devices is the zoomhash image without webinterface.

Great great work!!

Michellem: it would be great if there was a client/host version of minera that made it possible to run minera on a dedicated server to control a couple of Raspberries in a LAN network. What do you think? Do you like the idea?

Cheers, Stefna

Good to know. I´m getting 100 Grids this week and just ordered some more Raspberries.

Is this a limitation of the Raspberry or general under Unix? Does a Unix PC handle more Devices?

Does someone know the max amount of grids under windows?

It's a software limitation, can easily be bypassed by increasing http://lxr.free-electrons.com/source/drivers/usb/class/cdc-acm.h#L22
Someone will need to compile Raspbian though.

Is there also some sort of similar way to prevent a rpi from only being able to connect two levels down USB hub wise? Because then I'm going to see whether I can compile some updated version

[edonkey]

i just wanted to say OMFG you rock, i tried every rasberry image and Minera is THE BEST.

one quick question, on your rasberry image, ( i know i'm a newb when it comes to linux ) how do i change the
time zone to show the correct time on my minera web page.

thanks

You'll have to SSH into the device, and run "sudo raspi-config". That'll give you some options that you can change. Under Internationalisation Options should be where the time zone option is.

i did that and the rasberry keeps the proper time zone (i'm in the EST but Minera still shows GMT zone, any clue. ?

Did you try my suggestion posted earlier?

Code:

sudo -i
dpkg-reconfigure tzdata

[toxic0n]

Anyone trying to mine at BetaRigs with this distro?  Keep getting this error:

stratum_recv_line failed

when trying either of these two pools:

stratum+tcp://r3026.g63.rigs.eu.betarigs.com:6192
http://eu.betarigs.com:3333

No shares attempted at all...

These work just fine in my cgminer instances, but this is the first time I've used cpuminer.

EDIT: I enabled debug mode, but it didn't give me much more info...

https://i.imgur.com/qOENVdR.png

Is your rig rented out when this happens? Mine will do that as well when the rig is not rented. Once someone rents it, the pool becomes alive and cpuminer switches over to it after a few minutes.

[quakefiend420]

Anyone trying to mine at BetaRigs with this distro?  Keep getting this error:

stratum_recv_line failed

when trying either of these two pools:

stratum+tcp://r3026.g63.rigs.eu.betarigs.com:6192
http://eu.betarigs.com:3333

No shares attempted at all...

These work just fine in my cgminer instances, but this is the first time I've used cpuminer.

EDIT: I enabled debug mode, but it didn't give me much more info...

https://i.imgur.com/qOENVdR.png

Is your rig rented out when this happens? Mine will do that as well when the rig is not rented. Once someone rents it, the pool becomes alive and cpuminer switches over to it after a few minutes.

Yes, it is.

EDIT: Would you mind posting your config file?  Maybe there's some small difference...

[fivejonnyfive]

Speaking of problems when renting hashpower - I'm noticing that If I set a minimal price via Nicehash when it switches back and forth it usually causes one of the gridseeds to stop hashing.

the way nicehash does minimal price is you specify it in the password field, so if the price is too low it comes back "authentication failed" and then moves to the next failover pool. We can all agree that isn't ideal, but it also shouldn't make a miner stop hashing. I think it has more to do with the failover implementation in sandor's miner so I'm going to try and gather more info and post over in that thread, I was just curious if anyone else had seen this.

[walther1337]

Speaking of problems when renting hashpower - I'm noticing that If I set a minimal price via Nicehash when it switches back and forth it usually causes one of the gridseeds to stop hashing.

the way nicehash does minimal price is you specify it in the password field, so if the price is too low it comes back "authentication failed" and then moves to the next failover pool. We can all agree that isn't ideal, but it also shouldn't make a miner stop hashing. I think it has more to do with the failover implementation in sandor's miner so I'm going to try and gather more info and post over in that thread, I was just curious if anyone else had seen this.

I'm curious how you calculate the minimum price.

[fivejonnyfive]

I'm curious how you calculate the minimum price.

Not very scientifically. A few notches above the trailing 24 hour averages of the higher of clevermining or wafflepool?

[LexMine]

I'm curious how you calculate the minimum price.

Not very scientifically. A few notches above the trailing 24 hour averages of the higher of clevermining or wafflepool?

So why should someone rent it then?

[li_gangyi]

Tried solo mining today. Can't connect to my wallet/server. Seems like its coded to use stratum and no way to run getwork.

[azzman372]

Been using Minera for a week now and I love the gui and graphs for stats.
One issue I'm encountering is the failoverpool display. I know this is a nonissue but it would be nice if I could get this troubleshooted.

My main pool is nicehash and I've set a minimum payment that is above the current profitability rate. Nicehash rejects my connection and I failover onto my backup. When I ssh in and look at the session I can see that I am connected to my backup pool however on the Minera page it is still saying that my main pool is alive and hashing away.

Thank you for your time and effort that you have put into Minera