Topic: phoenix miner in solo question (Read 2989 times)

Well after a long time solo mining and seeing 2 rejected blocks i gave up on solo mining since that moment btc prices kept dropping from $15 to under $2.
Then i choose to join pools and have been changing pools alot.
But basically for me mining is and will be allways a loss no matter how much the bitcoin will be worth in the next 5 years
I am currently paying 0.33 euro per Kwh and in januari is announced another 2,5 % eco taxes will be added to this amount now that means that the total amount will be even more again since at the end of the bill they count 20% extra tax over the total amount.
I think Germany will follow the netherlands in their foot steps soon as well, so if you do not have "free" electricity mining is not an option.
The worst is also that we do not have sepa payements implemented before 2013 so trading bitcoins is a pain in the ass as well.
Especiually since i do not want to use the thiefs from paypal.
I have been trying to buy bitcoins since it was under 1,80 euro but have not been able todo so, yes i can make a international money order but the cost is not a joke.
They count 2,5% or a minimum amount 7 euro + and the other side 10% with a minimum of 10 euro does not make it a thing you want todo. I allways thought that netherlands was an early adopter of these things but lol germany uses sepa already since last 2009
So i fear for me i have to say goodbye to bitcoin till 2014 or so, if sepa has been introduced or some other way for trading has been implemented, or bitcoin has become $100 each. but to be honest i think their is for me nothing to gain on bitcoin.

This is the intended behavior. Phoenix only submits shares that meet the target specified by the server. For solo mining this will be the full network difficulty. If you want to verify that difficulty 1 shares are being found then you can use the -v option to enable verbose logging.

I was just experimenting with this to see how it works (as a backup maybe).

I ran "bitcoind -daemon" on my notebook with suitable rpcuser/rpcpassword in bitcoin.conf.
This seems fine for a test. I don't really want to start a new wallet on the miner.

Now on my miner machine I add a "pool" to cgminer with ip/user/pwd for my notebook.
However, it can never connect and go. Always dead.

Checking netstat on my notebook it appears the rpc server only listens on localhost. So I guess it's not listening externally on my LAN. Is there a way to do that? I don't see a config option.

I guess I could open a tunnel to my miner and tell cgminer to connect locally...

Just asking here because it looks like DeathAndTaxes knows about this stuff.
Thx.

Edit: Oh, using a tunnel seems to work, I think. cgminer says it's alive now and it appears to be doing work. But no progress messages. I guess cgminer doesn't report shares since there aren't any. Or something like that.

Is there some way to confirm it's really mining solo?

If you didn't get a reward you didn't get the block.  That simple. You can check the block explorer and see who got the block (if you know the last block completed when you were minning).

There is no such thing as "left off".  Each hash a COMPLETELY NEW TRY.  It is like a lottery ticket.  If you buy 100 lottery tickets and lose and then take a break (turn if off) when you come back and buy a new lottery ticket are you "closer to winning"?  No.  Each lottery ticket has exactly the same chance of winning.

At current difficulty if you perform 1 hash your chance of "winning" the block is exactly 1 in 7,255,634,282,020,860.  It doesn't matter how many hashes you have done in the past your next hash has exactly a 1 in 7,255,634,282,020,860 chance of winning.  If it wins you sign the block.  If it doesn't win it is COMPLETELY WORTHLESS and your next chance is no higher than 1 in 7,255,634,282,020,860.  Your computer keeps playing this lottery until you find a block or give up.  If your card is 100MH/s you are making 100 million lottery attempts per seconds.

If anyone on the network finds a block (which happens roughly once per 10 minutes) then you can no longer try to solve the current (which is now old block) and instead must solve the new block.


If the block counter increases it means SOMEONE has found a block not that you have found a block.  If you have no generated transaction then you didn't solve the block.


Solo mining isn't hard however it sounds like pool mining is better for you.  Honestly given there are pools with 0% fees (some with negative 5% fees) I have no idea why anyone wants to solo mine.  I have 10GH and wouldn't solo mine with less than double that.  Even then I wouldn't solo mine without a reason (I might if there were no non-hoppable pools w/ fee less than 3%.

Honestly I just don't see that this appeal is with solo-mining.  In the long run your reward is exactly the same (minus any fees) and in the short run you can have some awesomely bad luck and go months or even years without finding a block.   

I have never considered solo mining for bitcoin, I just can't take the risk getting nothing after months of hard mining.
My earlier records on btcguild showing that I have earned 122BTC from the pool, and I have found 2 blocks. If I went solo it could be either 0 or 5 blocks.

However I have done lots of namecoin solo mining in the past, the difficulty was low and I can always see progress within a day.

To me solo mining is still a mistery untill now i had 3 times a block but it never got rewarded and there is hardly any explaining about how and what is happening when mining with cgminer and such
Does it really stop finding the correct block when you stop a miner or does it continue where it left off
If so why does the solo server gets new blocks assigned to it ?
And yes the client never reports anything other then if it found some kinda valid block, but again i never got any reward for it.
Something else i noticed is that the block count on the local bitcoin server i run changed a after it found the so called block, but again there is so little support on this local mining that its almost sad.
I have found till today no explaining about nor has there been any improvement to see on this what your doing
Thats why i switched to mining in a pool, cause this whole local mining is full of misteries and lack of info
I wonder if anyone really know what it is really doing or where that hidden guy is 

Yeah pretty cool stuff the high end servers can do.  Asked the guys in hardware dept (I am database developer) and they have never had to use that feature but they could if necessary and without bringing the server offline.  Was kind of a letdown when he told me they wouldn't since my server is actually a fault tolerant cluster of servers.  They would just take the affected server offline make the switch and then rebuild the cluster.

I see the point but the gadget lover in me says how boring. 

roger that... 

btw,  switch a cpu on the fly..... wow.

Based on what you describe I am more confident it is a simultaneous connections issue.

Two clients using the same port shouldn't be an issue because the router "should" handle the NAT mapping just fine.  Also the client searches for peers on local subnet meaning that any blocks/transactions found by one client on your LAN will quickly be shared to all clients on the LAN.

Still if you have found that one client works better I can at least understand why you are running it that way.  Personally working w/ enterprise grade database servers in my job (redundant power supplies and electrical circuits, hotswappable cpu/ram, realtime fan monitoring, etc)  makes me always thinking about redundancy and it would just bug the crap out of me to have all my rigs vulnerable to a single point of failure.   

might have been the rrouter..  i have since bot a new one and will be upgrading the firmware so I can run cgminer.  I have to keep using the gui till I get it set up

I can run cgminer on 1, maybe 2 rigs, but when I started to bring them all over, the router could not handle it..

I thought it was just the fact of 2 bitcoin clients both using the same port to communicate

Once the block chain is up to date the amount of bandwidth to keep the chain up to date is trivial.   While the block chain is incomplete I could see issues.  Just use one client to "jump start" the block chain of other clients by copying over the blockchain .dat files.  

I have 6 rigs all with seperate clients which I was running solo during the DDOS attacks and didn't experience any issues.

On edit:  Just thought of something.  Some routers have problems with number of connections they can sustain at one time.  This can cause problems in bittorrent.  I wonder if you were experiencing something similar to that.  Newer routers shouldn't have an issue even with hundreds of simultaneous connections though.  You could try googling the name of your router and "bittorrent".  If you see threads about poor bittorrent performance it might be # of simultaneous connections "gimped".

if I run 2 clients on my network (i only tried once.) things did not go well, because both clients where trying to download the block chain.  I cant remember the actual symptoms, but I know I told myself not to do this in the future.

Is there something about the port the client runs on that my router couldn't handle both clients on the network?

I will chime in because nobody has helped.  Haven't use phoenix in a while (use cgminer now) but IIRC that is correct.

Phoenix will only submit valid blocks to bitcoind when running in solo mode.  Thus:

accepted = valid block found and accepted by bitcoind
rejected = valid block found but rejected by bitcoind (likely invalid due to you getting beat by another node solving same block faster)

everything else stays internal to phoenix and isn't reported.

Now I am not 100% sure on that simply because I haven't used phoenix in a while and haven't used it in solo mode for even longer. 


Personally I would just run a separate client (w/ separate wallet) on each machine.  For some reason this is the new trend lately to try and make a solo client-server relationship. 

KISS.  Keep It Simple Stupid.

There is no benefit to doing this.  You simply create a single point of failure.  If the machine running bitcoind goes down you lose 100% of hashing power until you detect it and fix it.  The first time that happens when you are alseep or somewhere you can't detect/fix it for 8 hours you are going to realize the folly of running it this way.  If you run each machine independently and a machine goes down you simply lose one rig's worth of hashing power until you get that machine up again.

Care to explain?   I do agree "someone" is confused.

maybe i'm doing something wrong, but i'm definitly not confused.

bitcoin client is runing in server mode on one of my machines.  the other rigs are connected to it and are hashing.  Do you know the answer to my question?

boy are you confused or what
You Don't run a miner if your mining solo
DUA

i'm running a few rigs in solo mode using phoenix.
everything is setup, it's connected and hashing, but the accepted and rejected counters remain at 0.
is this normal?