Tittiez
|
|
February 13, 2012, 10:08:23 PM |
|
Oh p2k because I am only going to have 2-5 people mining every night, is it best to lower Max. Cache & Size Max. Work Age a bit lower? If so what do you recommend? Edit: [2012-02-13 12:11:20.860][info] ltc_pool: Cache hit by richard/X.X.X.X - Queue size: 19/20 [2012-02-13 12:11:22.700][info] ltc_pool: Cache hit by kitteh/X.X.X.X - Queue size: 19/20 [2012-02-13 12:12:06.745][info] ltc_pool: Cache hit by richard/X.X.X.X - Queue size: 19/20 [2012-02-13 12:12:08.004][info] ltc_pool: Cache hit by kitteh/X.X.X.X - Queue size: 19/20
So it would be optimal to leave it at 20? Or change it to 19?
|
|
|
|
p2k (OP)
Newbie
Offline
Activity: 33
Merit: 0
|
|
February 16, 2012, 08:22:00 AM |
|
Hi,
sorry I've been busy the last days and will be away the following days too.
Cache size and work age aren't that important for ecoinpool because it can create work so fast that the cache never really gets empty.
Max work age is the time that workunits stay in the cache until they are discarded it mainly has to do how "good" you want to support new transactions coming in. On larger pools, this is never really hit, because work is requested faster than cached workunits are discarded; this could even be true for your small pool.
As for the cache size: It is not printed as an info message how the cache is refilled, you'd have to set the server logger to "debug" to see that. The fact that the number never gets below 19 is because the cache is being refilled to 20 only milliseconds after it was accessed. The value of 20 has been determined by watching the cache on a relatively big pool where the cache got below ~15 in only 2% of all cases (that's mostly on block changes, where the cache has to be drained anyway). Maintaining a cache of 20 workunits should be ok for any server as they just take a couple of cycles to make and only use a few bytes of ram.
The essence of all that I said: It's not really worth changing the default settings for those two values as what you gain from tweaking the numbers is pretty much marginal and will probably never impact work delivery at all. If you're really a perfectionist and want to save ~4kb of RAM and 100ms CPU time, then set the cache size to 10.
p2k
|
|
|
|
EnergyVampire
|
|
February 16, 2012, 09:14:44 PM |
|
Hello, I received this error a few times today on the namecoind debug.log: ERROR: ConnectInputsHook() : name_firstupdate Cannot be mined if name_new is not already in chain and unexpired Have I setup the auxchain wrong somehow? I installed namecoin version 32464 to try and do some merged mining. I haven't found any blocks but I am getting valid shares from couchdb. Thanks in advance for your replies, Cheers!
|
|
|
|
Tittiez
|
|
February 16, 2012, 09:31:21 PM |
|
Sigh, unlucky lately. No blocks since that first one found, and currently at 119800 shares.
@EnergyVampire I got no idea about your error...
|
|
|
|
marks1976
|
|
February 21, 2012, 04:21:05 AM |
|
I would like to post a BTC bounty for anyone who can assist me in the final stages of getting this all working. I think I have about 95% of the pool set up but I just can't seem to get the last little bit figured out. I have installed all the files onto a dedicated linux server. I have followed all directions and just can't seem to get the results of the pool running. Somewhere the setup of the pool has failed and therefore will not start. Please PM me if interested.
|
|
|
|
Tittiez
|
|
February 21, 2012, 11:35:56 AM |
|
Sweet, two ltc blocks in a row for us last night!
|
|
|
|
hashking
|
|
March 01, 2012, 04:12:41 PM |
|
Is anyone interested in setting this up for a bounty. Please pm me if you are. I'm interested in mining on it with a few friends. Thanks.
|
|
|
|
p2k (OP)
Newbie
Offline
Activity: 33
Merit: 0
|
|
March 02, 2012, 08:13:27 AM |
|
It's kind of funny that there are now bounties for setting ecoinpool up while I'm working on the bugs and features which will make setting up so easy that everyone should be able to do it But well, I'm not fast enough due to real life work load.
|
|
|
|
marks1976
|
|
March 02, 2012, 09:51:19 AM |
|
It's kind of funny that there are now bounties for setting ecoinpool up while I'm working on the bugs and features which will make setting up so easy that everyone should be able to do it But well, I'm not fast enough due to real life work load. I hope so. I still haven't had success finding someone to lend a hand or a stiff bounty wanted.
|
|
|
|
hashking
|
|
March 04, 2012, 05:56:37 AM Last edit: March 04, 2012, 07:25:11 AM by hashking |
|
I made it to the part where I'm trying to add a worker and I get While requesting a new user ID, the pool server did not respond in time. This usually means that the pool server is down or currently unreachable. As you are an admin user, you are advised to check your firewall settings so that this URL can be reached: http://localhost:8888/. Could this be because the chains haven't finished downloading.
|
|
|
|
hashking
|
|
March 04, 2012, 06:11:43 PM Last edit: March 04, 2012, 07:40:40 PM by hashking |
|
Was able to get this running and have a miner connected to showing that shares are being accepted in http://localhost:5984/ecoinpool/_design/site/_show. I'm not getting any output on the command prompt line except for New aux block, new main block every once in a while. Just wondering if there should be something showing up. Also should there be an ebitcoin client for namecoin? When I try and set this up I get document update conflict.
|
|
|
|
p2k (OP)
Newbie
Offline
Activity: 33
Merit: 0
|
|
March 04, 2012, 08:16:35 PM |
|
Yes, ecoinpool still crashes if the blockchain wasn't loaded by the local bitcoind/namecoind as mentioned in some posts before.
There won't be any console output except warnings, so this is normal. Try to follow the logfiles in the log folder for more detailed messages.
Yes, there should be an ebitcoin client for namecoin or the system will fall back to block change polling 5 times per second. The document update conflict seems to be a bug. Try reloading the page before you fill out the form and click save.
If only I had more time at the moment. Now I know how shadders felt back then... Got work up until night; free time is rare and when I finally have some, I rather use it to relax a bit. Looking at my deadlines, I need at least one more week before development can continue.
|
|
|
|
hashking
|
|
March 04, 2012, 08:32:16 PM |
|
Awesome project so far. Are you working on it yourself? Are you planning on allowing more then one chain for merged?
|
|
|
|
p2k (OP)
Newbie
Offline
Activity: 33
Merit: 0
|
|
March 04, 2012, 08:50:34 PM |
|
Yes, until this project matures a bit more (i.e. beta 0.4 is reached) it will remain a solo project.
There is a lot of complexity involved with multi chain merged mining, most of it related to shares logging, configuration, correct handling of stales etc. However, ecoinpool has partially be designed to support multi chain MM at some point in the future without having to rewrite the entire server component.
|
|
|
|
hashking
|
|
March 04, 2012, 09:43:07 PM Last edit: March 05, 2012, 02:48:00 AM by hashking |
|
My cgminer is showing over 100 effeciency. Its at about 147%, is this normal. Also when a block is found will it show it on the output screen or do we have to just check the bitcoin daemon.
|
|
|
|
hashking
|
|
March 05, 2012, 04:28:45 AM |
|
Found a merged namecoin block.
|
|
|
|
marks1976
|
|
March 05, 2012, 06:32:58 AM |
|
I have been mining mine for a little over 30 hours and no luck yet. 3 5970's on it so far. Awesome luck hashking.
|
|
|
|
hashking
|
|
March 05, 2012, 07:00:53 PM |
|
Nice setup P2k so far have found 2 namecoin blocks, and 1 bitcoin block solo mining. So I can say that your pool backend works great.
|
|
|
|
hashking
|
|
March 15, 2012, 09:53:55 PM |
|
What does it mean when its branching for 1 block
|
|
|
|
rjk
Sr. Member
Offline
Activity: 448
Merit: 250
1ngldh
|
|
April 22, 2012, 04:07:43 AM Last edit: April 22, 2012, 04:38:25 AM by rjk |
|
So I decided to give this a shot, just for fun. I see in your examples that you are using ubuntu, is there any particular version and/or architecture that works best? Also, does anyone know how to properly set up a paravirtualized pygrub loader for an encrypted LVM on Citrix XenServer?
I usually use Centos for my work stuff, so if that would be compatible I would prefer that.
|
|
|
|
|