Amoveo โฅ๐งฟ
Previous messages
Maybe if you download a fresh copy completely from github, and delete the old copy, then it will work.
Keep your keys if you have any VEO.
The internet is slow at download dependencies from github in my part of the world. It would take me a long time to rebuild the code after deleting _build.
I set it up so your node can only sync with one peer at a time, which is helpful because we use less bandwidth. but it is harmful because if you get stuck syncing with one peer, then you wont contact any other
it should sync automatically
Next messages
24 February 2018
G
02:16
Gonzalo
solo vs "shared" pool, nothing to do unless you run a node on a GPU rig with ubuntu
02:16
Deleted Account
I assume pool do distribute the rewards to all their contribute miners
G
02:16
Gonzalo
Maybe if the shared pools offer lower fees to the miners that come from a full node IP... ??
Z
02:16
Zack
no, the pool right now only pays the one worker who found the block.
02:16
hopefully someone will make a better mining pool soon.
G
02:17
Gonzalo
In reply to this message
This is a shared pool, like Mandel's... but not Zack's I believe
MH
02:18
Mandel Hoff
Pool has a strong "block propagation" advantage currently too. That delay may make up for the fee my pool takes. Yes, I think I have the only shared pool currently.
02:19
Deleted Account
But wait, you would still have more chance of finding the block in youre in a pool right?
Z
02:19
Deleted Account
the benefit in running node and miners at the beginning is collecting veo before the masses come. build a large amount of veo now and then wait for the price to appreciate
02:19
and keeping the network alive helps
MH
02:20
Mandel Hoff
Yes, I'm paying out shares based on difficulty of each miner's work.
02:21
What's the rewards per block?
02:21
Deleted Account
1
Z
02:21
Zack
It looks like you aren't sharing the code for this?
02:23
Deleted Account
1 veo per block?
02:23
Wow, it's a rare element indeed.
02:23
Deleted Account
thatโs what iโm getting right now
02:24
Deleted Account
That's just on testnet
02:24
Not sure what Zack has in mind for mainnet
MH
02:24
Mandel Hoff
No, I'm probably keeping my share-pool closed source. I wouldn't expect the 1st Amoveo-betting site to be open sourced.
Z
G
02:25
Gonzalo
In reply to this message
Not really, what you have more chances to get paid in proportion of your CPU/GPU power
Z
02:25
Zack
I open sourced the first betting site already. the browser light node.
MH
02:27
Mandel Hoff
That's very true. You did that.
Z
02:27
Zack
I think that I can program faster than Mandel.
If you keep writing closed source stuff, then I will just rewrite it as open source even faster
If you keep writing closed source stuff, then I will just rewrite it as open source even faster
02:28
Deleted Account
๐คฃ
02:29
Deleted Account
There is no place for closed source system/software here.
Z
02:30
Zack
well, Mandel could make a lot of money this way. but the rest of us would make less money.
Rivalry and competition is important for the development of Amoveo too.
Rivalry and competition is important for the development of Amoveo too.
02:30
Deleted Account
In that case, I'd rather you delay the launch until you get the gpu mining stuff correct so certain people don't have the unfair advantage.
Z
02:30
Zack
It is a game
G
02:30
Gonzalo
I dont like closed source soft but, at the end, this is market rules
M
02:34
Mike
Amoveopool node is at block 7028, while my node is at 7024. How to force sync?
02:34
Deleted Account
i donโt mind anyone connects to my node / pool and if i can change the fee then that would be great 30% is high
Z
M
02:34
Mike
yes
02:35
gitpulled several minutes ago, up-to-date
Z
02:35
Zack
and
sync:start(). doesn't sync?
M
02:35
Mike
it synced up to 7024, which was highest block at that point
02:35
and just stays there
02:35
while some other nodes are at 7028 already
02:36
I'm in quick syncmode now
Z
02:36
Zack
if you are mining then your node should keep syncing automatically
M
02:36
Mike
not mining
02:36
just running full node
Z
02:36
Zack
then it wont sync automatically.
02:37
Deleted Account
Does sync:start forces it to sync?
M
Z
02:38
Zack
if you have a publicly available IP address, then the other nodes will send you blocks.
It is more efficient that the blocks are pushed instead of pulled. they propagate faster this way.
It is more efficient that the blocks are pushed instead of pulled. they propagate faster this way.
M
02:38
Mike
of course I have it
Z
02:38
Zack
looks like the peer you tried to sync with went offline.
02:38
you deleted them from your list of peers, so you wont waste time contacting them again.
M
02:41
Mike
hmm... make prod-restart does not delete chain?
Z
02:42
make prod-clean is for deleting the blocks.
M
02:42
Mike
understood
02:43
resyncing from scratch
02:45
did it up to 7034
02:45
switching to normal mode is necessary?
Z
02:46
Zack
normal mode is needed if you want to mine or make transactions.
M
02:47
Mike
what if I want to mine with external miner to this node?
02:47
then quick or normal is preferred?
Z
02:48
Zack
if you want to mine, then you should use
otherwise you wont include transactions in your blocks correctly.
sync_mode:normal().otherwise you wont include transactions in your blocks correctly.
02:48
you want to include transactions because you get paid a tx fee for including them.
M
02:48
Mike
ok
02:49
so... it looks like it syncs to top block at the moment, but does not absorb any more blocks after it syncs
02:49
strange
Z
02:50
Zack
looks like you have 1 more block than how many headers you have.
02:50
are you mining?
M
02:50
Mike
light wallet is yours
02:50
vps terminal is mine
Z
02:51
Zack
if you are not mining, then it doesn't sync automatically. you have to type
sync:start(). if you want to sync.
M
02:51
Mike
I have to type it again and again?
02:52
each time when network finds a block? :)
02:52
I am not mining yet
Z
02:52
Zack
oh, this is a VPS? then the other nodes should be sending new blocks to you
M
02:52
Mike
yes, this is a VPS
02:53
199.247.26.222 it is
Z
02:53
Zack
I see you in my list of peers
02:54
I told my node to sync with you specifically. lets see if it works
02:54
it says it is sending you blocks.
M
02:58
ouch, it's happening!!!
Z
02:58
Zack
ok, I think I fixed it.
Thank you Mikhail for this useful error report. Amoveo is better because of it
Thank you Mikhail for this useful error report. Amoveo is better because of it
M
02:59
np )
Z
02:59
Zack
Are you in Russia?
M
02:59
Mike
yes
Z
02:59
Zack
cool
M
02:59
Mike
=)
Z
M
03:06
Mike
we'll see in a day
03:08
Zack, why don't you run a shared pool?
Z
03:08
Zack
I don't have software for that yet
03:10
Deleted Account
It's probably not that difficult considering Mandel could get it in a week with your existing code. Don't quote me as I am just going on a limb when it comes to these functional languages
Z
03:12
Zack
Yes. I'll probably write it soon.
03:12
Deleted Account
How long did it take you to learn erlang and become proficient at it. I may try.
03:13
?
Z
03:13
Zack
About a week.
03:14
Deleted Account
Your background is in C?
Z
03:15
I was studying physics before I started working on blockchain. I didn't have programming skills at the start.
03:27
Deleted Account
9 veo nice. at this rate iโll have a good supply then poof mainNet haha
G
04:26
Deleted Account
Gonzalo my node is 3 block behind you, but seems to be working fine.. giving and receiving blocks
Z
04:28
Zack
my node looks like it crashed
04:28
Deleted Account
i just synced to 7075
04:29
i'm caught up now
04:29
7076
04:30
i mined 3 blocks when i was 3 behind Gonzalo
G
04:30
Gonzalo
7078
04:30
Deleted Account
yep me too
G
04:30
Gonzalo
In reply to this message
probably you have lost them, I mean you are not getting rewarded for them
04:31
Deleted Account
yes happened earlier
04:31
i've lost at least 6 veo today due to propagation
G
04:32
Gonzalo
I believe Zack will improve propagation times
04:33
Deleted Account
yep.. I lost everything I mined
04:33
back to 8 veo
Z
04:33
Zack
propagation times do keep improving
04:33
Deleted Account
my node and miners are stable for the first time
04:35
my node just is having a hard time staying current
04:36
Gonzalo is at 7081 and I'm at 7073.. I mine and loose everything
Z
04:37
Zack
my cpu is locked at 100%. I am trying to find out why.
G
04:40
Gonzalo
In reply to this message
This is what happened to me all day long... Maybe Zack can improve the syncing, maybe is possible for the miner to ask all the peers height and ask blocks only to the updated ones
Z
G
04:42
Gonzalo
Yes, but if you are connected to node that is not updated, you wont get fresh blocks. Dont trust me, Im quite noob
04:42
Ive seen that my nodes tend to sync always with the same peer, for a long time
04:43
so if the peer is not updated...
Z
04:43
Zack
I found a problem with pushing blocks.
We only push the one most recent.
So if someone finds 2 blocks in a row, then they only tell others about the 2nd one, so no one can sync.
We only push the one most recent.
So if someone finds 2 blocks in a row, then they only tell others about the 2nd one, so no one can sync.
04:43
ill fix this, and then block propagation might be fixed.
G
04:44
great!
Z
05:17
Zack
I pushed the fix. once everyone updates, the blocks should propagate better.
05:36
Deleted Account
ok
Z
05:46
Zack
Less than 7 days till mainnet launch
ๅฎๆ้ปๆ joined group by link from Group
IP
05:49
I P
could we have better algo for diff adjustment?
05:50
the bitcoin and eth just does not seem good
05:50
for example new coin with 200ghs on eth and low diff. i point 400ghs from nicehash
Z
05:51
Zack
In reply to this message
I take the median of the first 1000 blocks, and the median of the next 1000 blocks. I take the difference of the 2 medians, and divide by 1000.
This is my estimate of the block time for the last interval of 2000 blocks.
This is my estimate of the block time for the last interval of 2000 blocks.
IP
05:51
I P
get blocks before diff rises
05:51
and leave and people can't crack blocks=)
Z
05:51
Zack
I don't understand you IP
05:52
I am still having troubles syncing.. any ideas ?
IP
05:52
I P
for example, diff is 2.5T nethash is 200GH on a small coin
05:52
i point nicehash order of 400gh for 1 hour
05:52
deff rises to 3.3T slowly
Z
05:52
Zack
In reply to this message
the peer you tried to sync with might have gone off line. they will be removed from your peer list, so you wont try syncing with them again
IP
05:52
I P
and i get blocks like from machinegun
05:53
then i leave and people are left with 3.3T diff and 200gh network
Z
05:53
Zack
oh, you are talking about when miners will take lots of cheap blocks, and leave us with a too-high difficulty.
05:54
yes, this can be a problem. It has killed blockchains before.
The long-term solution is ASICS. The short-term solution is hard-forks.
The long-term solution is ASICS. The short-term solution is hard-forks.
IP
05:54
I P
well yeah=) it is easy to do for a small coin with such diff adjustment like now
05:55
is there any kind of better more advanced diff adjustment algos?
Z
05:55
Zack
I have written some experimental diff adjustment algorithms before in other blockchains I have made
05:56
there are many attacks that we need to be careful of when it comes to retargetting.
The attack you describe is just one.
The attack you describe is just one.
05:56
It is risky to use experimental retargetting algorithms.
05:57
Bitcoin's retargetting algorithm is good enough to keep Bitcoin alive. So that is why I copied bitcoin.
05:57
For example, there are attacks where the miner writes the wrong time onto the block
05:57
Deleted Account
Hey Zack, so what't the best way to fix the bad peer. Restart ?
Z
05:58
Zack
In reply to this message
the bad peer should have already been removed from your list of peers.
So if you try syncing again, this time you are more likely to succeed.
So if you try syncing again, this time you are more likely to succeed.
05:58
The code has changed a lot recently, you should probably update
05:59
Deleted Account
My code is up to date , master branch ?
Z
05:59
Zack
yes
06:00
if you still have problems, then maybe the database got corrupted and you need to start over from block 0.
make prod-clean
06:07
Deleted Account
make prod-clean got me to block 2898 before it crashed again . Stuck like before
Z
06:08
Zack
it crashed on the same block twice?
OK
06:09
O K
is your hard drive full
06:09
๐
Z
06:10
Zack
git pull doesn't update the dependencies.Maybe if you download a fresh copy completely from github, and delete the old copy, then it will work.
Keep your keys if you have any VEO.
_build/prod/rel/amoveo_core/keys/keys.db
06:10
Deleted Account
Lots of hard drive space. YES it did crash on the same block
Z
06:11
Zack
ill look at block 2899 to see if it has anything different that could make you crash
06:14
it is a normal looking block. just a spend tx.
06:17
Deleted Account
A fresh git clone, fixed the issue, feels like
make prod-clean should blow away _build ?
Z
06:19
Zack
make prod-clean just deletes the blocks. It doesn't delete the compiled code or the dependencies.The internet is slow at download dependencies from github in my part of the world. It would take me a long time to rebuild the code after deleting _build.
06:24
Deleted Account
Anyway , thanks for the fix !
06:26
Zack, should I be able to connect to http://amoveopool.com/work with the amoveo c miner ? Or is the protocol their own ?
Z
06:26
Zack
I think his miner works a little differently, and they aren't compatible
06:52
Deleted Account
i updated.. now waiting to fully synch
Z
07:07
Deleted Account
i updated, but didn't zero the database.. it should update to head right?
07:07
I'm way behind
07:08
7148 v 7162
Z
07:13
Zack
It should update to head, but the old code was causing some database failures.
if
if
sync:start(). isn't updating your blocks, then you probably need to zero the database.
07:15
Deleted Account
says eventually will synch with peer, so lets see if that eventually happens
Z
07:16
Zack
what is your ip? Ill check if you are in the list
07:16
oh, is it VPS or do you own hardware?
07:17
vps
07:19
iโll check back later and see if it fixes itโs self
Z
07:20
Zack
you aren't in the list of peers. and I can't send you blocks.
07:21
looks like the database got corrupted.
M
07:22
Seems pretty expensive
Z
07:23
Zack
I think I figured out the thing that was making the cpu go to 100%
07:24
I am using a broken hack in the block_absorber
07:24
I think once I fix this, the databases wont be getting corrupted so often.
07:24
Deleted Account
redown load database, but we really need to get to a point when every update doesn't require a database 0 reboot
Z
07:29
Deleted Account
ok 7179 block is me.. if this worked
07:30
still behind by 3
Z
07:31
Zack
yes, I think until I fix this problem, all the nodes will have a lot of problems. it will take me a couple hours at least.
Z
07:57
Zack
The openCL miner is faster. It isn't clear if it is running on my CPU or GPU, but either way it is faster than the C-miner.
MH
08:09
Mandel Hoff
My updated Windows miner is much faster than my previous Windows Cpu miner from yesterday. It can be built on Ubuntu too. The Ubuntu versus Windows speeds should be similar:
https://github.com/Mandelhoff/AmoveoMinerCpu
https://github.com/Mandelhoff/AmoveoMinerCpu
08:11
As Zack said, the openCL Gpu miner is almost certainly the current fastest public Amoveo mining tool. I haven't tried it out.
08:20
Deleted Account
08:20
Hr is 0. Working?
MH
08:23
Mandel Hoff
Yes that's working. S is shares. S/HR is shares per hour. H/S is hashes per second. You are running about 0.49mh/s
08:25
Deleted Account
Thanks. For now you are running the only shared pool. Your fee is 9.9%. Will it be the same when mainnet?
MH
08:26
Mandel Hoff
Probably, yes.
08:26
Deleted Account
Thanks
08:31
How about open a discord channel? Similar to slack. But much better and convenient.
OK
08:32
O K
what's wrong with the telegram
08:36
Deleted Account
you either keep the telegram or move but donโt do both. telegram is fine
lw
G
09:33
Mandel I'm testing you Ubuntu miner but I get this, I guess that pool is not working:
09:33
ERROR: GetWork: 500 Sleep and retry...
MH
09:37
Mandel Hoff
Thank you for reporting it. Yes, node's pool died. It should be up now.
G
09:38
Gonzalo
Right! Mining now ๐
MH
09:38
Mandel Hoff
๐
G
09:42
Gonzalo
Mandel BTW, is it possible to pay different rewards to users. I mean to lower the fees to the miners that own a full node... It's just an idea to incentive people maintaining nodes on the network
MH
09:44
Mandel Hoff
Might be a good candidate for a smart contract like proof of stake rewards node participants
G
09:44
Gonzalo
๐
lw
09:44
lijun wang
has veo market?
Z
09:45
Zack
In reply to this message
https://www.reddit.com/r/AmoveoTrader/
This is the closest thing to a market so far.
This is the closest thing to a market so far.
G
09:45
Gonzalo
I believe it's important to incentive nodes
Z
09:46
Zack
I made a fix so that the CPU wont go to 100%. Hopefully this will stop the database from getting corrupted so often.
G
09:46
Gonzalo
great!
Z
09:46
Zack
In reply to this message
nodes are incentivized by lightning fees and market fees that they can collect from customers.
G
09:50
Gonzalo
In reply to this message
thanks. Apart from running a node, is there anything I should do to get this fees?
09:50
Mandel ERROR 500 again
Z
09:51
Zack
you have to put some money on the node, that way people can make channels with you.
And if you want to make a market, you have to do that manually. You have to give the time-limits, and tell it which oracle to connect the new market to.
And if you want to make a market, you have to do that manually. You have to give the time-limits, and tell it which oracle to connect the new market to.
G
09:56
Gonzalo
ok I see. I expected also another easyer way to earn some VEO with a node. In comparision just mining sounds easy and gives little work
09:57
Is important is to have as much nodes as posible?
Z
09:57
Zack
As long as there is 1 publicly available node, then we are fine.
If there are 0 for too much time, that would be bad.
If there are 0 for too much time, that would be bad.
G
09:58
Gonzalo
ok thanks for the clarification ๐
Z
09:59
Zack
A more important number to think of is this: What is the minimum cost to launch 1 more node?
G
10:00
Gonzalo
5$ a month
Z
10:00
Zack
for now. hopefully it stays fairly low.
MH
10:01
Mandel Hoff
Pool fixed. Node was alive, but pool went down. I did a git pull now - there were some updates.
Z
10:01
Zack
I did a small change to the pool so it wont crash so often.
MH
10:02
Mandel Hoff
Sounds great. ๐
G
10:03
Gonzalo
In reply to this message
if you arent mining, you are fine with the cheapest one right now
10:27
Deleted Account
i restarted my node, and synched, but now my miner is throwing an error connecting to the node
10:28
i guess I"ll try delete database again
10:31
felt good to have a working node and miner for about an hour today.. nothing but problems all afternoon
Z
10:32
Zack
yeah, there is a problem with the mining pool api now. thanks for mentioning.
10:32
Deleted Account
ok.. great..
10:32
not great, but thought it was on my end...
10:38
ok that fixed it.. reinstall database
Z
10:39
Zack
oh, my mistake. mining was working.
10:40
Deleted Account
prod-clean and it's working now
Z
10:40
Zack
cool
10:41
Deleted Account
i'm finding lots of issues.. running node/pool and 2 miners connected to my node..
10:41
now lets see what happens
10:49
i mined a block, and everything seems to be working, but nothing came across my node
Z
10:49
Zack
is the miner connected to your node?
10:50
Deleted Account
yes
Z
10:50
Zack
or is it on someone else's pool?
10:50
Deleted Account
it's mine
10:50
no fee paid, unlocked, like it didn't exist
10:51
tx_pool_feeder died
10:51
sync died!
10:51
alot
10:52
ok worked on my other miner.. just got another block
10:52
7266 is me
11:17
Deleted Account
Noob question. What kind of hardware is enough for a full node and pool? Do I need to setup a pool or just direct all my miners to a full node is enough?
Z
11:26
Zack
if you have a miner on a different computer from the full node, then you need to install a pool on the full node.
you need about 200 mb of ram available.
you need about 200 mb of ram available.
11:27
Deleted Account
iโm using a $5 VPS and itโs fine
11:28
1 gig RAM
11:31
Deleted Account
Thanks.
11:32
I setup a pool. Any way to check its status? Like how many miners connect to the pool and pool hr?
Z
11:33
Zack
no. this pool is very simple.
11:33
It stores some work, and updates the work every 30 seconds or so by contacting the full node.
11:34
if a solution is given to the pool, it immediately tells the server.
11:34
this is all it can do.
Deleted joined group by link from Group
11:36
Deleted Account
11:36
Node working?
Z
11:37
Zack
probably not. I pushed a major fix a couple hours ago. you probably need to restart from block 0 with the newest software
11:38
Blocks are syncing faster than ever. We can process like ~100 blocks per second.
G
11:38
Gonzalo
wow!
11:39
Deleted Account
In reply to this message
Zack , can you open a pool , you charge it we feel great as you develop most things
Z
11:39
Zack
I did.
11:40
the c-miner, the javascript miner, and the openCL miner will all connect to my mining pool by default.
11:41
Deleted Account
i had that same message, and had to restart from block 0
11:46
Deleted Account
i'm still mining blocks and not getting any credit.. the pool sends the transfer, and then I guess my block gets overwritten
11:46
7270?
Z
11:47
Zack
I am on 7270
G
11:48
Gonzalo
7270 as well
11:48
Deleted Account
ok so at the head..lets see what happens at my next block
G
11:48
Gonzalo
did sync_mode:normal().?
Deleted invited Matthew Slipper
11:49
Deleted Account
yep
Deleted invited Jacky
11:49
Deleted Account
unlocked
G
11:50
Gonzalo
๐
11:51
Deleted Account
I'm getting blocks and block hash info, so eventually I'll mine another block, and see if it'll propogate
11:51
Deleted Account
In reply to this message
Agree with Zack , so either join pool by Zack ( I think we should reward core developer) or some other open pool with lower fee and more competition
11:52
Deleted Account
just mined a block
Z
11:52
Zack
The main net is already set up to give me a reward. Though I wont be able to spend it for a long time.
11:53
Deleted Account
now if anyone else will get it
G
11:53
Gonzalo
7271
Z
11:54
Zack
my node got the header, but not the block
11:55
Deleted Account
got this data
syncing with this peer now ["spend","BHMKEKtJHrSThav/b6TSoVcbHbvayWjb5b4qvfTR4HY1wGOCVGI/mjDlt6EjJjvZDJd3CFfciD69S52ivV90ILI=",70000000][-7,[-7,52,234,133,196],8080]
trade peers
give blocks
finished sending blocks
got this data
["mining_data"]
mining data block hash is eventually will sync with peer "a1vW5eNJLkzsCbVfwBjMy3cPsBSAF2G2uQppAoxUyf4="[-7,[-7,52,234,133,196],8080]
get their top header
get headers
give blocks
they are not accepting our blocks.tx_pool_feeder died
sync died!
many headers [-6,5001,0]
got this data
["mining_data"]
eventually will sync with peer [-7,[-7,52,234,133,196],8080]
syncing with this peer now [-7,[-7,52,234,133,196],8080]
trade peers
mining data block hash is "jSCGoppafAPpaDicmOuK20muLRU527GKg2tHlYkhLeM="
get their top header
already synced with this peer
tx_pool_feeder died
sync died!
many headers [-6,5001,5000]
many headers [-6,5001,7271]
got this data
["mining_data"]
eventually will sync with peer [-7,[-7,52,234,133,196],8080]
syncing with this peer now [-7,[-7,52,234,133,196],8080]
trade peers
mining data block hash is "NUlU4eBzXzDM2uo3m6FuXT4iVY4Ut+kE5R7c1HH0qjU="
get their top header
already synced with this peer
tx_pool_feeder died
sync died!
syncing with this peer now ["spend","BHMKEKtJHrSThav/b6TSoVcbHbvayWjb5b4qvfTR4HY1wGOCVGI/mjDlt6EjJjvZDJd3CFfciD69S52ivV90ILI=",70000000][-7,[-7,52,234,133,196],8080]
trade peers
give blocks
finished sending blocks
got this data
["mining_data"]
mining data block hash is eventually will sync with peer "a1vW5eNJLkzsCbVfwBjMy3cPsBSAF2G2uQppAoxUyf4="[-7,[-7,52,234,133,196],8080]
get their top header
get headers
give blocks
they are not accepting our blocks.tx_pool_feeder died
sync died!
many headers [-6,5001,0]
got this data
["mining_data"]
eventually will sync with peer [-7,[-7,52,234,133,196],8080]
syncing with this peer now [-7,[-7,52,234,133,196],8080]
trade peers
mining data block hash is "jSCGoppafAPpaDicmOuK20muLRU527GKg2tHlYkhLeM="
get their top header
already synced with this peer
tx_pool_feeder died
sync died!
many headers [-6,5001,5000]
many headers [-6,5001,7271]
got this data
["mining_data"]
eventually will sync with peer [-7,[-7,52,234,133,196],8080]
syncing with this peer now [-7,[-7,52,234,133,196],8080]
trade peers
mining data block hash is "NUlU4eBzXzDM2uo3m6FuXT4iVY4Ut+kE5R7c1HH0qjU="
get their top header
already synced with this peer
tx_pool_feeder died
sync died!
Z
11:55
Zack
mined by BKF9
11:56
Deleted Account
that's my node address
G
11:56
Gonzalo
maybe try to do a sync:start
Z
G
11:56
Gonzalo
52,234,133,196 is Mandel's and I believe is down right now, at least the pool web is down for me
Z
11:57
Zack
sync:stop(). can be useful.I set it up so your node can only sync with one peer at a time, which is helpful because we use less bandwidth. but it is harmful because if you get stuck syncing with one peer, then you wont contact any other
11:58
there was a problem before were we would download the same blocks from 4 nodes at once, and waste lots of bandwidth
G
11:58
Gonzalo
good to know ๐
11:59
Deleted Account
i stopped and started
11:59
(amoveo_core_prod@amoveo)15> sync:start().
eventually will sync with peer [-7,[-7,199,247,26,222],8080]
syncing with this peer now [-7,[-7,199,247,26,222],8080]ok
(amoveo_core_prod@amoveo)16> trade peers
get their top header
get headers
give blocks
give blocks
finished sending blocks
eventually will sync with peer [-7,[-7,199,247,26,222],8080]
syncing with this peer now [-7,[-7,199,247,26,222],8080]ok
(amoveo_core_prod@amoveo)16> trade peers
get their top header
get headers
give blocks
give blocks
finished sending blocks
11:59
giving blocks to someone
Z
11:59
Zack
cool, now they are synced too
G
11:59
Gonzalo
Zack maybe you can push it up to 2 nodes, just in case it gets stuck
Z
12:00
Zack
you think I should launch a second node? but there are like 6 other people running nodes now
12:01
Deleted Account
we got down to 2 node functioning earlier today
G
12:01
Gonzalo
no, no, let me explain
Z
12:01
Zack
I was thinking of making something so if your database gets corrupted, you can re-read the same blocks on your machine already instead of having to download them again
D
12:02
David
i get this error if i want to spend veo. is this a problem on my node?
(amoveo_core_prod@ubuntu)17> block:height().
7271
(amoveo_core_prod@ubuntu)18> keys:status(). unlocked
(amoveo_core_prod@ubuntu)19> api:balance(). 20000000
(amoveo_core_prod@ubuntu)20> api:spend(base64:decode(ยซ"BLD6ruAVSDHBc4VPASxki9SWEVP/2iLRMEj0KS7So1GZ4Yobtl/SqgIm91UR1rcrbUdXlDIjeaKxms7VoB2CAsY="ยป), 500000).
tx_pool_feeder died
** exception exit: {{{badmatch,false},
[{tx_pool_feeder,absorb_internal,1,
[{file,
"/root/amoveo/_build/prod/lib/amoveo_core/src/consensus/tx_pool_feeder.erl"},
[...]
(amoveo_core_prod@ubuntu)17> block:height().
7271
(amoveo_core_prod@ubuntu)18> keys:status(). unlocked
(amoveo_core_prod@ubuntu)19> api:balance(). 20000000
(amoveo_core_prod@ubuntu)20> api:spend(base64:decode(ยซ"BLD6ruAVSDHBc4VPASxki9SWEVP/2iLRMEj0KS7So1GZ4Yobtl/SqgIm91UR1rcrbUdXlDIjeaKxms7VoB2CAsY="ยป), 500000).
tx_pool_feeder died
** exception exit: {{{badmatch,false},
[{tx_pool_feeder,absorb_internal,1,
[{file,
"/root/amoveo/_build/prod/lib/amoveo_core/src/consensus/tx_pool_feeder.erl"},
[...]
12:02
Deleted Account
I was thinking the node should read block height and api height and if there is a difference then sync.. caused lots of issues being 3 to 50 blocks behind today
G
12:02
Gonzalo
In reply to this message
I was refering this message. Before you downloaded blocks from 4 nodes, now just one. What if you download from 2?
Z
12:03
Zack
<<"abc">>
12:03
In reply to this message
It looks like you are using a different character for the <<>>><<<>> things.
12:04
Deleted Account
how much are trying to send.. I get something like that if I don't leave at least 1 VEO in the account
Z
12:04
Zack
In reply to this message
It would be good to download from more nodes.
But we should download different blocks from each node. no reason to download the same blocks twice.
But we should download different blocks from each node. no reason to download the same blocks twice.
G
12:05
Gonzalo
true
Z
12:05
It is almost like I am writing a torrenting client
12:05
Deleted Account
I meant leave behind in the account..
Z
12:06
Deleted Account
i had to leave at least 1 VEO or got error
Z
12:06
Deleted Account
not 1.00000000.. but 0.00000001
12:07
1 satoshi, or 1 zach i guess
Z
12:07
Zack
oh. if you want to completely spend all your Veo, you should use a delete_account transaction type
12:07
Deleted Account
ok
G
12:07
hehe
Z
12:07
Zack
delete_account actually pays you a little reward beyond how much money you have
12:08
to reward you for freeing up space
ac
12:08
alex c
how are you all mining? never dealt with cpu mining before. Can I run it on a vps?
12:09
Deleted Account
i'm running 2 VPS miners, but for mainnet that won't be profitable with GPUs active but for now works fine for testnet
Z
12:09
Zack
In reply to this message
I tested it for a few minutes on VPS, and it seemed to work fine.
The GPU code is here https://github.com/zack-bitcoin/amoveo-opencl-miner
The GPU code is here https://github.com/zack-bitcoin/amoveo-opencl-miner
ac
Z
12:10
Zack
CPU is possible, but it is less likely to be profitable in comparison to GPU.
12:10
it takes more electricity to do CPU mining
ac
12:10
alex c
any idea which GPU I should get?
ac
12:12
alex c
cool. thanks
12:13
Deleted Account
ok mined another block..
12:13
anyone get it
G
12:13
Gonzalo
7275?
Z
12:14
Zack
no.
12:14
Deleted Account
yep 7275
Z
12:15
Zack
7275. but I had to type
sync:start(). manually a couple time.
12:15
Deleted Account
In reply to this message
Zack , when Mainnet launch there will be gpu miner ? Or just cpu miner
12:15
Deleted Account
got this data
["work","pUpfiZ/6zNLj2RmgHCq9cDza3vE792Jwlug7ZAIAAAA=",0]
work block hash is "c1GEu5V8NkOF0kgxMEh1oihXQ16xpepo8XIT7wW/dFY=""kYnAHMO7+bYOeUoXhZnhsboYAaihoXyT8yKKPIycLfY="
block absorber
absorb block 7275
eventually will sync with peer [-7,[-7,52,234,133,196],8080]
got this data
syncing with this peer now ["spend","BHMKEKtJHrSThav/b6TSoVcbHbvayWjb5b4qvfTR4HY1wGOCVGI/mjDlt6EjJjvZDJd3CFfciD69S52ivV90ILI=",70000000][-7,[-7,52,234,133,196],8080]
trade peers
give blocks
finished sending blocks
got this data
["mining_data"]
mining data block hash is eventually will sync with peer "xG94aJUabDL8m2XDyKQKCko+91qNKbHVJ/97lVYfNXM="[-7,[-7,52,234,133,196],8080]
get their top header
get headers
give blocks
they are not accepting our blocks.tx_pool_feeder died
sync died!
(amoveo_core_prod@amoveo)16> sync:stop().
got this data
["mining_data"]
eventually will sync with peer [-7,[-7,52,234,133,196],8080]
syncing with this peer now [-7,[-7,52,234,133,196],8080]
trade peers
mining data block hash is "iHT+Vg1NqfCT5+5yobcg0fuXFwmU91PDaf8+Y2lzrkI="
get their top header
get headers
give blocks
they are not accepting our blocks.tx_pool_feeder died
sync died!
api:height().
* 1: syntax error before: '.'
(amoveo_core_prod@amoveo)16> api:height().
7275
(amoveo_core_prod@amoveo)17>
got this data
["mining_data"]
eventually will sync with peer [-7,[-7,52,234,133,196],8080]
syncing with this peer now [-7,[-7,52,234,133,196],8080]
trade peers
mining data block hash is "2iabJmLlnzNXOb8knDUw5ue3Ua1rddp37uSk5lZ4f98="
get their top header
get headers
give blocks
they are not accepting our blocks.tx_pool_feeder died
sync died!
(amoveo_core_prod@amoveo)17> api:height().
7275
["work","pUpfiZ/6zNLj2RmgHCq9cDza3vE792Jwlug7ZAIAAAA=",0]
work block hash is "c1GEu5V8NkOF0kgxMEh1oihXQ16xpepo8XIT7wW/dFY=""kYnAHMO7+bYOeUoXhZnhsboYAaihoXyT8yKKPIycLfY="
block absorber
absorb block 7275
eventually will sync with peer [-7,[-7,52,234,133,196],8080]
got this data
syncing with this peer now ["spend","BHMKEKtJHrSThav/b6TSoVcbHbvayWjb5b4qvfTR4HY1wGOCVGI/mjDlt6EjJjvZDJd3CFfciD69S52ivV90ILI=",70000000][-7,[-7,52,234,133,196],8080]
trade peers
give blocks
finished sending blocks
got this data
["mining_data"]
mining data block hash is eventually will sync with peer "xG94aJUabDL8m2XDyKQKCko+91qNKbHVJ/97lVYfNXM="[-7,[-7,52,234,133,196],8080]
get their top header
get headers
give blocks
they are not accepting our blocks.tx_pool_feeder died
sync died!
(amoveo_core_prod@amoveo)16> sync:stop().
got this data
["mining_data"]
eventually will sync with peer [-7,[-7,52,234,133,196],8080]
syncing with this peer now [-7,[-7,52,234,133,196],8080]
trade peers
mining data block hash is "iHT+Vg1NqfCT5+5yobcg0fuXFwmU91PDaf8+Y2lzrkI="
get their top header
get headers
give blocks
they are not accepting our blocks.tx_pool_feeder died
sync died!
api:height().
* 1: syntax error before: '.'
(amoveo_core_prod@amoveo)16> api:height().
7275
(amoveo_core_prod@amoveo)17>
got this data
["mining_data"]
eventually will sync with peer [-7,[-7,52,234,133,196],8080]
syncing with this peer now [-7,[-7,52,234,133,196],8080]
trade peers
mining data block hash is "2iabJmLlnzNXOb8knDUw5ue3Ua1rddp37uSk5lZ4f98="
get their top header
get headers
give blocks
they are not accepting our blocks.tx_pool_feeder died
sync died!
(amoveo_core_prod@amoveo)17> api:height().
7275
D
12:16
David
12:16
hmm...the ยซยป character seems correct
D
12:16
David
Z
Zack 24.02.2018 12:03:04
<<"abc">>
12:16
It looks like you are using a different character for the <<>>><<<>> things.
12:16
Deleted Account
7276 now
Z
12:17
Zack
looks good. your mining pool gave a reward to whoever mined that block.
12:17
Deleted Account
that me
12:17
as long as you get the block then seems to be working
G
12:17
Gonzalo
absorb block 7276
Z
12:18
Zack
In reply to this message
line 41 of tx_pool_feed verifies the signature. For some reason, the signature on your tx is invalid.
12:19
In reply to this message
did you decrypt your key?
keys:status(). should say "unlocked".keys:unlock("").
G
12:20
Gonzalo
๐๐
D
12:20
David
yes keys:status().
unlocked
unlocked
12:21
i used also keys:load(base64:decode(ยซ"BJ/lUOPjDgaITKrp4CmGQMtuj....
Z
12:22
Zack
oh. when you did keys:load it probably broke your keys
12:22
so now you cannot sign
D
12:23
David
oh ok :)
Z
12:23
Zack
I should probably get rid of keys:load. it has confused multiple people, and I don't think anyone needs it.
12:29
Deleted Account
In reply to this message
Closed sourced , look fishy , see Zack open sourced Amoveo betting site , that is the different
Z
12:35
Zack
I did an update so that we will share new blocks even more widely. You keep giving blocks to peers until it statistically looks like 2/3rd of your peers already know.
So now everyone tells at least 2 others about the new block.
So now everyone tells at least 2 others about the new block.
G
12:35
Gonzalo
good idea
Deleted joined group by link from Group
13:08
Deleted Account
Hey
13:08
Can someone here provide me your bitcoin talk announcement forum post?
Z
13:08
Zack
I never made a bitcoin talk forum post about Amoveo.
13:09
Deleted Account
Okay do you have a white paper
Z
13:09
Zack
yes. see the pinned message
13:10
Deleted Account
Thanks
Z
15:06
Zack
I read a bunch of documentation, and now I got the opencl code to compile for GPU.
I can tell it works because my graphics stop working while it is running.
I can tell it works because my graphics stop working while it is running.
15:08
The opencl code doesn't work well yet. It is written in a single-threaded way.
lw
15:08
lijun wang
# make prod-go
# make prod-attach
Node is not running!
make: *** [attach] ้่ฏฏ 1
why?
# make prod-attach
Node is not running!
make: *** [attach] ้่ฏฏ 1
why?
Z
15:09
Zack
make prod-restart
lw
15:10
lijun wang
make prod-restart
===> Verifying dependencies...
===> Skipping pink_hash (from {git,
"https://github.com/BumblebeeBat/pink_crypto",
{tag,"master"}}) as an app of the same name has already been fetched
===> Compiling amoveo_http
===> Compiling amoveo_core
===> Starting relx build process ...
===> Resolving OTP Applications from directories:
/root/amoveo/_build/prod/lib
/root/amoveo/apps
/usr/lib64/erlang/lib
/root/amoveo/_build/prod/rel
===> Resolved amoveo_core-0.1.0
===> Including Erts from /usr/lib64/erlang
===> release successfully created!
make[1]: ่ฟๅ ฅ็ฎๅฝโ/root/amoveoโ
Node is not running!
make[1]: * [attach] ้่ฏฏ 1
make[1]: ็ฆปๅผ็ฎๅฝโ/root/amoveoโ
make: * [prod-restart] ้่ฏฏ 2
===> Verifying dependencies...
===> Skipping pink_hash (from {git,
"https://github.com/BumblebeeBat/pink_crypto",
{tag,"master"}}) as an app of the same name has already been fetched
===> Compiling amoveo_http
===> Compiling amoveo_core
===> Starting relx build process ...
===> Resolving OTP Applications from directories:
/root/amoveo/_build/prod/lib
/root/amoveo/apps
/usr/lib64/erlang/lib
/root/amoveo/_build/prod/rel
===> Resolved amoveo_core-0.1.0
===> Including Erts from /usr/lib64/erlang
===> release successfully created!
make[1]: ่ฟๅ ฅ็ฎๅฝโ/root/amoveoโ
Node is not running!
make[1]: * [attach] ้่ฏฏ 1
make[1]: ็ฆปๅผ็ฎๅฝโ/root/amoveoโ
make: * [prod-restart] ้่ฏฏ 2
15:10
error too
15:10
centos 7
Z
15:12
Zack
I have never installed with centos.
The installation instructions are for ubuntu or mac osx
https://github.com/zack-bitcoin/amoveo/blob/master/docs/getting-started/turn_it_on.md
The installation instructions are for ubuntu or mac osx
https://github.com/zack-bitcoin/amoveo/blob/master/docs/getting-started/turn_it_on.md
lw
15:12
lijun wang
thks
15:13
follow this guide, and get error
Z
15:13
Zack
but when I compile for gpu, the work it produces is invalid.
lw
15:14
lijun wang
no...
Z
15:14
Zack
the guide is only for ubuntu.
15:14
or Mac OSX
lw
15:14
lijun wang
๐ฐ๐ฐ๐ฐ
Z
15:14
Zack
it is not for centos 7
15:15
่ฏฅ่ฏดๆๆฏ้ๅฏนubuntu็
lw
15:15
lijun wang
I am erlang developer too...
Z
15:15
Zack
cool
15:17
what if you do
make prod-attach now?
lw
15:17
lijun wang
which cmd can start ?
Z
15:17
Zack
I use
make prod-restart to turn it on.
X
15:26
YDX is a semi-automated escrow service that enables the trading of supported cryptocurrencies through a dedicated Slack team
15:26
It works well for iota before the exchanges
15:45
Deleted Account
How to unlock keys? It seems my miner doesnโt work.
15:53
Wired. My keys unlocked. But the miner connected to my pool got nothing for 3 hours. It used to get some coins at least.
Deleted joined group by link from Group
MH
18:31
Mandel Hoff
In reply to this message
Seems reasonable to remove it. It was something I tried several times early in the process.
18:31
In reply to this message
Sounds good. I just updated and sync'd from 0. It went well. Node and pool are up now.
18:54
Deleted Account
Will GPU miner be ready before launch?
MH
19:03
Mandel Hoff
Yes
Z
19:05
Zack
In reply to this message
Hard to know.
The opencl code is getting better quickly, but it still doesn't work and there are only 6 days left. I haven't written gpu code before.
The opencl code is getting better quickly, but it still doesn't work and there are only 6 days left. I haven't written gpu code before.
MH
19:07
Mandel Hoff
I've started looking at CUDA, because I have that on my nvidia boxes already. Seems very similar to OpenCL.
Z
19:07
Zack
I think all gpu can use opencl, but only some can use cuda
MH
19:08
Mandel Hoff
Yes, CUDA is nvidia only probably.
M
19:09
Mike
CUDA is easier to write, as far as I know
Z
19:09
Zack
Cgminer for example is written in opencl
M
19:10
Mike
basically you should be able to feed C miner code to CUDA device and it just executes it
19:10
in parallel on all its cores
19:10
but there are definitely some hidden obstacles which only gpu coders know
Z
19:11
Zack
You need a different nonce for every instance of the miner.
MH
19:11
Mandel Hoff
Ha! Sounds like click bait headline.
M
19:12
Mike
=) that's what I heard of it, but never tried myself.
Z
19:42
Zack
It looks like the nodes are staying synced better today.
19:44
I think OpenCL and Cuda are similar in that the number of executions of each for-loop needs to be hard coded.
19:45
I just fixed a infinite loop error in the mining pool. It was wasting lots of CPU power.
Anak Dagang joined group by link from Group
Z
20:48
Zack
I tried closing that oracle we made around height 2300.
This should increase the block reward.
This should increase the block reward.
20:52
oh, it settled in type 3.
I guess some more people made bets, so it wasn't in the state I expected.
This means that the block reward should be left unchanged.
I guess some more people made bets, so it wasn't in the state I expected.
This means that the block reward should be left unchanged.
MH
21:11
Mandel Hoff
Sounds great that it worked as expected.
Z
21:12
Zack
it did work as expected, but it didn't test what I wanted. So I wrote more single-node tests, and I found out that governance isn't updating correctly.
21:17
nevermind. my test was broken. The governance values do update correctly.
MH
21:17
Mandel Hoff
Good job testing extra cases.
21:59
Deleted Account
Zack what happens a freshnode tries to sync with corrupted node? They both get corrupted then ?
Z
22:04
Zack
In reply to this message
no. That would be a security vulnerability.
The good node can probably download some blocks from the corrupted node, but the corrupted node wont have all of the blocks.
The good node can probably download some blocks from the corrupted node, but the corrupted node wont have all of the blocks.
22:08
I made syncing faster again.
G
22:09
Gonzalo
๐ trying git pull
22:20
All done! Synced up to block 7450, it was pretty fast :-)
22:22
And so lucky! Started mining with just 1 core and bam!, found a block
Z
22:23
Zack
cool
G
22:23
Gonzalo
๐
22:31
Zack is it possible to do some calculations on erlang, like api:balance - 100000000 ??
Z
22:32
Zack
yes.
22:32
5 * 5.
G
22:33
Gonzalo
but how do I get my balance minus 1 VEO?
Z
22:34
Zack
api:balance() - 100000000.
G
22:35
Gonzalo
great, thank you. I test it before but probably I was missing a point from the command before and it didnt work
22:37
not "point" but missing a "dot" haha
Z
22:37
Zack
yes. erlang commands end in a period.
G
23:14
Gonzalo
Zack both my 2 nodes are syncing most of the time with 64.137.240.136 that is stuck at 7473, while Mandels pool is at height 7482
23:14
Deleted Account
i just updated my node, and I think it's all messed up again
23:15
7473 also
G
23:16
Gonzalo
maybe would be a good idea to do faster change/roll peers to sync with?
23:17
Deleted Account
could not get block 7585 from peer [-7,[-7,64,137,240,136],8080]tx_pool_feeder died
23:18
block:height().
7365
7365
Z
23:19
Zack
ok, I removed that peer from the list. thanks for mentioning.
23:20
Deleted Account
what block height is correct? My node was running all night, and I reset this morning "git pull" and now it's messed up again
G
23:22
Gonzalo
also I've seen that there two peers "requesting too often", 23.96.187.38 and 181.120.161.194 . Maybe they running old software?
Z
23:22
Zack
I think Mandel's node has the right height now
G
23:23
Gonzalo
ah ok haha
Z
23:23
Zack
I don't know about 23.96
But if you see that message, it means you are successfully blocking them from spamming you
But if you see that message, it means you are successfully blocking them from spamming you
23:24
Deleted Account
this isn't consensus, it's following the leader, we just automatically take the highest blockheight since all the nodes quite propagating
Z
23:25
Zack
my node was receiving blocks well for a long time. It just got messed up a few minutes ago because I was doing tests.
23:27
Deleted Account
this is interesting, little annoying but still interesting.. I'm going to clear my database
Z
23:27
Zack
it is important to do
sync_mode:normal(). after syncing. otherwise the blocks wont propagate correctly.
G
23:27
Gonzalo
I had to do sync:start on both my 2 nodes to change the peer syncing with and catch up the actual height
Z
23:28
Zack
yeah, I just added a bunch of peers to the initial default list. I guess some of them were no good
G
23:30
Gonzalo
In reply to this message
Maybe if you hardcode this would be better. I mean that a sync:start implies "quick mode" + "sync" + "normal mode"
Z
23:30
Zack
sync:start is different from sync_mode:quick
23:31
sometimes you only want to sync the 1 most recent block. so you need to do sync:start, but you don't want to be in sync_mode:quick.
because if you are in sync_mode:quick, then you won't propagate that 1 block after finding out about it.
because if you are in sync_mode:quick, then you won't propagate that 1 block after finding out about it.
G
23:33
Gonzalo
got it. Well, maybe you can code that, after doing sync:start, the mode would change to normal automatically
Z
23:34
Zack
but sometimes we want to stay in sync_mode:quick after doing sync:start. Like if you are on block 0 and want to download all the blocks.
G
23:35
Gonzalo
ok
Z
23:35
Zack
I got rid of a bunch of bad peers from the initial list.
23:36
Deleted Account
7492
G
23:36
Gonzalo
7494 ;-)
Z
23:36
Zack
me too 7494
G
23:37
Gonzalo
Mine is bigger ๐
23:38
ok Zack ๐
23:39
Deleted Account
ok i'm with you again
Z
23:39
Zack
maybe we should measure the rate at which you learn about new blocks, and if the rate should ever go below 1 block per minute, then we automatically switch to normal mode
23:41
Deleted Account
i lost one block, so I was mostly up to date last night
Z
23:42
Zack
All the nodes now
[{{199,247,25,21},8080},
{{209,250,250,137},8080},
{{159,65,182,117},8080},
{{162,218,209,18},8080},
{{51,15,69,135},8080},
{{52,234,133,196},8080},
{{51,15,212,91},8080},
{{176,9,84,75},8080},
{{199,247,26,222},8080},
{{159,89,106,253},8080},
{{34,229,146,193},8080},
{{209,250,248,93},8080}]
[{{199,247,25,21},8080},
{{209,250,250,137},8080},
{{159,65,182,117},8080},
{{162,218,209,18},8080},
{{51,15,69,135},8080},
{{52,234,133,196},8080},
{{51,15,212,91},8080},
{{176,9,84,75},8080},
{{199,247,26,222},8080},
{{159,89,106,253},8080},
{{34,229,146,193},8080},
{{209,250,248,93},8080}]
Z
23:43
Zack
you aren't in the list, so my node wont give you new blocks
23:44
Deleted Account
i always get kicked off.. unless you update your node near the same time
Z
23:44
Zack
so I guess it is kicking people out too easily
23:45
Deleted Account
yes. running this node is basically the only thing I've been doing for the past 2 days, and I get kicked off constantly
23:46
and sleep :)
Z
23:46
Zack
thanks for all the support :)
23:47
Deleted Account
no problem.. i'm in blockchain, and was looking for a protocol level project to run hardware that wasn't a money grab
Z
23:51
Zack
I synced with you a bunch of times in a row. I am having trouble reproducing the bug
23:52
Deleted Account
In reply to this message
This is the problem i was refering. When you download all blocks from one node it should change 2 3 peer to confirm if you are same height and if not then it should download more blocks from other nodes. If all nodes follow this rules no one will be left behind interms of blocks (just a opinion)
Z
23:53
Zack
I don't understand
23:54
after you download the blocks from one peer, what should happen?
23:58
Deleted Account
I read mssges many nodes were left behind like 10 blocks or more. If a new node syncs with you he is will be at same height as you. Mandel's node was 10 blocks ahead of yours so if new node communicates with mandels node after syncing with yours it should download 10 more blocks.
Z
23:59
Zack
yes
25 February 2018
00:01
Deleted Account
But the node only syncs with one node . It doesn't communicate with other nodes to confirm block height after downloading the blockchain from one node.
Z
00:03
Zack
right now
You suggest we try syncing with 3 peers, one after another instead.
Yes, this is a good idea. Thank you for sharing.
sync:start(). only tries syncing with 1 peer, then stops.You suggest we try syncing with 3 peers, one after another instead.
Yes, this is a good idea. Thank you for sharing.
00:03
Deleted Account
Yes.
00:05
But only after it downloads the blockchain from one peer. That way it will not use as much bandwidth.
Z
00:06
Zack
yes, good idea
00:06
Deleted Account
are we cross validating blocks?
Z
00:07
Zack
what does that mean?
Deleted joined group by link from Group
00:11
Deleted Account
you have the miner, who creates the block, and then multiple other nodes who are accepting it, but since we are using bitcoins protocol I was just wondering if we had a six confirmation limit like bitcoin or not
00:12
i guess this is a propogation issue which is being worked on
Z
00:13
Zack
the number of confirmations needed to do certain things are set up as variables in the config file.
00:14
Deleted Account
ok i'll take a look..
00:16
i'm not a dev, but more of a blockchain padawan dev haha
G
00:29
Gonzalo
In reply to this message
Yes this is somewhat I was trying to explain you... Something that you download all the blocks from one peer, then from another, etc. You will stop doing this when you find a peer at the same height than the last one, for example
Z
00:30
Zack
you can sort of do this now by doing
sync:start(). a few times in a row quickly.
00:30
it will wait to finish with the first peer before doing the next
00:30
the problem is that if the code should crash while talking to the first peer, then it doesn't try talking to any more peers.
G
00:34
Gonzalo
In reply to this message
yeah this is what Im doing. But you will want this to run alone, mainly unattended
Z
00:34
Zack
yes, you can do
sync:start(). sync:start(). sync:start(). and then leave your computer and it will run unattended.
G
00:35
Gonzalo
great, will do that for now. But hope this will not be forever! ;-)
00:39
Deleted Account
i think next time i have issue synching I'll do sync:start(). sync:start(). sync:start().
00:40
maybe that will get me to the top without clearing my database
Z
00:40
Zack
How about this: if you have more than 1 block, and it has been more than 120 seconds since you found out about a new block, then you switch to sync_mode:normal()
00:40
In reply to this message
yeah, it will likely fix it.
It is also possible that the first sync:start() will crash, and then sync will dump it's mail box, so it wont do the next 2 syncs.
It is also possible that the first sync:start() will crash, and then sync will dump it's mail box, so it wont do the next 2 syncs.
00:41
I think it is more important to teach the nodes to push blocks well, instead of teaching them to pull blocks.
00:42
Deleted Account
yes i agree, but short term fix so I don't dump my database multiple times a day while it's worked out will help me greatly
Z
00:42
Zack
if it says
sync died!then that means it dumped the mailbox
00:42
Deleted Account
ok I get that constantly right now
G
00:43
Gonzalo
In reply to this message
And also in the long term I believe it has to be as automated as possible
00:43
Deleted Account
due to peers not accepting me i think
Z
00:44
Zack
whether or not you are in the list of peers makes no difference for wheter or not
sync died!happens.
00:44
Deleted Account
for me personal win would being able to walk away from my node for 24 hours, come back, do a git pull, and the node doesn't break on me
Z
00:45
Zack
that would be nice
G
00:45
Deleted Account
it's the little things in life
Z
00:45
Zack
ideally you wouldn't have to do a git pull either
00:46
pretty soon we will split the code into stable and development branches. If you stay on the stable branch, then it shouldn't crash
00:46
Deleted Account
you do alot of updates, so when i wake up first thing i do is git pull and then fix my node
G
00:46
Gonzalo
Yeah, for sure!! I hope a few weeks after mainnet everything will run pretty smooth
Z
00:46
Zack
only 6 days left
00:47
Deleted Account
and stable releases planned ahead, so we can all update near the same time would be nice
G
00:48
"work" :)
00:48
Deleted Account
I don't mind running testnet node and then a production node
00:49
cheap VPS
G
00:49
Gonzalo
mee to
00:49
me too
00:49
I need to sleep a bit more
G
01:00
Gonzalo
Zack what If you open a read-only Telegram group only to make announcements like new releases, etc?
Z
01:01
Zack
yeah, that is a good idea. I didn't know read-only groups were possible.
G
01:01
Gonzalo
yes they are possible
M
G
01:03
Gonzalo
so if I leave my matrix desktop for a while I don't have to read hundreads of messages to find out a new release
Z
01:03
Zack
for now we can just assume that the announcement is constantly "there is new code, everyone needs to update."
M
01:03
Mike
Discord is more convenient than Telegram for managing project communications
01:04
Telegram is "one channel for all"
01:04
Deleted Account
i've setup a discord chat for development before, but we ended not using it.. i like it
M
01:04
Mike
I like both
G
M
01:04
Mike
but when community grows, Telegram turns into a mess
01:05
just saying )
01:05
Deleted Account
i would say discord is over kill for a couple hundred people, but we grow to thousands, and hundreds of contributors to the code, then telegram wouldn't work so well
G
01:05
Gonzalo
In reply to this message
that's why I propose a read-only chat. I don't care if it's here or whereever
M
01:05
Mike
also in Discord you can create one more separate channel "official links" or "useful info" and get all the links there, sitting forever
01:06
Deleted Account
a moveo news channel would be fine for me
M
01:06
Mike
for now Telegram for amoveo is perfectly fine
G
01:06
Gonzalo
๐
M
01:06
Mike
in future, we'll be forced to move somewhere else
01:06
Deleted Account
i set up discord in 2 hours with everything a software team would need, it isn't hard
01:08
Deleted Account
@Mandelhoff are you using Zacks mining pool in your pool (serverside) or did you write your own?
Z
01:09
Zack
His pays shares, so he had to re-write at least some of it. it i closed source, so we don't know how it works.
01:11
Deleted Account
zach..post the instant invite link
Z
01:12
Zack
right
01:12
Deleted Account
make it permanent
Z
01:13
good idea. it is permanent.
Z
01:31
Zack
I did an update based on @usmanspall and Gonzalo 's suggestions.
Now the node keeps trying to sync with peers until you go 3 minutes without finding a block. at that point it switches to sync_mode:normal and stops automatically syncing.
Now the node keeps trying to sync with peers until you go 3 minutes without finding a block. at that point it switches to sync_mode:normal and stops automatically syncing.
01:39
Deleted Account
i did git pull. prod-restart it's in quick now.. so it'll automatically go to normal?
Z
01:39
Zack
it should
01:40
is it automatically trying to sync with peers?
01:41
Deleted Account
7576 now
Z
01:41
Zack
7577 for me
01:41
7579
01:42
Deleted Account
7579
01:42
sync:start().
Z
01:43
Zack
it seems like blocks are happening faster than once every 3 minutes. so maybe it wont be able to switch to normal mode
01:44
Deleted Account
7581
01:44
i don't expect it'll switch.. too many new blocks
Z
01:45
Zack
I reduced it down to 60 seconds
01:45
from 3 minutes
01:46
maybe we should switch back to the original idea. If it finds less than 1 block per 5 seconds in the last 2 minutes, then switch to normal mode.
01:47
my node successfully switched to normal mode with the 60 seconds interval
01:48
Deleted Account
ok.. i did synch, and now at normal
01:48
7585
Z
01:48
Zack
cool
01:48
Deleted Account
7586
Z
01:50
Zack
It also attempts to sync with 3 peers when you do sync:start. but it doesn't always work.
01:54
Deleted Account
my node is running fine, now my miner is throwing If this error happens while connected to the public mining node, then it can probably be safely ignored.
If you are running a solo-mining node, then this error may have happened because you need to turn on and sync your Amoveo node before you can mine. You can get it here: https://github.com/zack-bitcoin/amoveo
If you are running a solo-mining node, then this error may have happened because you need to turn on and sync your Amoveo node before you can mine. You can get it here: https://github.com/zack-bitcoin/amoveo
01:57
Deleted Account
block height 7592 api block height 7597
01:59
block:height().
7597
(amoveo_core_prod@amoveo)24> api:height().
7604
I'm doing synch start a bunch of times to see if this will help
7597
(amoveo_core_prod@amoveo)24> api:height().
7604
I'm doing synch start a bunch of times to see if this will help
02:04
my block height and api block height aren't synching 7600 and 7609
02:05
(amoveo_core_prod@amoveo)33> sync:start().
eventually will sync with these peers [-6,[-7,[-7,159,89,106,253],8080],[-7,[-7,52,234,133,196],8080],[-7,[-7,199,247,26,222],8080]]
syncing with this peer now [-7,[-7,159,89,106,253],8080]ok
(amoveo_core_prod@amoveo)34> trade peers
get their top header
get headers
server crashed.
blocks
get blocks 2 failed connect
false 3
potential block died!block absorber died!
eventually will sync with these peers [-6,[-7,[-7,159,89,106,253],8080],[-7,[-7,52,234,133,196],8080],[-7,[-7,199,247,26,222],8080]]
syncing with this peer now [-7,[-7,159,89,106,253],8080]ok
(amoveo_core_prod@amoveo)34> trade peers
get their top header
get headers
server crashed.
blocks
get blocks 2 failed connect
false 3
potential block died!block absorber died!
02:05
my node can't connect and get blocks
02:05
happens every time I git pull
Z
02:06
Deleted Account
i always have to start with 0 block
Z
02:07
Zack
hmmm... I wonder what is wrong with the mining api
02:07
do the error messages continue once you are in normal mode?
02:08
is it an occasional message, or continuous messages that block you from syncing?
02:09
Deleted Account
i'm in normal mode now but block height and api height aren't the same.. although api block height is caught up as far as I can see
Z
02:09
Zack
oh, it is because mining_data is trying to run api:sync. but it wont let you sync with 2 peers at once.
This should be fairly easy to fix.
This should be fairly easy to fix.
02:10
but api:sync is running in the background... so I am still lost. this shouldn't be causing those error messages.
02:10
i can request mining data repeatedly really fast, and they all succeeed.
02:12
Deleted Account
the only thing I can think of is block height and api height not being the same is causing the miner to throw the error.. but no matter how many times I try synch.. it won't fix itself..
02:12
on the node
02:13
i'm going to prod-clean
02:13
everytime i git pull I have to get a fresh database
Z
02:13
Zack
I've got to stop for food for a bit. But when I come back I will do more tests to figure this out
02:16
Deleted Account
as soon as I started synching a fresh database my miner started working again.. so I stopped it while I'm synching
Z
02:17
Zack
Oh, so it does seem like the database is getting corrupted when you reboot.
02:18
Deleted Account
I'm going to give you the commands I've do..
Z
02:18
Zack
If you do
sync:stop (). and wait a bit before turning it off, does that help?
02:18
Deleted Account
api:off().
halt().
git pull
make prod-restart
halt().
git pull
make prod-restart
02:18
i don't do sync:stop()..
02:18
i need to do that first?
02:19
before api off
Z
02:19
Zack
I am not sure. Maybe it will help.
Since this problem seemed to start around the same time we made syncing parallelism
Since this problem seemed to start around the same time we made syncing parallelism
02:20
Deleted Account
ok I added sych stop as command 1
02:23
(amoveo_core_prod@amoveo)4> block:height().
7060
(amoveo_core_prod@amoveo)5> api:height().
7642
7060
(amoveo_core_prod@amoveo)5> api:height().
7642
02:23
trying synch start
02:28
i'm not receiving any blocks
Z
02:29
Zack
I guess this didn't fix it
02:30
Deleted Account
from database resynch.. 1 peer gave me up to api height 7646 block height 7060 and now just errors and failed connections to other nodes
Z
02:31
Zack
I think the node is shutting down while mid way through processing a block.
If the block doesn't finish processing, then we record it as being invalid and never try it again.
If the block doesn't finish processing, then we record it as being invalid and never try it again.
02:31
Deleted Account
7653 api height, but this block height of 7060 that comes from headers right?
02:31
doesn't move
02:32
(amoveo_core_prod@amoveo)19> trade peers
get their top header
get headers
server crashed.
blocks
get blocks 2 failed connect
get their top header
get headers
server crashed.
blocks
get blocks 2 failed connect
02:36
resynching database from scratch again
02:42
working now.. blockheight and api height aren't the same though.. only issue now
Z
02:43
Zack
it is failing on a fresh sync? I guess the new code for automatically syncing and changing modes is broken
02:43
Deleted Account
nodes not accepting my blocks
Z
02:43
Zack
yes, if your blockheight is lower then that means you are mining at that lower height.
02:43
Deleted Account
i did fresh synch.. and they are not accepting our blocks.block:height().
7652
(amoveo_core_prod@amoveo)7> api:height().
7665
7652
(amoveo_core_prod@amoveo)7> api:height().
7665
02:44
miner is working
02:44
i'll let it go until you push new changes
02:47
.api:balance().
90669906 < how does everyone feel about putting a period in the return value 0.90669906..??
90669906 < how does everyone feel about putting a period in the return value 0.90669906..??
02:51
:( so sad.. i resynch twice and still 20 blocks behind Gonzalo
02:51
my blocks won't count
Z
02:52
Zack
my node synced up. it took a few tries.
When it connects to someone with a dead node, it crashes sync. so it doesn't do any of the next peers in the mailbox
When it connects to someone with a dead node, it crashes sync. so it doesn't do any of the next peers in the mailbox
02:53
Deleted Account
to what block?
02:53
my node keeps getting blacklisted
Z
02:53
Zack
the server stays synced easily.
Maybe I should just add your ip address to the initial default list. Then you wont disappear, and people will send you blocks.
Maybe I should just add your ip address to the initial default list. Then you wont disappear, and people will send you blocks.
02:53
Deleted Account
or deleted which is basically the same
Z
02:53
Zack
it blacklists you for like 2 minutes if you get deleted
02:55
Deleted Account
that would be great.. i just keep getting deleted and no one will accept my blocks and like Gonzalo server stays way ahead of me
Z
02:55
Zack
maybe I should tell it to give your ip address to lots of people first, and then start syncing with some
02:56
you are using ubuntu linux like the rest of us?
02:56
Deleted Account
anyone who connects to my node won't ever get any money including me for mining
02:56
yep
02:56
i only use ubuntu
Z
02:56
Zack
I was able to publish a tx from my laptop today, but it did take a couple tries.
02:57
if you are in quick mode, then you probably can't publish your tx
02:58
Deleted Account
i see a big push to my node
02:59
they are not accepting our blocks.block:height().
7683
(amoveo_core_prod@amoveo)12> api:height().
7685
7683
(amoveo_core_prod@amoveo)12> api:height().
7685
Z
02:59
Zack
blocks are happening very fast
03:02
Deleted Account
still way behind.. (amoveo_core_prod@amoveo)18> block:height().
7684
(amoveo_core_prod@amoveo)19> api:height().
7703
and my miners stopped working
7684
(amoveo_core_prod@amoveo)19> api:height().
7703
and my miners stopped working
03:02
restarting again
03:05
still 3 behind Gonzalo
03:07
nope miner error
03:07
I'll have to delete database again
03:07
no point, nodes won't talk to me
Z
03:07
Zack
I think when the blocktime is < 1 minute, only the biggest mining pool will work.
03:08
Deleted Account
that's not decentralized, and when their database gets corrupted we are screwed
Z
03:09
Zack
my database hasn't been getting corrupted.
03:09
the server has been up all day
03:09
Deleted Account
i would hope not. :)
Z
03:09
Zack
it syncs pretty fast.
I have probably synced the blockchain 100 times today
I have probably synced the blockchain 100 times today
03:11
after 2000 blocks, it will retarget. then the blocktime will be slow enough for us all to mine.
This is how it works for bitcoin too.
This is how it works for bitcoin too.
03:11
I turned on my miner, and I am not getting errors.
03:12
Deleted Account
i'm trying again to synch the database
G
Z
03:13
Zack
now my server is broken
G
03:20
Deleted Account
trade peers
get their top header
get headers
server crashed.
blocks
get blocks 2 failed connect
get their top header
get headers
server crashed.
blocks
get blocks 2 failed connect
03:20
api:height().
7718
7718
03:21
block:height().
7050
7050
03:21
the same no matter how many times I resync from 0
03:21
nothing but connection errors after synching from this 1 peer
03:24
7737 I think maybe the head
03:27
Deleted Account
Which peer are you getting connected to ?
03:29
Deleted Account
[{{209,250,250,137},8080},
{{51,15,69,135},8080},
{{64,137,240,136},8080},
{{52,234,133,196},8080},
{{51,15,212,91},8080},
{{199,247,26,222},8080},
{{128,199,97,207},8080},
{{159,89,106,253},8080},
{{34,229,146,193},8080}]
these are my peers, not sure where I'm getting those blocks from though
{{51,15,69,135},8080},
{{64,137,240,136},8080},
{{52,234,133,196},8080},
{{51,15,212,91},8080},
{{199,247,26,222},8080},
{{128,199,97,207},8080},
{{159,89,106,253},8080},
{{34,229,146,193},8080}]
these are my peers, not sure where I'm getting those blocks from though
03:30
i get alot of connection errors trying to connect to peers
G
03:48
Gonzalo
Recently I found 1 blocks on both nodes, at like 775X height, but I dont see the rewards. I guess that If block propagation is not fast enough then biggest miner, Mandel's Pool right now, will have an unfairly advantage since he is always ahead. Am I right?
Z
03:49
Zack
for now his mining pool will get all the blocks
G
03:50
Gonzalo
I mean, every second I mine when being below the height I'm just waisting time
Z
03:50
Zack
yes
G
03:50
Gonzalo
:-(
Z
03:51
Zack
this is partially my fault for adding a bunch of bad peers to the list about 10 hours ago. I think people installed that code, and now they keep pushing the bad peers to us over and over
G
03:53
Gonzalo
And what if you practice a hard fork while in the testnet? Or maybe start from 0, like a mainnet test?
03:53
that would be funny
Z
03:55
Zack
there are a few consensus type changes I want to test before main net. so there probably will be a hard fork in the next 6 days.
03:55
today I want to focus on getting the nodes to stay in sync.
Deleted joined group by link from Group
G
04:28
Gonzalo
great!
G
04:45
Gonzalo
eventually my miner is stoping with this:
04:45
ยดยดยด
server gave confusing response
server gave confusing response
** exception error: bad argument
in function list_to_binary/1
called as list_to_binary(ok)
in call from miner:unpack_mining_data/1 (miner.erl, line 20)
in call from miner:start_c_miners/1 (miner.erl, line 61)
2>
ยดยดยด
server gave confusing response
server gave confusing response
** exception error: bad argument
in function list_to_binary/1
called as list_to_binary(ok)
in call from miner:unpack_mining_data/1 (miner.erl, line 20)
in call from miner:start_c_miners/1 (miner.erl, line 61)
2>
ยดยดยด
Z
04:52
Zack
is it connected to a mining pool, or to a full node?
04:52
Thank you for this helpful error message.
G
04:53
Gonzalo
full node on the same server
Z
04:53
Zack
cool, thanks
G
04:53
Gonzalo
I happened today like 3 times
04:55
wait, just happened on the other node:
04:55
server gave confusing response
server gave confusing response
** exception error: bad argument
in function list_to_binary/1
called as list_to_binary(ok)
in call from miner:unpack_mining_data/1 (miner.erl, line 20)
in call from miner:start_c_miners/1 (miner.erl, line 61)
2>
Z
04:56
Zack
the code says it tried contacting your full node 1000 times, waiting 1 second between each try. and your full node never responded.
G
04:56
Gonzalo
both nodes are stuck at 7781
Z
04:56
Zack
im at 7899
04:56
pretty soon the difficulty will increase.
I think Mandel might be writing the wrong times in the blocks to stop the difficulty from going down.
I think Mandel might be writing the wrong times in the blocks to stop the difficulty from going down.
G
04:58
Gonzalo
Well don't know but I have seen some kind of problem with Recent Blocks Timestamp on Mandel's Pool
Z
05:02
Zack
well, the time on the top block looks perfect. so I guess he isn't messing with the times on the blocks
Z
05:23
Zack
I found a couple bugs in the block organizer that were making the database get corrupted.
06:05
Mandel's pool is down i think
MH
06:10
Mandel Hoff
Azure gave me an unwanted "fail over" reboot. Syncing now.
G
06:15
Mandel btw, is there any minimum payment threshold on your pool?
MH
06:17
Mandel Hoff
Currently, it's set to 1.0 but that's configurable if 1.0 veo ever becomes too high.
G
06:18
Gonzalo
ok nice
MH
07:04
Mandel Hoff
Having trouble syncing. Stalled out in the 7k-7.7k range three times. Will try again later.
Z
07:05
Zack
im working on that now. it will probably work better for you when you try later
MH
07:06
Mandel Hoff
๐
Z
07:59
Zack
if the node starts looking like it froze, you can fix it by doing
Wait like 20 seconds in between for everything to cool down.
I increased the 'request_frequency' value in the config file, so once more people update, hopefully it wont freeze like that any more.
sync:stop(). to stop the syncing process, and then sync:start(). to start again.Wait like 20 seconds in between for everything to cool down.
I increased the 'request_frequency' value in the config file, so once more people update, hopefully it wont freeze like that any more.
08:00
in order to reduce the problem now, until they update, I made it so you only download 3 batches of blocks at time, and I made each batch bigger.
This way you do less requests in total, so you are less likely to exceed the request frequency.
This way you do less requests in total, so you are less likely to exceed the request frequency.
MH
08:06
Mandel Hoff
I pulled updates. Worked. I'm up to 8019.
Z
08:06
Zack
great
MH
08:07
Mandel Hoff
Great work.
Z
08:08
Zack
I feel like my own worst enemy lately.
I spent 2 days making the node download blocks super fast.
But it wasn't reliable and caused crashes. so then I sent 2 days making it reliable.
And now it is almost as slow as before I started, but over 1000 lines longer.
I spent 2 days making the node download blocks super fast.
But it wasn't reliable and caused crashes. so then I sent 2 days making it reliable.
And now it is almost as slow as before I started, but over 1000 lines longer.
08:08
I feel like I learned a lot. Which I guess is more important in the long run.
MH
08:09
Mandel Hoff
Diff only went up ~4x. That's probably correct, but it seems low.
G
Z
08:09
Zack
weren't we finding blocks multiple times a minute?
It seems to me it should have gone up like 20x.
I guess they aren't mining all day long.
It seems to me it should have gone up like 20x.
I guess they aren't mining all day long.
MH
08:09
Mandel Hoff
Yes, reproducing bugs is usually most of the way to solving them. Great attitude. Sleep is good too.
08:11
Yes, multiple per minute, so I was also expecting about 20x diff
Z
08:11
Zack
maybe the retargetting algorithm is broken.
08:23
According to the time stamps, the average block took 220 seconds during the last 2000 blocks.
08:23
I guess there were short periods of fast mining, and long periods of no mining.
08:24
Deleted Account
I've mined many blocks which were deleted
08:24
I updated and now syncing node
G
08:31
Deleted Account
ok. I'm updated and node/pool with 2 miners going all looks good
08:31
8029
Z
08:33
Deleted Account
i just aborted and checked.. remembered the miner doesn't messup like the node at restart
08:36
Until we do an update.. my node looks like it's going to work fine
Z
08:36
Zack
great
08:36
Deleted Account
i had to prod-clean twice
08:36
didn't take the first time
Z
08:36
Zack
that doesn't sound good
08:38
Deleted Account
i think because my node detached, so I had to start up, then shut down, prod-clean, and restart.. then no problem
08:38
whatever updates you did, my node didn't like at all
08:40
I'm still not talking to some peers, but the ones that will talk to me are at the top of the block so I'm staying in sync
Z
08:41
Zack
yeah that makes sense. if you try to build it while a zombie process is still allive, the zombie process might mess with the files.
prod-restart includes the prod-stop command, which deletes and zombie processes. https://github.com/zack-bitcoin/amoveo/blob/master/Makefile#L268
prod-restart includes the prod-stop command, which deletes and zombie processes. https://github.com/zack-bitcoin/amoveo/blob/master/Makefile#L268
MH
09:26
Mandel Hoff
Reporting a Pool Underpayout Bug:
I was using the node's api height to know which miners to credit for which pool blocks. The api height was giving me stale values, so some blocks had "no miners remaining to be credited" on those stale block heights.
I fixed have a fix for this in place to work around when the node's api height is not accurate, so that miners are credited and paid correctly.
Upside - I audited my payouts. Everything else looked good. Several payout txs still get "lost", and the resend payments system works great to send those payments again after they get confirmed as lost.
I was using the node's api height to know which miners to credit for which pool blocks. The api height was giving me stale values, so some blocks had "no miners remaining to be credited" on those stale block heights.
I fixed have a fix for this in place to work around when the node's api height is not accurate, so that miners are credited and paid correctly.
Upside - I audited my payouts. Everything else looked good. Several payout txs still get "lost", and the resend payments system works great to send those payments again after they get confirmed as lost.
Deleted joined group by link from Group
09:42
Deleted Account
In reply to this message
Hi Zack look into making things more understandable when you making an ICO and a website. If not people might not understand what you planning to do.
09:44
Is Amoveo going to support Atomic swaps?
MH
09:47
Mandel Hoff
Hopefully Zack when to bed. He worked hard today and hasn't slept much lately
G
09:49
Gonzalo
yeah, really
Z
09:54
Zack
In reply to this message
There is no ICO for Amoveo.
I don't want too many people involved yet. We need to figure out a good starting use case first.
I don't want too many people involved yet. We need to figure out a good starting use case first.
09:55
In reply to this message
It already has lightning payments, which is a type of atomic swap.
Amoveo is has turing complete smart contracts, so you can program in sha-256 type atomic swaps.
Amoveo is has turing complete smart contracts, so you can program in sha-256 type atomic swaps.
09:56
Maybe I should give a few more hash types to the smart contract language.
Z
09:58
Zack
you can buy Amoveo tokens from someone else.
10:00
Deleted Account
In reply to this message
Oh. A desktop pc won't get any profit from the mining right?
Z
10:01
Zack
In reply to this message
the gpu mining software doesn't work so well, so a desktop might be able to make a profit yes.
Z
10:07
Oh, Mandel wrote one for Windows. Maybe that is easier for you
10:10
Deleted Account
ok so here the million dollar question. are we at 8067?
10:11
i left for like 2 hours and looks like my node is staying in sync
G
10:11
Gonzalo
YES
10:11
Both Mandel's node and Zack are at 8068
10:12
Deleted Account
great.. now if I get some VEO for those blocks..
G
10:12
Gonzalo
also both my two nodeas are at 6068
10:12
8068
10:12
Deleted Account
what mandels and zach's ip? I will add and check against them
G
10:13
10:14
Gonzalo 1 http://51.15.69.135:8080/wallet.html
10:14
Gonzalo 2 http://51.15.212.91:8080/wallet.html
10:14
Deleted Account
im on 8068
G
10:14
Gonzalo
great!
10:14
Deleted Account
i think
10:15
G
10:16
Deleted Account
why does block:height(). return 0
10:17
Deleted Account
i mined 2 blocks just recently, and don't think I got credit for them
10:18
hard to know.. i don't know how to check who mined which block.. is there a command to see that info?
G
10:19
what's your pubkey?
G
10:20
Gonzalo
you have lots with BKF9
10:21
8066, 8065, 8063, 8062
10:26
Deleted Account
ok great. thanks
10:29
Deleted Account
why does that say network hask rate 11 MH/s but pool hash rate 330 MH/s ?
MH
10:37
Mandel Hoff
network hash rate is estimate based on difficulty and target 10 min blocks. If network hash rate was actually the value shown, blocks would be at the 10 min target time.
10:39
Deleted Account
ok
10:39
makes sense
MH
10:44
Mandel Hoff
Pool difficulty is an estimate too based on share difficulty and share solutions received.
12:00
Deleted Account
so all my miner quit working, and throwing errorsound a block. 1
read nonce n is 2
nonce failed 2
read nonce n is 1
nonce failed 1
nonce 0 error
Found a block. 1
read nonce n is 2
nonce failed 2
read nonce n is 1
nonce failed 1
nonce 0 error
read nonce n is 2
nonce failed 2
read nonce n is 1
nonce failed 1
nonce 0 error
Found a block. 1
read nonce n is 2
nonce failed 2
read nonce n is 1
nonce failed 1
nonce 0 error
12:00
Found a block. 1
read nonce n is 2
nonce failed 2
read nonce n is 1
nonce failed 1
nonce 0 error
did not find a block in that period
did not find a block in that period
read nonce n is 2
nonce failed 2
read nonce n is 1
nonce failed 1
nonce 0 error
did not find a block in that period
did not find a block in that period
12:02
mandell pool site went down, and all my miners are throwing errors like above
12:03
i was wondering why i wasn't finding any blocks for over an hour..
12:03
i found 2 right now.
12:03
8118 and 8119 should be me
12:04
Deleted Account
are you cpu mining
12:04
Deleted Account
yes
12:04
Deleted Account
what kind fo hash rate are you getting?
12:05
my miner seemed to have gone down recently too
12:05
but it might be the pool
12:05
Deleted Account
don't know that command
12:09
hope the node doesn't mess now.. lots of errors in miner
12:09
we'll see in the morning
12:24
Deleted Account
8119 for a while
13:33
Deleted Account
looks like i got most of the last blocks
Psdev joined group by link from Group
Z
18:09
Deleted Account
Zack what is your pool address
Z
18:24
Deleted Account
Im trying to mine on your pool from windows miner
Z
18:24
Zack
that probably wont work.
Mandel changed the communication protocol
Mandel changed the communication protocol
18:26
Deleted Account
In reply to this message
Yes it isn't working. His pool is down and with his miner seems like we cant change pools.
Z
18:26
Zack
Mandel wont share the code for his mining pool, so I can't fix any errors in it.
18:28
I am thinking I should update my mining pool so that it can work with Mandel's miners
18:29
Deleted Account
Good idea. If you do that it will give options to switch pools.
18:31
That way mandel won't take all the money as well. Like you said rivalry/compitetion is important for amoveo's growth.
Z
18:33
Zack
Testing random peers, 4/5 are still alive and synced.
18:36
it looks like his miner works in linux. so I can modify his miner to be compatible with old pools
18:41
it looks like Mandel might have made it already compatible
18:43
AmoveoMinerCpu.cpp:44:10: fatal error: 'cpprest/http_client.h'I get this error if I try to build it on my mac
file not found
#include <cpprest/http_client.h>
^
1 error generated.
clang: error: no such file or directory: 'AmoveoMinerCpu.o'
18:44
@Mandelhoff Could you document the api somewhere so that I can make my mining pool compatible?
18:48
Maybe I should re-write the linux miner to use python instead of erlang.
Then maybe it will be easy enough to use on windows or inux
Then maybe it will be easy enough to use on windows or inux
18:49
How many miners are there:
javascript
Mandel Hoff's
C-miner
openCL-miner
erlang miner in the full node.
That is 5 different miners.
javascript
Mandel Hoff's
C-miner
openCL-miner
erlang miner in the full node.
That is 5 different miners.
18:50
the opencl miner in CPU mode is about 3x faster than the C-miner.
G
19:28
Gonzalo
Zack I made a transaction about 8 hours ago but I did not receive the VEOs and also transaction is not showing on http://amoveopool.com/transaction
MH
19:28
Mandel Hoff
In he bottom section of this page, I show json and protocol details:
http://amoveopool.com/
1) Get work sends address with request
2) Get work response has no nonce and two diff values.
http://amoveopool.com/
1) Get work sends address with request
2) Get work response has no nonce and two diff values.
G
19:29
Gonzalo
How can I check what was wrong. My balance went down on senders address
MH
19:29
Mandel Hoff
In reply to this message
Yes, txs can get "lost" if they don't propagate or get mined in to orphan blocks. I think that's known - probably you did nothing wrong.
G
19:30
Gonzalo
Can I check something? Is there a tx id?
MH
19:30
Mandel Hoff
I had to write a re-payout system based on confirmed txs due to that same thing. Pretty much the whole reason that tx explorer exists, because I needed it for confirming pool payouts.
19:31
I also plan to add lots of dashboard and explorer features, but I have "needed" enough them yet.
G
19:32
Gonzalo
In reply to this message
So I guess that when this happens, the VEOS should "go back" to the senders wallet after a while? Just guessing, I don't know how this things are solved
MH
19:33
Mandel Hoff
Probably could use some more testing, but I think they do "go back".
G
MH
19:37
Mandel Hoff
My node went down last night (no prod-attach available). Got corrupted. Git clone and sync'd from 0. It's up now. Sync worked first time.
G
19:39
Gonzalo
Both my 2 nodes still synced after 8-10 hours as well
19:39
Height 8141 right now, as well as your pool
19:40
and Zacks node
19:40
that's great
Z
G
Z
19:42
Zack
In reply to this message
This is the ideal way software gets written. Someone solving a problem they personally experience.
19:43
In reply to this message
no.
You can look up a block to see the txs in it.
You can look up a block to see the txs in it.
block:get_by_height(1234)
19:44
There is a way to look up your historical account balance, but the configuration file by default says to only save your balance data for the recent about 30 blocks.
19:45
We could build a tx-id system as an add-on to amoveo. the same way the c-miner and the mining pool are add-ons.
I don't want to integrate something like this into the core code, I think most Amoveo users don't need it.
I don't want to integrate something like this into the core code, I think most Amoveo users don't need it.
19:47
The headers store the root of a merkle tree that contains the txs. so you can even use merkle proofs to prove that the tx existed in the block you say.
To make it trustless.
I haven't written this merkel proof library yet.
To make it trustless.
I haven't written this merkel proof library yet.
19:49
It is better to have lots of little tools that are connectable, rather than making Amoveo even bigger.
If we use erlang for the add-ons, then I think we can set up the Amoveo configuration file to automatically install the add-ons you want as dependencies. And to run them all while you are running Amoveo.
If we use erlang for the add-ons, then I think we can set up the Amoveo configuration file to automatically install the add-ons you want as dependencies. And to run them all while you are running Amoveo.
G
20:30
Greg
In reply to this message
maybe you're mining with the intel integrated graphics, it is an opencl device
22:35
Deleted Account
Zack how can i find out if my node has gotten credit for any blocks mined in the last 12 hours?
Z
22:35
Zack
The spend tx got included in the very next block 8204
22:36
or you can look up each block with
block:get_by_height(8001). and look in the coinbase tx to see who mined it.
22:40
Deleted Account
feature request: add node public address and list blocks mined in the explorer.. but the balance check is what I'm doing now thanks... mined all night and didn't get credit for any blocks... I accidently has multiple instances of the miner running on my miner.. so that might have been crediting an error.. or my node isn't working again.. checking now
22:41
block:height().
8235
(amoveo_core_prod@amoveo)10> api:height().
8235
8235
(amoveo_core_prod@amoveo)10> api:height().
8235
22:42
node is working still ๐
Z
22:43
Zack
maybe you are mining to someone else's addess?
22:44
Deleted Account
I'll check.. i restarted my miners and waiting to see a couple blocks get mined.. I increased my zombie hoard of miners to 4
Z
22:44
Zack
I don't understand the feature request
22:45
Deleted Account
you put the public address in the field, and it lists blocks processed by that node...
Z
22:46
Zack
I still don't understand.
Every full node processes all the blocks.
Every full node processes all the blocks.
22:57
Deleted Account
ok. i just processed block but i donโt think my blocks are counting. mandels node never showed it but Iโm fully synced.
Z
22:58
Zack
oh, you want an interface to see who mined each block? the explorer has that.
22:59
Deleted Account
yah. just trying to see my last block mined
M
23:12
Mike
Zack considering opencl miner: recheck what opencl-platform and what opencl-device it's using
23:13
it really can use intel gpu or smth
23:23
Deleted Account
i restarted my node after git pull.. all good so far..
23:23
I changed my miner to opencl-miner
23:24
lets see what happens ๐
Z
23:26
Zack
I studied opencl more. I am confident that the current opencl miner is using cpu.
On my computer it is about 3x faster than the c-miner.
On my computer it is about 3x faster than the c-miner.
Vincent B. joined group by link from Group
26 February 2018
P
00:08
Psdev
to configure a full node to mine to my address, I modify
master_pub in sys.config.tmpl ?
00:21
sorry, missed the "Everything above this line is needed for blockchain consensus" comment
Z
01:10
Zack
In reply to this message
False. The full node will mine to whatever address is in your keys.
api:pubkey().
Deleted joined group by link from Group
Deleted joined group by link from Group
03:24
Deleted Account
Started a fresh node on mac, running into a blocker:
"could not get block 250 from peer"
Any ideas? Appears to be for multiple peers
[-7,[-7,52,234,133,196],8080]
[-7,[-7,51,15,212,91],8080]
Full log:
trade peers
tx_pool_feeder died
get their top header
get headers
could not get block 250 from peer [-7,[-7,176,9,84,75],8080]syncing with this peer now [-7,[-7,176,9,84,75],8080]
"could not get block 250 from peer"
Any ideas? Appears to be for multiple peers
[-7,[-7,52,234,133,196],8080]
[-7,[-7,51,15,212,91],8080]
Full log:
trade peers
tx_pool_feeder died
get their top header
get headers
could not get block 250 from peer [-7,[-7,176,9,84,75],8080]syncing with this peer now [-7,[-7,176,9,84,75],8080]
Z
03:32
Deleted Account
Is this normal?
03:32
Z
03:33
Zack
No. Are you using the newest code from github?
03:33
Deleted Account
Yup, fresh clone
Z
03:33
Zack
Which OS?
03:34
Deleted Account
Mac
Z
03:34
Zack
I tried restarting from block 0 on my mac just now, and it is not happening.
If I can reproduce the error, it will be possible for me to solve it.
If I can reproduce the error, it will be possible for me to solve it.
03:35
Deleted Account
OK thanks Zack. I'll try
make prod-clean
Z
03:36
Zack
The error says you are on block about 100, and you keep requesting more blocks from different people, but none of them respond.
03:36
Deleted Account
Way smoother now
Z
03:36
Zack
you got the first 100 blocks, so I wonder what stopped you from getting more.
03:36
Thanks for testing and submitting this error report, even though I failed to fix anything.
03:39
Deleted Account
Looks like I'm somewhere around 8370 now ๐
Z
03:40
Zack
great. I am at 8379, so you are probably all synced up
03:40
Deleted Account
Z
03:42
Zack
looks good.
Z
04:03
Zack
A lot of nodes are trapped on block 8068, but they are syncing headers.
If your
and/or send me the error messages appearing at the terminal.
If your
api:height(). and your block:height(). are different, please send me the bottom of your log file at _build/prod/rel/amoveo_core/log/crash.logand/or send me the error messages appearing at the terminal.
G
Z
G
04:06
Gonzalo
yes, and this node its not updated, but its at 8368
Z
04:06
Zack
ok, thank you for confirmation
04:06
8392 is the current height
G
04:07
Gonzalo
the other one is updated, now at 8392
04:07
yeah
04:07
both nodes are running the same version of amoveo
04:08
will try to sync:start
Z
04:12
Zack
Could you update the nodes?
We are more interested in testing the new software, so we can uncover new problems to solve
We are more interested in testing the new software, so we can uncover new problems to solve
G
04:13
Gonzalo
Sure! I just did a git pull on the first one, without a prod-clean, but its not updating the height
Z
04:13
Zack
api:off().
halt().
git pull
make prod-restart
it should sync automatically
G
04:14
Gonzalo
I will restart again
Z
04:14
Zack
oh, git pull didn't work, maybe the database got corrupted from old broken code
04:14
yeah,
make prod-clean should fix it
G
04:14
Gonzalo
In reply to this message
this is wat I've done, but not syncing, height is stuck at 8368
04:15
gonna try prod-clean now
04:15
and aupdate the other node as well
04:15
Deleted Account
I am mining- testing from a windows in a small PC to my keypub in my node and i am getting what i thinks is an enormous h/s:
04:15
### NEW WORK ###
H/S: 1438692 S:0 S/Hr:0
H/S: 1439381 S:0 S/Hr:0
H/S: 1440055 S:0 S/Hr:0
H/S: 1441813 S:0 S/Hr:0
H/S: 1443532 S:0 S/Hr:0
H/S: 1438692 S:0 S/Hr:0
H/S: 1439381 S:0 S/Hr:0
H/S: 1440055 S:0 S/Hr:0
H/S: 1441813 S:0 S/Hr:0
H/S: 1443532 S:0 S/Hr:0
04:15
is this ok?
Z
04:16
Zack
In reply to this message
I don't know. I didn't write the windows miner. Maybe try sending Mandel Hoff a private message.
04:16
Deleted Account
ok thanks
Deleted joined group by link from Group
MH
05:24
Mandel Hoff
1.4mh/s is a normal speed. Looks fine to me from the screen shot.
05:25
Deleted Account
ok thanks.
Z
05:25
Zack
oh, hashes per second. H/S. I should have understood.
05:26
I tried making a tx from the server, and it got included in the next block.
But I noticed that my laptop did not sync the tx. So they aren't propagating as well as they could be.
But I noticed that my laptop did not sync the tx. So they aren't propagating as well as they could be.
G
05:40
Greg
Is it a requirement to open ports to run a full node?
Z
05:41
Zack
if you open port 8080, then we can send you blocks, that way you will stay in sync.
05:42
or you can pull blocks with the command
but that doesn't run continuously, it just syncs you up that once.
sync:start().but that doesn't run continuously, it just syncs you up that once.
05:42
if you wanted to make a transactions, you would first sync up, and then make it.
05:42
If you are running a miner, it should automatically stay in sync.
G
06:00
Greg
thanks, I guess this piece of information could improve the doc
Z
06:00
Zack
yes, you are right. it should be in the docs.
06:01
thank you for the suggestion
06:07
Deleted Account
I think my sync is corrupted. Only syncs to 8446
Z
06:08
Deleted Account
Hope mainnet will be much more stable than this.
06:08
Last week
06:08
I know you've been updating
Z
06:09
Zack
well, there were major changes about 16 hours ago, so anyone with code older than that should update
06:10
Deleted Account
Lol. ๐ค
Z
06:12
Zack
We will probably need at least 2 more main net launches until we have something that survives
06:16
Deleted Account
Well, I hope your baby will do well.
Z
06:35
Zack
I made a tx on the server, and when I told my laptop to sync with the server, it immediately got the tx.
07:09
Deleted Account
is there any update for the website and whitepaper?
Z
07:10
Zack
I did a bunch of some small changes to help txs propagate faster and more reliably.
All the tests I made showed that txs were already propagating fairly well.
All the tests I made showed that txs were already propagating fairly well.
07:15
Deleted Account
thanks
Z
07:59
Zack
We have been at the same height for a while, but everything seems normal. I think mandel's mining pool is down.
MH
08:02
Mandel Hoff
Yes pool looks down. I'm not home to fix it. I'll fix it in an hour or two.
Z
08:18
Zack
I updated the readme for the opencl CPU miner. now installation should be easier, and you can run a speed test to see how fast it is on your system. https://github.com/zack-bitcoin/amoveo-opencl-miner