2018-06-10 00:02:46
wait... does the pool miner not print out it's hashrate?
Tyler Boone
2018-06-10 00:03:22
ok, there it is
Tyler Boone
2018-06-10 00:05:19
ok, going to take a break for now
Tyler Boone
2018-06-10 00:05:56
if anyone wants to try the hybrid miner it's in tster/hybrid-miner for now
Tyler Boone
2018-06-10 00:06:34
INFO: Mining rate: 130666.289/sec
that's with my 64gb memory desktop
Tyler Boone
2018-06-10 00:07:01
memfield_precache_gb=52
Tyler Boone
2018-06-10 00:22:54
From the pool side the hash rate is a wild estimate since it can only infer from shares
Fireduck
2018-06-10 00:23:14
And it is 20 second rate period
Fireduck
2018-06-10 00:23:21
So really nonsense
Fireduck
2018-06-10 00:23:32
@Fireduck Pool diff increases if there are more than 3 shares per minute?
Protovist
2018-06-10 00:23:52
Holy shit someone looked at my code
Fireduck
2018-06-10 00:23:55
Yeah
Fireduck
2018-06-10 00:53:33
@Protovist are you the silent pool operator?
Fireduck
2018-06-10 01:05:40
Yes, I'm working on adding some UI.
Protovist
2018-06-10 01:06:00
cool. what UI are you looking to do?
Tyler Boone
2018-06-10 01:06:05
Cool
Fireduck
2018-06-10 01:06:49
I was thinking of adding a get status Rpc to MrPlow so explorer can get info and show it
Fireduck
2018-06-10 01:08:19
Just the basic miner/pool stats to start. Seeing if I can get rocksdb-node to read from plowdb.
Protovist
2018-06-10 01:09:53
You are better off making a wrapping class that starts a node and MrPlow in the same process.
Fireduck
2018-06-10 01:10:20
Otherwise you'll have strange rocks read issues if something else it writting to it
Fireduck
2018-06-10 01:13:10
Hmm, OK, I'll keep that in mind.
Protovist
2018-06-10 01:15:25
If you want a review if whatever you are planning I'm happy to help but of course you can do whatever
Fireduck
2018-06-10 01:17:04
Thanks. I'd like to just get something simple working ASAP, but we'll see how it goes...
Protovist
2018-06-10 01:18:14
How does the pool determine which address is used for the reward in the coinbase tx?
Protovist
2018-06-10 01:18:43
It doesn't seem to be the configured pool_address.
Protovist
2018-06-10 01:19:28
Heh, that is the entirety of the magic
Fireduck
2018-06-10 01:19:55
It passes a weighted map of addresses and share counts to the node to build the coinbase
Fireduck
2018-06-10 01:20:04
Which directly pays the miners
Fireduck
2018-06-10 01:20:23
So easy, no pool wallet and payment processing
Fireduck
2018-06-10 01:21:38
The shares are managed by ShareManager
Fireduck
2018-06-10 01:24:12
So is the address shown in the block explorer the highest weight?
Protovist
2018-06-10 01:24:37
I think the explorer just shows the first one
Fireduck
2018-06-10 01:24:55
That isn't fancy
Fireduck
2018-06-10 01:25:04
Having the payments in the coinbase directly is very slick..
Protovist
2018-06-10 01:25:19
The snowday pool always seems to have the same address shown.
Protovist
2018-06-10 01:26:13
Probably same first address each time
Fireduck
2018-06-10 01:28:07
yeah, it is a treemap inside somewhere so ends up being alphabetic
Fireduck
2018-06-10 01:28:13
so it is that miner with the low address
Fireduck
2018-06-10 01:35:08
if you look at the actual block you see all the outputs:
https://explorer.snowblossom.org/?search=0000000001c39d489a2d251f2356b3ca0550f4759edc15cf4c0320d2ade1ab4f
Tyler Boone
2018-06-10 05:49:03
[snowblossomcoin/snowblossom] Issue closed by fireduck64
GitHub
2018-06-10 06:11:21
*https://github.com/snowblossomcoin/snowblossom/compare/83f20a8e7fd4...e0fd69f5a556*
https://github.com/snowblossomcoin/snowblossom/commit/e0fd69f5a556aac1800bdb67ab051b85e83835c5 - Correct snowmerkle build typo
GitHub
2018-06-10 06:14:39
So...we have an issue
Fireduck
2018-06-10 06:15:05
I meant to make block reward half every 4 years. I apparently wrote 2 years in comments but actually implemented 1 year.
Fireduck
2018-06-10 06:15:21
https://github.com/snowblossomcoin/snowblossom/blob/master/lib/src/PowUtil.java#L178 ```
public static long getBlockReward(NetworkParams params, int block_height)
```
Fireduck
2018-06-10 06:15:36
So we could roll with it, or suggest a protocol change for sometime in the next year
Fireduck
2018-06-10 06:16:06
If we make a protocol change, I highly suggest we do it sooner than later
Clueless
2018-06-10 06:16:32
If you're going to change it, a continuous reduction rather than a cliff would be nice.
Protovist
2018-06-10 06:16:38
Easier and more painless to do it now while everyone is more more capable of upgrading their systems and fewer people affected.
Clueless
2018-06-10 06:18:35
@Protovist that is an interesting idea
Fireduck
2018-06-10 06:18:50
@Protovist What doy ou mean cliff exactly?
Clueless
2018-06-10 06:19:15
there will be no cliff if we make the change quickly enough.
Clueless
2018-06-10 06:19:55
I think he means rather than a big jump from 50 to 25 we do a slow scaling
Fireduck
2018-06-10 06:20:10
like one block is 49.5, the next is 49.48
Fireduck
2018-06-10 06:20:17
Yes, precisely.
Protovist
2018-06-10 06:20:26
AH, yeah, I like that idea.
Clueless
2018-06-10 06:20:30
I prefer a formula.
Clueless
2018-06-10 06:20:52
Formula( Height ) => Current in Existence.
Clueless
2018-06-10 06:24:05
*https://github.com/snowblossomcoin/snowblossom/compare/e0fd69f5a556...6291ef1738b5*
https://github.com/snowblossomcoin/snowblossom/commit/6291ef1738b5b345013a21b38b2b566eb1e3297d - Update block reward halving comments to match the actual code
GitHub
2018-06-10 06:51:05
if you public docs promise X and your code has a bug and delivers Y, I'm not sure I consider a vote appropriate for change. I would say it is proper to fix the code to match the original promise. Any nodes which haven't updated by the halving block will then end up forking. Likely no one currently in the community would still be running it, but if they are oh well, then there are 2 snowblossoms.
Tyler Boone
2018-06-10 06:51:46
I agree that makes sense
Fireduck
2018-06-10 06:52:03
but at this point, the safe and respectful thing to do is to make it a vote
Fireduck
2018-06-10 06:53:14
voting works. heh: http://snowblossom.1209k.com:8080/?search=00000009ce03855e8fadffcb46d1f26350603fb253678729ffa1d4fcaeef0ae9
Fireduck
2018-06-10 06:53:52
that's not hittable from here
Tyler Boone
2018-06-10 06:55:24
*https://github.com/snowblossomcoin/snowblossom/compare/6291ef1738b5...099bf69f5d61*
https://github.com/snowblossomcoin/snowblossom/commit/efffe783e7827bf50e2a253b3dd6b0dc4dca2479 - Enable voting in miner
https://github.com/snowblossomcoin/snowblossom/commit/099bf69f5d6176cea96bf41148ff6cc903f9782f
GitHub
2018-06-10 06:55:32
ipv6 only of course
Fireduck
2018-06-10 06:55:38
Transaction: afc78aa2b9feeb5c38d0b9107a5c61ae662749079e41249fa8f462c88532c163 size: 123
Coinbase - height:3617 remark:pool test 4
Motions Approved: 1 3
Motions Rejected: 2 4
Output: snowtest:emcpfwr7m5u6zmg3n4wmnz3lf4lz9h5ecarth54a 49.500000
Output: snowtest:yt60p2097jyf9prp2qd5v3p8cxl7673u2n7pwrfh 0.500000
Fee: 0.000000
Fireduck
2018-06-10 06:56:46
Why are you only allowing ipv6
quantumblockz
2018-06-10 06:57:01
just my internal network stuff on testnet
Fireduck
2018-06-10 06:57:23
I only have one ipv4 address, so most of my test and dev systems are ipv6 only from the outside
Fireduck
2018-06-10 06:58:07
Is that just a obscurity measure to limit access?
quantumblockz
2018-06-10 06:58:43
Just curious because I haven't seen that before
quantumblockz
2018-06-10 06:59:29
not at all
Fireduck
2018-06-10 06:59:53
on my home network, every VM I spin up has an ipv6 address that is reachable from the outside automatically
Fireduck
2018-06-10 06:59:58
so mostly I just use those
Fireduck
2018-06-10 07:00:40
so if I am testing something, it is likely to be ipv6 only
Fireduck
2018-06-10 07:00:52
and I tease anyone who can't access it
Fireduck
2018-06-10 07:01:04
Lol I understand
quantumblockz
2018-06-10 13:12:39
is there a pool operator quickstart somewhere or up to me to figure it out?
Rotonen
2018-06-10 13:13:40
@Tyler Boone when you do memory mapping of snow fields for the miner, if you can have the maximum per thread chunk match the maximum NUMA block on a symmetric multiprocessing system that'd help as well
Rotonen
2018-06-10 13:16:00
@Fireduck FYI shackleton still does not build for me on the 1.0.6 tag - seems like some build config issue as it's trying to write to a temp file which already exists
Rotonen
2018-06-10 13:17:18
```
ERROR: /Users/rotonen/koodaus/git/snowblossom/BUILD:38:1: Creating runfiles tree bazel-out/darwin-fastbuild/bin/Shackleton.runfiles failed (Exit 1)
_bin/build-runfiles (args bazel-out/darwin-fastbuild/bin/Shackleton.runfiles_manifest bazel-out/darwin-fastbuild/bin/Shackleton.runfiles): mkdir '__main__/shackleton': File exists [17]
```
Rotonen
2018-06-10 13:17:51
for heterogenous systems it's a touch more complicated to figure out, but there's ample literature out there, snowblossom is sorta the same problem as any modern physics experiment is dealing with
Rotonen
2018-06-10 13:57:59
and figured this out from the source code as well
Rotonen
2018-06-10 14:04:24
@Rotonen it's basically the same: https://github.com/snowblossomcoin/snowblossom/wiki/7.-Configs#pool-mrplow
Clueless
2018-06-10 14:04:34
the Pool
Clueless
2018-06-10 14:04:51
about, but i'm still not sure as to what the double for the pool fee means
Rotonen
2018-06-10 14:05:05
`1.0 == 100%` or `1.0 == 1%`
Rotonen
2018-06-10 14:06:42
pff reading comprehension, never mind
Rotonen
2018-06-10 14:08:11
@Rotonen Reward * 0.01 => 1%
Clueless
2018-06-10 14:08:19
1.0 is 100%
Clueless
2018-06-10 14:15:31
i suppose yours is the only currently public pool as it's in the docs and in the pool list
Rotonen
2018-06-10 14:23:09
namely now i also have one at `http://snowplough.kekku.li:23380` and unsure on what's appropriate for advertising
Rotonen
2018-06-10 14:40:22
@Rotonen Feel free to mention it in general.
I can put it : https://github.com/snowblossomcoin/snowblossom/wiki/9.-Mining-Pools if you want.
Clueless
2018-06-10 14:40:52
your HTTPS seems broken
Clueless
2018-06-10 14:41:24
it does belong in the wiki, but yours will get most of the users as it's in the default config snippet section
Rotonen
2018-06-10 14:43:03
so rather have a pointer in that section to a pool list?
Rotonen
2018-06-10 14:43:13
ultimately there'd be a dynamic one with health statuses etc.
Rotonen
2018-06-10 14:47:20
@Rotonen I believe in being fair and those are valid points.
I will add health statuses and remove the default config
Clueless
2018-06-10 14:48:37
rather than health, continent or so would probably be useful
Rotonen
2018-06-10 14:49:40
@Rotonen continent? ...
Clueless
2018-06-10 14:49:52
man, I barely recognize the word.
Clueless
2018-06-10 14:49:55
agreed
Clueless
2018-06-10 14:50:20
miners usually like to use a pool near them, for various reasons, which is not very important for snowblossom for now, but that'll cater to their expectations
Rotonen
2018-06-10 14:50:58
@Rotonen How would you feel about me creating another url above yours for eastern hemisphere?
Clueless
2018-06-10 14:51:19
legitimately asking. I'm considering how I should enact policy on it for people putting pools up and wanting to group them together.
Clueless
2018-06-10 14:51:51
indifferent, people who actually care just care of shortest-path ballpark
Rotonen
2018-06-10 14:52:06
so just group by country?
Rotonen
2018-06-10 14:52:23
That works.
Clueless
2018-06-10 14:52:33
I would've figured they'd group by some other affiliation
Clueless
2018-06-10 14:52:56
boils down to speed of light + crap at national borders
Rotonen
2018-06-10 14:54:31
@Rotonen What are your pool fees? (including potential donation to fireduck)
Clueless
2018-06-10 14:54:45
0.5%, 0 to the duck
Rotonen
2018-06-10 14:58:28
playing dirty I see
Clueless
2018-06-10 15:00:09
@Rotonen Do you want any comment/remark on your info?
```
#SNOWPLOUGH
#fee 0.5%
pool_host=http://snowplough.kekku.li:23380
```
Clueless
2018-06-10 15:00:35
none, and dunno if the port needs to be there if it's the default one
Rotonen
2018-06-10 15:00:47
that's default, I'll make it pretty
Clueless
2018-06-10 15:02:56
I avoided putting any pool in the windows release config
Fireduck
2018-06-10 15:04:33
Yeah, I only had it there because it was the only one in existence and a bunch of people wanted to jump in.
Clueless
2018-06-10 15:06:04
@Rotonen Sweet.
https://github.com/snowblossomcoin/snowblossom/wiki/9.-Mining-Pools
and I fixed it: https://github.com/snowblossomcoin/snowblossom/wiki/7.-Configs#poolminer
Let me know if you have any other ideas/thoughts on fairness, even small ones, be petty.
Clueless
2018-06-10 15:06:36
That's not sarcasm, be petty. I prefer to identify things now rather than have people feel slighted later.
Clueless
2018-06-10 15:07:55
[snowblossomcoin/snowblossom] Issue opened by fireduck64
GitHub
2018-06-10 15:08:48
@Rotonen OH, better yet, I can set that entire codeblock to whatever you want in those 3 lines. :) Do you want to be named as pool operator?
Clueless
2018-06-10 15:09:21
no, simple minimalism is my thing
Rotonen
2018-06-10 15:10:56
secondly
Clueless
2018-06-10 15:11:10
As a matter of policy, should we require pool operators to be known/public?
that's the sort of thing I prefer not to decide for everyone.
Clueless
2018-06-10 15:11:35
Up to them I guess
Fireduck
2018-06-10 15:11:51
Well, my concern is fraud.
Clueless
2018-06-10 15:12:06
If you're going to represent people, be known to them, sort of thing.
Clueless
2018-06-10 15:13:07
I think people should be suspicious of unknown pool operators but I don't think we should have any policy about it
Fireduck
2018-06-10 15:14:38
sounds fair.
Clueless
2018-06-10 15:15:21
agreed
Shoots
2018-06-10 15:18:32
all my data is public via RIPE, for one
Rotonen
2018-06-10 15:18:40
@Fireduck I like your comment on Pool failover. I was thinking about suggesting that myself given earlier discussion.
Clueless
2018-06-10 15:21:15
if anyone actually wants to argue for non-anonymity as an inclusion criterion for pool operators, i'd like to see all listed pool operators chip in for a neutral 3rd party law firm arrangement to actually ID individuals across nationstates, but somehow that seems a bit heavyhanded
Rotonen
2018-06-10 15:28:23
I was just slightly concerned about people complaining that we advertised someone that might scam them
Clueless
2018-06-10 15:28:34
(not you, but future postings)
Clueless
2018-06-10 15:33:38
I guess if someone well know claimed to run a pool we should verify that it is them
Fireduck
2018-06-10 15:33:52
Known account on Reddit it Twitter or gpg
Fireduck
2018-06-10 15:34:01
But other than that..meh
Fireduck
2018-06-10 16:14:48
@Rotonen Btw, my standard is, you can post it in the main channels on occasion, just don't spam.
Discord, slack, etc. Maybe we should make an announcement channel.
Clueless
2018-06-10 16:22:40
@Clueless I think fraud is less of a concern with all the funds being paid out in the coinbase transaction.
Protovist
2018-06-10 16:24:15
I don't want to gatekeeper the community. just compartmentalizing my roles.
Clueless
2018-06-10 16:25:37
A dishonest pool operator could still alter the transactions, I suppose.
Protovist
2018-06-10 16:26:58
I mean down the line when snow is worth $1000 apiece.
Clueless
2018-06-10 16:29:11
`INFO: Error in send work: io.grpc.StatusRuntimeException: CANCELLED: call already cancelled`
Clueless
2018-06-10 16:29:22
pool thing, seems ominous
Clueless
2018-06-10 16:30:18
Just a broken connection
Fireduck
2018-06-10 16:30:23
Not a problem
Fireduck
2018-06-10 16:32:43
specifically, there's hundreds of entries/constant
Clueless
2018-06-10 16:33:03
just flooding logs, looks concerning
Clueless
2018-06-10 16:54:10
Hum. Might need exception isolation for the notify loop
Fireduck
2018-06-10 16:54:22
Might be running out of connections
Fireduck
2018-06-10 16:54:35
But I am pretty busy today
Fireduck
2018-06-10 17:04:09
Any error from your miners?
Fireduck
2018-06-10 17:28:16
I'm no longer mining personally, until I get more ram
Clueless
2018-06-10 17:28:29
hashrate seems lower than I'd expect.
Clueless
2018-06-10 17:28:38
but it's been fluctuating enough it's hard to tell
Clueless
2018-06-10 17:28:57
72,872,405 h/s is the average of the last 3 values in the last 5 minutes.
Clueless
2018-06-10 17:31:21
Ok. Shouldn't get those errors more than occasionally
Fireduck
2018-06-10 17:31:45
I'll try to take a look
Fireduck
2018-06-10 17:34:37
You take all the time you need
Clueless
2018-06-10 17:44:31
someone else really needs to set up a pool
Tyler Boone
2018-06-10 17:44:37
I guess I can do it
Tyler Boone
2018-06-10 17:45:03
We have two
Fireduck
2018-06-10 17:45:09
oh sweet
Tyler Boone
2018-06-10 17:46:40
There's 2 public, and I think 1 private
Clueless
2018-06-10 17:46:45
nice, no fireduck donation on it
Tyler Boone
2018-06-10 17:46:46
pending 1-2 more private
Clueless
2018-06-10 17:46:50
screw that guy
Tyler Boone
2018-06-10 17:47:01
lol, right?
Clueless
2018-06-10 17:47:36
I like that there are secret private pools
Fireduck
2018-06-10 17:48:15
there's a bit of concern about people objecting to raising the supply
Clueless
2018-06-10 17:48:45
I recommend early as possible to avoid polarization and potential splits
Clueless
2018-06-10 17:49:19
you are asking early adopters to voluntarily dilute their blossoms by 4x
Tyler Boone
2018-06-10 17:49:28
of course, I think without doing it snowblossom is destined to fail
Tyler Boone
2018-06-10 17:49:45
Exactly
Clueless
2018-06-10 17:49:57
I think the early adopters are smart enough to understand that.
Clueless
2018-06-10 17:51:00
also, would voting work based on current miners, or amounts of snow?
Clueless
2018-06-10 17:51:08
Miners
Fireduck
2018-06-10 17:51:17
hm.
Clueless
2018-06-10 17:51:30
Well miners may be inclined to allow the expansion so they can mine more
Clueless
2018-06-10 17:59:06
We need someone to write up the proposal
Fireduck
2018-06-10 17:59:14
Give it a number
Fireduck
2018-06-10 18:00:33
Write up proposal on wiki?
Clueless
2018-06-10 18:00:44
Can you also link me the line number for reference?
Clueless
2018-06-10 18:02:40
Not at the moment, PowUtil getBlockReward or something
Fireduck
2018-06-10 18:02:55
thanks will lookup
Clueless
2018-06-10 18:03:18
another perspective is that we could just add it in. As that's what's originally advertised. That's honestly a fix.
Clueless
2018-06-10 18:05:26
Yeah, true
Fireduck
2018-06-10 18:05:34
But open is good, I think
Fireduck
2018-06-10 18:06:10
how long does the vote take?
Tyler Boone
2018-06-10 18:07:05
depending on the length, someone could spend a couple thousand to rent enough hashing power to win the vote
Tyler Boone
2018-06-10 18:08:15
easily, that's my concern.
Clueless
2018-06-10 18:08:52
I think it's fair to stick to the intended design. Someone trying to exploit it should probably be ignored.
Clueless
2018-06-10 18:08:56
doesn't sound like fireduck is as concerned about this problem as I am, which is fine
Tyler Boone
2018-06-10 18:08:56
it's not like they were mislead
Clueless
2018-06-10 18:09:05
he's busy
Clueless
2018-06-10 18:10:23
no, he doesn't consider the reduced halving time as big a deal as I do
Tyler Boone
2018-06-10 18:11:09
different priorities. it's fine
Tyler Boone
2018-06-10 18:37:20
@Tyler Boone I'm planning to make my pool public as soon as I get some stats working.
Protovist
2018-06-10 18:43:36
@Fireduck The time it takes for difficulty to adjust is somewhat concerning. There's a very real potential for large mining pools to push up the difficulty and then abandon it once it's no longer profitable. This happened recently with VEO.
Protovist
2018-06-10 18:44:39
It's less of a concern since this isn't a GPU PoW, but still possible...
Protovist
2018-06-10 18:44:44
It adjusts pretty fast. Hash power dropped hard yesterday and it is already steady
Fireduck
2018-06-10 18:45:59
Oh. I guess I hadn't looked at the logs to check the rate.
Protovist
2018-06-10 18:47:07
Anyone here familiar with Angular?
Protovist
2018-06-10 18:58:16
What for? (but no)
Clueless
2018-06-10 18:59:38
Was trying to use it for the UI. :slightly_smiling_face:
Protovist
2018-06-10 19:19:59
Perhaps a configuration option to dictate the amount of recent history to cache.
GitHub
2018-06-10 19:33:42
*https://github.com/snowblossomcoin/snowblossom/compare/099bf69f5d61...366f80af7197*
https://github.com/snowblossomcoin/snowblossom/commit/366f80af719752ddbd1b1ceb2387381c76bc734b - Add worker count outputs to pool
GitHub
2018-06-10 19:36:15
@FxxNIHAO has joined the channel
FxxNIHAO
2018-06-10 19:51:14
Jello
Fireduck
2018-06-10 19:51:49
hi
FxxNIHAO
2018-06-10 19:51:59
how are you doing?
FxxNIHAO
2018-06-10 19:54:00
Alright
Fireduck
2018-06-10 21:24:28
@Clueless I think those messages you were seeing about miners breaking connections is due to a bunch of miners connected by without the snowfield
Fireduck
2018-06-10 21:24:41
so they keep thinking they are stalled and restarting the connection
Fireduck
2018-06-10 22:03:11
This is definitely a lot faster adjusting than veo. Is every 2000 blocks isn't it? It would be best to be instantly variable. Similar to what bismuth switched to.
Shoots
2018-06-10 23:15:42
It actually shifts between blocks as well
Fireduck
2018-06-10 23:16:02
Diff drops slightly each new get block template
Fireduck