Thanks, that makes sense.  However... over the long term, luck should equalize out, since by definition its just statiscal noise.

You need a LOT more hashrate.  I'm using a 40PH (more or less) pool and on back-luck weeks its variance is barely tolerable.  I fear the only way this will happen is if several major pools die.

I know that mathematically, variance does not matter - in theory, but that theory does not take into account the changing difficulty, does it?

My p2pool node appears to be healthy, not seeing anything in the console output that looks weird.  The http://ngspoe.carpenter-farms.us:9332/static/ page looks fine, but I'm not seeing my node in http://poolnode.info/

It was there for the past day, just not seeing it now.  Any ideas?

Update:  next day, still not showing up.

Update2:  Seeing my minors fail over to their backup pool, briefly, several times an hour.  This ROUGHLY aligns with namecoin blocks being found by the failover site (e.g. block change requests).  I've also become concerned about the variance issue - I know that in theory it doesn't matter, but with multiple days per block, it feels like the difficulty change will unduly impact us.  Not sure I have the skills to run this pool effectively.

When it goes live, however those that crowdsourced originally, and those that donate their 10K ($160 worth of SJCX) will likely, and note I said ""likely" because it hasn't happened yet and we live in a world of change, will have priority access to paying customers.  Under the current test-group-B status you can test for free.  If you want to get paid SJCX, then you need the 10K investment.  Note that there are no paying customers yet - that won't happen during test-group-B, and I'm unsure if it will happen under test-group-C.  No paying customer == paid coins are coming from the original crowdsources and the developers.  Payments today are to basically help work out the payment system - not to make people rich!

Ah, that makes more sense.  Thanks!

Good to know, thanks.  Over the past 24ish hours I've seen it vary from 6 out, 0 in to 7 out, 1 in.  No idea what drives that, but it sounds like those numbers are somewhat healthy.

The http://ngspoe.carpenter-farms.us:9332/static/ page is confusing me a bit.  As of this post, I've found 41 shares, 8 have been orphaned (which I kind of understand), but my payout should a block be found is forecasted at 0 btc?  Shouldn't I expect to get some small payout (my shares/totalshares * BTC reward)

Thanks.  Somehow I missed that port reference from the guide I was using to install.  I had 9332 open, but not 9333.  It is now opened and dnatted to my p2pool VM.  Not sure it matters but I'll restart the p2pool just to be safe.  I'll drop max connections some while its down.

Hi all.  Thought I'd check out p2pool for my local 16.5TH of mining.

Happy to say it all went smoothly for doing straight bitcoin mining.  Saw a few python exceptions during startup, but once it got initialized, it appears to be working fine.

I'm not having as much luck with merged mining bitcoin and namecoin.

Syntactically all appears fine, and initialization proceeds as above.  However I see:



Reoccurring frequently in the console output.  All my miners are also showing a Diff of 1, which is a change.

Any thoughts?

Edit:  Also seem to be stuck at 8 connections, even though I'm starting up with " --max-conns=256" and my bitcoind has 19 connections.

Pool stats at:  http://ngspoe.carpenter-farms.us:9332/static/

Update:  Found my stupid mistake, was specifying namecoins coin port instead of the rpcport on the --merged line.  The Diff of 1 also cleared itself up after a few minutes.

Still concerned about the number of connections being low.  Any thoughts on how to encourage more of them?

As my boss says:  "Bad news early is good news"

I've seen a number of income projections that seem to bounce anywhere from $3/TB/month to $30/TB/month.

I wouldn't plan on making a million anytime soon, at least not via this product <smile>.

Suspect simply looking at Amazon prices (https://aws.amazon.com/s3/pricing/) will give you the best idea of what you can expect to earn.  Per the Founder chat this afternoon, startup pricing is expected to be around half of Amazon, with that inching up to about 90% of Amazons price if the market supports that.

So if you start with Amazons $0.003/GB/month, and presume 50% of that for a 10TB address allocation, it comes to $30/TB/Month (base 10 TB), or $15 at 50% of Amazon's rate.  That is $1.50/TB/month.  But Amazon charges for access as well, which StorJ isn't doing (that I'm aware of), so a higher rate is more than likely.

Bottom line:  Great use for extra storage, but likely not profitable if you needed to buy more.

That said... next time you upgrade maybe that next size up drive would be worthwhile?

November payouts have been calculated and are being reviewed:  https://storjcommunity.slack.com/files/pthalodezin/F0GU3BEP7/payout__5__preliminary_-_not_from_official_source_.xlsx

They have their temporary payout script running again.  Next run will be in early January for December farmers. 

Thanks for the positive feedback, its truly appreciated.

Regarding the 10K, that is an easy one:  Yes.  Only people with at least 10K per address will be eligible for testing payouts, whenever they decide to do that.  Currently you can have up to 25TB associated with an address, but I hear that is going to drop to 10TB in the future.  No idea why.

Minor update per the current live chat with the Storj Founder:  

The 10K cost has a couple of rationals... its consistent with the investment made by the original crowd-sourcing folks...  It also prevents having to deal with, at least at this point, a lot of low end payout people (someone that just joined for a brief period or only added a MB of storage).

Payouts will come monthly when they get the scripts fixed (working now) and will improve over time (twice a month, weekly, daily, ... to realtime)

The 0.5 GUI will have some of the basic issues fixed, like auto-starting upon reboot.  That code is being worked now.

Per the chat board last night, it looks like payouts are done manually once every couple of weeks, from the pool of coins originally defined.  Presumably that will be done based on storage being tested and time that storage was online.  That said, clearly this is not a "get rich quick" scheme, and you shouldn't expect to run for an hour and see deposits.

Heard there was some type of chat-with-the-developers scheduled for tonight (Thursday, Dec. 17th), so we may learn more then.

Actually, you can't "mine" SJCX in the traditional sense of the word, although you can trade it - which is a lot more than you can do with a lot of alt-coins.

Regarding being "safe" by storing on the cloud vs. via this method, well:

This method is new, so has inherent risk (like developers getting bored and leaving the project half done).

"the cloud" is hardly safe.  Office 365 from Microsoft backs your e-mail up over in China because its cheaper there.  You don't think the Chinese are trying their best to hack those e-mails?  Suggest reading the fine print and seeing just where your cloud data is replicated.  I have international clients who won't use any cloud where some of the data is stored in the USA out of fear of US Government access...

Can't argue that... alas I didn't start the thread.

Or at least "Deposit crypto to earn a bit more crypto".  Still trying to figure out what my TB's are going to earn during the Beta... I know I should earn some more SJCX, but unclear if that is 100, 1, or 0.001.

Kindof.  Its very easy to transfer funds both ways.  If I got tired of testing, I could move my 10K over to Poloniex and sell it.  I'd be at the mercy of market whims of course.

Just my sense... but believe they are now working on usability features vs. raw functionality.  There is still some work to do there.

A couple of examples:  The Windows GUI does not register itself, nor restart after a reboot.  It also doesn't sanity check the entered Counterwallet address to make sure its valid.
For linux users, there is some little cleanups, like adding the --workers= parameter into the internal help.

However, I think the real reason is that they simply wanted to test using higher volumes of data.  They are now up to 1.7PB, which was a landmark for them.  Nothing worse than launching a system and having it die unexpectedly due to its own success.  Better to be sure it can scale.

Everything on your hard drive is encrypted.  That should help I would imagine.

Also, at this point in time, there is NO user data being hosted, just test files.

StorJ is in its 2nd Beta period now.  I've been following it for awhile and just started participating yesterday.  Until recently it was hard to get setup (you had to be part of the original crowdsourcing effort), but now between Counterwallet.io and exchanges like poloniex.com its fairly easy.

1)  Go to Driveshare.org and grab the app.

2)  Go to Counterwallet.io and generate an address.

If you just want to fool around, just launch the app, insert your counterwallet address, allocate some space through it, and let it run.  (actually, they don't currently validity check the address, so anything will do).

If you want to earn some SJCX, then it is a little bit more complicated:

First:  you need to use a valid address.

Next:  The beta period itself is not renting out space, but is rewarding contributing members with SJCX.  By "Contributing" it means you have at least 10,000 SJCX associated with your address.  Spot-checking Poloniex as I write this, that is about 0.3 BTC worth.

So... you would need to buy 10k of SJCX at your favorite exchange and deposit that on your counterwallet.io address.
You also need to deposit a small (0.01 BTC is way more than enough) amount of BTC on your counterwallet.io address.  Since counterwallet rides on the bitcoin blockchain, just send your counterwallet address that BTC from you wallet.  That BTC dust will be used to pay bitcoin transaction fees (presumably only when sending SJCX, but that is unclear)

It is my understanding that you can associate upwards of 25TB with any individual counterwallet address, but rumors are that will drop to 10TB (not sure why).

Note that space can be allocated through a number of "drives", with each drive referencing a folder on your machine.  You can even use network folders, like \\fileserver\myspace, if you wish  - but this is pretty inefficient.  I've been testing an 8TB segment over on my fileserver and I see my network card spike every 7-10 seconds to max speed.  Running a 2TB local segment shows no such activity.

Also note that during the original population of your allocated space (which takes some serious time), the software will eat a good fraction of a modern core and all of an older one.

We keep trying to convince Poloniex to add us, but no luck so far.