On the control screen, go to option 4) Edit Settings
Double check the settings for:
AMD/ATI SDK installation location
Phoenix installation location
for example, my AMD/ATI SDK location looks like:
/home/jondecker76/AMD-APP-SDK-v2.4-lnx32/lib/x86
and my phoenix installation location looks like:
/home/jondecker76/phoenix/
Make sure yours look correct
Also, if you could get me the exact error message that keeps flashing, that would be helpful!
Thanks, this prompted me to jump from SDK 2.1 to 2.4 just in case and it seems to have cleared everything up.. I can't say exactly what caused it at this point but it's good now. |
|
|
|
Ok, the launch strung looks correct
What does your status screen show? Is it making all workers as down?
Yes, every few seconds it refreshes with some file not found errors, looks like it's trying to hit the screen sessions. I can manually run the miner with the same string copied and pasted without issues (tokens replaced with test values) but it doesn't want to run them with smartcoin, Ubuntu 11.04 64 bit. Added a bit of logging to try and get a feel for what's happening, only first miner attempts to even launch, and it doesn't look to make it.. 07/02/11 15:24:58 Starting main smartcoin screen session...
07/02/11 15:24:58 Creating tab for each machine...
07/02/11 15:24:58 localhost
07/02/11 15:24:58 Starting status monitor for machine 1
07/02/11 15:24:58 NEW PROFILE DETECTED!
07/02/11 15:24:58 Switching from profile: to profile: 2
07/02/11 15:24:58 Killing Miners....
07/02/11 15:25:00 Starting miner Miner.9!
07/02/11 15:25:00 Command: /home/braymond/smartcoin/smartcoin_launcher.sh
1 1 1 1
07/02/11 15:25:00 Miner Path: /home/braymond/mining/phoenix/
07/02/11 15:25:00 Launching miner with launch string: python /home/braymon
d/mining/phoenix/phoenix.py -v -u http://vUUHwB:PxtQKk@bitcoins.lc:8080/ -k phat
k device=0 worksize=256 vectors aggression=11 bfi_int fastloop=false
07/02/11 15:25:02 Starting miner Miner.10!
07/02/11 15:25:02 Command: /home/braymond/smartcoin/smartcoin_launcher.sh
1 2 1 2
07/02/11 15:25:02 Starting miner Miner.11!
07/02/11 15:25:02 Command: /home/braymond/smartcoin/smartcoin_launcher.sh
1 3 1 3
07/02/11 15:25:02 Starting miner Miner.12!
07/02/11 15:25:02 Command: /home/braymond/smartcoin/smartcoin_launcher.sh
1 4 1 4
07/02/11 15:25:02 Starting miner Miner.13!
07/02/11 15:25:02 Command: /home/braymond/smartcoin/smartcoin_launcher.sh
1 5 1 5
07/02/11 15:25:02 Starting miner Miner.14!
07/02/11 15:25:02 Command: /home/braymond/smartcoin/smartcoin_launcher.sh
1 6 1 6
07/02/11 15:25:02 Starting miner Miner.15!
07/02/11 15:25:02 Command: /home/braymond/smartcoin/smartcoin_launcher.sh
1 8 1 8
07/02/11 15:25:16 Exit option selected
07/02/11 15:25:16 Killing Miners.... |
|
|
|
Padrino- what revision or smartcoin are you on?
depending on revision, you should find a log in either ~/smartcoin.log or ~/smartcoin/smartcoin.log
Almost every time i've seen miners not launching correctly, I have found that the launch string in the database is incorrect.
Can you go to Configure Devices -> Edit -> Localhost -> Phoenix and tell me what is entered for the launch string? (also, make sure its listed as the default miner)
Also, I would still suggest deleting all of your duplicate worker entries, and using the auto profile - it will have the same effect as your manual profile, without having to have 8 workers set up!
The latest r290 python <#path#>phoenix.py -v -u http://<#user#>:<#pass#>@<#server#>:<#port#>/ -k phatk device=<#device#> worksize=256 vectors aggression=11 bfi_int fastloop=false I used separate workers because bitcoin.lc has issues with a single worked (username and password) handling all of my GPUs. This is actually the reason I decided to move from Diablo to smartcoin, I needed something to setup a separate worker per GPU and smartcoin is really slick. |
|
|
|
Kennel:
This is correct behavior. The auto profile launches 1 instance of each worker that is set to allow auto profiling on each GPU. Really, what you want, it to only define unique workers. To properly assign workers to cards, use the profile system.
For example, lets say you have a worker, named "My Worker"
The auto profile would start an instance of phoenix on each GPU, against "My Worker"
If you want to create your own profile (instead of the auto profile), then you have full control over which workers receive their own instance on each specified GPU.
Thanks, this was the same issue I had, moved to my own profile and it starts one worker per instance. Now I have an issue where it will not create or start the miners, file not found in /tmp/smartcoin..... How can I kick logging up to see what might be failing? |
|
|
|
|
I just grabbed smartcoin to give it a shot, love the post on the features. In any case I have 8 GPUs and I configured a worker for each, it appears it's trying to launch 64 miners
|
|
|
|
Jine,
Any word on the stale issue?
The more I throw at it the worse it is, over 4.3 Gh/s pointing at the pool now, the faster the worker the higher the % of stales produced. I've read around a few times, if you have many cards on the same worker, 2 cards will end up hashing the same thing, resulting in stales! Try setting up a different worker for each 5970! I run 7 total GPUs, and I get <.3%. Each has it's own worker! It depends on the implementation of the Miner, if I were using multiple instances pointed at the same worker you are exactly right. I use Diablo which has a single set of threads of pushing work to and from all of the GPUs it handles. It's one of the oldest most mature miners out there and has always worked this way. The pool itself started having issues after the patch was applied and I happily use Diablo on any pool expect this one without any issues. |
|
|
|
This is mostly true .. in newer versions of Xen there is options for PCI passthrough, where it maps the physical device to a specific VM. So it is possible, but not very easy, and I wouldn't count on it not taking the whole box down if something went wrong.
Last time I looked, PCI passthrough didn't work with GPUs. It works fine, parallels even sells their Extreme Workstation product specifically for the task. You do however need to make sure your chipset supports VT-d (Intel) or IOMMU (AMD) for hardware level support of direct IO. |
|
|
|
Jine,
Any word on the stale issue?
The more I throw at it the worse it is, over 4.3 Gh/s pointing at the pool now, the faster the worker the higher the % of stales produced. |
|
|
|
|
Anyone else battling any issues trying to get a Dwolla withdrawl from Mt. Gox? I see a note on Liberty delays but I'm having similar issues with a Dwolla withdrawl, no more then a "few hours" per their site but nothing. With the pattern on the site I wonder if a number of people are working to pull cash out.
|
|
|
|
|
Still haven't been able to get anything from bitcoins.lc with regard to the hgih rate of stales but other pools are looking really good. Added 2 more GPUs, nothing like one instance of Diablo churning at 3.2Gh/s
|
|
|
|
Hi : please research market rates before thread crapping. Goto ebay and do a search on completed auctions for 6990's. They've been selling for upwards of $1300. A huge benefit of the 6990 is simply condensing mining gear. I have 8 5830's running, but have to have them spread across 2 mobo's and then 1 in a stand alone system.
I still enjoy my 5830's, as my 4x5830 system was maybe $700 total and gets 1GH without oc'ing the cards to their max.
Hi I think it is stupid to pay $999 from a scalper. Why reinforce the behavior. 6990 for mining is utterly stupid now as it will not pay itself off for quite some time. Nothing wrong in letting n00bs know that they are getting suckered. Hi I have no part in this but I need to weigh in because of the very poor etiquette being shown by some of the members. You THINK it's stupid, that's your opinion and it belongs in a thread you create or contribute to that is focused on the topic, not on a goods thread. You are attempting to "protect" a "n00b" as you put it that wants a 6990 for one reason or another and needs to pay to get it, that's not your problem and constructive feedback on the alternatives belongs in other topics. |
|
|
|
|
Jine,
Any word on the stale issue?
|
|
|
|
Agreed, Many benefits to be had by running 6990 vs 5830's but it's not worth paying the extra, Retail is pricey enough
and those "problems" your talking about dont exsist anymore, The worst you'll get is a crappy oc card that only give 760Mh/s that runs hot (95°)
where as shappires sometimes cant even push 25mhz past stock on core
In your opinion then, which brand of 6990 allows for best OC? XFX? From what I've read and heard, Sapphire is one of the better 6990 cards out there... no?  Suggest you take your discussion to your own thread and leave the OP on topic, a seller's thread isn't the place to debate the pricing they set nor the alternatives. |
|
|
|
I think you mean keepalive support. DiabloMiner supports keepalive, and the use of it does not effect rejections either way.
It sounds like bitcoins.lc screwed something up.
Running with debug now, seeing "Forcing getwork update due to nonce saturation" in bursts of 8-10 at once, does that provide any hints? Nope, only that you seem to have 4 or 5 GPUs. 6 GPUs on this one, holing off turning on other 2 GPUs until I get this under control.... bitcoins.lc is looking at it to see what they can figure out, thanks for your help. It seems it's not an issue if I run a worker per GPU keeping the per worker hash rate lower. While I I'm posting quick note on another topic, I remember seeing a post stating -f 0 was not a good idea, I always run that on my poclbm miners, any insight on what it should be for dedicated cards on a headless system? |
|
|
|
I think you mean keepalive support. DiabloMiner supports keepalive, and the use of it does not effect rejections either way.
It sounds like bitcoins.lc screwed something up.
Running with debug now, seeing "Forcing getwork update due to nonce saturation" in bursts of 8-10 at once, does that provide any hints? |
|
|
|
Just posed this for Diablo: I didn't see this coming but it seems since bitcoins.lc patched their bitcoind to add support for socket reuse my primary rig running Diablo started getting 15% rejects. It seems my miners running 400Mh/s or so are alright but my main rig running at 2.3Gh/s is having a really hard time, other pools are no problems but bitcoins.lc is hating that speed on one worker. The backend is pushpool.
How would I go about running an instance per GPU or something similar, I hate the idea of doing it but it looks like that might be the only option? |
|
|
|
Any update on the stale shares? I threw another 2.2Gh/s back at it and I'm currently at 10% and climbing, this started with the patch yet other pools are still <1%.
Didn't take long, leveling off at 15% stale shares, hoping it can be sorted out because 15% is too much to give up to the pool if alternatives are much lower.
The strange thing is that this only seems to affect a few users. Could you please provide a full description of the issue, including what miner and hardware you're using - and what internet conneciton (and ip). I'll take a look at it then... Jine, Thanks for the reply, I just took a look at my other miners and see somthing, I run an aggregate of about 3.8Gh/s right now across a couple of machines. Have the following stale percentages: 5770: .5% 5870: .25% Primary 6 GPU 5970 rig: ~15% The only one affected is my primary miner running 2.2Gh/s. Using DiabloMiner the 2.2Gh/s is on one worker and that is the only one having issues with rejects, with the patch you added it seems there is something about a single worker pushing through data that fast. |
|
|
|
|
I didn't see this coming but it seems since bitcoins.lc patched their bitcoind to add support for socket reuse my primary rig running Diablo started getting 15% rejects. It seems my miners running 400Mh/s or so are alright but my main rig running at 2.3Gh/s is having a really hard time, other pools are no problems but bitcoins.lc is hating that speed on one worker. The backend is pushpool.
How would I go about running an instance per GPU or something similar, I hate the idea of doing it but it looks like that might be the only option?
|
|
|
|
|
Any update on the stale shares? I threw another 2.2Gh/s back at it and I'm currently at 10% and climbing, this started with the patch yet other pools are still <1%.
Didn't take long, leveling off at 15% stale shares, hoping it can be sorted out because 15% is too much to give up to the pool if alternatives are much lower.
|
|
|
|
Slush, I just found this, awesome work btw... but MtGox's data doesn't update, I have to close the feed and then open it back up again to get it to update, using 0.2, TradeHill works fine
Mt. Gox has piles of issues with persistent connections, suggest you try polling (-w) |
|
|
|
|
Show Posts
