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

[soulreafer]

does the autotune function also decrease the frequncy if the unit makes to much hw errors?

[1715200909]

1715200909

[1715200909]

1715200909

[1715200909]

1715200909

[ADutchGamer]

does the autotune function also decrease the frequncy if the unit makes to much hw errors?

Yes it does

[n00bminer]

I have only "/etc/php5/cli/conf.d/apc.ini" but none "/etc/php5/cgi/conf.d/apc.ini" path on my RPi

Same here.

Sorry guys, I left off installing fast cgi, which installs the file we modify: sudo apt-get install php5-cgi

[just some dude]

Feature request:

Federated UI

It would be nice to have a single window to manage multiple Minera instances running across several Raspis.  Switching pools, restarting miners, etc. would be good to see from one page.

[kaltar]

I tried using Minera with www.miningrigrentals.com
doesn't matter what i do, it just won't connect,

if i put the address in first, it's replaces it by a number like 6da23413, then spits out invalid url.
if i put it as a fail over and choses it after, on the top line ( putty and screen-r ) where it says connected, it will put a bunch of ascii charater and fail to connect.

and even more weird.

if i leave in in the pool selection somewhere
have betarigs in primary
have another pool like Coinmine.pw in failover

it will skip the betarigs since no rental is active
connect to coinmine and won't submit anyshares.

but if i remove miningrigs rental from the list and reboot
all is fine. just in case you want to reproduce this or investigate, the mininrigrental server i;m using is: us-east01.miningrigrentals.com

good luck   

[vm_mpn]

So, nobody else is having issues with pool info getting corrupt? I even went directly to command line and assigned just one pool to see if it gets messed up.... It does within an hour.





Do not know what to do here... Seems like this problem is getting worse... Could this be related to minis and blade hooked up to single Ras.pi?

[michelem]

So, nobody else is having issues with pool info getting corrupt? I even went directly to command line and assigned just one pool to see if it gets messed up.... It does within an hour.





Do not know what to do here... Seems like this problem is getting worse... Could this be related to minis and blade hooked up to single Ras.pi?

This is really weird, another guy has the same.

Can I ask you more info about this or how can I reproduce it? Please explain, you set up some pools, and without touching anything you get those messed up after an hour? What kind of devices you have connected?

Can you run this in the terminal before and after that happens:

Code:

redis-cli get minerd_pools

Thanks

[vm_mpn]

Thank You for reply Michele... Let me see if I can retrace my steps.

 I have downloaded and burned Minera Rasp.pi SD card image from your GitHub page which was version 1.1 last Friday if I remember correctly. Hooked up 10 GridSeed Minis (50 chips) and 1 Gridseed Blade (80 chips) to my single Rasp.Pi. Configured basic pool and credentials info, etc. I remember that it started to mine Ok but one my blades (1 half) weren't fully recognized (again, do not quote me on this) so I went ahead and updated to Minera Ver.2 which I was planning to do anyways from the help menu... Rebooted everything and all my units got recognized and worked beautifully for the next 20 - 22 hours (NOTE: I have not turned ON auto tune feature yet) .

Later, I have noticed that I have 0 hashrate at my pool account page, went to look at Minera dashboard and confirmed my dashboard primary pool info shows no info at all and miner switched to secondary (failover) pool. I have looked at configuration page and everything looked just fine there - all pools have been configured and displayed correctly exactly as I have left them. I did not pay much attention thinking it's just a fluke, restarted miner (under Minera menu)  and everything went back to normal. Later on I have enabled auto-tune feature, rearranged my failover pools and I think locked half of my blade to lower frequency due to a high amount of HW. Everything was working fine for another 12 hours or so and then again I have noticed 0 hashrate at my pool account page... This is where I have realized something was wrong and that the same pool information corruption has happened again... At this point I was kind of throwing everything I could at it - deleted pools and re-entered them, set both primary and secondary pools to the same address thinking that may be only first pool info gets corrupted and others will hold... To make long story short (lol) eventually my second pool information has disappeared and Minerera switched to a third one, etc. Finally, I have tried to avoid Json pool config and entered single pool info directly into minerd config, as you know with no success.

By the way - is SSH logon uses password minera? I could not logon on my Mac.

[michelem]

Thank You for reply Michele... Let me see if I can retrace my steps.

 I have downloaded and burned Minera Rasp.pi SD card image from your GitHub page which was version 1.1 last Friday if I remember correctly. Hooked up 10 GridSeed Minis (50 chips) and 1 Gridseed Blade (80 chips) to my single Rasp.Pi. Configured basic pool and credentials info, etc. I remember that it started to mine Ok but one my blades (1 half) weren't fully recognized (again, do not quote me on this) so I went ahead and updated to Minera Ver.2 which I was planning to do anyways from the help menu... Rebooted everything and all my units got recognized and worked beautifully for the next 20 - 22 hours (NOTE: I have not turned ON auto tune feature yet) .

Later, I have noticed that I have 0 hashrate at my pool account page, went to look at Minera dashboard and confirmed my dashboard primary pool info shows no info at all and miner switched to secondary (failover) pool. I have looked at configuration page and everything looked just fine there - all pools have been configured and displayed correctly exactly as I have left them. I did not pay much attention thinking it's just a fluke, restarted miner (under Minera menu)  and everything went back to normal. Later on I have enabled auto-tune feature, rearranged my failover pools and I think locked half of my blade to lower frequency due to a high amount of HW. Everything was working fine for another 12 hours or so and then again I have noticed 0 hashrate at my pool account page... This is where I have realized something was wrong and that the same pool information corruption has happened again... At this point I was kind of throwing everything I could at it - deleted pools and re-entered them, set both primary and secondary pools to the same address thinking that may be only first pool info gets corrupted and others will hold... To make long story short (lol) eventually my second pool information has disappeared and Minerera switched to a third one, etc. Finally, I have tried to avoid Json pool config and entered single pool info directly into minerd config, as you know with no success.

By the way - is SSH logon uses password minera? I could not logon on my Mac.

mmm, first of all I would like to suggest you to reinsert every your pools in the pool settings instead via command line.

Then please add this to "Extra options" field:

--retries=1 --gc3355-timeout=600

then try without any autotuning.

(to ssh into Minera from your Mac, open terminal and use user "minera" with pass "minera", so something like this: ssh minera@<ip>)

[vm_mpn]

As you have suggested I have done complete Minera reconfig this time using my PC with Firefox (I know it's a long shot) instead of MacBook Pro / Safari combo. SSH command shows correct pool info as my dashboard so far... I will attach before and after output if Minera pool info gets messed up again. Running Ok for over an hour now.

Thank You for All Your Help!

Vadim

[fivejonnyfive]

As you have suggested I have done complete Minera reconfig this time using my PC with Firefox (I know it's a long shot) instead of MacBook Pro / Safari combo. SSH command shows correct pool info as my dashboard so far... I will attach before and after output if Minera pool info gets messed up again. Running Ok for over an hour now.

Thank You for All Your Help!

Vadim

If it keeps acting up you might try a different SD card.
Some sd cards can be weird like this but for me a sandisk is rock solid

[vm_mpn]

No luck... Took close to 8 hours but pool info has disappeared again... SSH output below, and I do not see any difference between before (top command) and after (bottom one). Settings page still maintains the same info and miner went back up after restart with no issues.



Removed auto-tune and added line you have suggested --retries=1 --gc3355-timeout=600

Thank you for SD card suggestion J5 - it did not have any issues with other mining software so I did not take that route but I may get desperate enough to try and re-download / re-burn with new Minera image.

[edonkey]

The Minera/cpuminer combo is really amazing. I moved my main rig to it this weekend. Excellent work!

One thing I noticed is that if you have logging on and you stop and start the miner, the log file gets overwritten. I had logging on in debug mode with auto tune on so that I could tell when auto tune was done. Once I harvested out the per chip frequencies and added the command line, I disabled debugging and restarted the miner. It zapped my existing log, which was a bit disappointing.

Maybe this is a cpuminer issue and not Minera? I guess what I'm asking for is that cpuminer append to the specified log file instead of replacing it. If this is a cpuminer issue, then I can post on the corresponding support thread.

Also, it looks like log rotation is enabled with Minera, but I have yet to see the cpuminer.log get compressed and rotated. I think it's been 24 hours since I set up the rig, so I would have expected the log to rotate by now.

My only concern is that I'm using ramlog to minimize thrashing the flash card and as a result I have a relatively small (50 meg) mount point for /var/log. I find mining logs very useful, but don't want to exceed the space that I have.

Note that I installed Minera from the instructions to my existing Pi. I'm not using the Minera Pi image, in case it's relevant. Maybe something is missing in the manual install instructions for log rotation?

Thanks in advance for suggestions.

[MobileWill]

Hi all,

So far Minera is has been the best RPi mining manager! Love it. I actually don't have to worry about checking for dead miners every 5min. One thing I noticed that maybe someone else has experienced.

I have one pool blackcoinpool.com that the way Minera detects if a pool is alive doesn't work with this pool. So sometimes it shows alive for a min and I can add it. But then quickly shows dead. But i realized I can just select the pool and it will work anyway even though it shows as dead.

I can send log files or w/e you need if that helps.


Thanks!


Edit:

Forgot to ask if Minera(CPUMiner) has plans or will support new miners such as the Fury's  (aka Zeus Miners) from GAWMiners?

[tp30st]

A great user in our pool [ poochiboy ] offered these tips to enhance the speed of minera lighttp refreshes:
Most of the lighthttpd optimization pulled from here:http://c-mobberley.com/wordpress/2013/05/18/raspberry-pi-lighttpd-optimisation-for-wordpress-owncloud-and-more/

Install Fast CGI

Code:

sudo apt-get install php5-cgi
sudo nano /etc/php5/cgi/php.ini

Uncomment the line where it says:

Code:

;cgi.fix_pathinfo=1

to show

Code:

cgi.fix_pathinfo=1

Install PHP APC

Code:

sudo apt-get install libpcre3-dev php-pear php5-dev build-essential
sudo pecl install apc
sudo nano /etc/php5/cgi/conf.d/apc.ini

Add the following lines to the file.

Code:

extension=apc.so
apc.enabled=1
apc.shm_size=30

Now restart lighttpd

Code:

sudo service lighttpd restart

These optimizations sped up my refreshes like night and day.  

EDIT: forgot the fast CGI install, many applogies: sudo apt-get install php5-cgi

I have only "/etc/php5/cli/conf.d/apc.ini" but none "/etc/php5/cgi/conf.d/apc.ini" path on my RPi

Same here.

Sorry guys, I left off installing fast cgi, which installs the file we modify: sudo apt-get install php5-cgi

Excellent. I did it, I SSH in, really not sure If I noticed a decrease since I am using a class10 sd card. It might be a little bit faster but for me not much? I am truly not sure.

I did, do a back of my img afterwards to keep the settings if I need to reimage the sd card. Glad I got a 1TB hdd, it is very handy. 15 gig images, the size of my sd card.

I think this update/install/compile process took roughly around 10 minutes or so. It had to compile a few things.

Everything went smoothly. Thanks!

I'm getting a "503- Service Not Available" error after following these steps. Any suggestions?

[n00bminer]

I'm getting a "503- Service Not Available" error after following these steps. Any suggestions?

Try doing a

Code:

sudo service lighttpd restart

503 is probably the lighttpd not running.

[tp30st]

I'm getting a "503- Service Not Available" error after following these steps. Any suggestions?

Try doing a

Code:

sudo service lighttpd restart

503 is probably the lighttpd not running.

I did a restart of lighttpd and followed the guide exactly. I do have a backup image but would like to fix the web interface without downtime.

[emanzer0]

Would really like getwork so wallet mining is possible. Love the os and miner btw. as a temp solution is it possible to install bfgminer or cgminer to mine wallet? I'm not sure if they support a mix of 5 chip seeds and a blade? Is there a way I can setup a stratum to getwork proxy in minera or a getwork to stratum proxy in windows pointing my wallet so I can mine my wallet?

Thanks!

[Lucko]

For some reason I see disconnects from my pool all the time. I have only one in(scryptguild) and I see dead a lot of time...

[n00bminer]

I'm getting a "503- Service Not Available" error after following these steps. Any suggestions?

Try doing a

Code:

sudo service lighttpd restart

503 is probably the lighttpd not running.

I did a restart of lighttpd and followed the guide exactly. I do have a backup image but would like to fix the web interface without downtime.

When you restart do you see:
[ ok ] Stopping web server: lighttpd.
[ ok ] Starting web server: lighttpd.

?