Comkort are delisting LTB on Nov 30.

I'm using sgminer v5 with the ability to switch between multiple algorithms on the one rig.

I would like the ability to create three rig profiles on Betarigs, x11, x13 and x15 and advertise them all at once and at different rates.

As soon as one of them is hired the others could easily be flicked to "offline" or "rented".

Is this possible at the moment?

E.g. my config file would contain three betarigs pools, one for each algorithm.  Once my rig is hired only one pool should be alive.

At the moment if a second of the above profiles is hired then they would be enabled in sgminer based on priority which is a problem unless there is some way I can deactivate the other profiles once the rig is hired.

Thanks,

Here's how I confirmed it.

Get Catalyst (fglrx) version number:

dmesg | grep fglrx | grep module

outputs:
      fglrx: module license 'Proprietary. (C) 2002 - ATI Technologies, Starnberg, GERMANY' taints kernel. 
      fglrx] module loaded - fglrx 14.20.7 [May 22 2014] with 4 minors                                   
      
Date corresponds with Catalyst 14.6 beta linux version anyway.
      
Googling "catalyst fglrx 14.20.7" confirms it.

Just installed Pimp 1.5.51 AMD 14.6, the beta version zip file on the downloads page.

However, according to the terminal it's not running 14.6 drivers.

Terminal displays: PiMP Version:    pimp1.5.51-amd-14.4

Is there a way to check which drivers are actually loaded?

+1 for sgminer 5 multi-algorithm support.  I'd love to have my rig available to rent across multiple algorithms simultaneously.

From the documentation:

## Pool-specific configuration

If you use any of these options for a pool, then **you must** set that option
for every pool. This is necessary due to current poor implementation of
sgminer config parser.

Options that can be configured have a `pool-` prefix and work the same as the
global settings:

* pool-algorithm
* pool-nfactor
* pool-intensity
* pool-xintensity
* pool-rawintensity
* pool-gpu-engine
* pool-gpu-memclock
* pool-gpu-threads
* pool-thread-concurrency
* pool-gpu-fan

Ah balls.  Just tested this by renting my own rig.  I can add up to 8 pools as a customer once the rig is rented but as soon as I add the third pool I get locked out of the admin screen with the miner not responding error.  This is useless.  I can't rent out the rig like this.

I can manually edit it to fix things but that's no use when a customer is adding pools.  For the record, the syntax error does not occur when using cgminer.

The customer can only add a max of two pools by the looks of things so if I'm careful not to add any extra pools then the customer should never be locked out (hopefully) which was my main concern.

How is nobody else having this problem? 
Am I doing something different?

I've also noticed "Add Pool" in LeaseRig creates a syntax error which means sgminer subsequently can't even load the conf file!!?

It puts a comma at the end of the pool password line which throws an error:
"pass" : "x",

Still can't get LeaseRig back up and running smoothly.

I've installed BAMT and PIMP but am getting the same problem.

Everything works fine with one or two pools.  If I add a third pool using the LeaseRig admin panel I get a miner not responding error.

Does anyone have a fix for this?

The pools link in the admin panel gives this error:
Error: Mining rig is not responding!

Maybe LeaseRig is saving to a different conf file or can't access my conf file?!

I'm coming back to LeaseRig after a break but I can't get it running this time.

Rigtest.jsp passes for "summary" and "addpool" but fails at the "pools" command.

I can see in sgminer that the pool has been added but the configuration file /root/.sgminer/sgminer.conf has not been modified.

This is the file I specify as the --config parameter.

Any ideas?

Thank you.  That's perfect.

So, in theory, if my pool is running properly my accepted shares on NiceHash should equal my accepted shares on the pool otherwise I should switch pools right?

Are you guys seeing much variance in your testing?

Hi, can you please explain the fees?  They're not 2%.

I've tested the service with two orders 0.01 BTC each:

2014-04-17 14:56:24   Comment: Fee order #677   -0.00069000   
2014-04-17 14:56:23   Comment: Payment order #677   -0.00931000   

2014-04-16 11:17:42   Comment: One time fee order #327   -0.00010000   
2014-04-16 11:17:42   Comment: Payment order #327   -0.00990000   

How can someone looking to pay more than the current rate for hash power see the total available?

E.g. I want to see all available hash power including that which will kick in at a higher rental rate due to some rig owners waiting for the rate to rise above their minimum setting.

Great work.  Thanks.

Small bug in /opt/pimp/viewgpu the temp iterator should be k, not i.

Bump.

Has nobody had similar problems?

Cryptorush.in LTB is down for maintenance.  Glad I got mine out.

Hi,

what's the best way to start LeaseRigProxy after a reboot?

I added the following line to /etc/crontab:
            
@reboot         root    /root/LeaseRigProxy/leaserigproxy_bamt 3333 ***LRP ID*** 0 >> /home/user/LeaseRigProxy.log

However, last night LeaseRigProxy got stuck in a loop and I ended up with a 450Mb text file log filled with this:

WARNING: Hostname ***pool***.com resolved to zero!
--127.0.0.1:60171->connecting to pool ***pool***.com
--127.0.0.1:60172->rig connected
WARNING: Hostname gbs.desync0.com resolved to zero!
--127.0.0.1:60172->connecting to pool ***pool***.com
--127.0.0.1:60173->rig connected
WARNING: Hostname gbs.desync0.com resolved to zero!
--127.0.0.1:60173->connecting to pool ***pool***.com
--127.0.0.1:60174->rig connected
WARNING: Hostname gbs.desync0.com resolved to zero!

cgminer was running without any errors just no hashing. 

It didn't even go to the secondary pool.  Seems a bit flaky.

The pool works fine when I add it directly to cgminer.conf and bypass the proxy.

I'm running Litecoin BAMT 1.4.

Thanks.

Hi,

a few of times now I've woken up to a dead system with disk errors.  I click F to fix the errors and everything is fine.  I guess the last time the errors were too much as this didn't work.  Manually running fsck used up all memory and crashed.  The same thing happened when I ran a live cd and attempted to fix the disk.  So I re-installed the OS

I'm wondering if there is a more graceful way to shut down the system and avoid this or are we using best practices already?

Ok thanks.  This also messes up my rig stats.  It would be nice to auto revert to my pool.