Show Posts
|
Pages: « 1 2 [3] 4 »
|
Heres the code for minecoin. Is it in your pools.conf? minecoin mineco.in:3000 https://mineco.in/stats.json $time=~s/^.*"duration":(\d+).*$/$1/;$shares=~s/^.*"shares_this_round":(\d+).*/$1/;$rate=~s/^.*"hashrate":(\d+).*/$1/;$rate = int($rate/10**9); utility_1
I've never used minecoin, but wouldn't be opposed to adding it in. However, I don't want to add yet another pool that won't convert its shares. If you feel up to writing a rewards function too, I'll add it into the rotation. Otherwise, I might get around to adding the pool eventually ![Smiley](https://bitcointalk.org/Smileys/default/smiley.gif) The auto-refresh sounds reasonable, whenever I get some time for some more tweaking.
|
|
|
Is the server down? I just tried using it as everywhere seems to be getting dos'd at the moment and had very little joy.
Should be up, and seems to be working better now than it was earlier / over night.
|
|
|
The new thread for Multiclone is a good idea -- I had intended to start one once I had worked out the remaining kinks with Multiclone. I'll probably start my own Multiclone thread (so I can own/edit/etc the main post) in a few days.
Anyway, I think I figured out the issue with the recent flakyness -- it had to do with a partial removal of the solo pool I tried to do, while maintaining pool protection.
I've reverted that tweak, so let's hope things are more solid now.
|
|
|
Just wanted to clarify for the 100th time: btcmine and bitcoins-lc will pay out and it's not a problem with your personal shares that they're still pending. I just need some time to update the pool's rewards code to process those rewards. I'll try to get to it this weekend.
Since the code is open source, anyone else is free to update the code as well and I can easily incorporate that into Multiclone to get this payout done faster.
The Multiclone pool has received payments from both of these pools, it just doesn't know how to split them up among the users yet. You will eventually get paid.
If you all want, I can remove these pools as targets until this is resolved. But I think this will reduce the pool hopping benefit quite a bit....
|
|
|
Received payment too, for which I thank you, Nick. Multiclone is a bit shaky today, miner idle frequently.
I'll restart the server. I just hopped on the logs and noticed that there's only 3 active users. Figured something was flaky =/
|
|
|
Does bash-4.1$ bitcoind sendmany "main" '{"17x6iXEGfE94VnDwUMS1E5fii3FEqXLrEf":Decimal("0.323")}' "payout" work? I think I saw that somewhere, but I'm not sure if this could be the reason/fault. Unfortunately, that doesn't work either ![Sad](https://bitcointalk.org/Smileys/default/sad.gif) I just applied a little regex magic to turn my sendmany command into 40-something individual sends. Cost me more in transaction fees, but oh well. I tried to turn down the transaction fee in my client ("bitcoind settxfee 0.00001", which returned 'true'), but it stayed at 0.0005 for all the sends. *shrug* That means the first round of payments for multiclone is out! Everyone with a balance over ~0.03 should have received their payout. Sent out ~20BTC in total. I'm hoping to have some time this coming weekend to write the needed code for the two pools that aren't scraping/converting right.
|
|
|
FWIW, I brought multiclone back up again earlier this morning. I'm not sure what happened this time -- the server just stopped responding entirely. I couldn't even ctrl+c to kill it... In other news, I was all set to do a payout (which would include everything except btcmine and bitcoins.lc), but am getting an error when trying to do a sendmany, even when I reduce it to its simplest case, and use way fewer digits after the decimal: bash-4.1$ bitcoind sendmany "main" '{"17x6iXEGfE94VnDwUMS1E5fii3FEqXLrEf":0.323}' "payout" error: type mismatch
I'm using the newest release of bitcoin. Any ideas? I tried building bitcoind from source, but apparently did something wrong when selecting the berkeley db version. It compiles, but crashes on start: bash-4.1$ ./bitcoind -daemon bitcoin server starting bash-4.1$
************************ EXCEPTION: 11DbException DbEnv::open: Invalid argument bitcoin in AppInit()
terminate called after throwing an instance of 'DbException' what(): DbEnv::open: Invalid argument
If all else fails, I can do individual sends (rather than sendmany), but ... that's going to be a headache ![Smiley](https://bitcointalk.org/Smileys/default/smiley.gif)
|
|
|
I'm getting almost 1/3 of my shares rejected. That probably means a net loss for me, even with the hopping.
On Multiclone or with your own local pool? On Multiclone. <2.2% global rejection rate since the server restart earlier. Doesn't look like a server issue...
|
|
|
The pool software was still running, but something caused the getwork thread to die, so Multiclone had nothing to give to the clients.
Restarted the server a few hours ago, and will try to investigate the cause of the failure later this week. Don't have much time today, and forgot that Monday's a holiday in the US when I mentioned Monday as a payout target. I'll try to keep the server up, but I might not get around to making the payout until Tuesday.
In the meantime, I'd recommend running a backup miner in parallel with Multiclone, in case my server crashes again. Anyone have a good link that explains how to do that with a variety of miners/configurations?
|
|
|
I'm not running with much power but only getting an eff of 0.766 ![Sad](https://bitcointalk.org/Smileys/default/sad.gif) The cumulative stats at the top can be misleading, particularly if you have a bunch of pending shares still. And given that two very profitable pools aren't converting pending to confirmed (an issue I will fix, just don't have time for a few days), that could be dragging your average down. Aside: I'm going to be out of town/busy for the weekend, so please don't panic when I stop responding ![Smiley](https://bitcointalk.org/Smileys/default/smiley.gif) I'm tentatively planning to make the first payout either on Sunday or Monday If the pool goes down entirely, email me at: nick D0T multipool PLUS alert at gmail and I'll do what I can to bring it back up. If that alert email address gets abused/overused, it's going to be disabled. So please don't mail me there unless it's critical ![Smiley](https://bitcointalk.org/Smileys/default/smiley.gif)
|
|
|
Use the balancecurrent function for the balance from the current block. Sure, but that is just a wild guess based on current hashrate, isn't it? I'd like to see my estimated balance for finished rounds that are not yet confirmed, to provide some bit of assurance that I'm actually producing useful results with Continuum. Especially since my confirmed balance is still showing up completely empty. Is this expected behavior??
|
|
|
Do the pending shares eventually convert on Multiclone? The 3 pools that I've got the most shares on don't show anything other than pending after mining for about 18 hours.
The rewards scrapers for bitmine and bitcoins.lc are currently inoperational, so stats (and payouts) of these pools will be delayed. The multiclone pool is earning at them, though, so I promise you'll get your BTC eventually ![Smiley](https://bitcointalk.org/Smileys/default/smiley.gif) All shares here will show 'pending' for about the first 24 hours, then will start converting in the interface (for pools except bitmine and bitcoins.lc) shortly after that.
|
|
|
There is still a lot of solo going on though.
I tried to tweak some things to decrease the amount of solo, but it's hard to tell if it helped. I'm hesitant to remove solo entirely, as it's used to help prevent 'naughty' miners from trying to launch DOS attacks. Overall since the pool reboot last night, the total number of shares to solo is only ~1.4% of the shares submitted. It just looks like a lot in the interface, since it seems like I get just a small handful of solo shares each round, so there are a lot of table rows. Whereas with the 'real' pools, you tend to get lots of shares for each round Multiclone decides it's profitable to participate in. 1. Can one use more than one worker on the same payoutaddress without issues? I'm not sure about this one. Let's let Multipool reply. 2. All my shares on bitcoins.lc and bitmine are still pending after 36h. Bitcoins.lc pays out immedieatly after a block is found, howcome its so slow on multiclone? As someone pointed out earlier, Bitcoins.lc doesn't have a rewards function written for it yet, but it's been pretty profitable so I kept it in the rotation. So all the Bitcoins.lc shares will stay pending until I get around to writing the rewards function It looks like there's an error in the log indicating that Bitmine's rewards function is failing too, so that'll need to be fixed as well. Hopefully Multipool can help with this when he gets a chance, since bitmine's rewards function looks a little less straightforward than some of the other pools. 3. How big is multiclone now, is there a way to see the total hash power? No way to see the hash power from the user interface, but we've been ranging from 10-15 GHash/sec for the past day or so with 15-20ish simultaneous users.
|
|
|
Thanks for the info. For the time being, I took eligius out of the rotation until I can make sure it's going to work right (and, from the sound of it, adjust the utility function). It appeared from eligius stats that despite pointing to eligius-eu, the very few shares Multiclone submitted were being routed okay to the new system.
|
|
|
BTW, is Eligius-EU going to the new eligius pool?
Can you update me on what's changing with Eligius (or point me somewhere that summarizes)? I have on my to-do list: "Something's changing with Eligius??" :-)
|
|
|
I've been doing some testing of Continuum to consider using it as a backup pool over the last day or so. When I try to check my balance, it just returns an empty string. Not zero, just a totally blank page: http://www.continuumpool.com/balance.php?worker=1HEPamMMbwnd6XjXHsfqQ8smrNi77vyqPbI get the same thing for balance (a blank line) when doing a JSON call from a local script: #!/usr/bin/perl use JSON::RPC::Client; my $client = new JSON::RPC::Client; my $uri = 'http://rpc.continuumpool.com:8330/rpc'; $client->prepare($uri, ['hashrate','roundstart','monitoradd','dropmonitor', 'balance','balancecurrent']); $worker = '1HEPamMMbwnd6XjXHsfqQ8smrNi77vyqPb'; print $client->hashrate($worker); print "\n"; print $client->balance($worker); print "\n"; print $client->balancecurrent($worker); print "\n"; print "Done\n";
'hashrate' and 'balancecurrent' both return meaningful values. Is there a way to check my 'pending' (ie, non-confirmed) balance?
|
|
|
This also looks very strange at slush: (look at the efficiency compared to utility/shares)
133995 Thu Jun 30 09:02:05 2011 1 1.241 0.00013483 3.719 0.00000519 133992 Thu Jun 30 08:52:06 2011 5 8.216 0.00000225 0.012 0.00000000 133991 Thu Jun 30 08:17:52 2011 6 5.891 0.00006748 0.310 0.00000000 133986 Thu Jun 30 07:52:21 2011 3 4.946 0.00008378 0.770 0.00000000
This comes from Slush (and btcmine) being score-based pools, designed to discourage pool hopping. Actual efficiency doesn't seem to match very well with expected utility. I've seriously considered removing them both as target pools for this reason.
|
|
|
Bitcoinpool will accept shares even if your login is incorrect. Credit is given to pool donations. They have seperate logins for the pool and the forums as well.
ding ding ding, we have a winner! This is entirely my fault; I had the login credentials wrong for this pool. I just did a pool restart to fix it. If Multiclone doesn't end up re-calculating the bitcoinpool shares with proper crediting and thus automatically include these shares in the pool payout (I don't think it will, but lets wait and see), I'll pay out for these bitcoinpool shares from my personal wallet. Either way, bitcoinpool won't be paying me/the Multiclone pool, but I'll do what I can to make it right. Once new rounds start to get confirmed from bitcoinpool (so I know where the brokenness starts/stops), I'll tally up your 'broken' shares and pay out at Deepbit's PPS rate. Sound fair?
|
|
|
Up to 2750 shares in with 2730 pending.
The 20 non-pending are from slush and report a utility of 19.422 and efficiency of .576.
Please remember that just like with the original Multipool, it's going to take a while for the stats to become non-pending. It has to wait until the parent pool has fully validated and released stats for a given round. As I understand it, there'll typically be about a 24h delay for the stats to go non-pending. Interestingly even the solo shares report a utility of less than 1. Their utility is .9.
I tweaked the static solo utility to lower it below 1.0, thinking that would further decrease the chance of the solo pool being selected (since it's quite unlikely we'd find something in solo). I think, however, that the solo pool is always Multipool's last choice anyway. I should probably just make the tweak to remove the solo pool entirely. While we're doing that, I'll try adding deepbit and mt. red back in if it seems like they behave in local testing. Is a high reject rate normal on multiclone? Is this because of the long queue depth? I have 106 accepted and 25 rejected. That's like 19%
If you joined right before a round ended, you might see those kind of results. Overall, Multiclone has had a rejected rate of only about 2.1% in the past 18 hours.
|
|
|
|