@ pmconrad: Very cool, thanks very much!
Very nice, indeed. Just noticed some builds are failing 'cause the machine's OOM. Also, I think we should inform the exchanges (AFAIK only cryptsy and bter) to change to the new version before the deadline. No need for them to enable trading just yet as the risk of double-spends is kinda increased now IMO; but it never hurts to let them know well ahead of time. |
|
|
|
Errrm... So today I decided to investigate the performance issues of my block explorer. You know how some pages take literally ages to load, eg. the one for the richest address? Although the poor VPS has only 590MB available RAM (with another 1GB as swap), the bottleneck is actually the CPU... So I sifted through the bitcoin-abe source code for hours, just to stumble upon this: http://gregsramblings.com/2011/02/07/amazon-ec2-micro-instance-cpu-steal/TL;DR: FUCK YOU AMAZON! I have yet to decide whether I'll switch to a bigger instance or ditch amazon altogether. Do you have any good VPS provider recommendations? I tried chunkhost for a bit, but my VPS got restarted several times during a one month period, which is unacceptable. The EC2 instance has not had a single unwanted reboot since I started it, quite impressive. Sux. Still, as you mentioned it is pretty reliable and it will be hard to find something better in that price range. I've heard a lot of good things about this place lately but never used them so have no clue if good or not. http://www.dreamhost.com/servers/vps/Try lowendbox, or get a low-end dedicated. Don't want to give specific recommendations, but delimitervps offers (when it's not sold out) Intel Atom servers for just $5/month (quarterly) and freaking dual Xeons (Core 2's I think) for $20/month, which I have one of. Search lowendbox or webhostingtalk for the coupons. Offset it with a bit of CPU mining (don't think you're going to get $20 a month, but it's something.) Others that I can recommend (mostly unmanaged): ramhost.us (if they're not sold out) ramnode buyvm.net (has great storage servers) PLEASE don't mine on VPSes (except on-demand instances such as amazon or digitalocean). Recipe to get you kicked from your webhost, and violates every TOS imaginable. With a dedicated, it's yours, so no one cares. Thanks for the list! Any of them accepting BTC? |
|
|
|
As of last block: 0.52066266 BTC 97.67% Estimated Change: +0.00960806 BTC +0.11% Estimated Total: 0.53027072 BTC 97.77% Estimated Position in Payout Queue 885.13878168 BTC are ahead in queue, putting this user's payout after a 35 block delay. Stats back online, let's see where's the troll now  I bet he doesn't even dare to show up anymore I'm here. As of last block: 500.52066266 BTC 97.67% Estimated Change: +500.00960806 BTC +0.11% Estimated Total: 500.53027072 BTC 97.77% Estimated Position in Payout Queue 885500.13878168 BTC are ahead in queue, putting this user's payout after a 35500 block delay. It's means digits on some site not BTC in my wallet. Pool still don't pay ! rofl dude u got it all wrong with that hand-editing  |
|
|
|
Sux. Still, as you mentioned it is pretty reliable and it will be hard to find something better in that price range. I've heard a lot of good things about this place lately but never used them so have no clue if good or not. http://www.dreamhost.com/servers/vps/Thanks. Or maybe I should redo the whole server to handle reboots gracefully (erm... at all  ). I'm surprised it even works as is TBH. So we still think transactions will have a reasonable chance of being minted in 90 days? If so - that voids all of my concerns above.  Well, it the short/mid-term, yes. No clue if it will hold long-term. But even then, this aspect can be changed without a hard-fork if most of the nodes will be honest (at least short-term ) by increasing the minimum amount of YAC required for PoS and bumping the coin split threshold up a little - thus decreasing the number of PoS blocks required by a single address to get interest on all of its coins. I thought if the update prevented two POS blocks from being being consecutive, this would reduce the number of POS blocks to 720 per day. With less POS blocks, I was concerned that our transactions may sit for months (or years) before having a chance of being minted. Transactions would quickly hit their coin-age cap (90 days) and only a static few (~720) transactions would be minted each day. I have this issue with PPC - I have several transactions that haven't been minted in over a year and they are no longer gaining coin-age. They are all reasonably sized and have hundreds of coin-years in age. Never the less, they aren't minting - making them less appealing to hang on to (not to mention the full year I've wasted holding on to them). I just wanted to make sure we didn't introduce a bottleneck like this to YAC.
I've rewritten the code to actually allow consecutive PoS blocks, but only give non-zero trust to the first one. Also, if my math is correct, YAC is supposed to have 144 PoS blocks/day (10 minute interval), not 720. |
|
|
|
Errrm... So today I decided to investigate the performance issues of my block explorer. You know how some pages take literally ages to load, eg. the one for the richest address? Although the poor VPS has only 590MB available RAM (with another 1GB as swap), the bottleneck is actually the CPU... So I sifted through the bitcoin-abe source code for hours, just to stumble upon this: http://gregsramblings.com/2011/02/07/amazon-ec2-micro-instance-cpu-steal/TL;DR: FUCK YOU AMAZON! I have yet to decide whether I'll switch to a bigger instance or ditch amazon altogether. Do you have any good VPS provider recommendations? I tried chunkhost for a bit, but my VPS got restarted several times during a one month period, which is unacceptable. The EC2 instance has not had a single unwanted reboot since I started it, quite impressive. |
|
|
|
Created pull requests for the PoS fix for both master and testing branch. Changed the block number for switch to be #420000 as to give users more time to inspect code and start using new version - this should occur around Feb 11. I've disabled the clock drift change for now as it's not quite needed (that reverting commit could be reverted to enable it). Lemme know what you think. In other news, old c coder has been able to compile a native MSVC++ version of yacoind ( https://github.com/yacoin/yacoin/pull/17) and I will merge that to testing (or maybe YAC-MSVC++) once the outstanding issues are resolved. Anyone that is able to test it out, please do so. This (I believe) makes YAC the first coin that can be compiled in MSVC++ since the early days of Bitcoin. I'm not a fan of windows and neither is sairon, but there are a lot of extremely talented people out there (like Ron and Satoshi  ) who are masters on that platform, and I believe it will bring YAC to a whole new group of people. Wow, seems like it's been quite a lot of work. The #ifdef's don't look nice, but dunno if it can be done any other way. Thumbs up! Our block target for YAC is one minute. This would imply that there would be a maximum of 720 POS blocks per day. As YAC continues to grow, this may not offer many opportunities for people to mint their transactions. Therefore, is it possible to....
1) Mint more than one transaction at once/per block? (i.e. Five transactions of 10 YAC w/ one coin-year each, in an unlocked wallet would mint in the same block, and be returned a single transaction (50.50 YAC)
2) Increase/remove the 'coin-age' cap? (I believe transactions stop gaining 'coin-age' after 90 days)
This would help reduce the number of PoS blocks generated by a single user and ensure they wouldn't loose any coin-age / 'interest' on their unspent transactions. PoS target interval is 10 minutes. Setting it too low, however, would somewhat reduce the security of the network (lowering the active stake volume 10 times faster and we don't yet have enough PoS minters to negate this effect). Problem with #1 is that it would kinda defeat the PoS minting pseudonymity if you would merge multiple addresses into one PoS block. #2 would IMO also be bad as it would allow someone to absolutely dominate PoS minting for quite some time (as his coins would have a huge chance to be selected for PoS). You too late, you lose. 90 days is a plenty of time. Is the N change enforced by timestamps? If so, would it open a path for some kind of fake timestamps attack? (Like faking older timestamps -> smaller N allowed -> same hardware get much higher hashrate, difficulty and trust score in blocks.)
Yes, Nfactor changes are based on timestamps. However, it's not much of an issue as timestamp of a block can't be earlier than that of the previous block (simplified - it's calculated from median time and clock drift, etc). |
|
|
|
That's the original proposal. I think sairon has a newer one (in a pastebin he deleted) which mainly uses POW difficulty as trust and let POS block inherit the trust score from previous POW block. I have a habit of setting quite short expiration periods for my pastebins. |
|
|
|
So I couldn't just send a YAC to the address and include a message with that transaction? Sorry to be such a newbie.
Nope you can't. That's one of the misconceptions that blockchain.info is spreading (un-)intentionally. |
|
|
|
No joy. I tried all of your suggestions. Would there be some way to use the "sign message" functionality to see if any (perhaps honest) person has received the coins?
First you need to find that person, though. |
|
|
|
I don't really know how this happened, Sairon. I've got the YAC wallet locally, but my address is different. I'm guessing that the transaction is unredeemed means no-one is using that address. The coins were sent from Cryptsy, but the customer support there is not entirely helpful; they just closed the ticket with an "issue solved!" sort of message.
It only means no one has spent these coins yet. Unless you tend to copy-paste someone else's addresses, I think you have the private key to this address somewhere (or it's an exchange deposit addr). Double-check all your exchange accounts for deposits of 1371.497218 YAC. Did you use more wallets? (Do you have more wallet.dat files?) If so, try to find these and see if the coins are there. You could also try typing "validateaddress Y9MYeV2WcNbYJtMzE9Qs8dgS5yyAyvs9YF" into the debug window of YAC client. If you see something like "ismine: true" - then good news, the key to that address is in your wallet (but the client missed the transaction, somehow). Other than that, dunno. |
|
|
|
There ain't such thing as an invalid address... Is it your deposit address on some exchange? Your wallet's change address? Someone else's address? |
|
|
|
It's C++11 already.
Bitcoin was written during 2008-2009. Sure it was. Maybe even earlier. Just saying that C++0x was already standardized (and renamed) as C++11. Oh well, at least it works, somehow...
C++11 doesn't work sometimes. Just because there are only a few compilers supports it and this support is still under active development. It's easy to write C++11 code for g++-4.8, which will be impossible to compile with 4.6, for example. I don't see the problem here. g++ 4.8 is already available even for that windoze crap. BTW, libbitcoin has a much nicer/cleaner codebase. I'd like to see new alts based on that. |
|
|
|
Maybe Satoshi need to learn some features from C++0x and boost to write readable code without portability issues. But it's definately not a crap.
It's C++11 already. Oh well, at least it works, somehow... |
|
|
|
Hm... Just checked your sourcecode. It seems too overcomplicated for me, actually. This issue could be resolved with less than 10 additional lines of code. Maybe even 4-6 lines...
Whatever. The whole Bitcoin codebase is crap anyway. |
|
|
|
2 all
Guys, I suppose that you need block hashes caching , this approach would allow you to reduce startup time significantly (30х-100x times at least). This could be done through merging with NVC blockindex code and addition of your network rules.
I think this will resolve all your performance issues.
Already done. Didn't know it was in NVC, though. Would have been easier to pull the changes rather than write it from scratch. |
|
|
|
|
I would leave the target spacing for PoW and PoS as is. It really doesn't matter anyway with this scheme, so why bother.
|
|
|
|
Hmmm its appears to be something todo with the wireless. I've started it on a wired machine here and now its working straight from the get go with no messing of the downloaded files.
What exchanges are you guys using for YAC??
BTer.com has a yac exchange that appears to be active.
cryptsy.com is another, though it's disabled there ATM. |
|
|
|
Nothing happens without that file, the client just sits there with no connections.
Are you 100% sure you don't have any weird firewall that blocks outgoing connections? 'Cause the irc server is up with 131 nodes joined. |
|
|
|
[
{
"addr" : "194.190.198.22:7688",
"services" : "00000001",
"lastsend" : 1389286790,
"lastrecv" : 1389286790,
"conntime" : 1389286368,
"version" : 60005,
"subver" : "/YACoin-WM:0.7.3/",
"inbound" : false,
"releasetime" : 0,
"startingheight" : 371354,
"banscore" : 0
}
]
So far its only found one node from the three you gave me, and doesnt look from the debug that it can find anyone else. try removing the silly yacoin.conf altogether |
|
|
|
|
Show Posts
