Amoveo β₯π§Ώ
Previous messages
I successfully bet against myself in the oracle.
When you restart the node, it will use the default list of peers to start with.
Next messages
4 February 2018
MH
22:59
Mandel Hoff
π
Z
23:00
Zack
The tx fee that you write on the tx comes from the person who signed the tx, otherwise I could send you 1 satoshi to make you lose a tx fee.
23:00
and it goes to the miner.
Denis Repp joined group by link from Group
5 February 2018
I
03:01
Iridescence
Is anybody able to solo mine on a full node in their own machine?
Z
04:19
Zack
In reply to this message
I am able to mine.
Someone is mining blocks so fast that it takes too long for my blocks to propagate through the network. So the fast-miner is getting 100% of the block rewards.
Once it retargets at height 2000, then the blocks will come slower, and we will be able to mine again.
Someone is mining blocks so fast that it takes too long for my blocks to propagate through the network. So the fast-miner is getting 100% of the block rewards.
Once it retargets at height 2000, then the blocks will come slower, and we will be able to mine again.
04:20
I wonder if this means someone wrote GPU code. This would be a lot of CPU energy to throw at a testnet.
I
04:31
Iridescence
I am running the c miner and finding blocks but not getting rewards
04:31
I made a thread about it on reddit
04:32
I'm using just my cpu, an i7 6700k
04:34
The c miner is showing "Found a block. 1 , Found a block. 2" I assume this means that I mined a block
Z
04:34
Zack
Did you put your address in the miner.erl file?
I
04:34
Iridescence
Yeah
04:35
In the Pubkey define
Z
04:36
Zack
is your address BGDX.... ?
I
04:36
Iridescence
Yes
Z
04:37
Zack
it looks like your balance is 518 Veo.
04:38
when I copy-paste your address into the block explorer, it says you have over 600 tokens:
BGDXGmovuTEr3CFfzY0xr0zXS3purkIPPZsljZv4gvQe87K6W8be1IpwDwmqhgcSIlhTvb9Q7pgrl+h2dmXGEXM=
http://159.89.106.253:8080/explorer.html
BGDXGmovuTEr3CFfzY0xr0zXS3purkIPPZsljZv4gvQe87K6W8be1IpwDwmqhgcSIlhTvb9Q7pgrl+h2dmXGEXM=
http://159.89.106.253:8080/explorer.html
04:40
oh, my blockchain only said 518 because I hadn't synced all the blocks.
I
04:40
Iridescence
Oh wait that's not mine
04:41
BGDxLP9PUnumTNtHDAZmE4Dgx1BR9mkc5ocRJGLpL9FiZYYtgsWukSzbrbOl4/H1l1Xu+5wLAfDRObBWRvRc4Uw=
Z
04:41
Zack
so then, the problem is that this person is mining so fast, and the blocks aren't propagating fast enough. So they don't include any of your blocks.
All your blocks are getting "orphaned" as they say in the Ethereum community.
All your blocks are getting "orphaned" as they say in the Ethereum community.
04:41
haha they both start with BGDX
I
04:43
Iridescence
Ah I see
04:43
How pernicious
Z
04:44
Zack
it says you have 5.6 Veo. so I guess you managed to get like 8 blocks mined.
I
04:44
Iridescence
In the true chain, probably
MH
04:45
Mandel Hoff
It's not me mining. Clearing out my full node now. It was kind of dead.
Z
04:46
Zack
This has probably drained all the money from the mining pool.
Since the mining pool pays rewards even if the block doesn't get included.
So you are probably finding blocks and giving them to the mining pool, and then the mining pool doesn't have funds to pay you the reward.
And then the block also doesn't get included.
Since the mining pool pays rewards even if the block doesn't get included.
So you are probably finding blocks and giving them to the mining pool, and then the mining pool doesn't have funds to pay you the reward.
And then the block also doesn't get included.
MH
04:47
Mandel Hoff
Pool says "node is not running!" again. I just created it fresh and synced it. So strange.
Z
04:47
Zack
nope, the pool has lots of money
04:48
In reply to this message
did you change miner.erl for the type of mining you are doing? like, are you connecting to your own node, or to a server?
MH
Z
04:50
Zack
It is weird, the mining pool is telling me that it is sending money to BGDxLP9PUnumTNtHDAZmE4Dgx1BR9mkc5ocRJGLpL9FiZYYtgsWukSzbrbOl4/H1l1Xu+5wLAfDRObBWRvRc4Uw= @IR
MH
04:52
Mandel Hoff
Is that similar to what I reported here?
https://github.com/zack-bitcoin/amoveo/issues/90
Found blocks by miners were failing to pay out.
https://github.com/zack-bitcoin/amoveo/issues/90
Found blocks by miners were failing to pay out.
Z
04:56
Zack
I see part of the problem.
There are txs sitting in the mempool, and the miner is ignoring them while mining.
Another part of the problem is that spend_tx isn't factoring in the mempool when making txs, so the nonce is too low when the server tries making rewards for mining.
There are txs sitting in the mempool, and the miner is ignoring them while mining.
Another part of the problem is that spend_tx isn't factoring in the mempool when making txs, so the nonce is too low when the server tries making rewards for mining.
MH
05:01
Mandel Hoff
Is add_peer still necessary or does the latest node not need that to publish blocks?
Z
05:03
Zack
The network is healthier if you add the peer. It will reduce the load on the server I am maintaining, and it makes the testnet more effective at testing what real conditions will be like.
I think we changed it so it syncs correctly while mining even if you dont' add_peer.
I think we changed it so it syncs correctly while mining even if you dont' add_peer.
MH
05:05
Mandel Hoff
Do I do add_peer(myNode) or add_peer(yourNode)?
Z
05:06
Zack
The way they ignore all the txns and only mine empty blocks is concerning to me.
If they turn off their miner right after the retarget at height 2000, this might freeze the testnet.
If they turn off their miner right after the retarget at height 2000, this might freeze the testnet.
05:07
In reply to this message
The first one.
Everyone else is already on the list. you want to add yourself.
Everyone else is already on the list. you want to add yourself.
05:07
This way when people find out about blocks, they will tell you about it
05:12
is it just me, or did pruning get significantly slower recently?
MH
05:12
Mandel Hoff
Commands not responding. Any suggestions?
05:12
Z
05:13
Zack
I am guessing you have some sort of open parenthasis or quote or something above.
See how it says 16 on every line, it is still accepting the same command.
See how it says 16 on every line, it is still accepting the same command.
MH
05:13
Mandel Hoff
Z
05:13
Zack
"like this
is never closed
is never closed
MH
05:15
Mandel Hoff
I get a ton of spam like this on my node - not sure if it's "slower".
05:15
That was it. Not sure why or how but typing "()." fixed it.
05:15
Good advice.
05:19
05:19
Why is my balance going up and up and up? It started way under 34, and it keeps going up and up. I swear I'm not mining. I don't even know the pub key for my full node. I have the keys.db but not the address.
05:20
I'm nearly sure no one is mining on my node, because it's been down/broken for a couple days. I think it's valid now, but still likely no one is mining in to my node.
Z
05:21
Zack
In reply to this message
You are syncing. It shows what your balance was at the point in time that you synced to.
We are passed block 1200 now.
We are passed block 1200 now.
05:21
1344
MH
05:24
Mandel Hoff
05:24
I think I'm synced:
Z
05:26
Zack
looks like they are finding a block every 10 seconds.
MH
05:27
I swear I'm not mining, and I have no idea why my node balance is increasing. π
Z
05:27
Zack
So we will retarget in about 1 hour 40 minutes.
At that time we will know whether this is an attack, or someone experimenting.
At that time we will know whether this is an attack, or someone experimenting.
05:27
is your address BGDXGmovuTEr3CFfzY0xr0zXS3purkIPP... ?
05:28
that is the address getting all the block rewards
05:28
you only have 97, so that isn't yoru address.
MH
05:28
Mandel Hoff
How can I know my node's address? I saved the keys.db but have never known its address.
Z
05:28
Zack
api:pubkey().
MH
05:29
Mandel Hoff
Ok - not me then. Still +7 veo in 26 blocks. How/why?
05:30
Nice! That got me this:
base64:encode(keys:pubkey()).
base64:encode(keys:pubkey()).
Z
05:30
Zack
if you have a mining pool, you would get like 0.3 per block mined.
ill look at the txs in the recent blocks.
ill look at the txs in the recent blocks.
MH
05:30
Mandel Hoff
Oh, ha! - yours is better
Z
05:33
Zack
In reply to this message
This is so weird.
I looked at every block in that range, and there isn't any tx sending money to you.
I looked at every block in that range, and there isn't any tx sending money to you.
MH
05:35
I'm so rich. π
05:38
1 Block per 10 sec is about 4mh/s. I think I was mining about 2mh/s on 7 threads, so that might just be two faster computers.
Z
05:38
Zack
The attacker has 868 tokens.
I
05:39
Iridescence
How do you check mh/s?
Z
MH
05:43
Mandel Hoff
I added a couple lines to the c code when I was trying to verify nonce duplication and difficulty versus expected blocktime.
05:44
05:44
BGhNv807cWJClkAadBzCeX23JPC0NFHuvAu0AgG7tV8uFxkhj+EdkGPMgR2gz8ZjNuB/VIztWAhoMACkm1MXdeY=
05:46
Light wallet says NaN for my address.
Z
05:47
maybe you are running a version of Amoveo from a long time ago?
05:47
or the merkle trie library might be old
MH
05:47
Mandel Hoff
My keys.db is old but I did a fresh clone.
Z
05:47
Zack
git pull doesn't update the merkel tree dependency
MH
05:47
Mandel Hoff
If I start from scratch, do I just need to rm -rf amoveo? or are there other things I need to clear out?
05:48
I did a full folder delete before a fresh clone.
Z
05:48
Zack
if you rm -rf amoveo, then you are fine.
You only really need to get rid of the
You only really need to get rid of the
amoveo/_build/ subfolder
MH
05:49
Mandel Hoff
Yes, rm -rf amoveo and then make prod-build, then copy keys.db back, and then make prod-go and make prod-attach. Then sync:start(). and RICH!
05:50
With a fresh clone after the rm and before the prod-build (of course).
Z
06:09
Zack
ok, I am syncing a fresh clone too, to see if I can reproduce the bug.
06:24
BGDXGmovuTEr3CFfzY0xr0zXS3purkIPPZslj....
is the address for the mining pool I am running.
is the address for the mining pool I am running.
06:28
ok, it seems like there is a problem with potential_block.erl if the blocks are being mined too quickly. It isn't including the txs from the mempool when it makes blocks.
06:28
I was able to crash whoever was mining so quickly by messing with the potential block on the miner.
MH
06:29
Mandel Hoff
Didn't realize the pool had an address separate from the node and miner.
Z
06:29
Zack
It doesn't
That is the address for the amoveo full node that is connected to that mining pool.
That is the address for the amoveo full node that is connected to that mining pool.
06:44
In reply to this message
I just finished syncing a fresh clone, I could not reproduce the bug you described.
Z
07:04
Zack
I think it is fixed. If you mine with the mining pool, it should be paying out rewards again.
Even if we are mining so fast as 10 seconds per block.
Even if we are mining so fast as 10 seconds per block.
G
07:04
Gonzalo
Hi, it's not me as well. I have a full node on a cheap vps that is stopped at least 2 weeks ago, and also mining at home on the light wallet, but was stopped hours ago. 10 minutes ago I started mining again on the light wallet and found 2 blocks but the balance is not increasing.
07:05
ok I'll see if I find a new block
Z
MH
07:07
Mandel Hoff
Please let me know when the tx are being rolled in to blocks instead of mining the empty blocks. I think I got my "remove spend" working (in the most embarrassing design) way possible. Balances haven't updated yet, but looked like success on the request.
Z
07:08
Zack
In reply to this message
txs are being included in blocks.
I don't know what you mean by "remove spend".
I don't know what you mean by "remove spend".
MH
07:11
Mandel Hoff
Typo - meant "remote spend" external to erlang or light wallet (but effectively uses the js from the light wallet). Couldn't figure out how to "sign" without the js. All other "sign" libraries were giving me different results.
G
07:12
Gonzalo
ok, just found a new block and got paid
Z
07:13
Zack
In reply to this message
it uses sha256 to hash the data.
Probably the difficult part is to serialize the tx to binary exactly the same way that JS does.
Probably the difficult part is to serialize the tx to binary exactly the same way that JS does.
MH
07:15
Mandel Hoff
My first spend of 0.001 didn't seem to go through yet, but my second 1.0 spend went through. Blances updated 1.0 but not the 0.001 yet.
G
07:21
Gonzalo
the "attack" has stopped
07:23
ah, or maybe Zack made some changes because light wallet is not mining
Z
07:24
Zack
does anyone still have the blocks?
07:28
ive got the first 1589
MH
07:30
Mandel Hoff
I'll check my node.
G
07:30
Gonzalo
I've 2 open browsers, one is stuck at 1889. I've refresehd the other one and is updating but *very* slowly, goes from 10 to 10 and I've to push the button "get more headers" every time
07:30
380
07:30
390
07:30
420
07:30
440
Z
07:31
Zack
that is how many blocks are on the server.
07:31
I am feeding blocks to the server from my laptop
G
07:32
Gonzalo
oh ok I understand
MH
07:32
Mandel Hoff
My node is back at 1453. π₯
Z
07:32
Zack
but I only have 1589
G
07:34
Gonzalo
ok, and now 1589
07:35
I see all my balance that was earned before block 1589, but off course not what I got after that block
Z
07:36
Zack
right
G
07:36
Gonzalo
π
I
15:07
Iridescence
Blockchain stuck at height 2000?
Deleted joined group by link from Group
Z
17:47
Zack
In reply to this message
It says 2229 is the height on the block explorer, so it looks like it did not get stuck at 2000.
2000 is when retargetting of difficulty happens, it should be much harder to mine now.
2000 is when retargetting of difficulty happens, it should be much harder to mine now.
G
20:24
Gonzalo
Well, It's at 2238 right now on the explorer, but the light wallet only updates up to 2000
Z
G
20:25
Gonzalo
π
Z
20:38
Zack
the light node is calculating that the new difficulty should be 2 higher than the difficulty calculated by the full node.
There must be some rounding error in the retargetting algorithm in javascript.
There must be some rounding error in the retargetting algorithm in javascript.
6 February 2018
Z
00:40
I set up the erlang full node so it has 2 sync_modes now.
in
in
The default mode is normal.
in
sync_mode:quick() it is able to download blocks faster, but you cannot mine or do channel stuff.in
sync_mode:normal you can mine and do channels stuff, but it downloads blocks more slowly.The default mode is normal.
00:40
I rewrote the part of the api that the light node uses for downloading the headers. Depending on where you are located and your internet connectivity, the light node could sync faster for you now.
00:46
Now full nodes can share headers even if they have not yet downloaded the blocks.
Deleted joined group by link from Group
Z
04:20
Zack
Hi Tom Bennett.
Tom is working on a project that involves an LMSR market maker. So he is thinking about similar problems as us.
Tom is working on a project that involves an LMSR market maker. So he is thinking about similar problems as us.
04:20
Deleted Account
hey Zack, thanks!
04:21
the project I'm working on is called Futuur -- you can check it out at futuur.com
G
04:22
Gonzalo
Hi Tom
04:22
Deleted Account
it's a play-money prediction market that we'll be launching officially later this month, but we are beginning to explore market maker algorithms as a replacement for our current (super-simple) algorothm, so looking to learn more about LMSR and variants...
Z
04:23
Zack
Does your market use single price batching?
04:24
like, matching a bunch of trades simultaniously in periodic time increments
04:24
and all the trades that match at the same time match at the same price
04:25
Deleted Account
no, we use play money so have built a simple alg that is not concerned with matching or losing money... so there's no matching whatsoever at the moment
Z
04:26
Zack
Do you have a leader-board for the people who are most accurate?
04:27
Deleted Account
yep
Z
04:27
Zack
cool
04:36
Deleted Account
we are in the very early days of exploring market maker and matching approaches, in preparation for adding real money down the road. but for the moment, we are just a game, and have focused on making a prediction market that is fast, social, and user-friendly. of course, any and all feedback is welcome!
Z
04:36
Zack
You have a lot of movies stuff to bet on. This is great. I bet it is easier to get people engaged when it is about movies.
04:37
Deleted Account
curious to hear more about how you guys have approached this. you mentioned that you are not using LMSR... what did you end up going with?
04:37
yeah, movies and sports are hot categories, even if not the most useful from a saving-the-world perspective π
Z
04:37
Zack
oh, your long term goal is to use money that has value. Great, at that point I will have lots of ideas that could be valuable to you.
04:38
It seems like channels are important for making blockchains scalable. Maybe you have heard of the lightning network?
04:39
Our main goal with Amoveo is to move markets into the channels.
But channels have weird limitations, so I built a programming language in the channels, that way we can try programming different kinds of markets.
But channels have weird limitations, so I built a programming language in the channels, that way we can try programming different kinds of markets.
04:40
Deleted Account
yep, the *likely* end scenario is a layer of monetized predictions using a blockchain-based approach, on top of the gamified/social version of the app; and progressively decentralized governance.
Z
04:40
Zack
If you use a centralized market, then you can just build a giant centralized computer to run the market and you wont need scalability tools like channels.
04:41
Deleted Account
right, we are 100% centralized right now so scalability etc not an immediate concern
04:41
re LN, yes I'm familiar with it, but I'm not a programmer myself and can't say I fully understand how it works π
04:42
do you see Amoveo as working together with other prediction markets out there, eg augur, gnosis, etc?
Z
04:42
Zack
The LN is built out of lots of small 2-party relationships called channels.
Both participants in the channel can know the state of their channel, but they can't prove it to anyone else.
Both participants in the channel can know the state of their channel, but they can't prove it to anyone else.
04:43
I think Amoveo will be much more affordable to use than Augur or Gnosis. I wrote comparisons of the various oracle projects here https://github.com/zack-bitcoin/amoveo/blob/master/docs/progress_reports/February_2018.md
04:44
It seems like you will eventually want to fork the software from some existing oracle blockchain project
04:44
or build on top of an existing blockchain project
04:45
Deleted Account
the comparison is super interesting, thanks
04:46
yes, we are in research mode in terms of the architectural approach on the blockchain front; our immediate focus is launching the gamified version and building out our userbase
Z
04:47
Zack
Maybe we can make a insured crowdfunding contract on Amoveo to pay you guys to hook up amoveo as a back end to your pretty betting interface. https://github.com/zack-bitcoin/amoveo/blob/master/docs/use-cases-and-ideas/insured_crowdfund.md
04:49
Deleted Account
hmmm, interesting idea
MH
05:07
Mandel Hoff
Welcome Tom
05:08
Deleted Account
thanks Mandel, and hello Gonzalo!
G
05:16
Gonzalo
:-) π
Z
05:59
Zack
I just made garbage collection about 100x faster.
Z
06:38
Zack
I set up the erlang full node so it changes sync mode automatically.
Blocks sync much faster than yesterday.
Blocks sync much faster than yesterday.
MH
06:40
Mandel Hoff
Sounds great
G
06:41
Gonzalo
Also it seems that light wallet updates height much faster
Z
06:42
Zack
In reply to this message
yes, I made the api for downloading headers faster, so the light wallet is faster for most people, depending on the quality of their internet connection.
G
06:43
Gonzalo
nice! For me is about 5x faster
Z
06:44
Zack
very cool.
For me it is less than 2x faster.
For me it is less than 2x faster.
06:44
The server is in Germany
G
06:45
Gonzalo
Ok, I'm in Spain
Z
07:53
it syncs like 3x faster
Z
08:13
Zack
I fixed
If you want to turn a node off while it is syncing, it is probably safest to stop the syncing first before you do
sync:stop().If you want to turn a node off while it is syncing, it is probably safest to stop the syncing first before you do
api:off().
Anson joined group by link from Group
A
14:45
Anson
Hi, I am new to this project
14:45
just found it from Aeternity project
14:46
actually what problems do Amoveeo aiming to solve ?
Z
Z
19:29
Zack
I made a governance oracle to increase the block reward.
19:34
We can start betting in it at height 2300
Z
20:13
Zack
For the last 100 blocks, we averaged at 900 seconds per block. (which is 15 minutes per block)
The target blocktime is 600 seconds per block. (which is 10 minutes per block.)
It looks like retargetting at height 2000 was successful.
Betting in the governance oracle will start in about 4 hours.
The target blocktime is 600 seconds per block. (which is 10 minutes per block.)
It looks like retargetting at height 2000 was successful.
Betting in the governance oracle will start in about 4 hours.
20:16
Deleted Account
Zack do you still believe yesterday there was somebody GPU mining? Are you going to code a GPU miner before launch?
Z
20:18
Zack
In reply to this message
no, Mandel did a calculation to show that it actually wasn't very much hash power for modern CPU.
My laptop is kind of old, so it seemed like a lot to me.
My laptop is kind of old, so it seemed like a lot to me.
20:18
Deleted Account
Good ππ»
Z
20:38
Zack
I want to make a gpu miner before launch, but I might not.
For now I am focused on these tasks:
1) encryption library in javascript for lightning payments.
2) when an oracle is closed, the channel manager in javascript should clean up old expired contracts.
Once I finish those things, then I will make the 2 week announcement, and start working on the gpu miner.
For now I am focused on these tasks:
1) encryption library in javascript for lightning payments.
2) when an oracle is closed, the channel manager in javascript should clean up old expired contracts.
Once I finish those things, then I will make the 2 week announcement, and start working on the gpu miner.
20:49
Deleted Account
Thanks for sharing
MH
23:44
Mandel Hoff
How do we participate in the "governance oracle to increase the block reward" test?
Z
23:45
Zack
oh, yes, the height is finally high enough
23:47
Deleted Account
Hey guys I've joined the group about a week ago just got around to reading all the messages :) liking the idea any way I can help? I'm fairly technical but still need to go through the github
23:47
I'm a fullstack and android dev
Z
23:47
Zack
https://github.com/zack-bitcoin/amoveo/blob/master/tests/market.py#L68
Here is an example of using the api to bet in an oracle.
Here is the api defined: https://github.com/zack-bitcoin/amoveo/blob/master/apps/ae_http/src/api.erl#L261
Here is an example of using the api to bet in an oracle.
Here is the api defined: https://github.com/zack-bitcoin/amoveo/blob/master/apps/ae_http/src/api.erl#L261
23:48
In reply to this message
http://159.89.106.253:8080/wallet.html
Any ideas on how to make this better?
Do you know how to make it into a secure android app?
Right now you are trusting the server who you download the javascript from.
Any ideas on how to make this better?
Do you know how to make it into a secure android app?
Right now you are trusting the server who you download the javascript from.
23:50
Deleted Account
Ill have a look this weekend for the site and need to go through the github because I've yet to fully figure out what that page does tbh
Z
23:52
Deleted Account
Ok thks Zack
MH
23:52
Mandel Hoff
Your Oracle link says "useful for testing market from light node" - is there a UI for the Oracle bet? I'm not seeing it. Or, is that something I should run with erlang commands on the full node?
I think Oracle bets are "on chain", so it's not part of the channel-markets. Do I need to "lock funds" or do anything before the bet?
I think Oracle bets are "on chain", so it's not part of the channel-markets. Do I need to "lock funds" or do anything before the bet?
Z
23:55
Zack
In reply to this message
The full node api lets you make any of the txs.
There are also commands to make and sign any tx seperately
There are also commands to make and sign any tx seperately
23:55
The bet is locking funds.
7 February 2018
Z
00:01
Zack
I've got to sync my node, then I'll try it out
00:09
This is the command I used to make a bet
api:oracle_bet(<<3:256>>, 1, 33826072).
00:10
the
I bet about the minimum amount of money possible.
If you want to bet in the same direction as me, you need to do at least twice as much money.
If you want to bet in the opposite direction, then any amount should work.
1 means I am betting for true, which I think means the governance variable will increase.I bet about the minimum amount of money possible.
If you want to bet in the same direction as me, you need to do at least twice as much money.
If you want to bet in the opposite direction, then any amount should work.
00:12
Here is the current state of the oracle
(ae_core_prod@zacks-MacBook-Pro)12> trees:dict_tree_get(oracles, <<3:256>>).
trees:dict_tree_get(oracles, <<3:256>>).
{oracle,<<0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,
0,0,0,...>>,
0,
<<227,176,196,66,152,252,28,20,154,251,244,200,153,111,
185,36,39,174,65,228,100,155,147,76,164,...>>,
2300,1,0,
<<31,49,201,13,172,247,17,36,69,4,231,217,145,177,123,
69,115,82,19,115,100,...>>,
<<4,200,213,238,176,0,178,197,62,12,103,124,195,238,151,
179,32,210,118,10,...>>,
3304,1,50}
00:13
It looks like we can't close this oracle until 3304.
00:15
and if someone bets against me, the timer will restart.
1000 blocks is about 1 week.
Maybe this was too long of a period of time.
Luckily this 1000 block waiting period is a governance variable. So if 1 week is too long to wait, we can shorten this waiting period:
1000 blocks is about 1 week.
Maybe this was too long of a period of time.
Luckily this 1000 block waiting period is a governance variable. So if 1 week is too long to wait, we can shorten this waiting period:
(ae_core_prod@zacks-MacBook-Pro)13> trees:dict_tree_get(governance, minimum_oracle_time).
trees:dict_tree_get(governance, minimum_oracle_time).
999
MH
00:18
Mandel Hoff
What is the <<3:256>> value?
Z
00:19
Zack
it is a 32 byte binary that uniquely identifies this oracle.
in Hex it is like ....00000000003
in Hex it is like ....00000000003
MH
00:19
Mandel Hoff
Ok, is 0 or 2 false? Test case link shows a "2".
Z
00:20
Zack
I think 1 is highier, 2 is lower, and 3 is "no change".
3 is if you think the governance variable should stay the same.
3 is if you think the governance variable should stay the same.
MH
00:21
Mandel Hoff
Is this a gamble or a vote?
Z
00:21
Zack
gambling.
00:22
The 6th thing in the oracle state is currently a "1". This means that if the oracle doesn't get any more bets, it will make the governance variable higher once it is closed.
MH
00:22
Mandel Hoff
Who/what determines the result if these are just bets?
Z
00:22
Zack
If you put lots of bets in for state "2", then that value will change from a "1" to a "2", and the timer resets.
MH
00:23
Mandel Hoff
That sounds like a "pay for votes" scenario rather than a gamble of chance.
Z
00:23
Zack
miners can censor transactions. The same consensus mechanism that maintains blocks is also doing work here.
00:24
The gambling mechanism is embedded inside of a bigger game. You need to take the bigger game into consideration when analyzing it. I wrote some docs about the oracle in the docs folder.
MH
00:25
Mandel Hoff
How much higher or lower will the value go after this closes? +/-10%, +/-100%, other?
Z
00:26
Zack
https://github.com/zack-bitcoin/amoveo/blob/master/apps/ae_core/src/consensus/trees/governance.erl#L30
this governance value limits how much it can change by at a time.
this governance value limits how much it can change by at a time.
MH
00:26
Mandel Hoff
Can I adjust my bet after making it? Add more to it or cancel it?
Z
00:26
Zack
The last value in the oracle definition is a 50, so the current governance oracle will increase the block reward by 50 units
00:27
You can make more bets.
If your existing bet is next to be matched, then you could bet against yourself and it would cancel out your risk. But you would have twice as much money locked into the oracle until it is closed.
If your existing bet is next to be matched, then you could bet against yourself and it would cancel out your risk. But you would have twice as much money locked into the oracle until it is closed.
MH
00:28
Mandel Hoff
50.0 veo or 50 satoshi?
Z
00:28
Zack
https://github.com/zack-bitcoin/amoveo/blob/master/docs/design/transaction_types.md
These are all of the tx types.
There is no tx type for canceling orders in an oracle, so it is not possible to cancel your order.
These are all of the tx types.
There is no tx type for canceling orders in an oracle, so it is not possible to cancel your order.
00:29
Governance variables are stored in a sort of scientific notation. Increasing one by 256 makes it twice as big. So I guess increasing it by 50 makes it around 17% bigger?
00:31
(ae_core_prod@zacks-MacBook-Pro)15> governance:tree_number_to_value(1620).looks like it can get about 50% bigger.
governance:tree_number_to_value(1620).
100227592
(ae_core_prod@zacks-MacBook-Pro)16> governance:tree_number_to_value(1670).
governance:tree_number_to_value(1670).
157546925
(ae_core_prod@zacks-MacBook-Pro)17>
MH
00:31
Mandel Hoff
Z
00:31
Zack
I guess each unit makes it 1% bigger
MH
00:33
Mandel Hoff
I have 2.1 veo, and tried to bet 1 veo. Doesn't look like it worked, but not really sure.
Z
00:33
Zack
"bad return value 'check'"
really odd, I haven't seen this error.
really odd, I haven't seen this error.
00:33
did you sync all the blocks?
00:34
you are trying to make the block reward smaller
MH
00:34
Mandel Hoff
00:34
I think so - block:height(). says I'm at 2304.
00:35
Just betting against you for testing purposes.
Z
00:36
Zack
Usually it at least give a line number. the error is giving me nothing to work with.
MH
00:37
Mandel Hoff
Not this?
in function gen_server:call/2 (gen_server.erl, line 204)
in function gen_server:call/2 (gen_server.erl, line 204)
Z
00:38
Zack
that is an erlang otp built-in library that I am using in like 30 different erlang files.
00:41
api:oracle_bet(<<3:256>>, 2, 1).I successfully bet against myself in the oracle.
00:42
now I bet against myself twice.
00:44
ok, I was able to reproduce your error
00:44
once I bet against myself with an entire Veo.
MH
00:45
Sounds right. I think I made a small bet that worked
Z
00:47
Zack
I found where it crashes. https://github.com/zack-bitcoin/amoveo/blob/master/apps/ae_core/src/consensus/trees/orders.erl#L259
00:47
It looks like I added a line to make it crash there for testing purposes, and I forgot to remove it.
G
00:47
Gonzalo
FYI Light Wallet is not mining right know
Z
00:53
In reply to this message
I pushed a new version of the code to github to fix this problem. Now we can bet in the oracle even in large amounts
MH
00:55
Mandel Hoff
Did it take my oracle bets? Local node says my balance is around 1.3 after I mined 1 block. That makes sense if one of my 1 veo bets went through.
G
00:56
Gonzalo
Mandel would you accept a few VEOs from me? π
00:56
Free money!
MH
00:56
Mandel Hoff
You want me to send funds? What address?
G
00:57
Gonzalo
No, I want to send YOU funds
MH
00:58
Mandel Hoff
Sure -
BKajhhwIjvYIW79i0nT8Rx+6Yy5zSjgA+bHLFMoxQtr6pLdoIdBODzgpsAvXOeLDdCgoUfkXMX08T0PDb3v6MZ8=
BKajhhwIjvYIW79i0nT8Rx+6Yy5zSjgA+bHLFMoxQtr6pLdoIdBODzgpsAvXOeLDdCgoUfkXMX08T0PDb3v6MZ8=
Z
G
MH
00:59
Mandel Hoff
Yes, that was my test that went through after two failed 1.0 bets.
G
00:59
Gonzalo
Just sent 10 VEO to BKajh...
01:01
I suppose that you'll see them after the next block is minted, isn't it?
Z
01:01
Zack
I see the 10 VEO tx sitting in the mempool
MH
01:02
Mandel Hoff
My local node is reporting its balance is wrong - why?
01:02
01:02
That 2.6 is what I'd expect if the 1.0 veo bets failed.
Z
01:03
Zack
Maybe there is a tx stuck in your mempool that isn't being shared?
api:balance accounts for the txs in the mempool
api:balance accounts for the txs in the mempool
01:03
tx_pool:get(). tells you about your mempool
G
01:04
Gonzalo
ah, I was wrong... I CAN actually see it on the explorer.html, but not on the wallet.html. Explorer always goes like 1 block ahead wallet
Z
01:05
Zack
In reply to this message
right, the explorer.html includes the mempool, because you are just trusting them to tell you.
wallet.html uses merkle proofs and block headers to have cryptographic proof of what it says. So it can't tell you anything until it is in a block.
wallet.html uses merkle proofs and block headers to have cryptographic proof of what it says. So it can't tell you anything until it is in a block.
G
01:06
Gonzalo
Thanks Zack, now I realize this is the second time I'm asking the same question, sorry!
MH
01:07
Mandel Hoff
01:07
This is my mempool. I mined 2 blocks, but they aren't pushing over to your node. Why?
Z
01:08
Zack
I just updated the testnet using a soft fork so that we could make more types of bets in the oracle.
01:08
and while doing so I broke database on the server
01:09
I will have to resync it from block 0.
MH
01:09
Mandel Hoff
Oh - sure, I didn't update since yesterday. Makes sense. Want me to stay online so you can sync to me, or should I start over?
Z
01:11
Zack
could you push the blocks to the server before you reboot?
G
01:11
Gonzalo
In reply to this message
Nice! Best way to advance and go further is to break things. π
MH
Z
01:12
Zack
It is ok if you can't.
sync:start().
01:12
I can do it from my end eventually, but my internet isn't so great.
01:13
I should probably just launch a second server.
Or maybe I can run 2 full nodes from the same server.
Or maybe I can run 2 full nodes from the same server.
MH
01:13
Mandel Hoff
Ok, I typed sync:start(). It said "ok" here, but nothing else happening. I usually get a bunch of "absorb blocks" and "prune blocks" and "timeout" messages.
Z
01:14
Zack
my laptop is feeding blocks to the server. it says "give blocks" over and over.
Deleted joined group by link from Group
03:57
Deleted Account
when will this launch
Z
04:08
Zack
In reply to this message
The plan is to launch before the end of April.
I will give at least 2 weeks warning before we launch.
I will give at least 2 weeks warning before we launch.
Deleted joined group by link from Group
Z
04:18
Zack
In reply to this message
https://github.com/zack-bitcoin/amoveo start here to learn about Amoveo. Currently there is a testnet with mining.
Here is a web-wallet light node that connects to the testnet: http://159.89.106.253:8080/wallet.html
You can mine from the browser.
Here is a web-wallet light node that connects to the testnet: http://159.89.106.253:8080/wallet.html
You can mine from the browser.
04:19
Before launch I want to get the channels to work better in the light node. The blockchain is done.
04:21
Deleted Account
so i was looking through aeternity and it's nice project looks better then eos and cardano to me and upon research i found you on reddit so im noob please shed some light that why are you not working on improving aeternity and but started new project . this chat is very long can anyone summaries situation to me thanks.
Z
04:34
Deleted Account
i read the article and i wish i had the power to help you. i am gonna follow you as long as you work on blockchain. Without you aeternity is dead i guess although the project is so amazing.
04:49
so i need to do is to click on start mining and thats it..? is there any mining pools or nvidia gpu miner for this.?
MH
04:51
Mandel Hoff
Yes, that's it. Difficulty is low enough that your browser will hit blocks. No "share" pools yet. You are a "pool of one". No gpu miner is available.
Z
04:59
Zack
In reply to this message
There are mining pools, so you don't have to run a full node to mine. But for now you only get paid when you find a block.
Here is the software to run a mining pool: https://github.com/zack-bitcoin/amoveo-mining-pool
This is the fastest miner that I know about:
https://github.com/zack-bitcoin/amoveo-c-miner
It is much faster than the browser miner.
There is no GPU mining yet, but hopefully we will have it soon.
Here is the software to run a mining pool: https://github.com/zack-bitcoin/amoveo-mining-pool
This is the fastest miner that I know about:
https://github.com/zack-bitcoin/amoveo-c-miner
It is much faster than the browser miner.
There is no GPU mining yet, but hopefully we will have it soon.
06:12
Deleted Account
thank you i will install ubunto
06:17
these software are for linux right
MH
06:18
Mandel Hoff
Yes, you can stand up a full node and local mining pool. Any miner can point to any publicly facing mining pool. Yes, Ubuntu.
Z
06:20
Zack
I have only been doing testing using OSX for the last month.
I used Ubuntu before that, so it probably works on Ubuntu as well.
Erlang is supposed to make it easy to run on any system.
You probably don't want to use the mining pool software. Either connect to my mining pool, or else run Amoveo + C-miner on your own without a mining pool.
I used Ubuntu before that, so it probably works on Ubuntu as well.
Erlang is supposed to make it easy to run on any system.
You probably don't want to use the mining pool software. Either connect to my mining pool, or else run Amoveo + C-miner on your own without a mining pool.
06:20
oh, the server is running ubuntu, so I guess ubuntu is being tested too.
MH
06:25
Mandel Hoff
I've only used Ubuntu for testing so far. I tried to erlang on Windows but that didn't go well. I have no idea how to get that working, but I'd be happy if it did work that way.
Z
06:25
Zack
I fixed the javascript encryption library.
This was necessary for lightning payments.
This was necessary for lightning payments.
MH
06:25
Mandel Hoff
Nice! π
Z
06:26
Zack
In reply to this message
I can't use windows. I get too angry. Hopefully someone else can do it.
06:31
Deleted Account
In reply to this message
i will solo mine for a week on browser to see if it produces any results. If i don't see results i will join your mining pool with whatever software is there for ubunto .
MH
07:10
Mandel Hoff
Zack, would it make sense to open several browsers and have each one mine? Or, is there still that issue where each browser will start with the same nonce?
Z
07:12
Zack
In reply to this message
They start with the same nonce? that is really bad if it is true.
I think using multiple browsers, or mutiple tabs, might be an effective way to increase the hashpower of some machines. I have not tested it.
I think using multiple browsers, or mutiple tabs, might be an effective way to increase the hashpower of some machines. I have not tested it.
07:13
https://github.com/zack-bitcoin/amoveo/blob/master/apps/ae_http/priv/external_web/miner.js#L85
No, we ignore x[2], the nonce from the server, and instead we generate a random nonce.
No, we ignore x[2], the nonce from the server, and instead we generate a random nonce.
MH
07:20
Mandel Hoff
Ok, good. I just remembered the server passing out nonce identical to bhash, but sounds like that is ignored.
Z
18:09
Zack
Why amoveo will win:
I have the strength that only comes to a person who is protecting their friends.
Once my grandfather's friend fell into a grain bin of canola seeds, and began drowning. My grandfather alone lifted the walls of the bin, and all the canola spilled onto the ground, saving the man's life. The next day, my grandfather and 3 other strong men together could not lift the walls of the bin.
I have the strength that only comes to a person who is protecting their friends.
Once my grandfather's friend fell into a grain bin of canola seeds, and began drowning. My grandfather alone lifted the walls of the bin, and all the canola spilled onto the ground, saving the man's life. The next day, my grandfather and 3 other strong men together could not lift the walls of the bin.
r
18:51
raketralph
What the fuck
G
18:52
Gonzalo
wow
Z
19:25
Zack
A person who is working to protect someone they care about can accomplish much more than an average person.
G
Z
19:55
Zack
A friend of mine managed to escape Venezuela a couple months ago. He is working to save up money to help more of his family escape.
It is really hard for these people just to eat.
I want to build Amoveo to help people like this.
Amoveo will give us warnings if medicine or food will become unavailable in a part of the world. That way people can prepare or leave.
Amoveo can warn us which political leaders might cause food or medicine to be unavailable. This way it will be harder for the most dangerous people to gain the support of the masses.
Amoveo will let people buy insurance, so if food does become unavailable in their region, they will have enough money to afford to escape.
It is really hard for these people just to eat.
I want to build Amoveo to help people like this.
Amoveo will give us warnings if medicine or food will become unavailable in a part of the world. That way people can prepare or leave.
Amoveo can warn us which political leaders might cause food or medicine to be unavailable. This way it will be harder for the most dangerous people to gain the support of the masses.
Amoveo will let people buy insurance, so if food does become unavailable in their region, they will have enough money to afford to escape.
MF
20:01
Mr Fox
Sticker
Not included, change data exporting settings to download.
π, 12.2 KB
20:01
ae listed on binance
20:20
Deleted Account
i have mined 8 blocks on browser but when i click on check balance it doesnt show anything
MH
20:21
Mandel Hoff
Zack, node keys locked?
20:24
Post your address and I'll send you veo
Z
21:00
Zack
The node was already unlocked.
Maybe the javascript miner doesn't know how to deal with higher difficulty?
Maybe the javascript miner doesn't know how to deal with higher difficulty?
21:01
In reply to this message
Thanks for testing it out.
Maybe you need to sync the headers.
Are you sure you are still using the same key that you mined with?
Maybe you need to sync the headers.
Are you sure you are still using the same key that you mined with?
21:08
Deleted Account
I opened 2 browsers to mine n check balance didnt show anything and then i refreshed the browser it worked. What is current block reward ?
Z
21:10
Zack
the block reward in the mining pool is 0.7
refreshing causes it to download the headers again. So clicking "get more headers" probably would have solved your problem a little faster.
refreshing causes it to download the headers again. So clicking "get more headers" probably would have solved your problem a little faster.
21:11
The browser wallet is a light node that downloads and verifies the longest chain of headers.
MH
21:19
Mandel Hoff
Opening a new browser defaults to a new account, right? Did you load your account to the browser after you opened them and before mining?
21:21
Deleted Account
Yes i did load my public key and it showed my balance afterwards.
MH
21:27
Mandel Hoff
Are you mining on the 159.x node? I have a 13.x node but it's behind the soft fork currently and that won't mine right if you are using my node.
21:50
Deleted Account
yes i'm mining on 159.x node so im cool ? my mining speed is 1.19 mhash but its nowhere mining 1 block per minute its alot more.
Xortalus joined group by link from Group
EW
21:52
Eli W
Zack saw your message mentioned token distribution is exponential across time , where can find the distribution design/ graph ?
MH
22:01
Current difficulty is 0x1b42 which means 1 in 168,820,736 hashes. At 1.19mh/s, you'd get a block about every 141 sec.
22:02
Is that per browser and you have a few open? 3+ @ 1.19mh/s would be sub-minute block time at current difficuty.
Z
22:04
Zack
In reply to this message
False.
The block reward is set by the governance mechanism. I cannot predict what the community will decide is a good block reward.
The block reward is set by the governance mechanism. I cannot predict what the community will decide is a good block reward.
EW
MH
22:18
Mandel Hoff
Zack, I really want my priv key from my keys.db file. How can I get that?
Z
22:19
Zack
I think the easiest way would be to insert a print statement in keys.erl so that it gets printed out while it is decrypted.
22:20
or you could write something to have the unencrypted version written to a file
MH
22:22
Mandel Hoff
Is the keys.db format something other tools might be able to consume?
Z
22:23
Zack
no.
22:25
I wrote the encryption library we are using, it is customized for amoveo alone.
MH
22:26
Mandel Hoff
Is there an api to check current block reward? I browsed tests and api, but didn't see one yet.
Z
22:28
Zack
(ae_core_dev1@zacks-MacBook-Pro)6> trees:dict_tree_get(governance, block_reward).
trees:dict_tree_get(governance, block_reward).
100227592
(ae_core_dev1@zacks-MacBook-Pro)7>
22:29
You can also access it from the api
22:30
{proof, treename, id, hash}
22:30
the external api
MH
22:30
Mandel Hoff
Thank you.
22:31
I think I'm having trouble loading my keys.db. If I do this, should it get my updated keys.db?
22:32
sync:stop().
api:off().
halt().
make prod-stop
upload new keys.db
make prod-go
make prod-attach
api:pubkey().
api:off().
halt().
make prod-stop
upload new keys.db
make prod-go
make prod-attach
api:pubkey().
Z
22:32
Zack
yes, that will tell you the new pubkey
22:33
here is an example of getting a governance value from the external api:
MH
22:33
Mandel Hoff
Ok, maybe I lost the keys.db I was hoping to restore. That could be the case. π±
Z
22:34
Zack
good thing it is just a testnet
MH
22:36
Mandel Hoff
If I edit a .erl file, do I have to rebuild - or is Erlang interpreted?
Z
22:37
Zack
erlang is similar to java. The files need to be compiled to erlang byte code.
22:38
api:off().halt().make prod-restart
MH
22:42
Mandel Hoff
22:42
I just got a fresh clone for node and pool. I sync'd to 2387 and mined two blocks. Block found not advancing height. Any idea why?
Z
22:44
Zack
it looks like we are on 2388. are you sure it didn't work?
MH
22:45
Mandel Hoff
Yes, pretty sure someone else mined 2388. My node has 0 balance and I just absorbed block 2388 after calling sync:start().
22:46
Also, do I have to call sync:start() to get blocks? It doesn't appear to be syncing automatically.
Z
22:46
Zack
it should sync automatically if you are mining.
22:46
I am eating now, I will try reproducing this error later.
Thanks for the helpful error report.
Thanks for the helpful error report.
MH
22:48
Mandel Hoff
Sure, I'm not seeing that ["mining_data"] print on the node currently if that matters. My miner did receive a work item though, so that part worked.
22:48
Later is perfectly fine.
MH
23:25
Mandel Hoff
23:25
Check balance on the Light Wallet might be broken:
23:51
Deleted Account
can u provide a guide how to install c miner on ubunto and how to run it
MH
23:58
Deleted Account
when i type sh build_ubuntu.sh i get this error
23:58
build_ubuntu.sh: 10: build_ubuntu.sh: erlc: not found
build_ubuntu.sh: 12: build_ubuntu.sh: erl: not found
build_ubuntu.sh: 12: build_ubuntu.sh: erl: not found
8 February 2018
MH
00:02
Mandel Hoff
Need some dependencies updates:
00:02
sudo apt-get install erlang libncurses5-dev libssl-dev unixodbc-dev g++ git erlang-base-hipe make
00:32
Deleted Account
after attaching to the node mine:start(). this was my command now its absorbing blocks
00:52
Deleted Account
make prod-attach
Attaching to /tmp/erl_pipes/ae_core_prod@usman-GE70-2PC/erlang.pipe.3 (^D to exit)
Attaching to /tmp/erl_pipes/ae_core_prod@usman-GE70-2PC/erlang.pipe.3 (^D to exit)
00:53
is this ok..?
MH
00:58
Mandel Hoff
Is that on your full node or miner? Seems ok - looks like you are just attached to the erlang prompt.
01:01
Deleted Account
im attached to the node
01:02
now i want to mine and im confused
MH
01:03
Mandel Hoff
Check that your node is sync'd with block:height().
01:04
If good, then you should be able to startup the pool and then launch the miner pointing at your pool (config in miner.erl Peer maybe?)
01:05
There are 3 separate "services": Node, Pool, and Miner. Need to stand them up in that order. Sounds like your Node is good - double check it's sync'd with the network sync:start(). and block:height().
01:07
Deleted Account
usman@usman-GE70-2PC:~/amoveo$ make prod-attach
Attaching to /tmp/erl_pipes/ae_core_prod@usman-GE70-2PC/erlang.pipe.3 (^D to exit)
--> [Quit]
usman@usman-GE70-2PC:~/amoveo$ sync:start()
> bash: syntax error: unexpected end of file
usman@usman-GE70-2PC:~/amoveo$ api:height().
bash: syntax error near unexpected token `.'
usman@usman-GE70-2PC:~/amoveo$ api:height()
> bash: syntax error: unexpected end of file
usman@usman-GE70-2PC:~/amoveo$ mine:start()
> bash: syntax error: unexpected end of file
usman@usman-GE70-2PC:~/amoveo$ make prod-build
Attaching to /tmp/erl_pipes/ae_core_prod@usman-GE70-2PC/erlang.pipe.3 (^D to exit)
--> [Quit]
usman@usman-GE70-2PC:~/amoveo$ sync:start()
> bash: syntax error: unexpected end of file
usman@usman-GE70-2PC:~/amoveo$ api:height().
bash: syntax error near unexpected token `.'
usman@usman-GE70-2PC:~/amoveo$ api:height()
> bash: syntax error: unexpected end of file
usman@usman-GE70-2PC:~/amoveo$ mine:start()
> bash: syntax error: unexpected end of file
usman@usman-GE70-2PC:~/amoveo$ make prod-build
01:07
this is complicated
MH
01:07
Mandel Hoff
Type make prod-attach to get to the erlang prompt.
Z
01:12
Zack
In reply to this message
mine:start().
You can mine wihout the mining pool, and without the c- miner.
You can also mine from the javascript light wallet.
You can mine wihout the mining pool, and without the c- miner.
You can also mine from the javascript light wallet.
01:13
it*
01:14
usman@usman-GE70-2PC:~/amoveo$ make prod-attach
Attaching to /tmp/erl_pipes/ae_core_prod@usman-GE70-2PC/erlang.pipe.3 (^D to exit)
--> mine:start()
Unknown command
Attaching to /tmp/erl_pipes/ae_core_prod@usman-GE70-2PC/erlang.pipe.3 (^D to exit)
--> mine:start()
Unknown command
MH
01:14
Mandel Hoff
At the erlang prompt, type stuff like:
block:height().
api:balance().
sync:start().
You can't start the miner from the node's erlang prompt.
block:height().
api:balance().
sync:start().
You can't start the miner from the node's erlang prompt.
Z
01:16
mine:start().
MH
01:16
Mandel Hoff
Ok, I'm wrong then - I didn't think you can start the miner from the node's prompt.
Z
01:16
Zack
There are 3 different miners.
MH
01:17
Mandel Hoff
Ok, I think I've only tried the js and c miner then. Didn't know there was a 3rd.
Z
01:19
Zack
Trying to get Usman to install all this different software is no good. You are confusing him needlessly.
He specifically asked to run the c-miner. So he doesn't need the mining pool, and he doesn't need an amoveo full node.
He specifically asked to run the c-miner. So he doesn't need the mining pool, and he doesn't need an amoveo full node.
MH
01:20
Mandel Hoff
I agree. I linked the amoveo-c-miner and the dependency list.
01:24
Deleted Account
ok i installed the dependency
01:24
this is after i run c miner
01:24
/media/usman/72DE89DCDE89994D/amoveo-c-miner-master/amoveo-c-miner-master$ sh build_ubuntu.sh
miner.erl:86: Warning: variable 'E' is unused
miner.erl:97: Warning: no clause will ever match
Erlang/OTP 18 [erts-7.3] [source] [64-bit] [smp:8:8] [async-threads:10] [hipe] [kernel-poll:false]
Eshell V7.3 (abort with ^G)
miner.erl:86: Warning: variable 'E' is unused
miner.erl:97: Warning: no clause will ever match
Erlang/OTP 18 [erts-7.3] [source] [64-bit] [smp:8:8] [async-threads:10] [hipe] [kernel-poll:false]
Eshell V7.3 (abort with ^G)
Z
01:26
Deleted Account
so its mining now..?
Z
01:26
Zack
No.
Can you understand the directions that Mandel linked to?
Can you understand the directions that Mandel linked to?
01:28
If you cannot understand English well enough to read the directions, then just stick with the javascript miner.
Which language do you know better? Maybe I'll have the c miner instructions translated
Which language do you know better? Maybe I'll have the c miner instructions translated
01:33
Deleted Account
i will stick to JavaScript miner for now and wait for gpu miner. π
Z
01:36
Zack
In reply to this message
I didn't intend that the same machine would have all 3 at the same time.
Mining on the same machine that is supporting a public Api seems bad.
The Api would probably be all slow because your computer is busy mining.
Mining on the same machine that is supporting a public Api seems bad.
The Api would probably be all slow because your computer is busy mining.
01:38
In reply to this message
Node + pool makes a mining pool.
Node + miner makes a solo miner.
Miner alone is good for connecting to mining pools.
Node alone is useful.
Node + miner makes a solo miner.
Miner alone is good for connecting to mining pools.
Node alone is useful.
Z
02:00
Zack
In reply to this message
This is such a bad bug, I can't imagine how this happened.
The proofs data structure is starting with a binary, so weird. and it says it is empty even though it is not empty.
accounts:get(base64:decode("BCjdlkTKyFh7BBx4grLUGFJCedmzo4e0XT1KJtbSwq5vCJHrPltHATB+maZ+Pncjn\
fvt9CsCcI9Rn1vO+fPLIV4="), 155).
accounts:get(base64:decode("BCjdlkTKyFh7BBx4grLUGFJCedmzo4e0XT1KJtbSwq5vCJHrPltHATB+maZ+Pncjnfvt9CsCcI9Rn1vO+fPLIV4="), 155).
{<<132,149,109,44,108,189,1,161,236,245,211,205,3,174,112,
111,125,124,69,122,173,128,84,121,164,202,132,113,...>>,
empty,
[<<47,131,133,89,213,135,216,27,176,53,37,181,10,224,44,
75,152,69,26,172,149,116,49,198,21,...>>,
{<<217,134,71,58,8,59,95,195,64,56,64,229,218,58,41,211,
158,134,106,60,189,220,137,...>>,
<<0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,...>>,
<<0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,...>>,The proofs data structure is starting with a binary, so weird. and it says it is empty even though it is not empty.
02:01
ive only been working on the javascript light node for days now. I can't remember the last time I could have created this bug.
02:05
oh, I think maybe I deleted this account using the new "delete account" feature, and I forgot.
So it should be fine.
So it should be fine.
MH
02:20
Mandel Hoff
This is happening to several of my accounts. Like this one should have like 34 Veo:
BPA3r0XDT1V8W4sB14YKyuu/PgC6ujjYooVVzq1q1s5b6CAKeu9oLfmxlplcPd+34kfZ1qx+Dwe3EeoPu0SpzcI=
BPA3r0XDT1V8W4sB14YKyuu/PgC6ujjYooVVzq1q1s5b6CAKeu9oLfmxlplcPd+34kfZ1qx+Dwe3EeoPu0SpzcI=
02:20
How did all of my accounts get deleted?
02:21
(By "all", I mean like 4.)
Z
02:31
Zack
I think that quick_sync is causing this. The trees aren't being updated right
Z
02:59
Zack
ok, I fixed it.
block:top() was terribly slow. by rewriting sync to avoid that function, it goes super fast without corrupting the tree.
block:top() was terribly slow. by rewriting sync to avoid that function, it goes super fast without corrupting the tree.
02:59
ill download the blocks to my laptop, then restart the server with better code, then push the blocks to the server. It should be fixed up in 15 minutes or so.
03:02
probably this will undo some blocks. Since there are txs that are invalid.
Z
03:21
Zack
I updated the code and pushed the blocks, but it is still having problems.
03:28
there are no problems. Some blocks just got undone because they were built on an invalid database.
It is fixed.
It is fixed.
MH
03:48
Mandel Hoff
Was my account deleted during this bug, or should it be unaffected? I don't really care about the balance, but I'm a little worried that my account disappeared:
BPA3r0XDT1V8W4sB14YKyuu/PgC6ujjYooVVzq1q1s5b6CAKeu9oLfmxlplcPd+34kfZ1qx+Dwe3EeoPu0SpzcI=
BPA3r0XDT1V8W4sB14YKyuu/PgC6ujjYooVVzq1q1s5b6CAKeu9oLfmxlplcPd+34kfZ1qx+Dwe3EeoPu0SpzcI=
Z
03:48
Zack
do you remember what height this account was created? I think over 100 blocks were undone
MH
03:50
Mandel Hoff
Probably between block 1 and 400. That was my primary miner address from the start.
Z
MH
05:15
Mandel Hoff
π I was starting to think I was going crazy. Glad you were able to confirm some sanity.
Rob joined group by link from Group
Z
05:20
Zack
At some point between 2348 and 2363 it is breaking.
It is so weird.
I looked at all those blocks, none of them have a tx that should cause this.
When I look at the merkle tree, it seems like all the merkle data is set up as if your account still exists, but when you try to load the account data it says "empty".
It is so weird.
I looked at all those blocks, none of them have a tx that should cause this.
When I look at the merkle tree, it seems like all the merkle data is set up as if your account still exists, but when you try to load the account data it says "empty".
MH
05:24
Mandel Hoff
I'm confident you'll figure it out. I don't care if I lose the wallet, but it's a scary bug.
Z
05:52
Zack
I stopped the blockchain at height 2365, and looked at all the previous 20 blocks. They all make your account look empty.
05:55
There are less than 100 accounts. the merkle tree already has tests showing we can add tens of thousands of elements, and then create proofs for any of them.
05:58
I turned off pruning, and it still happens the same way.
MH
06:01
Mandel Hoff
I'd love to provide more clues, but I have no idea where it disappeared to. I'm sure that I didn't delete the account or spend it down to 0.
06:02
I used it for a few 0.001 to 2 spend tests and mostly used it as my miner account that my full node would pay out to.
Z
06:04
Zack
im going to see if it always disappears at the same height. Maybe it is some kind of race condition.
I am sure you didn't delete it, there are no delete account txs in the range where it disappeared.
Besides, when it disappears, it erases history too, so it looks like it never existed.
When you delete accounts with a tx, you can still see their history.
I only see txs that spend 0.7 at a time, and the amount of money you had isn't a multiple of 0.7, so I am sure you didn't spend the money to 0.
And besides, I think when you spend the money to 0, it doesn't delete your account.
You get a reward if you delete it instead of leaving it around. The reward is like 1/2 the cost of making an account.
I am sure you didn't delete it, there are no delete account txs in the range where it disappeared.
Besides, when it disappears, it erases history too, so it looks like it never existed.
When you delete accounts with a tx, you can still see their history.
I only see txs that spend 0.7 at a time, and the amount of money you had isn't a multiple of 0.7, so I am sure you didn't spend the money to 0.
And besides, I think when you spend the money to 0, it doesn't delete your account.
You get a reward if you delete it instead of leaving it around. The reward is like 1/2 the cost of making an account.
MH
06:06
Mandel Hoff
Sounds good. Again, I'm glad you don't think I'm crazy on this bug - very strange.
Kevin Roberts BWD invited Deleted Account
Z
08:31
Zack
I turned off pruning correctly this time, and it fixed the error.
I think the pointers to leaves aren't updating correctly in some edge case.
I will make more tests for pruning, to identify and fix it.
I think the pointers to leaves aren't updating correctly in some edge case.
I will make more tests for pruning, to identify and fix it.
MH
08:32
Mandel Hoff
π sounds great.
Z
21:07
Zack
It seems like I introduced this error when I made the quick-mode sync.
Z
21:26
Zack
ok, I fixed it for downloading, uploading is still a problem.
Oliver joined group by link from Group
Z
22:47
Zack
ok, it is all fixed.
The problem was actually in writing batches of data to the merkle tree.
There was a place where we were copying a pointer, when we should have copied the data.
This was a problem, because later when we deleted the data from the old block, it meant the newer block was pointing at nothing.
The problem was actually in writing batches of data to the merkle tree.
There was a place where we were copying a pointer, when we should have copied the data.
This was a problem, because later when we deleted the data from the old block, it meant the newer block was pointing at nothing.
22:48
you need to get a new version of the merkle tree library to run Amoveo.
git pull isn't enough.rm -rf _build followed by make prod-restart should get you the new version of the code, but be careful. This will delete your key.db. So keep a copy of your keys file first.
MH
23:10
Mandel Hoff
Wonderful! I'll update very soon. Great bug fix.
MH
23:33
Mandel Hoff
Did you push to github yet? Last update says 2 hours ago, which is well before you reported it's completely fixed.
G
9 February 2018
Z
00:14
Zack
In reply to this message
Yes. Since the server downloaded from github and the server says you have your money.
MH
01:17
Mandel Hoff
Awesome fix. Looks good.
G
01:41
Gonzalo
In reply to this message
I had to delete not only _build folder but also all the amoveo folder and start over. Maybe this is because I installed it many days (weeks) ago. Now syncing ok π
Z
03:23
Zack
In reply to this message
It is good you pointed this out. Otherwise who knows how long it would have taken me to realize.
G
18:44
Gonzalo
Zack FYI I'm mining on the light wallet but my balance is not updating: BBr9r23dAd/+dx2/xIBMGq4SwhJdKJEVUlfmEaD98orhHfaWr6YNErYPbYt+4EF1KkctQkZ3ysRWe03WT3aU16A=
Z
18:46
Zack
In reply to this message
yes, it seems like when I fixed syncing, I messed up block creation. So when blocks are made it isn't including txs. So the tx spending you money isn't being included.
I will fix this soon.
I will fix this soon.
G
18:47
Gonzalo
π
Z
19:49
Zack
In reply to this message
my mistake. The problem was that I didn't unlock the keys on the server.
I do this so often, I wonder how I can stop myself.
I do this so often, I wonder how I can stop myself.
G
19:50
Gonzalo
π
10 February 2018
Deleted joined group by link from Group
Deleted joined group by link from Group
Mihaly Barasz joined group by link from Group
Z
04:51
Zack
Today I fixed Amoveo so that when you are running the test version, you can modify the javascript without having to rebuild the code.
This will allow us to run tests in javascript much faster, so we can finish implementing the lightning payments and market bets sooner, and then do the main-net launch.
I am embarrassed I didn't do this sooner. I think Mandel even suggested it.
This will allow us to run tests in javascript much faster, so we can finish implementing the lightning payments and market bets sooner, and then do the main-net launch.
I am embarrassed I didn't do this sooner. I think Mandel even suggested it.
MH
05:10
Mandel Hoff
Sounds awesome! Keep up the great work.
Z
06:41
Zack
After that change, things are going more than 10x faster.
There is some code that I need to rewrite from erlang to javascript in order to deal with lightning payments in the channels.
There is some code that I need to rewrite from erlang to javascript in order to deal with lightning payments in the channels.
11 February 2018
KR
03:36
Kevin Roberts BWD
im i doing this right?
(ae_core_prod@BevoTest)23> mine:start().
trade peers
talk_helper weird response
talk_helper weird response
talk_helper weird response
talk_helper weird response
ok
talk_helper weird response
talk helper fail
[-7,13,92,84,57]
get their top header
(ae_core_prod@BevoTesmy height is less
(ae_core_prod@BevoTest)24> block absorber died!
(ae_core_prod@BevoTest)23> mine:start().
trade peers
talk_helper weird response
talk_helper weird response
talk_helper weird response
talk_helper weird response
ok
talk_helper weird response
talk helper fail
[-7,13,92,84,57]
get their top header
(ae_core_prod@BevoTesmy height is less
(ae_core_prod@BevoTest)24> block absorber died!
Z
03:37
Zack
it is better to sync before mining.
sync:start().
KR
03:37
Kevin Roberts BWD
will do thanks
03:39
(ae_core_prod@BevoTest)27> sync:start().
trade peers
ok
[-7,13,92,84,57]
[-7,159,89,106,253]
get their top header
my height is less
(ae_core_prod@BevoTest)28> block absorber died!
?
trade peers
ok
[-7,13,92,84,57]
[-7,159,89,106,253]
get their top header
my height is less
(ae_core_prod@BevoTest)28> block absorber died!
?
Z
03:39
Zack
Maybe you need to download a newer version of the code
KR
03:40
Kevin Roberts BWD
ok
KR
Z
04:21
Zack
yes
KR
04:23
Kevin Roberts BWD
new vm
cloned from link
===> Starting relx build process ...
===> Resolving OTP Applications from directories:
/root/amoveo/_build/prod/lib
/root/amoveo/apps
/usr/lib/erlang/lib
===> Resolved ae_core-0.1.0
===> Including Erts from /usr/lib/erlang
===> release successfully created!
root@BevoTest:~/amoveo# make prod-go
root@BevoTest:~/amoveo# make prod-attach
Attaching to /tmp/erl_pipes/ae_core_prod@BevoTest/erlang.pipe.1 (^D to exit)
(ae_core_prod@BevoTest)4>
(ae_core_prod@BevoTest)4>
(ae_core_prod@BevoTest)4>
(ae_core_prod@BevoTest)4> api:height().
2706
(ae_core_prod@BevoTest)5> sync:start().
trade peers
ok
talk_helper weird response
talk_helper weird response
talk_helper weird response
talk_helper weird response
talk_helper weird response
talk helper fail
[-7,13,92,84,57]
get their top header
(ae_core_prod@BevoTest)6>
cloned from link
===> Starting relx build process ...
===> Resolving OTP Applications from directories:
/root/amoveo/_build/prod/lib
/root/amoveo/apps
/usr/lib/erlang/lib
===> Resolved ae_core-0.1.0
===> Including Erts from /usr/lib/erlang
===> release successfully created!
root@BevoTest:~/amoveo# make prod-go
root@BevoTest:~/amoveo# make prod-attach
Attaching to /tmp/erl_pipes/ae_core_prod@BevoTest/erlang.pipe.1 (^D to exit)
(ae_core_prod@BevoTest)4>
(ae_core_prod@BevoTest)4>
(ae_core_prod@BevoTest)4>
(ae_core_prod@BevoTest)4> api:height().
2706
(ae_core_prod@BevoTest)5> sync:start().
trade peers
ok
talk_helper weird response
talk_helper weird response
talk_helper weird response
talk_helper weird response
talk_helper weird response
talk helper fail
[-7,13,92,84,57]
get their top header
(ae_core_prod@BevoTest)6>
Z
04:23
Zack
Looks like you downloaded the headers successfuly
04:24
block:height(). will tell you how many blocks you have
04:25
I just cloned a copy, and synced as the first command, and it worked correctly.
04:26
Maybe it is a missing dependency?
04:32
Maybe your internet connection is so much faster than mine, that it is causing a race condition when you try to sync.
04:32
Is your internet particularly fast? is this a server you are renting?
04:34
Install instructions are here: https://github.com/zack-bitcoin/amoveo/blob/master/docs/getting-started/linux_dependencies.md
KR
04:37
Kevin Roberts BWD
yes from Digital Ocean and yes i will share the file
04:42
i dont see it
04:42
oot@BevoTest:~/amoveo/_build/prod/rel/ae_core/log# tree
.
βββ erlang.log.1
βββ run_erl.log
0 directories, 2 files
root@BevoTest:~/amoveo/_build/prod/rel/ae_core/log# ls
erlang.log.1 run_erl.log
.
βββ erlang.log.1
βββ run_erl.log
0 directories, 2 files
root@BevoTest:~/amoveo/_build/prod/rel/ae_core/log# ls
erlang.log.1 run_erl.log
Z
04:44
Zack
When it said "block absorber died!" it must have created the crash.log .
If you rebuild the project, then it erases the crash.log
If you rebuild the project, then it erases the crash.log
KR
04:45
Kevin Roberts BWD
ill try and recrash it
12 February 2018
Phenofix joined group by link from Group
Deleted joined group by link from Group
Deleted joined group by link from Group
Deleted joined group by link from Group
Deleted joined group by link from Group
13 February 2018
Deleted joined group by link from Group
Z
16:54
Zack
In reply to this message
Someone contacted me today saying they will try writing it in Cuda.
17:20
Deleted Account
In reply to this message
Doesn't that mean this will only work on Nvidia cards? AMD guys will kill themselves hahaha
Z
17:22
Zack
In reply to this message
Sure, and we will have opencl miners soon enough.
I suggested opencl to him, but he seems to prefer cuda.
Since I am not paying him, it is not like I can order him around.
Cuda gpu miners is still a lot better than cpu.
I suggested opencl to him, but he seems to prefer cuda.
Since I am not paying him, it is not like I can order him around.
Cuda gpu miners is still a lot better than cpu.
17:26
He isn't very reliable, I think there is a 10% chance he will deliver a working miner.
Deleted joined group by link from Group
Dino Mihalopoulos joined group by link from Group
14 February 2018
Adam DAugelli joined group by link from Group
KR
00:08
Kevin Roberts BWD
Where is a good place to find someone to develope a gpu miner? and we cant sha-256 hash correct?
Z
00:09
Zack
Hiring someone would be more effort than writing it.
BK
00:13
Big Kurkowski
Nvidia first would be fine, most the rigs I'll point this way are going to be gtx.
Deleted joined group by link from Group
MH
03:33
Mandel Hoff
Zack, is there a function to go from a number to that 2 byte scientific format used in the difficulty value?
Z
04:05
Zack
In reply to this message
https://github.com/BumblebeeBat/pink_crypto/blob/master/src/pow.erl#L92
sci2int and int2sci
sci2int and int2sci
MH
04:18
Mandel Hoff
Thanks - int2sci. That's it.
04:32
I just rewrote it in c# and had to add a Math.Ceiling here to make it work:
var a = Math.Ceiling(Math.Log(x, 2)) - 1
var a = Math.Ceiling(Math.Log(x, 2)) - 1
04:33
Not sure if that's needed for the erl version or not.
04:33
Might be a bug, might not.
Z
04:36
Zack
I implemented the exponent and log functions using only deterministic operations. I did not use floating point. Probably you have a bug yes.
04:41
I can't believe I am calling lg() one line, and undoing it the next. such an obvious mistake
MH
04:41
Mandel Hoff
Yes, c# math.pow and math.log output double format, so floating points happen there.
Z
04:42
Zack
I think we can implement exponent and lg equivalent to mine much faster if we use bit-shifts.
04:51
In reply to this message
I guess it wasn't a mistake. I had to do this to make it the exact opposite of pair2int.
lune joined group by link from Group
fcu joined group by link from Group
Deleted joined group by link from Group
14:15
Deleted Account
Hello all
A friend invited me here as Iβm an enthusiast about the technology provided by BC. Anyway Iβm a total dick in computer and maths lol but Iβm very curious and I would like to know more about your project if you accept, of course to simplify it a lot in explanation for me :)
A friend invited me here as Iβm an enthusiast about the technology provided by BC. Anyway Iβm a total dick in computer and maths lol but Iβm very curious and I would like to know more about your project if you accept, of course to simplify it a lot in explanation for me :)
14:15
Many thanks
MH
17:44
Mandel Hoff
You'll probably want to start with this document. It's pretty low tech and big on concepts:
https://github.com/zack-bitcoin/amoveo/blob/master/docs/white_paper.md
https://github.com/zack-bitcoin/amoveo/blob/master/docs/white_paper.md
WL
21:20
Wilson Lau
I constantly got "talk_helper socket closed remotely" after i sync my node, it that normal?
Z
21:30
Zack
In reply to this message
Yes. Erlang uses fail fast technique. So as long as it is working, you can ignore all the errors.
WL
21:34
Wilson Lau
In reply to this message
i see. i also run the c-miner connecting to my node, after a while it return "server gave confusing response".
And the node return:
got this data
["mining_data"]
syncing with peer [-7,[-7,52,226,34,101],8080]
trade peers
potential block died!get headers
[-7,52,226,34,101]
[-7,159,89,106,253]
get their top header
And the node return:
got this data
["mining_data"]
syncing with peer [-7,[-7,52,226,34,101],8080]
trade peers
potential block died!get headers
[-7,52,226,34,101]
[-7,159,89,106,253]
get their top header
Z
21:35
Zack
Yes, the blockchain is frozen at a particular height.
I wrote a test which reproduces the problem, and I am working on a solution now.
I wrote a test which reproduces the problem, and I am working on a solution now.
MH
21:49
Mandel Hoff
Sounds great. I was just updating to test, but I'll wait for the fix.
WL
21:49
Wilson Lau
Thanks!
15 February 2018
scott_l | DeFi Pulse joined group by link from Group
Mike joined group by link from Group
Z
04:47
Zack
the test is passing.
I updated the server with the new code.
We should be able to mine blocks again.
I updated the server with the new code.
We should be able to mine blocks again.
04:59
Someone is already mining. Who is it?
MH
05:06
Mandel Hoff
Great. I'll update now.
05:20
Height went up +2 and my node absorbed blocks. Good fix!
05:29
Deleted Account
In reply to this message
I read the WP but I probably missed some points lol (about dev) anyway the scalability and LN are very interesting. :)
MH
05:31
Mandel Hoff
Zack, please do keys unlock on your pool if you haven't yet
Z
05:48
We have done this a couple times now. The testnet broke, and we managed to fix it without losing our history.
We will need to do this to keep the main net alive. So it is a good sign that we can do it now.
We will need to do this to keep the main net alive. So it is a good sign that we can do it now.
MH
05:57
Mandel Hoff
Yes. I was amazed balances were preserved when you fixed the previous bug where my balances were showing NaN.
Z
05:59
Zack
I wouldn't have even noticed that bug if it wasn't for you Mandel.
G
08:10
Gonzalo
π
16:12
Deleted Account
Zack may we have some info about when the VEO will be able to mine? And as Iβm a dick in mining is will be some easy way for such guy like me to proceed and mine VEO in an easy way? :)
Z
16:14
Zack
In reply to this message
You can practice mining on the test net now.
The easiest way to mine is from the browser light node.
There will be 2 weeks warning before main net launch. The goal is to launch before the end if April.
The easiest way to mine is from the browser light node.
There will be 2 weeks warning before main net launch. The goal is to launch before the end if April.
16:20
Deleted Account
Ok perfect so even a noob in mining like could succeed smt good with VEO??? Thatβs another very good point for your project ππΌ
Btw could you have some tuto or procedure to proceed? Maybe a link for the test mining net?
Btw could you have some tuto or procedure to proceed? Maybe a link for the test mining net?
Z
17:01
Deleted Account
Ok sorry for the dumb question Zack
17:01
Animation
Not included, change data exporting settings to download.
168.0 KB
G
17:51
Gonzalo
π
Z
18:48
Deleted Account
Thanks Γ lot zack
xcpdx joined group by link from Group
16 February 2018
Z
00:30
Zack
I was able to do a lightning payment from the javascript light node. I updated the server to support this.
MH
01:03
Mandel Hoff
Nice! When you do updates like this, is that why my node stops syncing? I left my node on all night and it hasn't updated since 2729. I tried a sync:start() and it says "already synced with this peer". If it's a versioning thing, that'd make sense.
01:04
"Lightning payment" is instant payment between channel participants, right? Or no?
Z
01:05
Zack
This is a problem that is happening because there is only one node in the network.
I set it up so that if a node becomes unavailable, you stop syncing with it an delete it from your list of peers.
Normally this isn't a problem, because when the node re-joins the network it shares it's ip address, which you can download from any of your other peers.
If there is only one peer, it breaks the way you describe.
I set it up so that if a node becomes unavailable, you stop syncing with it an delete it from your list of peers.
Normally this isn't a problem, because when the node re-joins the network it shares it's ip address, which you can download from any of your other peers.
If there is only one peer, it breaks the way you describe.
01:06
In reply to this message
Channels allow you to make instant payments to your channel partner.
A "lightning payment" is when 2 ore more channel payments are linked together. So if you and I both have a channel with the same server, then we can pay each other.
A "lightning payment" is when 2 ore more channel payments are linked together. So if you and I both have a channel with the same server, then we can pay each other.
MH
01:09
Mandel Hoff
Wow, that's very powerful. Sounds great.
01:10
Is there a way my node can recover? Sync:start() didn't update my height to yours. I'm fine with rm -rf if that's necessary.
Z
01:10
Zack
api:off().
halt().
make prod-restart
When you restart the node, it will use the default list of peers to start with.
MH
01:12
Mandel Hoff
Oh, I get it. The one peer in the list is its self and not your node.
Z
01:12
Zack
I should probably add your IP to the default list, so you don't have to manually add it so often
01:12
or maybe ill set it up so every node tries to add itself
MH
01:13
Mandel Hoff
Not necessary to add my IP. I'm in azure and don't use static IPs, so they can change when I reboot.
01:14
I may bring up a 2nd node in the next few days instead.
01:18
Is there a graceful way to shutdown the pool? api:off() is not a thing on the pool.
Z
01:20
Zack
Is there is a problem with the instructions in the readme? https://github.com/zack-bitcoin/amoveo-mining-pool
01:20
halt().
MH
01:24
Mandel Hoff
Thanks. That worked. I recalled that resulting in the "Node is not running!" behavior on a previous occasion, but worked today.
Z
01:25
Zack
yes. this is the command for turning it off. if you try to attach.sh while it is turned off, the correct message is "node is not running!"
G
02:12
Gonzalo
Zack I intend to run a full node at least for 1 year, if you want to add the static IP to the peer list: 85.217.170.211
Z
G
02:50
Gonzalo
π
MH
03:31
Mandel Hoff
Would it be ok for me to use the same keys.db file on 2+ nodes?
Z
03:31
Zack
yes.
MH
05:15
Mandel Hoff
Are payment txs being picked up and added to blocks? I have made several, and I don't think the recipient balance is getting them. (No rush. Can't test more until later.)
Z
05:20
Zack
I see that the top block only has a coinbase tx, and that it was mined by someone other than my mining pool.
The server's tx pool is currently empty.
The server's tx pool is currently empty.
05:22
I made a spend tx, and it synced to the different nodes correctly.
05:25
It is height 2898 now. So if 2899 contains my tx, then that means it works. I spent to this pubkey BCjdlkTKyFh7BBx4grLUGFJCedmzo4e0XT1KJtbSwq5vCJHrPltHATB+maZ+Pncjnfvt9CsCcI9Rn1vO+fPLIV4=
Z
06:23
Zack
Block 2899 does contain the tx I made. So this confirms that spend txs from the full internal api are working correctly.
MH
06:26
Mandel Hoff
Yes, I had one tx go through. Thanks for testing. Not sure where the rest are going.
Z
06:26
Zack
I got the markets working from the light node.
This is the final task I wanted to complete before making the 2-week announcement for the main net.
This is the final task I wanted to complete before making the 2-week announcement for the main net.
MH
06:28
Mandel Hoff
That's awesome!
G
06:29
Gonzalo
nice!
Z
06:29
Zack
I am looking at my schedule, I am not certain if March 1st will be a good time for me. I will think a little more before making the announcement.
06:35
In reply to this message
I also notice that block 2899 was mined by my mining pool. Maybe you found a bug where txs are getting dropped from the mining pool when someone else mines a block.
MH
06:40
Mandel Hoff
I'll test more later and watch with api:mempool().
G
06:43
Gonzalo
Right now my full node is stuck at height 2910 while yours is at 2913. Even if I do api:sync().
MH
06:51
Mandel Hoff
Sync:start ().
G
Z
07:27
Zack
In reply to this message
Are you using the most recent version of the code?
Is it possible that you shut it off incorrectly, or tried turning it on when it was already on?
Is it possible that you shut it off incorrectly, or tried turning it on when it was already on?
07:29
I just turned on the node on my laptop, and it is syncing correctly.
Z
07:47
Zack
I am thinking we need a forum to act as a market place for trading VEO with other cryptocurrencies.
Does Reddit seem like a good choice for this to you?
https://www.reddit.com/r/AmoveoTrader/
To limit risk: you either send each other small amounts of cryptocurrency back and forth over and over, or you can use a trusted third party.
Does Reddit seem like a good choice for this to you?
https://www.reddit.com/r/AmoveoTrader/
To limit risk: you either send each other small amounts of cryptocurrency back and forth over and over, or you can use a trusted third party.
G
07:50
Gonzalo
In reply to this message
Yes I made a new reinstall a few hours ago. Now I don't know what was wrong. I stopped the node and after a make prod-restart is working fine again, thanks
Z
G
07:56
Gonzalo
Thank you. BTW, are there any other commands on the c-miner apart from miner:start(). ??
Z
07:57
Zack
In reply to this message
yes.
After you have killed the c-miner, there are still some zombie processes using up your cpu. You can kill the zombies with
After you have killed the c-miner, there are still some zombie processes using up your cpu. You can kill the zombies with
sh clean.sh
07:57
also you can configure the c-miner by changing some text at the top of miner.erl
07:58
It is all in the readme https://github.com/zack-bitcoin/amoveo-c-miner
G
07:58
Gonzalo
ok, but anything else? Is it possible to ask for the balance? Or found blocks?
Z
07:59
Zack
the c-miner doesn't know what a balance is.
It is not keeping a record of how many blocks it found.
It is not keeping a record of how many blocks it found.
G
07:59
Gonzalo
ok thanks
Z
08:03
Zack
In reply to this message
Here is a good place to check your balance: http://159.89.106.253:8080/wallet.html
08:04
you can make a watch-only wallet if you don't want to put your private key in the browser.
MH
08:04
Mandel Hoff
I think I know why my txs were being lost... is the first payment to a new address is somehow different that a normal payment? Why/how? If I give you an address, how do you know which spend approach to use?
G
08:05
Gonzalo
oh, I had to change killall for pkill on clean.sh, killall is not found on my system (running Ubuntu 16.04.3 LTS)
MH
08:05
Mandel Hoff
I don't have a better idea than the Reddit Trade forum. I don't like or dislike that idea - no opinion.
Z
08:07
In reply to this message
I think I set up api:spend to do whichever is necessary.
create_account_tx is for making a new account.
spend_tx is for giving money to an existing account.
They are different because when you create an account you need to pay a fee for using up more of the expensive consensus state. When you spend you aren't using up any consensus state.
create_account_tx is for making a new account.
spend_tx is for giving money to an existing account.
They are different because when you create an account you need to pay a fee for using up more of the expensive consensus state. When you spend you aren't using up any consensus state.
08:09
In the light node it looks like I didn't make it automatically choose for you yet. you have to do either spend or create_account correctly.
G
08:10
Gonzalo
sure, it's just change "killall amoveo_c_miner" for "pkill amoveo_c_miner" on clean.sh. Anyway my system is running a minimal installation, probably others won't have any problem.
MH
08:11
Mandel Hoff
Ok, pretty sure that's my problem. I got confused because "new address miners" were being paid out fine, so I didn't realize spend was different.
Z
MH
08:21
Mandel Hoff
Yes, nearly done.
Z
08:21
Zack
cool, I hope you make it open source
MH
08:21
Mandel Hoff
Lots left to do of course but nearly beta tester ready.
Z
08:24
Zack
You should share your work so I can see it
MH
08:25
Mandel Hoff
Sure. Probably have it public facing by Tuesday
Z
08:26
Zack
are you going to make it compatible with the c-miner, or release a fork of the c-miner that is compatible?
MH
08:28
Mandel Hoff
Probably. Change is very small. Just need to give address with get work and accept share diff instead of nonce
G
09:02
Zack there's something wrong with explorer.html on the full node, maybe is that is not updated, is it possible?
Z
09:03
Zack
What is wrong with it? looks fine at first glance.
G
09:04
Gonzalo
When I visit localhost:8080/explorer.html from the node, and I put "localhost" on the "channel node IP", click on "update height" and get 2882
09:05
If I change and put your node IP, 159.89.106.253, I get the actual size: 2917
09:06
if I ask on erlang "api:height()." I will get 2917
09:06
quite weird
Z
09:06
Zack
that part is supposed to be invisible, it is supposed to be the same as what it says in the url.
It is because I am sharing a library with wallet.html
I should really clean it up.
It is because I am sharing a library with wallet.html
I should really clean it up.
09:06
if you put in "localhost", then it connects to the erlang full node running on your machine. If you have one.
G
Z
09:07
Zack
you can type "localhost" into this one: http://159.89.106.253:8080/explorer.html
and it will act the same way
and it will act the same way
G
09:10
Gonzalo
please have a look:
09:10
Z
09:11
Zack
the picture is too compressed, I can't see it.
09:11
oh, I can see now
09:12
yes. there is nothing wrong here. you are successfully running an explorer against the full node on your computer, and against the full node I am running.
G
09:13
Gonzalo
yes, and both nodes are at 2917. But when I ask the actual height to my node through the explorer.html it says 2882
Z
09:14
Zack
it sounds like you are running a full node that has 2917 headers, and 2882 blocks.
I should probably display both in the explorer.
blocks:height(). tells how many blocks.api:height(). tells how many headers.I should probably display both in the explorer.
G
09:18
Gonzalo
oh ok, block:height(). says 2882
Z
09:20
Zack
maybe your files got corrupted. or maybe it failed while syncing. or maybe it is an old version of the code.
09:21
the files get corrupted if it is shut off or turned on incorrectly.
G
09:24
Gonzalo
ok thanks, I'll check and restart
09:30
I've got another question... I'm running a c-miner against your pool but I believe I'm not geting enough VEO. I've found at least 5 blocks so far but when asking the wallet it says only 0.7 VEO
09:31
BPa9Rcorao7jhprTfFAUeM778lo0LxXDYexfS0DCvJXKPj2cxVjMl+ONOz7KBZC+IQwlIPMeGYys6xdNChk4BEs=
Z
09:31
Zack
did you update the pubkey in the miner.erl file?
G
09:33
Gonzalo
yes
09:34
-define(Pubkey, <<"BPa9Rcorao7jhprTfFAUeM778lo0LxXDYexfS0DCvJXKPj2cxVjMl+ONOz7KBZC+IQwlIPMeGYys6xdNChk4BEs=">>).
Z
09:35
Zack
did the block height go up when you found a block?
G
09:35
Gonzalo
no
Z
09:37
Zack
thanks for reporting this problem. I will run some tests soon and try and reproduce it.
G
09:38
Gonzalo
nice, thank you! Let me know if you need access to my server to check whatever
Z
09:38
Zack
I am guessing it is a timing error, because the blocks used to be 1 per minute, and now they take like 10 minutes
G
09:40
Gonzalo
aha, so something like I'll have to found 10 blocks to get the reward of just 1?
09:40
just found another block, but height is the same: 2917
Z
09:42
Zack
no, it is some error. maybe it is displaying the message of finding blocks even when it fails. or maybe the server has already updated the data before you publish your work, so your work is for a block that was deleted.
G
09:43
Gonzalo
ok. Is there any log that I can send you? Or do you want to check something inside?
Z
09:47
Zack
Or maybe the server got messed up
09:47
I don't think there is anything more. You have already been a big help.
G
09:48
Gonzalo
glad to help
10:43
Deleted Account
When is the launch time
Z
wlozz s joined group by link from Group
G
17:14
Gonzalo
Hi Zack, FYI:
17:15
My C-miner is not working:
17:15
17:15
Also my full node, on another server, is not minning:
Z
17:15
Zack
In reply to this message
Blocks are being found very quickly right now.
The C-miner must be working for someone.
The C-miner must be working for someone.
G
17:16
Gonzalo
Z
17:16
Zack
In reply to this message
I have never seen this error before. I am not sure what is wrong.
17:16
Deleted Account
When is the launch time?
Z
G
17:17
Gonzalo
Ok, just to informa you. Both my c-miner and full node are fresh installs from yesterday
Z
17:17
Deleted Account
OK
Z
17:18
Zack
In reply to this message
Even after I make the announcement there will be a 2 week waiting period before we launch. That way everyone can get their miners ready.
G
17:21
Gonzalo
In reply to this message
yes, but is not mining. When idle CPU is at 10-20%, when mining is at 90-100%. Now CPU is at 15% but mine:status(). says "go"
Z
17:28
Zack
In reply to this message
I am testing mining from the erlang node now. It appears to work. How long does it take for the error to happen?
G
17:31
Gonzalo
about 2-3 seconds after I launch with mine:start().
Z
17:32
Zack
I tested out the c-miner just now. it looked normal.
17:34
I am unable to reproduce any of the bugs you describe. I am guessing it is because you installed ubuntu differently.
Your screen shots look really weird. Is this a server you are renting? or is this on a computer you own?
Your screen shots look really weird. Is this a server you are renting? or is this on a computer you own?
G
17:35
Gonzalo
I've just restarted c-miner as well, looking good now. In c-miner the error was showing up after several minutes mining ok
17:38
In reply to this message
both are vps that I'm renting. On the full node I installed Xfce so you're watching at a terminal window on desktop
CB
17:40
CeloStarter Bandits
@zack also for us windows noobs hopefully
Z
17:40
Zack
In reply to this message
This is not the intended way to use Amoveo.
I don't have time to fix bugs on custom setups like this.
My priority now is to make the standard set up work well.
I don't have time to fix bugs on custom setups like this.
My priority now is to make the standard set up work well.
17:41
In reply to this message
I will not be testing anything on windows, but the browser light node will probably work.
G
17:46
I understand your priorities but I believe this is not any special custom setup. It's just ubuntu 16.04 running the Xwindows, so I'm watching the shell on a terminal emulator.
17:47
Anyway I'll restart and run the node from outside the Xwindows, just to be sure
17:47
Thanks
Z
17:47
Zack
you don't have killall, or libc.so.6 shared library.
I have never seen the libc.so.6 error before.
I have never seen the libc.so.6 error before.
17:48
It is clear that someone is able to use the C-miner. as blocks are being found very quickly.
17:49
running from outside of Xwindows probably wont fix your problem.
If you install ubuntu normally, not a minimal install, then you install all the dependencies from the docs, then it should work.
If you install ubuntu normally, not a minimal install, then you install all the dependencies from the docs, then it should work.
G
17:52
Gonzalo
Well, the fact is that I installed all the dependencies shown on the docs, so I suppose this has to be enough
Z
17:52
Zack
It looks like 11:00 AM GMT time on March 2 will be the launch of mainnet.
17:53
In reply to this message
those dependencies are for someone who already has a normal ubuntu installation.
The dependencies for a minimum ubuntu installation is probably different.
The dependencies for a minimum ubuntu installation is probably different.
G
17:53
Gonzalo
Both servers are running the same Ubuntu version, but are rented on different companies:
17:53
Distributor ID: Ubuntu
Description: Ubuntu 16.04.3 LTS
Release: 16.04
Codename: xenial
Description: Ubuntu 16.04.3 LTS
Release: 16.04
Codename: xenial
17:55
One server is from scaleway.com, the other is from vps.ag
Z
17:55
Zack
Distributor ID: Ubuntu
Description: Ubuntu 16.04.3 LTS
Release: 16.04
Codename: xenial
Me too.
but mine came with
Description: Ubuntu 16.04.3 LTS
Release: 16.04
Codename: xenial
Me too.
but mine came with
killall.
G
17:55
Gonzalo
Both are running the ubuntu version that they offer
Z
17:55
Zack
How come yours didn't have killall?
G
17:56
Gonzalo
Yes this is quite weird, the vps.ag one also comes with killall, but not the scaleway.com
17:56
don't know
17:57
root@amoveo-scaleway-01:/amoveo-c-miner# killall
-bash: killall: command not found
-bash: killall: command not found
Z
17:59
Zack
So are the miners only broken on the scaleway one?
It seems like scaleway is using a modified version of Ubuntu.
It seems like scaleway is using a modified version of Ubuntu.
18:02
https://superuser.com/questions/788304/where-to-download-missing-libc-so-6
It looks like libc.so.6 is pretty important library, and your system wouldn't be able to run without it.
Maybe this is a problem of permissions? Amoveo is not supposed to be run from
It looks like libc.so.6 is pretty important library, and your system wouldn't be able to run without it.
Maybe this is a problem of permissions? Amoveo is not supposed to be run from
root
G
18:07
Gonzalo
Thanks for the link. The fact is that both servers are running from the root account. Scaleway has only c-miner, vps.ag is running the full node
18:07
I'll check this
18:07
Thanks
Z
18:08
Zack
adduser and su are useful shell commands for this
G
18:09
Gonzalo
ππ
Z
18:09
Zack
In reply to this message
Thanks for sharing this useful bug report. It seems like the documentation isn't clear enough about this.
18:11
Deleted Account
would the mined on the test net be carried over to the main
Z
18:12
Deleted Account
thank you
MH
18:53
Mandel Hoff
Zack, the js fee value for spend and create account appear on light wallet to be the same. Did you say create account has a bigger fee somewhere?
18:56
I'm trying to calculate and estimating fees for payouts and create accounts.
G
18:58
Gonzalo
Have a look, I started a new server on scaleway.com and installed a fresh copy of Ubuntu. KillAll is missing. Maybe their Ubuntu image is customized in some way, I don't know
18:58
Z
18:59
Zack
In reply to this message
I am expecting create to eventually be more expensive.
I made the initial governance fee for each TX the same.
I cannot completely predict what the community will decide.
I made the initial governance fee for each TX the same.
I cannot completely predict what the community will decide.
19:01
In reply to this message
Cool. Thanks for this useful report.
Our current plan, which you suggested, is to use pkill instead
Our current plan, which you suggested, is to use pkill instead
MH
19:05
Mandel Hoff
Does both create account and spend fees scale with payment amount?
Z
19:07
Zack
In reply to this message
No.
The amount of money you use doesn't change the fee size for any of the TX types.
The amount of money you use doesn't change the fee size for any of the TX types.
19:09
A miner is incentivized to accept txs that have the highest fee per byte.
MH
19:10
Mandel Hoff
Thank you. I was remembering that incorrectly then.
Z
19:11
Zack
In reply to this message
My expectation is that create account fee will be higher, because it consumes consensus state.
I cannot predict what the community will decide.
I cannot predict what the community will decide.
MH
19:12
Mandel Hoff
I need to get better and accessing governance variables, so I can read this stuff dynamically. To do.
Z
19:12
Zack
Maybe the community will want to enforce more privacy, so they will make the spend fee super high to disable the spend TX.
That way you need a new address every time you accept VEO.
That way you need a new address every time you accept VEO.
19:16
(ae_core_prod@)4> trees:dict_tree_get(governance, block_period).
trees:dict_tree_get(governance, block_period).
5982
(ae_core_prod@)5> trees:dict_tree_get(governance, 1).
trees:dict_tree_get(governance, 1).
100227592
(ae_core_prod@)6> trees:dict_tree_get(governance, block_reward).
trees:dict_tree_get(governance, block_reward).
100227592
(ae_core_prod@)7>
19:17
Javascript uses the api to download a merkle proof of a governance value, then it verifies the merkle proof.
You can use the same api to download a value, and just ignore the proof part. Just trust the node.
You can use the same api to download a value, and just ignore the proof part. Just trust the node.
19:18
https://github.com/zack-bitcoin/amoveo/blob/master/apps/ae_http/src/ext_handler.erl#L171
It has an example of using curl to do the same.
It has an example of using curl to do the same.
MH
19:18
Mandel Hoff
Thank you. I'll try that governance API.
Z
19:27
Zack
In reply to this message
https://github.com/zack-bitcoin/amoveo/blob/master/apps/ae_core/src/consensus/trees/governance.erl#L130
Here you can see the integer codes for each governance variable
Here you can see the integer codes for each governance variable
MH
19:27
Mandel Hoff
π
G0zilla joined group by link from Group
Deleted joined group by link from Group
21:57
Deleted Account
HiοΌguys
MH
21:59
Mandel Hoff
Welcome
Petre joined group by link from Group
P
22:22
Petre
Hello, how to get amoveo?
G
P
22:24
Petre
Is there any guideline how to mine?
G
P
G
P
G
22:27
G0zilla
Good
Z
22:33
Zack
In reply to this message
The easiest way to mine is with this page: http://159.89.106.253:8080/wallet.html
22:33
In reply to this message
the fastest way to mine is with this tool: https://github.com/zack-bitcoin/amoveo-c-miner
P
MH
22:37
Mandel Hoff
Zack, the explorer seems stuck, and light wallet balance is having errors.
22:37
22:38
Explorer is at 2989 and block:height(). is about 3031 now.
22:44
I found another issue with my custom payments. I was using a fee that was slightly higher than the 0.00152050 value and was getting errors using a slightly higher fee.
Z
22:53
Zack
In reply to this message
yes, I confirm that this is happening. Thank you for pointing it out.
22:54
In reply to this message
the governance fee is as you say. there is also a miner fee that you can set in the config file. I think the default is 20 satoshis. The miner will only accept txs that pay more than this fee.
23:03
I think I accidentally corrupted the database earlier by truing to turn it on while it is already on.
It fails silently, which is dangerous.
It fails silently, which is dangerous.
MH
23:24
Mandel Hoff
Zack, when a new node stands up and makes a new keys.db - who/what pays fees to create account on that new keys.db?
17 February 2018
MH
00:33
Mandel Hoff
Zack, my node's network output is 80gb in under 1 day. Azure charges extra for network output. I'd guess most vps services do similar. Can that be reduced by a lot?
00:33
Z
01:23
In reply to this message
Yes, it isn't very efficient about sharing data. And the data isn't compressed.
01:23
In reply to this message
There is a wide range of prices for too much data.
I think with digital ocean it is like $5 a terabyte
I think with digital ocean it is like $5 a terabyte
MH
01:37
Mandel Hoff
Ok. I'll have to look around. I think this 10$ per day per node on my current azure nodes. I shut down my 3 nodes for now.
Z
01:37
Zack
In reply to this message
Wow, that is a lot. I think I am spending $5 a month to run the server.
MH
01:40
Mandel Hoff
I think they charge 1$ per 10gb output. Might be some special packages for more output
01:41
My home ISP has 1 TB month download limit. 100gb per day per node kind of alot.
Z
01:42
Zack
Yes.
Ideally we would use state channels to pay each other for data.
Ideally we would use state channels to pay each other for data.
01:43
If twice as many people download the blockchain from you, then it will necessarily transfer twice as much data
01:44
Hopefully we can get most people to use light nodes, as they don't use so much data
MH
01:44
Mandel Hoff
Agreed. Could exceed 100gb per day with a busy community easily.
P
01:44
Petre
Zack when veo will be available in any exchange?
01:44
I want buy as much as i will be abale
Z
01:45
Zack
In reply to this message
It is custom software, so it is hard for an exchange to integrate it.
I think there won't be a good exchange for months.
I made an informal exchange on reddit for people to find someone to trade VEO with.
I think there won't be a good exchange for months.
I made an informal exchange on reddit for people to find someone to trade VEO with.
G
01:46
Gonzalo
@ scaleway.com they offer unmetered bandwith @vps.ag the cheapest 3β¬ vps has 3TB included, then they charge about 1β¬/TB
Z
MH
01:47
Mandel Hoff
Thanks for the suggestions
01:48
Azure node cpu was 13$ per month and then was eating 10$ per day in bandwidth. π₯
Z
P
01:49
Petre
In reply to this message
Got it, thanks and one more question what is amoveo c miner? I never mine any coin so I donβt know about it.. i really like your vision and project and want to have some Veo :(
01:51
And how much will be total supply?
Z
01:52
Zack
In reply to this message
The community can use he governance mechanism to change the block period and the block reward at any time. I cannot predict how big of a market cap they will want.
P
01:54
Petre
Ok I understand.. thanks
MH
02:01
Mandel Hoff
Ugh - confirmed. At 100gb/day, Azure bandwidth is expensive: https://azure.microsoft.com/en-us/pricing/details/bandwidth/
Z
02:04
Zack
In reply to this message
I am looking at my node's data usage for the last 6 hours.
It says I was consistently using 14 mb/s for 3 hours, then it was less than 0.1 mb/s for 3 hours.
It was all inbound traffic.
It looks like my node was failing in some way, and it started downloading lots of unnecessary stuff from your node.
Maybe this was because my node looked like it wasn't totally synced. So you would keep trying to push the recent blocks to it.
It says I was consistently using 14 mb/s for 3 hours, then it was less than 0.1 mb/s for 3 hours.
It was all inbound traffic.
It looks like my node was failing in some way, and it started downloading lots of unnecessary stuff from your node.
Maybe this was because my node looked like it wasn't totally synced. So you would keep trying to push the recent blocks to it.
02:06
This is a bad failure mode, it is expensive for both of us
MH
02:07
Mandel Hoff
My network usage screen shot show my output was very steady for the time it was up - a bit under 24 hours.
02:08
Kind of funny failure problem too if that's what causes it.
02:09
My 3 nodes are down about 90 min now if that helps you diagnose your nodes behavior
Z
02:13
Zack
oh, my outbound traffic dropped from 1 mb/s to 0.05 mb/s about 90 minutes ago.
So that would make about 100 gb per month.
1 mb/s does seem a little heavy, but it isn't unreasonable.
I haven't put work into this before, so there should be easy opportunities for improvement.
So that would make about 100 gb per month.
1 mb/s does seem a little heavy, but it isn't unreasonable.
I haven't put work into this before, so there should be easy opportunities for improvement.
02:18
Deleted Account
would card mining be implemented after main net launch
Z
02:18
Zack
In reply to this message
I hope gpu mining happens soon. I don't know when it will be available.
02:20
In reply to this message
I think having gpus sooner is better. otherwise someone with closed-source gpu software will take all the coins.
I
02:55
Iridescence
Here's a thought : what incentive does somebody who has developed a GPU miner have to share it with the community?
Z
02:56
Zack
In reply to this message
We can use Amoveo to pay people to do this: https://github.com/zack-bitcoin/amoveo/blob/master/docs/use-cases-and-ideas/insured_crowdfund.md
Z
03:14
Zack
I got a tutorial openCL program to compile and run. Maybe I can write gpu code quickly.
03:39
Deleted Account
could c-miner be used to gpu mine
Z
03:42
Deleted Account
is the current mining test or main
Z
03:43
Zack
the main net will launch March 2. Right now it is a test net.
03:43
Deleted Account
thanks man
M
03:44
Mike
Damn. Exciting news
Z
03:49
Zack
https://github.com/Fruneng/opencl_sha_al_im
This is almost all the work for making a gpu miner
This is almost all the work for making a gpu miner
03:52
Deleted Account
i tried mining on light node and for some reason it doesn't mine
MH
03:53
Mandel Hoff
Yes, Zack's mining pool appears to be down currently.
MH
03:54
Mandel Hoff
Here's a preview of my mining pool UI:
03:54
03:54
03:54
Z
03:55
Deleted Account
that's great man
Z
MH
03:55
Mandel Hoff
Pool Stats, Mining Monitor, and Light Wallet Spend features.
Z
03:57
Zack
In reply to this message
Very cool. You should make it open source so that I can help you avoid mistakes. That way your time is used more efficiently.
03:57
Deleted Account
In reply to this message
yes if that happens gpu miner will be neccessary until then let people mine on cpu miner
JasonWW joined group by link from Group
03:58
Deleted Account
if someone start mining on closed source release the gpu miner and hash rate can be divided
04:29
Deleted Account
When will cpu miner like minerd launch on windows ?
Z
04:30
Zack
I wont be developing any windows software. Maybe someone else will.
I think that you can mine in windows from the browser:
http://159.89.106.253:8080/wallet.html
I think that you can mine in windows from the browser:
http://159.89.106.253:8080/wallet.html
04:31
Deleted Account
Yes i am but it won't mine if i open another tab on browser.
MH
04:37
Mandel Hoff
I will almost certainly make a windows tool available before launch
KR
Z
04:53
Deleted Account
Different window?
G
Z
04:54
Zack
I don't know if this will work. Just a suggestion
J
05:05
JasonWW
i am testing on my laptop. for a hashrate of 0.03mh/s, what's the average time to mine a block?
05:09
only for one tab though
Z
05:10
Deleted Account
but block which is .7 is very low
J
Z
05:12
Zack
In reply to this message
It looks like it takes 169 mh on average to find a block.
```pow:sci2int(element(6, block:top())).
168820736```In 33 seconds you would get a mh, so you would find a block every 90 minutes or so
```pow:sci2int(element(6, block:top())).
168820736```In 33 seconds you would get a mh, so you would find a block every 90 minutes or so
RL
06:20
R L
In reply to this message
Hey Mandel or Zack are either of you going to run a pool??or anyone else?
06:20
Guys, BTW you are rocking it
06:21
Thanks for all your hard work
Z
06:21
Zack
In reply to this message
It seems like at least both of us will be running pools.
I don't know who else will be.
I don't know who else will be.
RL
06:21
R L
Great
06:22
I want to try and build the cminer what environment do you suggest?
Z
RL
06:22
R L
I skimmed through and see that someone is working on a gpu miner??
06:22
Awesome thanks Zack
06:24
Zack are you personally going to setup a docker or something and mine on aws?
Z
06:26
Zack
no, I don't like aws, and I don't use docker.
I might mine on a vps, but maybe I will finish the gpu miner before we launch, in which case I would pay my friend who has lots of graphics cards to mine for me.
I might mine on a vps, but maybe I will finish the gpu miner before we launch, in which case I would pay my friend who has lots of graphics cards to mine for me.
RL
06:26
R L
OK cool
06:27
Thanks!
MH
06:30
Mandel Hoff
Yes. I had been using azure but the bandwidth for the node is very expensive on azure.
RL
06:35
R L
Hmm ok
06:36
Yeah.. Azure can be expensive
Deleted joined group by link from Group
MH
08:11
Mandel Hoff
Amoveo Amoveo, Amoveo Amoveo - Rock me Amoveo! https://www.youtube.com/watch?v=cVikZ8Oe_XA
08:13
Can't get that out of my head... going to be a long night.
MH
09:32
Mandel Hoff
Ok, got my pool hosted here:
09:32
Z
09:37
Zack
Looks great.
MH
09:49
Mandel Hoff
Thank you.
RL
10:55
I registered it awhile ago
10:55
Zack if you want it or want to use it just let me know
10:56
If not I will try and get something up and going or point to Mandel
M
10:57
Minieep21
In reply to this message
Love Love Love the design!
Is it currently setup for test net?
Is it currently setup for test net?
G
11:04
Zack while you decide what to do with www.amoveo.io, I've pointed it to your github amoveo account
G
11:40
Deleted Account
My bad, is working now.
11:41
And looks very good, thanks Mandel!
Z
G
17:48
Gonzalo
Mandel I've got 1 suggestion and 1 question for you:
17:50
- on the "wallet" side, after you send VEO between addresses, the "account details" is not updating with the new balances of both addresses. I would be nice if you update the info on here..
17:52
- on the "miner" page, after mining a few hours, this is my strange balance so far:
17:52
17:54
I've spent less than 1 VEO on BDTZg... so my actual balance should be way higher than 2.17 VEO, isn't it?
17:56
Thanks!
17:59
In reply to this message
Hi rw freshmore, have a look at your domain, maybe you have to confirm your email ;-)
Mike joined group by link from Group
D ZZ joined group by link from Group
MH
20:24
Mandel Hoff
In reply to this message
After you submit a "spend", your transaction sits in the mempool to be mined. It takes time for the balances to update, because block chain... not lightning payment yet.
20:29
In reply to this message
You're right. Payouts txs appear to be happening as scheduled but they are not all being received. Some are. Some aren't. This appears to be affecting all addresses. I checked my miners - same. Working on it.
RL
Z
21:18
Zack
In reply to this message
It is hard to know what sort of machines will be best. For now this is the best mining software: https://github.com/zack-bitcoin/amoveo-c-miner
I don't know which CPU is best for this software.
I am working on GPU mining software now, but I am not sure if I will finish in time.
I don't know which CPU is best for this software.
I am working on GPU mining software now, but I am not sure if I will finish in time.
M
21:20
Mike
as far as I know, if there is a working C miner, it shouldn't be difficult to use it in CUDA cores or OpenCL program...
21:20
sadly, I have no experience in it at all =(
Z
21:21
Zack
In reply to this message
yes, it seems like you are right. I think it will not take me very long, even though this is my first time programming for GPU.
M
21:22
Mike
launching without GPU miner gives control of mining to botnets and private closed source gpu miners since the first moment :) I really hope you can make it in time, Zack
Z
21:23
Zack
In reply to this message
If we only have CPU mining for 2 weeks or a month, I think it does not matter very much.
M
21:23
Mike
this project seems so fresh after all shitcoins I met last months
21:24
Just continue what you're doing ;)
Z
G
21:29
Gonzalo
In reply to this message
Totally agree. Genuine project, freshly coded from 0, clever developer. No ICO, no pre-ICO, no premine, no... etc, etc. 100% tech, 0% scam π
Z
21:41
Zack
I am glad you guys are happy with the results so far. I will continue giving my best effort to build a great product.
I?
MH
21:43
Mandel Hoff
The derivatves market is the trillion dollar feature. I didn't get that until after I placed a bet.
21:44
I got chills from that once I understood it.
Z
21:44
Zack
In reply to this message
yes, this is why I delayed main net launch until that feature was working in the browser.
RL
21:51
And you guys are building a front end to it right out the gate
21:51
THAT gives me chills
21:52
Honestly what the hell are gnosis and augur doing?
21:53
Millions of dollars and all this time and I still can't just go to a website and make a bet
21:53
Hahaha
Z
21:53
Zack
In reply to this message
If Amoveo can survive the first month, it will make Augur and other oracle projects look bad.
RL
21:53
R L
It already is making them look bad
Z
21:53
Zack
Maybe we will help drive them into action.
RL
21:53
R L
It's a joke
21:54
That's why i see even this simple Web interface you made as being such an integral part
21:54
It will bootstrap participation right out the gate
21:55
Doesn't matter if it's 100 % secure or whatever
21:55
Point is it can be used.
21:55
With all the warnings and disclaimers people can PLAY
21:55
they can mine they can bet they can transact
21:56
Want to really blow this whole thing out of the water? Add chat right in Web interface with wallet as the users identity
Z
21:57
Zack
If Augur had used my strategy and kept me employed, then I think we could have launched Augur in 2016, and it would have looked how Amoveo looks now.
RL
21:57
R L
I know building the infrastructure is key and it might seem like a misdirected effort
21:58
Good thing they didn't
21:58
Crypto is hard to use
21:58
When I visited the amoveo page the first time... awesome
Z
21:58
Zack
Chat sounds cool. but there are already lots of chat apps. I was thinking of putting an encryption tool into the browser, so you can encrypt a message, and it can only be decrypted by the person who controls the address that you send the message to.
RL
21:59
R L
Yeah I mean.. even integrating an existing chat platform
21:59
That would be cool for aure
Z
21:59
Zack
then you could copy/paste the encrypted message into a chat program.
RL
21:59
R L
Signed messages
Z
22:00
Zack
yeah, a signed message tool would be good too
RL
22:00
R L
Yeah that would be awesome
22:00
It would be cool if from the Web interface I could choose which pool I wanted to connect to
Z
22:01
Zack
you can. Just put the IP address into the part that says "channel node ip"
RL
22:01
R L
Duhh ha OK awesome
Z
22:20
Zack
Wow, blocks are being found so fast today. I wonder if someone has a GPU miner.
MH
22:38
Mandel Hoff
I have cpu pool miner and it's about 1 min per blk
22:38
Been running all night
Z
Denis joined group by link from Group
MH
22:57
Mandel Hoff
Let me know if you want thoughts on main net initial difficulty
Z
22:59
Zack
In reply to this message
Sure, feel free to share.
I think I can finish the GPU miner before main net, so maybe we should wait a little longer before making estimates.
I think I can finish the GPU miner before main net, so maybe we should wait a little longer before making estimates.
MH
23:01
Mandel Hoff
Yes that would change it a lot ok will hold off on estimates.
KR
23:08
Kevin Roberts BWD
Light wallet not connecting to my new node?
Z
23:09
Zack
In reply to this message
Can you be more specific about what you did, and what happened that you didn't expect?
KR
23:15
Kevin Roberts BWD
I pointed the light wallet at the node I started this morning and I see no blocks
159.65.44.147
159.65.44.147
Z
KR
23:28
Kevin Roberts BWD
Looks like itβs on my end. Something is blocking the 8080 port
Z
23:29
Zack
I am looking at why the bandwidth consumption is so high. It seems like someone downloads account data from the server every 1/2 second.
KR
23:33
Kevin Roberts BWD
Do you know when it started?
23:34
I launch my node about 1 hr ago
Z
23:35
Zack
midnight it started. things got twice as bad about 90 minutes ago.
KR
23:35
Kevin Roberts BWD
Hmm?
MH
23:36
Mandel Hoff
Account data json is about 80 bytes per request
KR
23:36
Kevin Roberts BWD
Iβll stop my node to see if it changes
Z
23:37
Zack
In reply to this message
I think the http overhead is a lot bigger than that, but yeah, you make a good point. i should weight the messages by how much data they move, and maybe then I will find the bottleneck is different from my first assumption.
KR
23:37
Kevin Roberts BWD
23:39
My node is offline
23:39
Any difference ?
Z
23:41
Zack
Things got way better about 10 minutes ago
23:42
I think it is just because I rebooted the server in order to do more measurements, and so I got dropped from the other server's lists.
23:44
the accounts and headers spam is still happening, but it costs almost 0 bandwidth. So I was focusing on the wrong thing.
The bandwidth cost must be from the full node servers communicating with each other.
The bandwidth cost must be from the full node servers communicating with each other.
23:45
I am the only node on the list of peers.
Did everyone else shut off? how come no one is feeding me any more peers?
Did everyone else shut off? how come no one is feeding me any more peers?
MH
23:48
Mandel Hoff
I'm all off
M
23:53
Mike
I was on a light node & trying to mine
23:53
and stopped it 10 or 20 minutes ago )
23:54
will do a full node on VPS tomorrow to play with it
18 February 2018
Z
00:08
Zack
I am launching a second server to figure this out.
MH
00:14
Mandel Hoff
Thanks for investigating the bandwidth issue.
Z
00:24
Zack
I synced the new node. It didn't use excessive bandwidth.
Maybe someone had a node up that had a corrupted database, so it was refusing blocks. And everyone else would keep trying to give them the blocks they are missing.
Maybe someone had a node up that had a corrupted database, so it was refusing blocks. And everyone else would keep trying to give them the blocks they are missing.
KR
00:29
Kevin Roberts BWD
Iβm off
Z
00:39
Zack
It seems like about 200 people are downloading headers from the server per second. so weird.
00:39
It seems like having a second node didn't cause any problems
00:40
Maybe someone has like 1000 computers, and they are using javascript miners in all of them?
like a botnet
like a botnet
00:42
ALL of the traffic is coming from this IP: 23,96,187,38
haha.
I guess I will write some code so each IP address has a limit on how much bandwidth it can access
haha.
I guess I will write some code so each IP address has a limit on how much bandwidth it can access
00:45
Someohe must have written custom code.
23,96,187,38 is asking about multiple different accounts in the same second
23,96,187,38 is asking about multiple different accounts in the same second
00:56
after more experimentation, I once again think it is because the full nodes are sending excess data back and forth.
MH
01:06
Mandel Hoff
Want me to bring up a node or would that hurt your investigation?
Z
01:09
Zack
I am unable to make the bandwidth spike using 2 full nodes. It was a coincidence before.