Amoveo ♥🧿
Previous messages
give blocks
finished sending blockstalk_helper weird response
talk helper fail
talk_helper weird response
talk_helper weird response
talk helper fail
get their top header
already synced with this peer
talk_helper failed_connect 1
then you restart with
Next messages
18 February 2018
Z
01:11
Zack
In reply to this message
I am not sure if it would help. I am still in the exploratory phase, I don't know what exactly is wrong.
01:12
This is the pattern I notice: the longer I leave the node on, more people start attaching it and they keep getting data for mining.
but there is always 1 person who is doing like 20x more requests than the rest
but there is always 1 person who is doing like 20x more requests than the rest
01:19
I think someone modified the mining software to go faster at the expense of the network.
01:20
It is so weird. the same person will request account data alternatively for 2 different accounts. over 1000 times in a row. multiple times per second.
MH
01:21
Mandel Hoff
My pool caches the work json so it would poll the mining api more frequently than once per X seconds
Z
01:22
Zack
I don't understand
MH
01:22
Mandel Hoff
My pool browser can ask for balance but again that should be once per 10 sec max due to cache
Z
01:26
Zack
different IPs keep spamming, but none of them have been yours
MH
Z
01:27
Zack
Maybe if your light-node has poor connectivity, it will keep spamming the message quickly hoping to get a response eventually.
Maybe I need to increase how long the browser waits for a response
Maybe I need to increase how long the browser waits for a response
01:30
you barely contact me at all. It is all those 2 spammers.
I think I just need to put limits, so you can only do so many requests in a 10 second period.
I think I just need to put limits, so you can only do so many requests in a 10 second period.
01:33
should I limit by bandwidth, or by number of http requests, or both?
MH
01:34
Mandel Hoff
Bandwidth for sure. Account lookups and low bandwidth should happen often.
Z
01:34
Zack
you only need at most 1 account lookup per block. and a block takes 10 minutes.
01:36
oh, it is my 2nd server causing all the spam. I managed to reproduce the problem
MH
01:37
Mandel Hoff
Sounds good.
Z
01:38
Zack
I think
sync:start(). is causing a zombie process
MH
01:39
Mandel Hoff
Oh - good catch! Sneaky bug.
Z
01:44
Zack
I notice that the public nodes aren't switching from sync_mode quick to normal.
01:50
it isn't a problem that they start in sync_mode quick. They switch as soon as it matters.
01:53
The same node is still contacting me 23,96,187,38 but now it doe it at a slow regular rate.
I wonder what changed
I wonder what changed
01:59
In reply to this message
I think your node has a corrupted database. are you at the same height as me?
02:00
or maybe mine is corrupted. i will check.
02:00
both mine are at the same block:height(). 3983
02:01
http://40.121.213.53:8080/explorer.html
it says you are at 3079, so it seems like your database is corrupted.
it says you are at 3079, so it seems like your database is corrupted.
MH
02:03
Mandel Hoff
Yes, I'm behind for some reason and not syncing correctly. When I shut it down, I'm sure I did sync:stop(). api:off(). and halt().
Z
02:03
Zack
and now my node is trying to send all the blocks to you repeatedly. I think we verified how this error is happening
MH
02:03
Mandel Hoff
Oh well. I'll git clone fresh again later.
02:03
Or, I can leave it this way if you want me to.
02:04
Sounds great if my bad node is helping. 🙈
Z
02:04
Zack
you should probably shut it down before you lose too much money
MH
02:04
Mandel Hoff
Thanks, will do that if you don't need it more. I'm getting many many "block absorber died" console messages on my end if that helps at all.
Z
02:05
Zack
I think you spend more on Amoveo than me in the last 24 hours
02:05
to fix your database
make prod-clean
MH
02:06
Mandel Hoff
Shutting down then. Good luck with the fix. I'll be happy to test more later, but probably busy the rest of today. Thank you for tracking this down.
Ezra joined group by link from Group
RL
03:24
R L
That ip you posted zack is a Microsoft ip in Redmond
03:24
Likely not azure. Could be bill Gates
03:24
:)
Z
03:25
Zack
cool
Z
03:45
Zack
I made a couple changes to solve the problem of too much data moving around.
1) if someone is not able to process blocks, then don't push infinite blocks at them.
2) only accept so many http requests per ip address per second. defined in the config file which is generated by the make file. by default it accepts 10 per second.
1) if someone is not able to process blocks, then don't push infinite blocks at them.
2) only accept so many http requests per ip address per second. defined in the config file which is generated by the make file. by default it accepts 10 per second.
03:46
In order for this update to work, you will need to delete the old config file: amoveo/config/prod/sys.config
MH
03:47
Mandel Hoff
All code done? Want me to clone and test?
Z
03:48
Zack
I pushed these updates to github. Sure, lets test it
MH
03:49
Mandel Hoff
Want me to spin up my old node to test a corrupt node attack on your good nodes or just get new node code?
Z
03:50
Zack
yeah, that is a good idea
03:50
lets try attacking it again
MH
03:50
Mandel Hoff
Ok. I'll spin up and attack sync.
03:54
I'm on 52.224.52.38. Sync:start(). and peer added.
Z
03:54
Zack
looks like I am still sending too much data to you. lets give it a few more minutes to confirm
MH
03:55
Mandel Hoff
Z
03:56
Zack
oops, I was syncing my node at the same time which messed up the test. so we should give it a few more minutes to confirm
03:56
In reply to this message
yes, I still have to push blocks occasionally, but hopefully it doesn't do it so often now
MH
03:58
Mandel Hoff
Also - I think I found why my pool wasn't paying out correctly. Pretty sure the problem was I was paying out from the wrong address and making payments from a broke account, so spends were failing. I'll run more tests, but that was likely the big problem with payouts.
Z
03:59
Zack
cool, good you figured it out.
It seems to me that the address that pays the prize is the same that wins the block reward, so I am not sure how this could have happened.
It seems to me that the address that pays the prize is the same that wins the block reward, so I am not sure how this could have happened.
03:59
It looks like the test is till failing.
03:59
way too much data is moving around
MH
04:02
Mandel Hoff
Yes, I had my payment manager trying to spend from the wrong address. Hopefully, just a bad configuration on my part. Certainly will test more.
04:02
Want me to keep my node up or take it down? I can leave it up a few hours if that helps or drop it. I won't be home much more today to change it though.
Z
04:03
Zack
If you can leave it up for another hour or so, that would be great. I have some ideas
MH
04:03
Mandel Hoff
Ok, I'll leave it up.
04:25
Deleted Account
In reply to this message
when i open the link is says who is verfication is pending. Is it just me finding this error.?
Z
04:28
Zack
my laptop full node froze at 3079 too, just like Mandel's node. this is concerning.
Z
04:47
Zack
it froze at 3079 because it was downloading blocks from Mandel. no problems here.
04:48
Deleted Account
mine shows 4004
Z
04:48
Zack
yeah, me too
04:51
about once per minute someone syncs a light node with the network.
Z
05:09
Zack
I backed off to 4 http requests per second. And now I only push 5 blocks at a time to your peer. That way if they don't download the blocks, you haven't wasted so much bandwidth by pushing unnecessary blocks.
It seems like this lowered the bandwidth requirements enough. but I need to do some more testing to know that these changes don't break anything else.
It seems like this lowered the bandwidth requirements enough. but I need to do some more testing to know that these changes don't break anything else.
05:15
In reply to this message
The new code is about 40x less expensive, but it is still vulnerable to griefing attacks, where the attacker and defender are both using up the same amount of bandwidth.
Please turn off the attack when you get a chance.
Griefing attacks are un-defendable in this context, as far as I am aware.
Please turn off the attack when you get a chance.
Griefing attacks are un-defendable in this context, as far as I am aware.
Z
05:41
Zack
I put a small delay into the server's side of the api. the delay is inversely proportional to the frequency limit of how fast people can do http requests.
This way, as long as you do one http request at a time, it will be slow enough that none of the requests will fail.
This way, as long as you do one http request at a time, it will be slow enough that none of the requests will fail.
05:56
Deleted Account
hey zack im interested in setting up mining pool i know its a complicated setup but i want to give it a go. Can you provide me any link for setup ?i don't wanna get scammed.
Z
05:57
Zack
Why do you want to set up a mining pool? Are you sure a solo mining operation wouldn't satisfy your needs?
05:59
Deleted Account
when the mainnet goes live and if its a success many ppl with gpu farms will increase the difficulty. I don't have a mining rig so pool will get me steady income.
Z
06:00
Zack
you are going to set up a server and let people join your mining pool?
06:00
Deleted Account
yes that's what i plan to
06:00
do.
06:01
well if you think its not a great idea i will drop it then
Z
06:01
Zack
ok, first install amoveo, and make sure all the blocks sync: https://github.com/zack-bitcoin/amoveo
Then install this on the same machine https://github.com/zack-bitcoin/amoveo-mining-pool
Then install this on the same machine https://github.com/zack-bitcoin/amoveo-mining-pool
06:01
I recommend renting a Ubuntu server.
06:01
Deleted Account
ok
Z
06:31
Zack
In reply to this message
Thanks for all the help today Mandel. We made it much more affordable to run an Amoveo full node.
06:33
Deleted Account
zack is your server down?
Z
06:34
Zack
MH
06:39
Mandel Hoff
In reply to this message
Awesome. I'll update my node in a couple hours. I'm not home now. Thanks for improving that.
06:42
Deleted Account
In reply to this message
yes its working. the headers weren't updating so it stopped. i'm mining on multiple tabs so maybe that's the reason they didn't update
Z
06:42
Zack
In reply to this message
I changed how the headers download. before it was 100 at a time. Now it does 5000 at a time.
This way is much faster, but you get less visual confirmations that it is working. So it is easy to think that it is not working.
This way is much faster, but you get less visual confirmations that it is working. So it is easy to think that it is not working.
06:44
Deleted Account
In reply to this message
no i checked through my task manager. well its working for now i will keep an eye on it this time.
G
06:51
Gonzalo
In reply to this message
Sorry I was out for many hours and my CPU miner was working. I had a full node and c-miner on the same machine, IP 51.15.69.135
Z
06:58
Zack
In reply to this message
cool. Could you update it to the most recent version of the code?
rm config/prod/sys.config
git pull
make prod-restart
G
06:59
Gonzalo
sure
Z
07:00
Zack
oh you need to turn it off before those commands, if it is already on
G
07:02
Gonzalo
By the way, this morning (CET) I reinstalled everything, fresh installation, full node and c-miner... everything went smooth for a while but then again I had the same error than yesterday: "/usr/lib/erlang/erts-7.3/bin/child_setup: error while loading shared libraries: libc.so.6: cannot open shared object file: Error 24"
Z
07:03
Zack
did you make a user account instead of using root?
G
Z
07:05
Zack
thanks for sharing. Unfortunately I am not sure how to help you. I am not able to reproduce this error, and I haven't seen it before.
If you have any more clues as to how I could reproduce the bug, please share.
If you have any more clues as to how I could reproduce the bug, please share.
G
07:09
Gonzalo
In reply to this message
Many thanks, but please don't care about me, I'm just sharing for you to know, maybe is something wrong on the code that should be fixed. If it's just me I will do whatever, no problem. 👍
07:10
Anyway, if you need to access my server to check whatever, just let me know.
Z
07:11
Zack
Oh, good idea.
I could log on and look at it
I could log on and look at it
G
07:12
Gonzalo
Sure! If you need it to improve something on amoveo, my server is your server! 😂
Anak Dagang joined group by link from Group
Z
08:12
Zack
Our best chance at survival is if we start small, and grow smoothly. This way we can upgrade pieces of Amoveo as grow. We will have live feedback to know what is important to upgrade. Once we are bigger, it will be able to afford specialized programmers. So it is better to do each upgrade as late as possible.
So don't tell too many people about Amoveo. If there are too many people on the first day, it will probably break.
So don't tell too many people about Amoveo. If there are too many people on the first day, it will probably break.
10:11
Deleted Account
OK
Lyndon Something joined group by link from Group
11:44
Deleted Account
So im trying to mine solo
11:45
Downloaded amoveo
11:45
And started the node
11:45
11:45
I dont get the next step
G
11:47
Gonzalo
Make prod-attach
11:49
And then, inside the node:
api:sync().
mine:start().
api:sync().
mine:start().
11:50
Deleted Account
Aparently its still communicating
11:50
I will wait
G
11:50
Gonzalo
👍
11:52
Sorry I can't help you more right now, I'm going to bed, it's 4:52AM here 😴
11:53
Deleted Account
Ok its still attaching here. Im gonna sleep as well it should be attached
11:53
By morning
G
11:54
Deleted Account
Ok
M
12:56
Mike
In reply to this message
I think you're underestimating modern cryptocommunity. It's almost impossible to "start small" now... once mainnet goes live, it will be smashed by huge wave of mining machines coming in
12:58
Just saying. Be prepared. I see that constantly.
Z
16:05
In reply to this message
Yes, I will prepare as much as I can.
If it breaks, I will do my best to fix it.
If I can't fix it, then we will start over with better code.
If it breaks, I will do my best to fix it.
If I can't fix it, then we will start over with better code.
Deleted joined group by link from Group
16:12
Deleted Account
Hi all
Z
16:25
Zack
It looks like the updates to decrease network traffic have been successful. After 5 hours, there is still little traffic.
Deleted joined group by link from Group
16:48
Deleted Account
hello
Z
16:49
Zack
Hello. Welcome Rwt and Ryan
16:49
Deleted Account
Hi, im setting up mining/nodes and getting familiar
Z
16:50
Zack
great
16:51
Deleted Account
I was reading some of the docs and noticed that block rewards are dynamic but am I correct that block time is not? Any reason for this?
16:52
Is it because the reward itself can be set up as a betting market
Z
16:55
Zack
In reply to this message
the block reward is set by the governance mechanism. The average block period is also set by the governance mechanism
16:55
Deleted Account
Okay, i will have to read more about that
18:09
Deleted Account
how is supply determined from genesis? It sounded like there was some frm of airdrop? I guess I assumed it was starting from 0 and only mined
Z
18:10
Deleted Account
ok
18:14
I will put myself out there to buy VEO from miners on day one
18:15
I will base the value on a ratio of total rewards and network hash power
Z
18:21
Deleted Account
It makes sense, though I'll need to look up costs for hash rates
Z
18:22
Zack
yeah. I will do some limited testing on hash rate costs.
The community should use the forums to discuss these things to make it easier to know what the cost is.
The community should use the forums to discuss these things to make it easier to know what the cost is.
18:29
Deleted Account
The average price people in the U.S. pay for electricity is about 12 cents per kilowatt-hour.
Z
18:32
Zack
It seems to me that the opencl kernel should run multiple hashes, and keep trying to find a block with many different nonces without returning anything to the c code for as long as possible.
But when I look at opencl kernels for bitcoin, it looks like they are only checking one nonce
But when I look at opencl kernels for bitcoin, it looks like they are only checking one nonce
18:32
Deleted Account
so if the network has 1000 kwh on average, and the reward is 100 per day, then the network would be using $120 of electricity to create 100 VEO, and so the VEO should be valued at $1.20 that day
18:33
presumably the miners are efficient, and so actually pay less than $0.12, but the buyers would not be efficient and would have to pay $0.12 if they were to do it themselves
Z
18:34
Zack
In reply to this message
yes, this is a great way to make a first-order estimate of how much to pay for Veo.
I don't have a very good processor to test with, but maybe we can look up statistics online for different processors and calculate how fast they would mine.
If I manage to write the GPU miner in time, then we will have to do some different research.
I don't have a very good processor to test with, but maybe we can look up statistics online for different processors and calculate how fast they would mine.
If I manage to write the GPU miner in time, then we will have to do some different research.
18:34
Deleted Account
so (network_KWh * 0.12) / hourly reward
18:35
yeah we could get more sophisticated too
CB
18:35
CeloStarter Bandits
can i create an Discord... for Trading.. I would give your @Zack full admin rights.. and you can takeover when ever you want
18:35
Deleted Account
But if people are overpaying for VEO, wouldn't the reward get adjusted to move back to equilibrium anyways
CB
18:36
CeloStarter Bandits
Would be Sell Buy Escrow
Z
18:37
Zack
In reply to this message
sounds good.
The difficulty stays constant for 2000 blocks at a time, so for each 2000 block period the amount of electricity consumed is proportional to the number of blocks mined.
2000 blocks is set to about 2 weeks with the initial governance constants.
The difficulty stays constant for 2000 blocks at a time, so for each 2000 block period the amount of electricity consumed is proportional to the number of blocks mined.
2000 blocks is set to about 2 weeks with the initial governance constants.
CB
18:38
CeloStarter Bandits
@zack would it be ok, with discord Trade?
18:38
Deleted Account
Couldnt you write a short term contract for deviations between block periods, so you could hedge any risk of changes
Z
18:38
Zack
In reply to this message
the governance mechanism can adjust the reward, but it takes at least 1000 blocks to do this. With the initial governance constants.
18:40
Deleted Account
so if the difficulty is set, and then suddenly mining increases materially such that rewards are coming in twice as fast, you can protect yourself from this by writing a contract that would have yourself paid out in the event that rewards exceed expectations at the time of the next purchase. So you overpay, but you also get paid out on your contract because the miner received extra rewards.
Z
18:40
Zack
In reply to this message
I have never used discord. I already have a lot of forums I am managing. Is there a reason you think we should use discord as opposed to https://www.reddit.com/r/AmoveoTrader/
18:41
In reply to this message
I was thinking the contract could just follow the value of bitcoin or USD. Derivative markets tend to be of simple things that can be understood quickly.
18:42
Error
18:43
EW
18:46
Eli W
I’m in some project discord , but most community still just used telegram chat telegram channel
Z
18:49
Zack
In reply to this message
That is an error I have never seen. It looks like erlang is opening up in a different mode.
Z
18:59
Zack
In reply to this message
I can get into the same mode as you if I hold control and press the "G" key.
19:00
http://learnyousomeerlang.com/starting-out
Here is documentation on the mode you are in. How to enter and exit it.
I have never seen the code automatically enter this mode.
Here is documentation on the mode you are in. How to enter and exit it.
I have never seen the code automatically enter this mode.
19:04
I made a discord. It is good to use different forums, in case any one of them should fail. https://discord.gg/a52szJw
MH
21:41
Mandel Hoff
I just sync'd a new node for <100MB. Great bandwidth fix!
Z
19 February 2018
Z
00:38
Zack
I set it up so Amoveo nodes will automatically share their ip to the network to help serve blocks.
It looks like 3 servers besides the one I launched are running now.
It looks like 3 servers besides the one I launched are running now.
G
00:53
Gonzalo
Zack, when asking the balance at the full node with
api:balance().I get 701593144. How many VEO is this??
Z
G
00:53
Gonzalo
👍
00:56
BTW I'm getting lots of
talk_helper weird responseand
talk_helper weird response
talk_helper weird response
talk_helper weird response
talk_helper weird response
talk helper fail
talk_helper failed_connect 1and a few
talk_helper failed_connect 1
block absorber died!
Z
00:57
Zack
Maybe you need a newer version of the code?
00:58
or maybe these are normal errors we can ignore. The internet is not perfect. Many messages get dropped.
01:00
In reply to this message
I am getting a lot of messages like this too.
It could be that someone running a public node is using old software that gives us invalid blocks and ignores the blocks we give it.
It could be that someone running a public node is using old software that gives us invalid blocks and ignores the blocks we give it.
G
01:33
In reply to this message
I'm pretty sure Im running the last version but anyway it would be nice if you keep us informed whenever a major update is available
Z
MH
01:40
Mandel Hoff
I'm updated but not syncing correct. I'll fix my node later. Stuck at 5280ish.
Z
01:43
Zack
In reply to this message
The config files were changed in a recent update.
You can manually delete them with
git pull isn't enough to fix them.You can manually delete them with
rm config/prod/sys.config and then they will be rebuilt according to the new rules
MH
01:47
Mandel Hoff
I think I lost sync when you rebooted and there were no other peers.
Z
01:48
Zack
there are 4 nodes up right now, so even if I reboot you should be able to sync with someone
01:49
when you use
Sometimes it even selects your own node, which is useless.
So you might have to call
sync:start(). it selects one peer at random to try and sync with.Sometimes it even selects your own node, which is useless.
So you might have to call
sync:start(). more than once in order to get it to sync.
02:14
Deleted Account
In reply to this message
5280 ? It was around 4200 yesterday who is mining ? Are you there is no gpy miner out there ?
Z
02:15
Zack
I haven't finished the gpu code yet. I am not sure if someone else wrote gpu code.
MH
02:23
Mandel Hoff
I cpu mined a lot at about 1 blk per minute
02:23
Difficulty is still pretty low for cpu
02:26
Deleted Account
Im trying to setup cpu miner on ubuntu but i keep getting errors
02:27
Can you guide me ? I want erase all data n start again
MH
02:34
Mandel Hoff
I haven't used the github miner in several weeks. It worked when I last used it but I forget all details
02:34
Deleted Account
Which miner are you using ?
MH
02:38
Mandel Hoff
I took the c miner code and built it on windows. I've been using that. Almost exactly the same code without the erlang wrapper around it. I remove the erlang part and added c++ Rest calls.
02:39
Deleted Account
Can you upload that miner and provide the link ?
02:39
It would help alot. I get many errors on ubuntu.
MH
02:43
Mandel Hoff
I will almost for sure before main net launch but working on some other bugs first.
02:43
Deleted Account
Ok thanks man
Deleted joined group by link from Group
G
03:11
BTW are you working on the pool?
Z
03:12
Zack
In reply to this message
I bet a lot of people would be interested in mining from windows. We should package this up as a torrent.
MH
03:26
Mandel Hoff
In reply to this message
Yes, my pool is "done" but it currently suffers greatly from the 1 min block time. My node experiences ~80% orphaned blocks, and I'm crediting miners for those blocks. My payout txs in those orphaned blocks also become "lost", so many payouts go "missing". Zack is pretty sure when block time goes to 10 minutes and sync fixes are implemented that this will mostly fix itself.
03:26
I will still need to update my pool to wait 10 blocks before crediting miners to avoid orphaned blocks that will still happen (but hopefully much less often than 80%).
Z
03:27
Zack
In reply to this message
the txs getting dropped wont get fixed by a slower block time. This is a different problem that I will try to fix.
MH
03:29
Mandel Hoff
Ok, I thought longer block time would let the txs propagate to other node's mem-pools, but sounds like you have some insight that I'm missing. Sounds great.
Z
03:30
Zack
Maybe I should make sync mode default to normal mode, and give instructions to manually switch in to quick mode if you want to sync lots of blocks.
MH
03:32
Mandel Hoff
Ethereum Geth client has a "fast sync" switch you have to opt-in to when you start a node.
Z
03:32
Zack
yeah, I should probably use the same strategy
Z
03:56
Zack
I pushed an update to fix the issue where the mempool was deleting all the txs.
Now you have to manually call
I updated the documentation to reflect this.
Now you have to manually call
sync_mode:quick(). if you are going to sync a lot of blocks.I updated the documentation to reflect this.
G
MH
05:02
Someone has a peer with a 10.x address which can't be valid:
Z
05:03
Zack
In reply to this message
yeah, those addresses look weird.
There is code to drop addresses that don't work, I guess it is failing to drop those for some reason.
There is code to drop addresses that don't work, I guess it is failing to drop those for some reason.
05:03
I started writing the GPU miner https://github.com/zack-bitcoin/amoveo-opencl-miner
M
05:05
Mike
When/if the original chain breaks, balances will not carry over when it restarts?
MH
05:06
Mandel Hoff
I just cloned new and sync'd. It got up to curent height good but it fails to stay sync'd with your node.
05:06
Z
05:06
Zack
In reply to this message
If the main net breaks, I will do everything I can to keep everyone's balances when we restart.
If I can't keep Amoveo alive for 6 months, then I wont get paid.
If I can't keep Amoveo alive for 6 months, then I wont get paid.
05:07
In reply to this message
Nodes do not stay synced by default. If you are mining, then they should stay synced.
MH
05:07
Mandel Hoff
My node tried to sync with that 10.x and then 52,234,133,196 which is not yours. Both says "already sync'd with this peer" but yours is many blocks ahead.
05:08
My mining pool is "on". You can see my node says that "mining with hash" message.
Z
05:08
Zack
yes, I guess those nodes got stuck at a lower height or something? If you fail to connect, then the error would be different and it would drop the peer from the list. You must be connecting to something.
I think the 52.235 one might be yours. so it is connecting to itself.
I think the 52.235 one might be yours. so it is connecting to itself.
MH
Z
05:09
Zack
So if you try syncing with my node, do you download the blocks then?
05:10
I am not sure what is going on with the 10.x nodes. maybe they are acting like localhost.
MH
05:10
Mandel Hoff
05:11
Yes, looks like when my sync chooses your node it absorbs blocks up to the height that matches your node height.
Z
05:12
Zack
great. We should probably update it so it doesn't connect to itself, and we should be more aggressive about dropping peers that stop functioning.
MH
05:12
Mandel Hoff
It just has a 1 in X chance of picking those bad nodes or self, so it seems unsync'd pretty often. I get many blocks behind.
Z
05:13
Zack
yes, this is probably part of the reason we keep mining on forks and ignoring each other's blocks.
Another way to fix this is by more forcefully sharing a block once you mine it.
Someone suggested an algorithm for this to me before.
Another way to fix this is by more forcefully sharing a block once you mine it.
Someone suggested an algorithm for this to me before.
MH
05:14
Mandel Hoff
That sounds good to prevent orphans. I'd much rather know about your blocks as you find them than be mining on something that's going to end up as an orphan.
Z
05:15
Zack
They said if you have a new block, you should keep trying to share it until more than 1/2 the people you connect to say they already know about it.
05:15
So new blocks would propagate very quickly.
MH
05:16
Mandel Hoff
That sounds great. Even on 10 minute blocks, I wouldn't wan't to be even a few seconds behind on the mining race for the next block.
Z
05:16
Zack
yes
MH
05:18
Mandel Hoff
Currently, we are roughly 1min/blk and I'm several blocks behind. 😢
Z
05:19
Zack
yes, it is good you realized this. Now Amoveo will be way better at launch than it would have been.
MH
05:19
Mandel Hoff
I'll leave my node running, but going out for a while.
05:19
👍
Z
05:20
Zack
I think Mandel provides more value per day than any programmer who has been paid to work with me at Augur or Aeternity.
MH
05:23
Mandel Hoff
😜 This much too much fun not to participate.
Jesse joined group by link from Group
RL
06:19
Thanks to both you guys. You are nailing it.
G
06:42
got this datafinished sending blockstalk_helper weird response
["mining_data"]
syncing with peer [-7,[-7,52,234,133,196],8080]
trade peers
mining data block hash is "KFMPuds5+1dDF6neUyH+DmauBDAoR0XeyI8lSG2vMUo="
talk_helper weird response
talk_helper weird response
they are not accepting our blocks.talk_helper weird response
talk_helper weird response
talk_helper weird response
talk_helper weird response
give blocks
give blocks
finished sending blockstalk_helper weird response
talk helper fail
talk_helper weird response
talk_helper weird response
talk helper fail
get their top header
already synced with this peer
talk_helper failed_connect 1
06:42
ups, sorry 😊
06:42
first time I see this *they are not accepting our blocks.*
Z
06:43
Zack
yes. I added a feature. If they don't accept blocks, we stop sending blocks to them. it prevents a failure mode where we were wasting bandwidth.
G
06:44
Gonzalo
ok nice
Z
06:47
Zack
In reply to this message
I did the 2 upgrades we talked about. Now you don't waste time syncing with yourself, and when you find out about a new block, you send it to people right away.
G
06:48
Gonzalo
nice
06:48
but what if nobody accept your blocks? Is that possible? Now I'm at height 5553 but you node is at 5552
06:49
so I guess I was the last one finding a block but your node is not aware of it, isn't it?
Z
06:52
Zack
I probably got dropped from your list of peers when I went offline. If you try pushing the block a couple more time, it will probably work.
06:52
once to download the new list of peers, and once to push the block to me
G
06:53
Gonzalo
syncing with peer [-7,[-7,159,89,106,253],8080]
06:54
this is your node
Z
06:54
Zack
yes
G
06:54
Gonzalo
so I suppose you are still in my list
Z
06:55
Zack
maybe I am now. it doesn't mean I was when you tried to send the block.
G
06:55
Gonzalo
anyway, how do I "push the block", no idea how to do it
Z
06:55
Zack
sync:start().
G
06:55
Gonzalo
ah, sorry 😊
Z
06:55
Zack
I have 5553 now, so I guess it worked.
G
06:56
Gonzalo
Is it possible to see how minted that block?
06:56
who
Z
06:58
Zack
BC8uGkC9tUhcBWcDh6wYuZh9Gyh5UDRy2v7ybXDERD22gMtFdYyhQQJCMXafzgfITP8CgNkllGvh1kac6NKOI90=
06:58
base64:encode(element(2, hd(element(11, block:get_by_height(5553))))).
06:59
If you use the new version of the code I just published, then you will be able to share blocks much faster.
G
07:01
Gonzalo
(ae_core_prod@scw-9653e4)46> block:height().
5553
(ae_core_prod@scw-9653e4)47> api:height().
5553
(ae_core_prod@scw-9653e4)48> base64:encode(keys:pubkey()).
<<"BC8uGkC9tUhcBWcDh6wYuZh9Gyh5UDRy2v7ybXDERD22gMtFdYyhQQJCMXafzgfITP8CgNkllGvh1kac6NKOI90=">>
07:01
This was me!!
Z
07:02
Zack
great
G
07:02
Gonzalo
👍
07:04
I forgot to do
sync_mode:quick().so it took ages to download the blocks
Z
07:04
Zack
yes, it is super slow without that.
07:05
I updated the docs already, I am not sure what else I should do.
G
07:06
Gonzalo
It would be nice if the node informs you every time the block height increases, so you don't have to ask the node
07:06
sometimes is not easy with all the messages showing quicky
Z
07:06
Zack
yeah, I should delete a lot of the unnecessary messages
G
07:08
Gonzalo
also when installing amoveo there are lots of warnings (i suppose that about variables not in use, not sure)
Z
07:09
Zack
yes, there are lots of compiler warnings it is true. Those are all things that should probably be fixed.
G
07:09
Gonzalo
anyway, nothing serious
Z
07:09
Zack
but it isn't important to do soon.
MH
07:11
Mandel Hoff
That was a fast fix. Updating now. Excited for this change.
Z
07:11
Zack
I just did another fix now, so you wont add IP addresses to the list if they are for private networks.
G
MH
07:15
Mandel Hoff
Love the bandwidth fixes. I'm back using azure again.
Z
07:16
Zack
Often times in software development the bottleneck is in realizing what needs to be done.
Once I know what to do, the actual programming is easy.
That is why it is so critical to have people like Gonzalo and Mandel.
Once I know what to do, the actual programming is easy.
That is why it is so critical to have people like Gonzalo and Mandel.
G
Deleted joined group by link from Group
G
07:19
Gonzalo
But as Mandel said before, "this is too much fun not to participate"
Z
07:21
Zack
In reply to this message
you are more valuable than a typical beta tester. You are here voluntarily. Early adopters are the life-blood of startups.
I hope to maintain this fun environment so that you will continue to help.
I hope to maintain this fun environment so that you will continue to help.
G
MH
07:25
Mandel Hoff
I'm all sync'd at 5555 now. How long does it take for it to auto-switch back to normal sync?
Z
07:26
Zack
it should switch back 20 blocks before you reach the top
MH
07:26
Mandel Hoff
MH
07:26
Mandel Hoff
Stays in quick sync, but I can change it manually.
Z
07:26
Zack
well for now you can manually do it
sync_mode:normal().
MH
07:27
Mandel Hoff
Works for me.
Z
07:28
Zack
I think ill just add instructions to the docs to do it manually both times.
Making it automatic when pulling blocks isn't too hard, but if someone else is pushing blocks to you, it seems impossible.
Making it automatic when pulling blocks isn't too hard, but if someone else is pushing blocks to you, it seems impossible.
G
07:28
Gonzalo
In reply to this message
This is not stated on the docs, maybe you have to update them
MH
07:36
Mandel Hoff
I'm all sync'd will check back later.
Z
07:36
Zack
cool
MH
07:37
Mandel Hoff
I'm going to point my miner at your pool and measure how many (average) seconds it takes for my pool to update its height to yours.
Deleted joined group by link from Group
G
07:38
Gonzalo
zack I'm still having 10.x.x.x peers. Best way to update the software without deleting the blockchain?
Z
07:39
Zack
good idea. My pool should give the block to you fairly quickly.
Though it looks like there are 3 nodes right now. if the 3rd peer is running old software, I might just give the block to them, and then my node will realize that 1/2 of the peers have the block, and then it might stop sharing.
Though it looks like there are 3 nodes right now. if the 3rd peer is running old software, I might just give the block to them, and then my node will realize that 1/2 of the peers have the block, and then it might stop sharing.
G
07:39
Gonzalo
thanks
Z
07:40
Zack
i just got 5562
07:41
Deleted Account
from the docs looks like CPU mining only? is this correct
Z
07:42
Zack
In reply to this message
yes. at this point in time there is only cpu mining, but I am working on making a gpu miner.
G
07:42
Gonzalo
I freshly updated from Zack's to 5563
07:42
Deleted Account
ok thanks
G
07:44
Gonzalo
Zack I did
git pullbut my node is still trying to sync with 10.x.x.x and 192.168.1.x
Z
07:44
Zack
did you turn the node off and on?
G
07:44
Gonzalo
yes, with api:off(). and halt().
Z
07:45
Zack
peers:all().
[{{162,218,209,18},8080},
{{52,234,133,196},8080},
{{159,89,106,253},8080}]
07:45
what is your
peers:all().?
G
07:46
Gonzalo
sorry I turned off. Now on again and says:
(ae_core_prod@scw-9653e4)1> peers:all().
[{{10,5,73,155},8080},{{159,89,106,253},8080}]
07:48
now, after sync:start().
07:48
(ae_core_prod@scw-9653e4)3> peers:all().
[{{162,218,209,18},8080},
{{52,234,133,196},8080},
{{10,5,73,155},8080},
{{159,89,106,253},8080}]
07:49
again, after a new sync:start().
07:49
(ae_core_prod@scw-9653e4)5> peers:all().
[{{10,0,2,15},8080},
{{10,0,1,4},8080},
{{162,218,209,18},8080},
{{172,17,0,1},8080},
{{52,234,133,196},8080},
{{192,168,1,107},8080},
{{192,168,0,16},8080},
{{10,5,73,155},8080},
{{159,89,106,253},8080}]
07:49
😳
Z
07:50
Zack
try the steps one more time.
turn it off, then do
Check if
Maybe you edited a file, and so
turn it off, then do
git pull.Check if
git pull is having some error, since it seems like you are stuck on the old version of the code.Maybe you edited a file, and so
git pull is failing and giving you an error asking if you want to save your changes.
G
07:50
Gonzalo
ok thanks
Z
07:51
Zack
git is a powerful tool, but it can also be annoying.
G
07:51
Gonzalo
This is what said _before_:
07:52
amoveouser@scw-9653e4:~/amoveo$ git pull
remote: Counting objects: 56, done.
remote: Compressing objects: 100% (20/20), done.
remote: Total 56 (delta 48), reused 44 (delta 36), pack-reused 0
Unpacking objects: 100% (56/56), done.
From https://github.com/zack-bitcoin/amoveo
6411857..735ce9d master -> origin/master
Updating 6411857..735ce9d
Fast-forward
apps/ae_core/src/consensus/chain/block_absorber.erl | 11 ++++++-----
apps/ae_core/src/consensus/sync.erl | 52 +++++++++++++++++++++++++++++++++++++++++++---------
apps/ae_http/src/ext_handler.erl | 4 ++--
apps/ae_http/src/peers.erl | 19 +++++++++++++++----
docs/api/commands_basics.md | 4 ++++
docs/todo.md | 9 +++++----
6 files changed, 75 insertions(+), 24 deletions(-)
07:52
This is what says now:
07:52
amoveouser@scw-9653e4:~/amoveo$ git pull
Already up-to-date.
07:52
Everything looks ok for me
07:54
Ahh, maybe I was supposed to do "make prod-build" after "git pull" ???
Z
07:54
Zack
The code was updated 3 times in the last hour. This
git pull might be from too long ago
07:54
make prod-restart is how I would do it.
07:54
I think you are right
07:54
you solved it without my help
G
07:54
Gonzalo
aha, thanks... Remember what I said before about me being a bit noob XD
Z
07:55
Zack
no, it is my fault for making it so confusing
cantabrew joined group by link from Group
Z
07:56
Zack
http://162.218.209.18:8080/wallet.html
This person is running a node, but I can't sync any headers.
This person is running a node, but I can't sync any headers.
G
Z
07:56
Zack
I wonder who it is
G
07:56
Gonzalo
hehe
07:59
Deleted Account
Nice
G
08:01
Gonzalo
Ok Zack, problem solved:
(ae_core_prod@scw-9653e4)10> peers:all().
[{{162,218,209,18},8080},
{{52,234,133,196},8080},
{{159,89,106,253},8080}]
Kevin Roberts BWD invited Deleted Account
G
08:03
Gonzalo
Question: this first 3 IPs are "harcoded" on amoveo?
MH
08:11
Mandel Hoff
No. I'm the 52.x now but I have dynamic azure IPs. They change every few days on reboots
G
08:13
So you should also see my node when typing "peers:all()." ?
08:13
I'm 51.15.212.91
Z
08:14
In reply to this message
no, currently only my server is hard-coded. you downloaded the other 2 from my server.
MH
08:14
Mandel Hoff
I don't have you. Just the 3 above you listed
G
08:15
Gonzalo
that sounds weird
08:16
Deleted Account
so i’m interested in running a node miner and understand cpu only at this time. is cloud hosting an option i use digital ocean for my servers but i suppose i could build a local server in my house
Z
08:16
Zack
In reply to this message
I don't have it either.
It is supposed to be automatic, but my software probably isn't perfect. It is also possible that your port is being blocked by a fire wall or something.
We can try adding it manually and see what happens.
It is supposed to be automatic, but my software probably isn't perfect. It is also possible that your port is being blocked by a fire wall or something.
We can try adding it manually and see what happens.
08:16
Deleted Account
ok great
G
08:17
Gonzalo
ok thanks. Any idea on how to check ports on Ubuntu?
MH
08:18
Mandel Hoff
I opened azure ports through the azure dashboard networking section. No idea how you would need to do your setup.
Z
08:18
Zack
In reply to this message
I added you manually, and you stuck in the list. So it seems like your port is already forwarded. Next I will try using the light node from your server.
08:19
http://51.15.212.91:8080/wallet.html
The light wallet does work. So it looks like Gonzalo has everything set up correctly, and my software failed to automatically add his ip to the list.
Maybe I broke this feature with a recent update.
The light wallet does work. So it looks like Gonzalo has everything set up correctly, and my software failed to automatically add his ip to the list.
Maybe I broke this feature with a recent update.
08:22
when i added you manually, you spread to the server correctly. and the server still recognizes you.
08:22
so the server is sending blocks to you when they are found
MH
08:22
Mandel Hoff
I still type add peer command out of habit. Forgot you tried to automate that
Z
08:23
Zack
I will turn my server off for a bit so you all delete me, then I will turn it on again.
toptickcrypto joined group by link from Group
Deleted joined group by link from Group
Z
08:27
Zack
it doesn't work. my ip isn't disappearing. I think maybe the blacklist isn't working right.
08:37
yes, the blacklist was broken. So I fixed it.
I still don't know why Gonzalo's ip didn't get added automatically.
I still don't know why Gonzalo's ip didn't get added automatically.
MH
08:46
Mandel Hoff
Zack, your node is up but is your pool off?
Z
08:54
I think i broke my mining pool
08:56
it seems to be working now. I am not sure what went wrong before.
MH
09:03
Mandel Hoff
Confirmed. I'm mining and measuring but so far block propagation looks very improved
Z
09:04
Zack
I think my VPS doesn't charge extra if I leave the CPU at 100% all the time.
The server seems to use less than 1% of the CPU on average to serve everything to everyone.
The server seems to use less than 1% of the CPU on average to serve everything to everyone.
09:04
So I am thinking I should just leave the miner turned on all the time
09:05
once everyone updates to the improvement, then blocks should propagate even better
MH
09:05
Mandel Hoff
Did you make fixes to the tx propagation? Should I test making txs on my node while only mining on your node?
Z
09:06
Zack
it should not be deleting txs when blocks are found.
MH
09:06
Mandel Hoff
Ok, testing making spends on my node while mining only on yours
Alex PSN:Ladimir_Lepin joined group by link from Group
MH
09:13
Mandel Hoff
Might have lost some txs I put two 0.1 veo in my mempool. A create and spend. They are not in mempool and balances unchanged.
09:13
09:14
Sender:
BD1rrLUjQO/jEm3yO2lfAHziKr3aY+F+4IZRRa3mYuczIvRMqj8XcxTQcu+9QJUsT+nHoAcKCxu7zOqixpoyRpo=
Recipient:
BMblRgpUyYDgvF1trmqR02Vobat/JoQTTPCL91jUeGKSDDXaGbAeIKrI0ftR9FxXenXK89/+tWn76Qy9RJlzKwM=
BD1rrLUjQO/jEm3yO2lfAHziKr3aY+F+4IZRRa3mYuczIvRMqj8XcxTQcu+9QJUsT+nHoAcKCxu7zOqixpoyRpo=
Recipient:
BMblRgpUyYDgvF1trmqR02Vobat/JoQTTPCL91jUeGKSDDXaGbAeIKrI0ftR9FxXenXK89/+tWn76Qy9RJlzKwM=
Z
09:15
Zack
and you are in sync_mode normal?
MH
09:16
Mandel Hoff
09:16
Yes
Deleted joined group by link from Group
Z
MH
09:21
Mandel Hoff
Want me to turn off my mining? Approaching difficulty recalc at 6k
09:22
I don't want to make testing take longer when diff increases.
Z
09:22
Zack
yes, good idea
MH
09:23
Mandel Hoff
ok, my miner is off.
Z
09:27
Zack
In reply to this message
after further inspection, I see that the tx never got deleted from the mempool, and the problem instead was that it wasn't shared.
Amadeo 🦇🔊 joined group by link from Group
Z
09:30
Zack
oh, I see a
died! in your log. I think I know what is wrong. one of the txs fails to get included, so the miner drops the rest from the list.
A
09:38
Amadeo 🦇🔊
I like the idea of Amoveo ... but why build your own protocol layer? Would it not be more effiecent to build ontop of an existing protocol for extra security?
Z
09:39
Zack
In reply to this message
Amoveo's oracle is deeply connected with it's consensus mechanism. Most blockchain communities are not so comfortable with updates that change the consensus mechanism. If the update has a mistake, it would put all the tokens at risk.
A
09:42
Amadeo 🦇🔊
I understand that but still with an seperate Dapp sidechain you would also have a lot of freedom
09:42
Where is more info on the Amoveo consensus?
Z
09:44
Zack
In reply to this message
it is the same consensus mechanism as bitcoin.
https://github.com/zack-bitcoin/amoveo/tree/master/docs
https://github.com/zack-bitcoin/amoveo/tree/master/docs
09:45
In reply to this message
Every blockchain I have looked at is unsuitable for building Amoveo. I would have to make such extensive modifications that it is easier to start over.
A
09:46
Amadeo 🦇🔊
In reply to this message
Ok cool but you say you gonna change that ...? In what way? This looks the same as BTC as of now
Z
09:46
Zack
For example, Amoveo has the same merkel tree benefits of Ethereum, but unlike ethereum it is also compatible with sharding.
RL
09:47
R L
Plus it's more fun to make your own thing from ground up
09:47
Every project is rushing for "big gainz" pump and dumps
09:49
Why not take a relaxed look at what needs improvement and start from scratch? I think it's great.
Z
09:51
Zack
Amoveo needed a custom virtual machine because the computation is happening inside the state channels, and this environment is specialized.
09:52
Amoveo's oracle can also be used as a governance mechanism, so I designed the consensus mechanism with this in mind.
A
09:54
Amadeo 🦇🔊
Ok cool guys
09:56
As I may give my feedback this part I would remove ... or make clear how the blockcain will say: The blockchain will eventually say something like "The movie will make about X more money with Bob as the main actor rather than Charlie".
09:57
I can not see without a verry complex orecal system and AI how this would work
09:57
Also some other examples lack technical backing
Z
09:58
Zack
In reply to this message
I was abused by a boss, and lost a lot of money because I didn't cover it up.
This is one of the most important use cases of Amoveo for me.
The oracle doesn't tell the blockchain the price until afterwards. The oracle doesn't predict the future, the prediction market does.
This is one of the most important use cases of Amoveo for me.
The oracle doesn't tell the blockchain the price until afterwards. The oracle doesn't predict the future, the prediction market does.
09:58
Deleted Account
Hi how to check my hashrate?
Z
09:59
Zack
In reply to this message
the browser miner says it in the browser. but the browser miner is also very slow.
There is no easy way with the c-miner yet.
There is no easy way with the c-miner yet.
10:00
In reply to this message
So the news papers will already say how much money the movie made. We just look it up in the newspaper, and then tell the blockchain. That is what an oracle is.
10:00
Deleted Account
Thanks
10:02
Keep showing “did not find a block in that period”. Normal?
10:02
Is it mining?
Z
10:02
Zack
yes, that means you are mining.
10:03
Deleted Account
Z
10:03
Deleted Account
Thanks. This means I minted 2 blocks, right?
Z
10:04
Zack
no, just one.
10:04
and it looks like it failed to share. probably you aren't synced with the network.
10:04
Deleted Account
Ah. Keeps showing like this.
10:05
My network connection issue?
10:05
I just downloaded and installed miner and dependency.
Z
10:07
Deleted Account
was thinking that we should have the cpu miner only for the launch so everyone can mine. then if we notice that someone is about to take over the network by closed source gpu miner, we can have gpu miner ready to run in that case. what do you think Zach?
Z
10:10
Zack
In reply to this message
I fixed it. Now the txs don't get dumped if one of them is invalid.
10:10
Deleted Account
Should be more convenient if we can check our hashrate
Z
10:10
Zack
In reply to this message
no, you are optimizing for the wrong thing.
It doesn't matter how many people can mine.
What matters is the profit margin of the average miner. The lower the profit margin, the better.
It doesn't matter how many people can mine.
What matters is the profit margin of the average miner. The lower the profit margin, the better.
MH
10:22
Deleted Account
10:22
Cpu usage like 1% . Miner not working.
10:22
Do I need to install any other software besides c-miner and dependency?
Z
10:23
Zack
In reply to this message
No. this is because I just restarted the server. it should work if you try again now.
10:40
Deleted Account
Thanks. Works now.
Z
10:40
Zack
great
Deleted joined group by link from Group
dimsim joined group by link from Group
KR
11:38
Kevin Roberts BWD
Node and miner up
11:41
11:57
Deleted Account
zack when do u think gpu mining would be ready after the mainnet is set up
Z
Paul N. | BAND (🅑) - I'll never DM you joined group by link from Group
P|
15:06
Paul N. | BAND (🅑) - I'll never DM you
when is the launch date again?
CB
15:36
CeloStarter Bandits
2 March
15:36
i think
15:36
expect forks bugs frustration.. so all good
Z
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
Long John joined group by link from Group
LJ
21:57
Long John
Are there any plans to keep snap shots in case everything goes to hell and Amoveo breaks?
Z
21:59
Zack
Well, there is no situation where amoveo would delete any of the blocks it downloaded.
As long as we have all the blocks, we can rebuild the state.
There are some failure modes where being able to rebuild the state isn't enough to save the blockchain. Some hyperinflation attacks for example.
As long as we have all the blocks, we can rebuild the state.
There are some failure modes where being able to rebuild the state isn't enough to save the blockchain. Some hyperinflation attacks for example.
LJ
22:03
Long John
Cool
22:03
Im going to have a play around with things tomorrow and just do some reading up today
Z
22:03
Zack
Cool. I'm here if you have questions.
LJ
22:03
Long John
Does mining take advantage of more cores?
Z
G
22:04
Gonzalo
Hi Zack, I'm minning now and I seems to me that I'm not getting any reward, is this possible?
22:06
Deleted Account
So whether a local PC or a VPS mine faster?
G
22:06
Gonzalo
And how do you know if you find and orphan block?
22:07
Deleted Account
I am think about buying a new PC or renting more VPS as many as possible
G
22:07
Gonzalo
In reply to this message
You can configure the c-miner to use as much cores as you want/have
Z
G
22:08
Gonzalo
In reply to this message
I suggest not to buy a new PC since CPU mining will no last for long time
LJ
Z
22:09
Zack
In reply to this message
Good question.
The only way now is by looking up the block in the chain to see who's address is written in the coin base TX.
The only way now is by looking up the block in the chain to see who's address is written in the coin base TX.
G
22:10
Gonzalo
In reply to this message
Well, it's up to Zack's hability to write the GPU miner, so I suppose it's only a matter of days or weeks ;-)
LJ
22:11
Long John
Guess Im going to have to unpack my mining motherboards
22:11
Havent touched them for 1 year since I quit ETH mining
G
LJ
Z
22:13
Zack
I am thinking of adding a thing to the explorer. to display some info about the 10 most recent blocks. like who mined them, and how many txs are in each one.
G
22:13
Gonzalo
This will be very usefull!
22:15
And also will be relief to you, probably less questions to answer...
LJ
22:16
Long John
Do you think a mining pool will be set up fairly quickly or better to mine on a local node?
G
22:18
Gonzalo
Zack´s pool is at 159.89.106.253:8085. Also Mandel is working on another one at amoveopool.com
22:18
Deleted Account
22:18
Am i mining correctly ?
G
22:19
Gonzalo
It looks ok to me, but you are not getting any reward like me
22:19
Deleted Account
Yes i aint
22:21
What is the current block height
G
22:21
Gonzalo
5699
22:22
You can verify here: http://159.89.106.253:8080/explorer.html click on "update height"
22:23
and here: http://amoveopool.com/pool
Z
22:27
Deleted Account
Synced again lets see if i get rewards now
22:30
But when i type api:height(). On node It shows me 5699
Z
22:30
Zack
maybe your miner is connected to my mining pool instead of to your amoveo full node?
22:31
did your balance go up once you sync?
22:31
Deleted Account
No
22:31
Its still zero
Z
22:32
Zack
what is your pubkey?
G
22:32
Gonzalo
is there any diff between api:sync() and sync:start() ??
22:33
Deleted Account
G
22:33
Gonzalo
you have to update the software!
22:34
Deleted Account
I did download new one last night
Z
22:34
Zack
In reply to this message
https://github.com/zack-bitcoin/amoveo/blob/master/apps/ae_http/src/api.erl#L373
looks like they are the same thing.
api.erl is all the functions that are exposed to an http interface to your local machine.
So any other software running on the same machine can access these commands.
looks like they are the same thing.
api.erl is all the functions that are exposed to an http interface to your local machine.
So any other software running on the same machine can access these commands.
G
22:35
try:
22:35
```api:off().
Z
22:36
Deleted Account
Can i update it from within the node
Z
22:37
Zack
first turn it off, then do
git pull then turn it on again.
G
22:39
Gonzalo
Zack I see lots of "block absorver died!" and I'm below the actual block height. I suppose I'm not getting the updates, isn't it?
22:40
Deleted Account
Whats the command for git pull
G
22:40
Gonzalo
git pull
22:40
Deleted Account
Z
22:41
Zack
anything like
this is something you can copy paste into your terminal.
22:41
images load really slow in my part of the world. It is better to copy paste, and put it between triple back-ticks, like
22:41
this
22:43
you used git when you installed amoveo, there is no way you don't have it now.
Z
22:44
Zack
well then, do that again
22:44
or install git
G
22:44
Gonzalo
that's why your folder is not "amoveo" I guess
Z
22:44
Zack
amoveo comes with instructions on dependencies. you install git as part of the process of installing amoveo
Z
MH
22:46
Mandel Hoff
22:46
Zack, I just did a fresh clone and my node tried to sync with itself:
G
22:47
Gonzalo
In reply to this message
I happens to me as well since we are in the peers list I suppose
Deleted joined group by link from Group
Z
22:50
Zack
In reply to this message
since you manually added your peer to the list, and it was also automatically added, I think it is now in the list twice.
My code removes the first instance of your ip, then randomly selects from the list.
My code removes the first instance of your ip, then randomly selects from the list.
MH
22:51
Mandel Hoff
Ok, I'll stop doing the add peer manually.
Z
22:52
Zack
oh nevermind, the code doesn't let you add a peer twice. my mistkae.
22:54
Deleted Account
Sync_mode:quick().quick is for faster syncing?
Z
22:54
Zack
yes.
22:56
Deleted Account
Im syncing with normal mode how can i stop and shift to quick mode
MH
22:56
Mandel Hoff
22:57
I got this on my sync. I typed sync:start(). a few more times and it resumed fine now. I'm up to 5200ish now.
Z
22:58
In reply to this message
yes. someone using that ip address is trying to access your server more frequently than the limit set in the config file.
G
22:59
Gonzalo
requesting too often: [-7,178,152,130,230] died!requesting too often: [-7,178,152,130,230] died!requesting too often: [-7,178,152,130,230] died!requesting too often: [-7,178,152,130,230] died!requesting too often: [-7,178,152,130,230] died!requesting too often: [-7,178,152,130,230] died!requesting too often: [-7,178,152,130,230] died!requesting too often: [-7,178,152,130,230] died!requesting too often: [-7,178,152,130,230] died!requesting too often: [-7,178,152,130,230] died!requesting too often: [-7,178,152,130,230] died!requesting too often: [-7,178,152,130,230] died!requesting too often: [-7,178,152,130,230] died!requesting too often: [-7,178,152,130,230] died!requesting too often: [-7,178,152,130,230] died!requesting too often: [-7,178,152,130,230] died!requesting too often: [-7,178,152,130,230] died!requesting too often: [-7,178,152,130,230] died!requesting too often: [-7,178,152,130,230] died!requesting too often: [-7,178,152,130,230] died!requesting too often: [-7,178,152,130,230] died!requesting too often: [-7,178,152,130,230] died!requesting too often: [-7,178,152,130,230] died!requesting too often: [-7,178,152,130,230] died!requesting too often: [-7,178,152,130,230] died!requesting too often: [-7,178,152,130,230] died!requesting too often: [-7,178,152,130,230] died!requesting too often: [-7,178,152,130,230] died!
Z
22:59
Zack
that same ip is trying to contact me too frequently too.
G
Z
23:01
Zack
In reply to this message
what happens if you do
does it give the ip address for your server?
peers:my_ip(). ?does it give the ip address for your server?
G
23:02
Gonzalo
peers:my_ip().
{10,5,73,155}
MH
23:03
I'm on an Azure instance.
Z
23:03
Zack
In reply to this message
oh, it looks like it is from a cellular network? I guess their phone is interpreting the light node code a little differently or something.
looks like
looks like
peers:my_ip(). is failing on both of your systems. those are internal network ip addresses.
23:04
maybe I should write it so you ask some other node what your ip address is.
23:05
😂😂
23:06
Its me
G
23:06
Gonzalo
😂
MH
23:10
Mandel Hoff
I just sync'd a fresh node from a new clone. Is this mempool record a tx from your pool that sync'd to me?
23:10
23:11
It looks like a 0.7 block payout, so might make sense that my node absorbed it.
Z
23:11
Zack
it looks like it came from someone's mining pool.
23:11
if it is spending from my address, then it must be from my pool
MH
23:11
Mandel Hoff
👍
23:15
Deleted Account
my node is stuck at 5705 block its not moving
MH
23:20
Mandel Hoff
23:20
23:21
I ran some tx tests, and something was strange. I used all 0.1 amounts. I made a create account for BMblRgpUyYDgvF1trmqR02Vobat/JoQTTPCL91jUeGKSDDXaGbAeIKrI0ftR9FxXenXK89/+tWn76Qy9RJlzKwM=
23:21
Sender was always this: BD1rrLUjQO/jEm3yO2lfAHziKr3aY+F+4IZRRa3mYuczIvRMqj8XcxTQcu+9QJUsT+nHoAcKCxu7zOqixpoyRpo=
23:21
I put several 0.1 spends in the mempool. Some got mined, and my wallet showed 0.1 and then 0.4.
23:22
Then I absorbed some more blocks and I'm back at 0.2 for that new wallet now.
Z
23:23
Zack
it sounds like only some of the txs from your mempool ended up in a block
MH
23:23
Mandel Hoff
All together, I should have put 0.7 in the new wallet via 0.1 txs. Strange that it went up to 0.4 and then down to 0.2 too.
MH
23:24
Mandel Hoff
Feels like some of the txs ended up in orphaned blocks, but that's sort of just a guess
23:25
Deleted Account
Z
23:48
Zack
In reply to this message
@Mandelhoff I just pushed a change. now you look up your ip by asking a peer. so this should be fixed for you guys.
MH
23:50
Mandel Hoff
Great!
Z
23:50
Zack
In reply to this message
I think your database got corrupted. We all keep sending you the blocks, but you wont accept them.
23:50
there is a proper way to turn it on and off so it doesn't get corrupted.
23:51
n then restart
Z
23:51
Zack
ok
G
23:53
Deleted Account
(ae_core_prod@usman-GE70-2PC)1> sync_mode:quick().
quick
(ae_core_prod@usman-GE70-2PC)2> sync:start().
syncing with peer [-7,[-7,159,89,106,253],8080]
trade peers
ok
get their top header
get headers
talk_helper tmimeout
quick
(ae_core_prod@usman-GE70-2PC)2> sync:start().
syncing with peer [-7,[-7,159,89,106,253],8080]
trade peers
ok
get their top header
get headers
talk_helper tmimeout
23:54
it doesn't sync and height is 0
Z
23:55
Zack
my node says it tried sending the first 4000 blocks 5 times in a row. I wonder why they didn't arrive.
G
23:56
Gonzalo
FYI I'm syncing right now
Z
G
23:58
Gonzalo
will do in a while, Im waiting the node to finish syncing, then I will do git pull and try
Z
23:58
Zack
it doesn't have to be synced to do this
23:58
you can pause in the middle of syncing without messing it up.
G
23:58
Gonzalo
how?
Z
23:59
Zack
api:off().
halt().
then you restart with
make prod-restart
G
23:59
Gonzalo
thanks, done
20 February 2018
Z
00:00
Zack
so what is your
peers:my_ip(). ?
G
00:00
Gonzalo
(ae_core_prod@scw-9653e4)4> peers:my_ip().
{10,5,73,155}
MH
00:00
Mandel Hoff
Need a git pull too
G
00:00
Gonzalo
I did git pull & make prod-restart
Z
00:01
Zack
what if you try syncing first, then get your ip?
00:02
if you have more peers to get your ip from, then maybe it will work
MH
00:02
Mandel Hoff
00:02
make prod-restart error. I'll start over
Z
00:02
Zack
no, wait
00:02
just do
make prod-attach
00:02
I should put a little delay in there, sometimes attach is running too soon
MH
00:02
Mandel Hoff
Attach attached.
G
Z
MH
00:04
Mandel Hoff
Z
00:04
Zack
still no good
G
00:07
Gonzalo
(ae_core_prod@scw-9653e4)11> sync:start().
syncing with peer [-7,[-7,159,89,106,253],8080]
trade peers
ok
get their top header
my height is less
(ae_core_prod@scw-9653e4)12>
(ae_core_prod@scw-9653e4)12> absorb block 4151
server crashed.
talk helper fail
server crashed.
talk helper fail
server crashed.
talk helper fail
absorb block 4152
server crashed.
talk helper fail
server crashed.
talk helper fail
00:07
lots of "server crashed"...
Z
00:08
Zack
yeah, I think that is happening if you contact them too frequently. I made the protocol more defensive recently, so now I need to slow down how quickly we talk at each other.
00:08
you should be able to safely ignore it for now.
G
00:08
Gonzalo
👍
Z
MH
00:15
Mandel Hoff
Works now
Z
00:15
Zack
great
MH
00:16
Mandel Hoff
G
00:18
Gonzalo
(ae_core_prod@scw-9653e4)1> peers:my_ip().👍
{51,15,212,91}
Z
00:18
Zack
great
00:20
Deleted Account
ok so I'm setting up the node, and it's giving me
block absorber died!
block absorber died!
00:23
I'm at sync:start(). in the instructions
G
00:24
server crashed.
talk helper fail
server crashed.
talk helper fail
absorb block 4259
server crashed.
talk helper fail
prune block 4230
prune block 4231
prune block 4232
prune block 4233
prune block 4234
prune block 4235
prune block 4236
prune block 4237
prune block 4238
prune block 4239
server crashed.
talk helper fail
server crashed.
talk helper fail
absorb block 4260
server crashed.
talk helper fail
server crashed.
talk helper fail
server crashed.
talk helper fail
absorb block 4261
server crashed.
talk helper fail
server crashed.
Z
00:25
Zack
In reply to this message
yes, there is a problem introduced with the change to
peers:my_ip(). it will be fixed soon.
G
00:25
Gonzalo
👍
MH
00:30
Mandel Hoff
Zack, I tried http post with ["block",5] but got server 500 error. How do I post to the api for a block?
00:33
Nevermind. That worked.
00:33
My json was poorly formed.
Z
00:33
Zack
great
00:37
I fixed the issue where new nodes were unable to sync.
G
00:42
Gonzalo
after git pull, without syncing before:
00:42
(ae_core_prod@scw-9653e4)2> peers:all().
[{{159,89,106,253},8080}]
(ae_core_prod@scw-9653e4)3>
(ae_core_prod@scw-9653e4)3> peers:my_ip().
[-7,[-7,159,89,106,253],8080]
talk_helper tmimeout
talk_helper tmimeout
talk_helper tmimeout
talk helper fail
{10,5,73,155}
00:43
Deleted Account
G
00:43
Gonzalo
now syncing, lots of error messages:
00:43
talk_helper tmimeout
talk_helper tmimeout
talk helper fail
[-7,[-7,159,89,106,253],8080]
talk_helper tmimeout
talk_helper tmimeout
talk_helper tmimeout
talk helper fail
absorb block 4463
[-7,[-7,159,65,134,21],8080]
server crashed.
talk helper fail
talk_helper tmimeout
talk_helper tmimeout
talk_helper tmimeout
talk helper fail
[-7,[-7,52,234,133,196],8080]
talk_helper tmimeout
talk_helper tmimeout
talk_helper tmimeout
talk helper fail
[-7,[-7,159,89,106,253],8080]
talk_helper tmimeout
talk_helper tmimeout
talk_helper tmimeout
talk helper fail
absorb block 4464
[-7,[-7,159,65,134,21],8080]
server crashed.
talk helper fail
talk_helper tmimeout
talk_helper tmimeout
talk_helper tmimeout
talk helper fail
Z
00:44
Zack
In reply to this message
yes. it doesn't work so well until you download the list of peers.
00:47
I did the upgrade Gonzalo suggested. Now the explorer lets you look up who mined each block.
G
MH
SS
01:08
S S
Zack reading through your code. Love your comments 'instead of getting the thing, we can build it while doing store' :)
Gregor joined group by link from Group
Z
01:51
Zack
https://github.com/zack-bitcoin/amoveo-c-miner
I added a speed test feature to the c-miner. So you can measure how fast your computer mines blocks.
I added a speed test feature to the c-miner. So you can measure how fast your computer mines blocks.
Z
Gal R joined group by link from Group
Z
03:21
Zack
My computer gets about 0.34 megahashes per second. According to the new c-miner speed test.
So for me the c code is about 7 times faster than Javascript.
So for me the c code is about 7 times faster than Javascript.
03:22
Deleted Account
so when i find a block i got two responses first server gave confusing response second is did not find a block in that period
03:22
is that normal..?
MH
03:22
Mandel Hoff
Is that 0.34 per thread? I get about 2.2mhs on 7 threads
03:23
how long did u check it ..?
Z
03:23
Zack
no, 0.17 per thread.
So each thread is 3.5x faster in c than javascript.
So each thread is 3.5x faster in c than javascript.
03:23
It takes about 3 seconds to run the speed test for me. because the test does 1 megahash, and my computer is 1/3rd mhps.
MH
03:24
Mandel Hoff
Ok
Z
03:24
Zack
In reply to this message
It looks like that block didn't get used. are you running the freshest code? did you sync before mining? did you switch to sync_mode:normal(). before mining?
03:31
Deleted Account
yes i am
03:31
03:32
how much is this
Z
03:32
Zack
looks like you mined about 6 blocks. That is about 6 VEO.
03:32
Deleted Account
Z
03:33
Zack
did you download the newest version of the miner?
03:33
yes
03:33
i also did git pull
Z
03:33
Zack
it should say "difficulty is 0" when you run the test
03:34
oh, it is doing the same for me. I must have broken it.
Z
03:36
Zack
In reply to this message
ok, I found the problem. the documentation was wrong, it skipped a step. I updated the directions, and pushed the fix to github.
Z
03:41
Zack
great
03:41
Deleted Account
per cpu it means per thread..?
Z
03:41
Zack
yes
03:42
your hardware has a finite number of CPU, if you try to run more threads than this it would make the mining slower.
03:42
my laptop has 2 CPU, so I mine with 2 threads.
03:44
Deleted Account
yes it has 4 core n 8 threads
03:44
how much do you recommend.?
03:45
many*
Z
03:46
Zack
oh yes, I have heard of this. Some cpu are designed to support parallel execution inside of them.
I think you should run 8 threads of mining.
Or only 7, if you want to use your computer for something else at the same time.
I think you should run 8 threads of mining.
Or only 7, if you want to use your computer for something else at the same time.
Z
03:49
Zack
In reply to this message
My computer has a tool that lets me look at the active running processes, and how much computational load each is consuming.
You could use a tool like this to easily know if you are running the right number of threads.
You could use a tool like this to easily know if you are running the right number of threads.
03:49
top is a command line tool for this, but I think most OS's today come with a GUI version
03:50
Deleted Account
is it possible to check balance while mining .? yes i m using htop
Deleted invited Alex the Trading Monk
Deleted invited Deleted Account
Z
03:51
Zack
yes. there are lots of ways.
The easiest way is probably to use the explorer to look up your balance: http://159.89.106.253:8080/explorer.html
The easiest way is probably to use the explorer to look up your balance: http://159.89.106.253:8080/explorer.html
03:51
Deleted Account
looks like my full node is synching.. lots of
prune block xxxx
died!absorb block XXXX
prune block xxxx
died!absorb block XXXX
03:52
i have to assume normal behavior
Z
03:52
Deleted Account
no...
Z
03:57
Deleted Account
ok great... quick working
03:57
thanks
Z
03:57
Zack
great
03:57
Deleted Account
how long to synch?
03:57
what's the block heigtht?
Z
03:58
Deleted Account
ok
03:58
can i run the miner on the same system as the full node?
Z
03:58
Zack
In reply to this message
you don't have to install anything to use Amoveo. There is a light node in the browser: http://159.89.106.253:8080/wallet.html
03:58
In reply to this message
yes.
You can also run the miner alone, and connect it to a mining pool.
You can also run the miner alone, and connect it to a mining pool.
03:59
Deleted Account
In reply to this message
it doesn't work in explorer. i loaded the same key in light node
03:59
it showed me balance
03:59
Deleted Account
i want to run a full node, and solo-mine
03:59
so 2 computers one for full node and one for miner?
04:00
might as well test everything
Z
04:00
Zack
In reply to this message
you could do that. or you could put miners on both computers, and an amoveo full node on just one.
04:00
Deleted Account
ok
Z
04:01
Zack
you could run a mining pool so that both of your computers are connected to the same amoveo full node
04:01
Deleted Account
sure.. I can do that
Z
04:01
Zack
In reply to this message
https://github.com/zack-bitcoin/amoveo-mining-pool
here is the mining pool software.
here is the mining pool software.
04:01
Deleted Account
I don't do mining pools.. if I can help it anyway
04:02
unless I'm running it i guess.. thanks I'll install it
04:03
i couldn't find the block reward anywhere for production in March.. is it 50 like bitcoin?
04:03
Deleted Account
its .7
Z
04:03
Zack
You would mine faster and use less bandwidth if you connected both your computers to a pool.
With just 2 computers it doesn't matter much, but if you had hundreds this would be more important.
With just 2 computers it doesn't matter much, but if you had hundreds this would be more important.
04:03
Deleted Account
ok
Z
04:04
Zack
In reply to this message
the block reward is about 1, but the mining pool is currently taking a 0.3 fee.
04:04
if you run your own mining pool, or if you solo mine without a pool, then you don't have to pay 0.3, so you get to keep the entire reward.
Z
04:05
Zack
In reply to this message
it didn't. you though that it was 0.7 because you have been paying 0.3 as a fee to the mining pool.
04:05
1 - 0.3 = 0.7
04:05
Deleted Account
I'll run my own mining pool then.. if the network is stable then I'll just upgrade as required
04:06
hardware upgrade that is...
04:07
once things are going
Z
04:07
Zack
sounds good
04:07
Deleted Account
i've been following you for a while
04:07
so looking for a good project to support - mining
Z
04:08
Zack
great
04:10
Deleted Account
oh yah. 1 block reward then what’s the max supply?
Z
04:13
I cannot predict what the community will decide.
04:13
maybe there will be no max.
04:14
Deleted Account
ok. i couldn’t find it defined. thanks
Deleted joined group by link from Group
04:26
Deleted Account
will this coin be gpu-mineable?
Z
04:26
Deleted Account
upon launch on March 2, just cpu mining?
Z
04:41
Deleted Account
ok
04:41
will check back in closer to the launch date
04:42
thanks and goodluck with the project
MH
04:56
Mandel Hoff
Zack, is there some part of the block that says how large the coin base tx was for the miner that made it?
04:57
I'm working on the pool block confirmation. I got miner address, but I'd like to also get veo value if possible too. Looking up governance value for "X blocks ago" isn't going to be always appropriate.
Z
04:58
Zack
I think not.
It wouldn't be expensive to preserve all the governance for every block. we could make it a configuration setting, if this is something people want.
It wouldn't be expensive to preserve all the governance for every block. we could make it a configuration setting, if this is something people want.
04:58
If several blocks use the same governance data, then you only store the data once. You don't have to re-store with each block.
04:59
governance data is rarely updated, so it would be very cheap to store it all.
MH
05:01
Mandel Hoff
Ok, would be nice to have history of what governance values were set to what value during which blocks for all sorts of historical queries. I can probably just work around this since I'm only looking up block-veo value ~10 blocks back.
Z
05:02
Zack
so if a node is storing historical data, would it want to store historical data for all the trees?
I think it should be set up so you can choose a shard of a tree, and only store the historical data for that shard.
So for example, you only store it if the first 3 bits of the key are 001.
I think it should be set up so you can choose a shard of a tree, and only store the historical data for that shard.
So for example, you only store it if the first 3 bits of the key are 001.
05:02
but we don't need to shard the governance tree, it is tiny
05:04
I will add this stuff to the todo list for now, as I think other things are more critical now.
MH
05:05
Mandel Hoff
I think you see this from many angles better than me. I'm just thinking "how do I calculate total current veo supply?" Veo per block changes, so I need to know Veo per block at the time each block was minted.
Z
05:05
Zack
it is a good suggestion. thank you for mentioning
05:06
I was thinking of storing the total amoveo supply in each block.
05:07
I could make a merkle tree of values like this, and combine it with the txs merkle tree to make the txs root hash that goes in the header.
Then it would be simple and quick to know values like this without doing calculationn.
Then it would be simple and quick to know values like this without doing calculationn.
05:07
Can you think of any other value worth keeping track of besides the total number of coins?
05:08
total number of: accounts, channels, oracles
05:08
maybe how many times each transaction has been processed?
MH
05:09
Mandel Hoff
Total veo locked in channels? Locked veo in bets?
Z
05:09
Zack
though we dont' want too many valus like this. it makes it more expensive to download blocks
05:09
oh yeah, the portion of veo inside channels is a good thing to track
MH
05:09
Mandel Hoff
Yes, right now I just want veo minted at block X, but this sort of thing can get wild quickly. 🙈
Z
05:23
Zack
In reply to this message
No I think this was a really important idea. Amoveo is going to be a lot better because you brought this up.
MH
05:24
Mandel Hoff
👍
05:30
Knowing how much veo is locked in channels and total supply is powerful information when whales may try to swing governance votes/bets. Could be double edged. Certainly powerful information.
Z
05:31
Zack
if it is expensive to know something, then the rich have the advantage
05:31
it is better if the info is free
MH
05:32
Mandel Hoff
True, if it's not public info, then large channel servers could estimate far better than the community users.
G
07:07
Gonzalo
My balance went from 41.x to 0, I guess Zack is doing something..
Z
07:08
Zack
In reply to this message
maybe the hard drive got corrupted? it is a little tricky turning it on or off.
07:09
my balance did not go to 0.
07:09
what is your address?
G
07:09
Gonzalo
maybe you are right, if I check my address on your explorer is not 0
07:09
BC8uGkC9tUhcBWcDh6wYuZh9Gyh5UDRy2v7ybXDERD22gMtFdYyhQQJCMXafzgfITP8CgNkllGvh1kac6NKOI90=
07:10
My miner continuosly says:
07:10
server gave confusing response
server gave confusing response
server gave confusing response
server gave confusing response
07:11
and my node seems to be spaming a lot
07:11
goes very fast
07:12
got this data
["mining_data"]
[-7,[-7,199,247,25,21],8080]
potential block died!get headers
server crashed.
talk helper fail
my height is less
server crashed.
talk helper fail
get their top header
talk_helper tmimeout
syncing with peer [-7,[-7,159,89,106,253],8080]
trade peers
server crashed.
talk helper fail
give blocks
get headers
syncing with peer [-7,[-7,159,65,134,21],8080]
trade peers
got this data
["mining_data"]
07:13
also, peers:my_ip() is not working properly, says 10.x.x.x:
07:13
(ae_core_prod@scw-9653e4)22> peers:my_ip().
[-7,[-7,199,247,25,21],8080]
talk_helper tmimeout
talk_helper tmimeout
talk_helper tmimeout
talk helper fail
[-7,[-7,159,65,134,21],8080]
talk_helper tmimeout
talk_helper tmimeout
talk_helper tmimeout
talk helper fail
[-7,[-7,52,234,133,196],8080]
talk_helper tmimeout
talk_helper tmimeout
talk_helper tmimeout
talk helper fail
[-7,[-7,159,89,106,253],8080]
talk_helper tmimeout
talk_helper tmimeout
talk_helper tmimeout
talk helper fail
{10,5,73,155}
(ae_core_prod@scw-9653e4)23>
07:14
my node got stuck at height 5880
Z
07:18
Zack
yes. sounds like the configuration is broken.
it says 10.x because you are spaming them so fast that they stopped responding.
it says 10.x because you are spaming them so fast that they stopped responding.
G
07:19
Gonzalo
aha
Z
08:19
Zack
https://github.com/zack-bitcoin/amoveo-opencl-miner
The gpu miner is pretty slow going. I probably can't finish it in time.
The gpu miner is pretty slow going. I probably can't finish it in time.
Z
08:21
Zack
It isn't hard, but I have never written gpu code before. So I have to look up everything.
MH
08:23
Mandel Hoff
I'm syncing a fresh cloned node, and it's pretty slow. Lots of "requesting too often".
08:23
Also, my IP is wrong:
08:23
08:23
I'm at 52.234.133.196
Z
08:24
Zack
In reply to this message
there is a message appearing which is unrelated to the request you made. you cut off the response.
MH
08:25
Mandel Hoff
Z
08:25
Zack
my mistake. maybe you didn't cut it off.
08:26
oh, yes you did cut if off.
the ip responds in this format {1,2,3,4}.
the ip responds in this format {1,2,3,4}.
08:26
without a port.
MH
08:26
Mandel Hoff
08:26
You're right. Very spammy. I missed it.
Z
08:28
I am unable to reproduce these error message.
08:28
s
08:29
now I am getting these error messages.
MH
08:32
Mandel Hoff
I seem to only be trying to sync with 37.211.237.126, which is in Qatar. Not sure if physical distance would be an issue. My node is in Eastern North America.
08:40
My cpu is at 100% through this sync too, and I have a very small azure instance. That might be causing trouble, but it was pretty fast sync and low cpu before.
Z
08:40
Zack
I just pushed an update that should fix peers:my_ip().
MH
08:42
Mandel Hoff
Ok, I'll update my code and try again.
Z
08:43
Zack
don't forget sync_mode:quick().
MH
08:44
Mandel Hoff
Yes, I had that. One of the screenshots above showed it.
08:46
Ok, this sync is going much faster. I'm already at the block height it took 15 min+ to get to on last try..
Z
08:46
Zack
great
MH
08:57
Mandel Hoff
Healthy fast sync takes 60-80% of a single cpu. Very good.
Deleted invited Deleted Account
MH
09:56
Mandel Hoff
09:57
Zack, might be some balance bug issue. Pretty certain I have the latest code and I'm all sync'd up. My node is a few blocks ahead of yours, but both our nodes show my address is 0/NAN balance.
09:57
BKyLqCvF5vt7KBiPLId7xlW64Bb/uV+icPnvn5mgEKncqlCKiaXUTkRcZv7iYK3GKxSspNh0leIhc4YA+DJAY/s=
09:58
Deleted Account
Check.ur balance on node
09:58
Light node didnt show my balance either
09:59
Neither block explorer
MH
09:59
Mandel Hoff
api:balance(). shows 0 too.
10:01
Deleted Account
Mine is showing
bigdestiny joined group by link from Group
David joined group by link from Group
Z
16:15
and your node keeps saying "block_abosrber died!" "potential_block died!"
Andy M. joined group by link from Group
Z
16:44
Zack
In reply to this message
I found a bug that was stopping the node from syncing. Hopefully now the server will be able to sync the blocks you mined and show your updated balance.
G
16:59
Gonzalo
Zack have you restarted the blockchain?
Z
17:04
Zack
yes
17:04
I mean, I restarted the server. it is syncing
17:06
oh, no more blocks synced this time than last. I am not sure why your balance was 0 before.
17:07
I did change how ext_handler:many_headers is programmed. It had a bug that was preventing syncing for some nodes
G
17:10
Gonzalo
👍
Z
17:11
Zack
I notice that Amoveo is consuming more than twice as much RAM as before
G
17:13
Gonzalo
I can't confirm, sorry
Z
G
17:16
Gonzalo
me too, thanks
Z
17:17
Zack
it looks like 200 mb of ram for Amoveo is still in the normal range. I guess nothing is broken.
G
17:33
Gonzalo
Is this normal??
17:34
got this data
["mining_data"]
mining data block hash is [-7,[-7,199,247,25,21],8080]
"AborK7SR6XTU7X/37J/Q/tgWQOG8chJ59H/AocLxhRw="
syncing with peer [-7,[-7,176,9,84,75],8080]
trade peers
give blocks
got this data
["mining_data"]
[-7,[-7,199,247,25,21],8080]
mining data block hash is "DSKND35SYqm6MIrbyVslBJmEve4OKrM+7LIfYebQjIk="
syncing with peer [-7,[-7,159,89,106,253],8080]
trade peers
give blocks
Z
17:34
Zack
yes. this means someone is using your node for mining.
G
17:34
Gonzalo
look at "mining data block hash is..."
17:35
one says
mining data block hash is [-7,[-7,199,247,25,21],8080]
Z
17:35
Zack
oh, it printed out 2 different unrelated things at the same time, so they are mixed inside each other.
G
17:35
Gonzalo
ah ok
17:43
I guess that difficult right now is pretty high and I'm the only one mining since I started mining more than 20 minutes ago and haven't find any block. Height stills at 6045...
Z
17:43
Zack
yes, we retargetted at height 6000
G
17:45
Gonzalo
oh I see... Is it possible to know the actual network's hashrate? Or is it possible to know how many miners are working?
17:46
I see 12 peers, so let's suposse there are more people mining right know... and nobody is finding blocks ¿?
17:47
peers:all().
[{{169,231,8,73},8080},
{{199,247,25,21},8080},
{{159,65,182,117},8080},
{{162,218,209,18},8080},
{{159,65,134,21},8080},
{{24,5,185,238},8080},
{{52,234,133,196},8080},
{{181,120,161,194},8080},
{{51,15,212,91},8080},
{{176,9,84,75},8080},
{{73,232,25,43},8080},
{{159,89,106,253},8080}]
Z
17:47
Zack
someone is mining with my mining pool
G
17:47
Gonzalo
Maybe diff is too high?
17:47
Wow I just found 1 block!
Z
17:47
Zack
I don't choose the diff. it is retargetted to maintain a 10 minute block time.
G
17:48
Gonzalo
6046
Z
17:48
Zack
this is a common problem with testnets. The tokens don't have value, so there is no incentive to mine.
G
17:48
Gonzalo
yes, true
Z
17:55
Zack
I found out why the memory was being wasted.
It looked like amoveo crashed at some point, and the crashed process was still living in the background, consuming RAM.
I used
It looked like amoveo crashed at some point, and the crashed process was still living in the background, consuming RAM.
I used
sh scripts/kill_all_erlang.sh to get rid of it. The memory consumption is down to 90 mb of RAM again.
17:55
there was a zombie emacs process consuming ram too.
cantabrew invited manu
scr1mp3r joined group by link from Group
MH
23:13
Mandel Hoff
Good fix Zack. My node and account balances are up and good again.
Z
23:13
Zack
great
MH
23:16
Mandel Hoff
I've been measuring block propagation delay. It's ~5 seconds average for your node and my node to sync to the same block height after a new one is found. I measured across ~100 blocks.
23:17
That's very much improved over a few days ago.
Z
23:17
Zack
cool. sounds like there is still opportunity for improvement here
23:18
It is hard to predict how it will act if there were 30+ nodes
MH
23:18
Mandel Hoff
Yes, it's very positive versus a few days ago and still room for improvement.
23:21
Did you add the "missing tx" report I provided yesterday to your to-do list? It was during a busy time yesterday so chat buried it.
Z
23:23
Zack
missing tx?
23:23
sorry, I must have forgotten. What was it again?
MH
23:25
I made seven 0.1 spends to a new account across a few blocks. The balance went up to 0.4 and then down to 0.2.
23:26
I was suspecting the tx either were lost or perhaps rolled in to orphaned blocks and then lost due to the balance saying 0.4 and then back down to 0.2 when querying the account-api for the recipient address.
Z
23:27
Zack
right, I remember you mentioning this.
23:28
I guess I should start by trying to reproduce it.
MH
23:31
Mandel Hoff
Blocks were mining pretty fast yesterday so probably lots more orphans and easier to reproduce. I haven't tried today.
Z
23:32
Zack
depends which step the txs are getting dropped at.
23:32
maybe you didn't finish making the first tx when you started making the second, so they had the same nonce and the second was invalid.
MH
23:36
Mandel Hoff
I got screenshots of the txs in the mempool posted in chat. Is the nonce in the mempool print out?
Z
23:36
Zack
if the tx gets into the mempool, then it is valid for sure
23:37
you were in sync_mode:normal?
23:37
quick mode drops txs
23:39
I think the easy way to fix this is to more aggressively push txs to peers right when you make them.
Maybe your node isn't perfect and will drop txs sometimes, but once the tx is in all of the mempools, then we keep sharing it around until it gets included. Even if some of us drop it sometimes, we will find out about it again right away through the gossip protocol
Maybe your node isn't perfect and will drop txs sometimes, but once the tx is in all of the mempools, then we keep sharing it around until it gets included. Even if some of us drop it sometimes, we will find out about it again right away through the gossip protocol
MH
23:40
Mandel Hoff
Yes, definitely normal sync. I think I was mining on your node and making the txs on my node intentionally to test this.
Z
23:41
Zack
I think I might get the first draft of the gpu miner operable today.
MH
23:41
Mandel Hoff
Any fix sounds good to me.
23:59
Deleted Account
In reply to this message
Hey zack i ws mining for night but my internet went off . There were 6048 block height on explorer but 6066 on my node 😂 i had more balnce on node as well.
21 February 2018
00:00
Deleted Account
Nw if i mine it isnt adding blocks to the chain
MH
00:01
Mandel Hoff
Current diff is 1.66B. At 2mh/s, you get 1 blk per 10 min.
Z
00:01
Zack
there is a limit to how big of a fork we can recover from
00:02
it depends on how much history you leave un-pruned.
00:02
in the configuration file.
00:03
Deleted Account
How can i fix this ?
KR
00:36
Kevin Roberts BWD
Z
KR
00:37
Kevin Roberts BWD
yes
Z
00:49
How many megahashes is a typical GPU?
MH
00:54
Mandel Hoff
Depends on the hash type. I think a low end 750 TI does ~80mhs on bitcoin, which is something like a double sha256.
00:54
Zack,
00:54
I'm submitting blocks to my pool, and the pool is responding with server 500 to my miner on blocks that I'm "getting".
Z
00:54
Zack
ok, my first draft is doing about 0.9 megahashes per second
MH
00:55
Mandel Hoff
6068 and 6067 are blocks that my pool got, but the miner received a serer 500 both times when it submitted the work solution.
Z
00:55
Zack
I haven't connected it to amoveo yet
MH
00:56
Mandel Hoff
I think the gtx 1050-1060s would do around double, so ~160mhs.
00:56
Yes, your block explorer shows my address got credit on the chain but the pool responded with server 500.
Deleted joined group by link from Group
Z
00:59
Zack
https://de.bitcoin.it/wiki/Mining_hardware
According to this page 1.9 megahashes is the best I could do if I was mining bitcoins with this gpu.
So once it is optimized, I am guessing it would do around 3.8 megahashes on Amoveo.
So I am about a factor of 4 away from being optimized.
According to this page 1.9 megahashes is the best I could do if I was mining bitcoins with this gpu.
So once it is optimized, I am guessing it would do around 3.8 megahashes on Amoveo.
So I am about a factor of 4 away from being optimized.
01:00
I bet I could use my GPU and CPU to mine at the same time.
MH
01:01
Mandel Hoff
Yes, I monero on cpu and gpu at same time. Shouldn't be any reason not to.
01:01
I got block 6069, but pool had this error when the miner submitted it.
01:01
01:02
More of the same error log:
01:02
Z
01:03
Zack
This just says that the server didn't respond to the message.
01:03
I don't know why the server didn't respond. maybe you didn't unlock your keys
01:03
?
MH
01:04
Mandel Hoff
Right, keys locked on purpose. I'm not worried about payout.
Z
01:05
Zack
the server isn't responding because it tries to sign a tx that would spend VEO, and it crashes because the key is locked.
That is why you have this error message.
I think you can safely ignore it.
That is why you have this error message.
I think you can safely ignore it.
MH
01:10
Mandel Hoff
Ok, good to know the reason.
KR
02:48
Kevin Roberts BWD
what GPU units are people using?
03:02
Deleted Account
5962
my node blockheight.. but I think this is not correct?
my node blockheight.. but I think this is not correct?
03:05
looks like blockheight is in the 6000's so reboot, and synch
03:06
Deleted Account
Would GPU mining be set up when the main net launches?
Z
03:08
Zack
In reply to this message
It is easy to mess up the database. Then it gets stuck at one height.
03:08
In reply to this message
Yes, I think so.
I am progressing quickly at writing it.
https://github.com/zack-bitcoin/amoveo-opencl-miner
I am progressing quickly at writing it.
https://github.com/zack-bitcoin/amoveo-opencl-miner
03:10
Deleted Account
Thanks man
03:19
Deleted Account
Hey zack my priv key is in keys.db file right ?i unlocked the keys from node copied it from build folder and tried to open it but its show couldnt display keys.db
Z
03:21
Deleted Account
When i want to load it in fresh nkde i need to copy priv key but file is nit opening
03:21
How can i open ?
Z
03:22
Zack
I don't know what you mean by "open".
It is an encrypted file.
It is an encrypted file.
03:24
Deleted Account
I want to load it in fresh node
Keys :load(my pubkey, my privkey, my pass) i know my pubkey n have encrypted priv key
Keys :load(my pubkey, my privkey, my pass) i know my pubkey n have encrypted priv key
03:24
How do i put encrypted priv key into command ?
Z
03:25
Zack
Don't use keys:load.
Just copy the keys.db file into the other node.
It is not recommended to move keys between nodes like this. It would be better to make new keys on the other node, and then spend your VEO to the new pubkey.
Just copy the keys.db file into the other node.
It is not recommended to move keys between nodes like this. It would be better to make new keys on the other node, and then spend your VEO to the new pubkey.
04:02
Deleted Account
04:03
I tried to sync this is what showed up
04:03
I dunt think i can spend the amount like this
04:03
Plus my balance in node is wrong
04:04
Its more then actual balance
04:05
Zack can you think of any solution ?
Z
04:09
Zack
There will be lots of error messages, you can pretty much ignore all of them. Erlang uses fail-fast strategy.
Why do you think your balance is wrong?
Why do you think your balance is wrong?
04:10
Deleted Account
I checked on explorer
04:11
It shows 200 on explorer n 240 on node
Z
04:11
Zack
are you mining?
04:12
oh, this is the exact same problem as you mentioned before.
did you reinstall?
did you reinstall?
04:14
Deleted Account
No not anymore
04:14
Im just trying to spend my balance to another account
04:14
So i can delete this node with wromg height
04:14
Wrong*
Z
04:15
Zack
there is no way to spend money from a broken node.
Z
04:18
Zack
The web wallet is much easier to use than installing Amoveo http://159.89.106.253:8080/wallet.html
04:19
You have 2 options to get your balance back
1) you can download Amoveo into a new folder and copy the keys.db file
2) you can shut down amoveo, delete the blocks with
1) you can download Amoveo into a new folder and copy the keys.db file
2) you can shut down amoveo, delete the blocks with
make prod-clean, update the code, and resync
MH
04:20
Mandel Hoff
Zack, is there anything in the block that has a time-stamp when it was mined?
Z
04:20
Zack
yes. it is recorded on the header too.
04:21
it is part of Nakamoto consensus
G
04:21
Gonzalo
In reply to this message
This is why I think it would be nice to have a command to export the privkey on the node
04:22
In reply to this message
I don't see how a command like that would improve Usman's situation. Also, a command like that would make it less secure.
04:22
it would encourage people to do insecure things.
G
04:24
Gonzalo
Hmmm yes I agree
MH
04:24
Mandel Hoff
What is the time format? Here's block 5:
["ok",["block",5,"ME51ZOrKAh1aYxyZVE+KxUUmGoJfaV/RlNVJShsHhV0=","4Zz2LyQkEvWOcgIeOotV5+VzY/RP3ba5p8EmzLEgtP0=",347504066
["ok",["block",5,"ME51ZOrKAh1aYxyZVE+KxUUmGoJfaV/RlNVJShsHhV0=","4Zz2LyQkEvWOcgIeOotV5+VzY/RP3ba5p8EmzLEgtP0=",347504066
Z
04:26
Zack
https://github.com/zack-bitcoin/amoveo/blob/master/apps/ae_core/src/consensus/chain/block.erl#L200
Uses
Which is defined here:
https://github.com/zack-bitcoin/amoveo/blob/master/apps/ae_core/src/consensus/chain/block.erl#L127
Uses
block:time_now(). to generate the number.Which is defined here:
https://github.com/zack-bitcoin/amoveo/blob/master/apps/ae_core/src/consensus/chain/block.erl#L127
04:27
Testing it in my terminal, it seems to be counting in 1/10ths of a second.
04:28
I put this in my terminal
(os:system_time() div (1000000 * constants:time_units())) - constants:start_time().