miningpoolstats.stream/monero added p2pool.

finally :D

don't ask why I don't sleep yet :D

Finnish flag? lol

technically correct

I chose Helsinki datacenter because of low ping to me

Why technically correct? Oh lol.

I talked to a fin this mornings so I chuckled when I saw it.

I think they show flags (or world icon) depending on where pool nodes are

My daughter refuses to believe that Finland exists since they're less than 0.07% of the world population. She says they might be a statistical anomaly.

Says someone who's 0.0000000125% of the world population all by herself ?

Did you double check you have a daughter ? :)

Lol

miningpoolstats.stream/monero still not showing "pool fee" or "min pay", you might have to email admin about that

most likely something on their end

also it's probably better to change the flag to world icon

anyone can e-mail, it's not like I'm pool admin :D

that's the whole point of p2pool

OK, I can reach out, will let you know if we need to change anything on api side

got 6 connections on 18083 @ p2pmd.xmrvsbeast.com, hope they are mining... will keep it running until we get a release with full p2pool compay.

*compat.

I've just sent the message

lol but forgot to fill in sender field, so I won't get a reply

make sure to let them know the pool is PPLNS

it shows fee and min payment there

shows pplns

they just need to browse stuff

How do sites like that even know that pools aren't lying about stats?

they use some heuristics

but it can't be 100% guarantee

I think they will just blacklist the pool if it is noticed

block!

You're up late tonight sech1. You usually aren't replying in here after midnight.

yep, and I need to wake up in 5 hours :D

Sleep and dream of p2p sheep.

lulz. p2pool has the most whack hashrate to # of miners

I don't see that listed on the site.

you gotta zoom out.

#10

Just wanted to introduce myself, and mention that I heard about this project from the xmrvsbeast mining pool. Got everything set up a couple of days ago and have been seeing how this works

crypto nick? interesting choice of username...

s/crypto/Cryptonick/, s/nick?//

XvB node stuck at height 2445449 it seems

which one exactly?

p2pmd.xmrvsbeast.com

was tracking heights for some

p2pool node would spit some warnings in the log in this case

Yep it did

My soul is now free and open source

monerod was stuck, kicked it...

miningpoolstats.stream/monero fixed the display, p2pool looks legit now

wow, even p2p logo

and 2 blocks with 30% effort to celebrate it :D

:)

I love people on github usercontent.irccloud-cdn.com/file/ki27QDPV/image.png

Any way to validate that my current p2pool can properly submit shares?

It's running, staying in sync, and no serious warn/err messages that I can see.

But after switching p2pool instances early yesterday I haven't found a share and just want to ensure all is well.

Only WARNING log of significance is quite a few instances of this:

`SideChain add_external_block: block is built on top of an unknown mainchain block dc7f5c396adfc801d107a52b95c54eb6c99b5829f8ed317b9442e1205942c5b3, mainchain reorg might've happened`

run one xmrig against 84.240.124.114:3333 and one against your own p2pool node, if they keep in sync then yours is working

I went a day or two without landing a block, played around with the difficult level a bit and I've landed 4 since midnight GMT

changing the difficult does not matter

just luck then?

yes

It's like gambling.

*gambling that actually follows the laws of probability and isn't tilted towards the house

:)

I'll take all the luck I can get, was about $50 away from being liquidated in the markets 2 days ago :P

sethsimmons: hmm, you might be a bit behind, maybe monerod not getting blocks quick enough

These are all of my stale block warnings:... (full message at libera.ems.host/_matrix/media/r0/do…1c4416a16ae7ae98537ad72bb5325db1f02)

Seems like more than I would expect, but my node is fairly well connected and has a priority node of my very well connected public node.

2021-09-09 12:20:40.458 I Monero 'Oxygen Orion' (v0.17.2.3-9c82593b5)

Height: 2445614/2445614 (100.0%) on mainnet, not mining, net hash 2.90 GH/s, v14, 52(out)+72(in) connections, uptime 0d 9h 53m 26s

Or maybe someone else is disconnected from main chain sethsimmons

Check those block hashes and see if you can find them

guys imagine getting your hands on 850'000 core Cerebras WSE-2 cpu and mining monero with it (assuming that you have 15x 1000w psu's)

it probably won't have enough l3 cache to get good hr

;P

nvm it categorizes as gpu instead

but should mine monero fast anyway

It might actually be able to run Windows Vista at a reasonable level (once it was recompiled for the architecture)

<DataHoarder> "Check those block hashes and see..." <- All exist on-chain.

So it's likely some lag on my side :/

Yep, or RPC / notify disconnected sethsimmons

I'm going to have an unstable Internet connection tomorrow, probably few 10 minute blocks of being disconnected. Is it worth using XvB's remote monerod so that I only need to sync up p2pool on reconnect?

DataHoarder: That would be odd as both are running locally and connected locally, but could be some sort of bug I guess.

Why do I sometimes have to restart my pc to recover the hashrate?

Think that's to do with the ability to allocate the large pages

Block found?

Is there a fix for it?

Happy 69 block day.

2021-09-09 13:07:32.4065 P2Pool submit_block: BLOCK ACCEPTED at height 2445630 and difficulty = 353797233526

nice

nice

17 miners paid, that's the highest I've seen

heck, I've hit the minimum threshold from XvB's pool already about 3 days sooner than I would have done on the pool (minus bonuses)

are you saying p2pool is better than xvb?

Announced my new guide, would love any more feedback you all may have!

.comfortably different?

well I think it's way better ;)

will miss the bonus HRs though!

I'm saying... I loved XvB, but p2pool is the future :)

yes it is, should be in top 3 pools in 6months time

its ideal for the smaller miners, as that minimum payout lets you see short term growth in your wallet :)

As soon as the binaries start rolling out, then I can see the pool jumping in size rather quickly

lazy miners not wanting to compile stuff themselves :/

saying that, I haven't rebuilt it since Tuesday

Posted the guide on Reddit as well: reddit.com/r/Monero/comments/pkxxar…rther_decentralize_monero_mining_by

<sethsimmons> the guide is pretty extensive, what is the benefit of using docker images? no compiling?

No compiling, auto-updates, simple spin up.

If you run this you'll get the latest code daily for p2pool, and monerod updates as-released.

so you will updating and rebuilding these?

I have CI in Github that re-builds p2pool daily from scratch, and rebuilds monerod weekly to update the base image + I push any new tagged releases.

I mean how do auto-updates work?

The Watchtower container that gets spun up as part of this checks for updates daily and re-deploys the container.

great job! will post it in my sub as well, thanks!

Glad to do it, hopefully will help simplify the process a bit for others!

yes, for sure, many people want to join but can't compile anything

or don't want to...

Honestly the p2pool Github lays everything out quite well to simplify it.

But even that is too complex for many people.

does anyone know why we using 3072 hugepages, when 1/2 that works fine?

The guide I wrote looks complex at first but you could go from fresh VPS to fully functional p2pool/monerod node with just copying and pasting.

XvB: In case of monerod + p2pool + xmrig all running locally?

3072 pages because both xmrig and p2pool need full dataset

Idk the exact requirements of each, though.

p2pool needs ~1300 huge pages

xmrig the same, monerod needs 256 pages

ok got it, not running xmrig on the node...

Sounds like a missed opportunity 👿

#EveryHashMatters

lol

maybe this should be noted in the readme, people with lower ram systems might be excluded not realizing they do not need all the HP with no xmrig

MoneroD reports no incoming conns, which is to be expected, I haven't forwarded on 18080 from the FW/router. Would this hinder mining? as in time to get new blocks?

How tolerant is p2pool to latency? I've got 92ms from my house to my VPS where p2pool and monerod are running. I only ask because I haven't gotten any shares since I moved p2pool to the VPS to hide my home IP.

QuickBASIC: AFAIK latency monerod<>p2pool is the important thing, and latency miner<>p2pool is only a nice to have.

Okay. I'm just unlucky. Thanks sethsimmons

(I could absolutely be wrong, but that's my understanding.)

latency from miners to p2pool is also important if you don't want to get many uncle blocks

<100ms latency is fine though

latency from p2pool to monerod and to other p2pool nodes is more important

I'm using a cheap cheap vps. Any way to check latency to other p2pool nodes?

ping them maybe?

lol of course unless they block ICMP.

or you can look for peer list request-response lines in the log and do the math

124 ms

Oh damn p2pool_peers.txt just literally just a list of everyone's node. Does this list get culled?

pinging it from my node gives 92-93 ms on average though

each node only sends their live connected peers which are confirmed working

that .txt file is used for initial connections

but a peer is removed from it when the node can't connect to it

Oh that explains it.

awk '{print substr($1, 1, length($1)-6)}' < p2pool_peers.txt | while read ip; do ping -c1 $ip >./p2pool_ping.txt 2>&1 && echo $ip IS UP || echo $ip IS DOWN; done

Many of these probably just have ICMP blocked though. But A lot of my pings to peers is sub 10ms

14 - 230ms

those Aussies

:P

Lol

Hi all

Just wanted to thank everyone for all their hard work on this.

btw XvB there is a Docker compose setup where you compile things yourself, so you don’t trust someone else binaries on Docker

DataHoarder: Yup, great work on that: github.com/WeebDataHoarder/p2pool-compose

Added that to a new section: sethforprivacy.com/guides/run-a-p2p…ode/#alternative-ways-to-run-p2pool

nice, thanks

27 miners!

pretty nice

If I found a share with 10% effort it mean that I was lucky, and for one with 700% I was unlucky or reverse?

That's correct.

100% is average, less is lucky, more is unlucky.

Finally got it to work lol. Looks like everyone is less than 120ms (a few sub 40ms and even less sub 10ms). Bonus is, I learned how to trim a string in bash with awk and to do do read lol. paste.debian.net/1211123

QuickBASIC: you might want to replace it with a tcp connect, to see it actually active and not depend on ICMP reply

Does nc give latency? Because that's what I was curious about. Or something else to do tcp connect?

awk '{print substr($1, 1, length($1)-6)}' < p2pool_peers.txt | while read ip; do nmap -p 18089 $ip | grep "Host is"; done

Seems to work.

QuickBASIC: Technically you'd want to do it against `37889/tcp`, though:

`awk '{print substr($1, 1, length($1)-6)}' < p2pool_peers.txt | while read ip; do nmap -p 18089 $ip | grep "Host is"; done`

s/18089/37889/

But works well for me :D

oh yeah.. wrong port number lol

It still works for most of them, because most are also running a node at the same address haha

sethsimmons DataHoarder As someone who's only just learned what docker is, can I ask what the benefit of running Monero and p2pool in docker is, over just running the binaries as a systemctl service is?

Much, much easier to run, containerized so in some ways more secure/limited in scope, easier to upgrade, more portable across platforms/architectures (in theory).

But the big one for me is easier to spin up and easily auto-updated.

It's a vm though right? So is there a performance impact, or is it more like a chroot?

No, it's not a VM, it runs on the local kernel natively.

I'm looking at the illustration on docker.com/resources/what-container and not getting it lol.

It also greatly simplifies development, FWIW.

Okay, so for a guy that has one VPS for this one purpose it's only really useful for ease of install and updates, but one example of how it could be useful is if I wanted to run one container per friend on that same VPS so they could have their own p2pool instance.

Yup, that could be a good use-case.

It's an approach to virtualization, but differs quite a bit from an entire VM.

if its not virtualized, do they break because of unexpected hardware/software? I'm also not very educated on docker but going to have to be soon

oh, it is virtualized

welp nvm

It's virtualized but uses a shared kernel with the base OS among other resources.

As long as the image is compatible with your arch, you can run anything on top of that without touching the base OS

it's an approach to *isolation*

I think my misconception came from Docker+WSL because the original version of docker in WSL ran a hyperv vm per docker container. Whereas my understanding of WSL2, it's only running one kernel now so the only time that I played with it previously it actually was virtualization.

containers don't impose a layer of virtualization, they run on the native system

hyc, I was explaining that my misunderstanding was coming from the wackadoodle way that windows does docker.

Docker on macOS/Windows is different than on Linux, and requires real virtualization (a VM on top of the OS kernel).

hyc: Yup

^ right

I believe you can get away with a very thin shim on MacOS since you're just mapping Linux syscalls to BSD ones

do containers have similar security benefits to qubes then?

or isolated vms in general I would assume

container isolation is a bit weaker than a full VM

buggy software in a container can still crash the entire machine

in a VM it should only be able to crash the VM

also, anyone successfully connected their p2pool node to their monerod outside of lan? I must be missing a monerod flag or something because even with no firewalls the p2pool node is saying its not getting the new main chain blocks and desyncing with monerod, but when I connect to xmrvsbeasts node for p2pool it works

you need to set --rpc-bind-ip and --zmq-pub to the public IP address of the monerod

abberant[m]: currently using this btw

./monerod --zmq-pub tcp://127.0.0.1:18083 --rpc-bind-ip 0.0.0.0 --confirm-external-bind --hide-my-port --restricted-rpc --limit-rate-up 65536

My understanding is that docker on Windows is slightly better now because you can run a full Linux kernel in WSL2 and each docker container uses that single virtualized kernel rather than running a full VM for each docker container like you used to, but yeah there's still performance hit on Windows.

127.0.0.1 is not gonna work for you

hyc: good to know

Windows itself is slower than Linux anyway, so yeah, everything else on top of it is even slower

hyc: ohh, that makes sense

let me try

<hyc> "you need to set --rpc-bind-ip..." <- --zmq-pub tcp://0.0.0.0:18083 --zmq-rpc-bind-ip 0.0.0.0 --rpc-bind-ip 0.0.0.0

Its not working with this, and I use 0.0.0.0 because it should auto bind and manually putting my ip hasnt worked in the past

Do you know the difference between zmq-pub and zmq-rpc-bind-ip? do I have the wrong type of setup

p2pool doesn't use zmq-rpc

it only uses regular rpc and zmq-pub

which is admittedly weird, but zmq-rpc doesn't fully implement the rpc command set

removing the zmq-rpc line its still not connecting to the jsonrpc

I port forwarded 18083 and allowed it through my firewall, and changed my p2pool line to connect to my external ip

s/line/flags/

ah, if that's going thru a router, then jusst use 0.0.0.0

same as your rpc-bind-ip

yeah, thats what I have

hmm

That config should work, must be network issues

Either local firewall or port forwarding

hyc: oh, thats why that exists to set as an ip other than 0.0.0.0, cause people need to bind it if going through a datacenter or somethign with a vps

No need for `--zmq-rpc-bind` of course.

right

Sure that monerod is running properly?

how do I do the little code box btw?

formatting

oh I can just select it

ufw rules... (full message at libera.ems.host/_matrix/media/r0/do…6e0183151e74e9915826e77cafe06fea531)

* ufw rules... (full message at libera.ems.host/_matrix/media/r0/do…660d8810d3ac7df59c4432bef038cf0e8a1)

> <@abberant:matrix.org> ufw rules... (full message at libera.ems.host/_matrix/media/r0/do…d68a521c7ab8984d5525fcd97b662c8b1f6)

`18081/tcp` or `18089/tcp` depending on your config.

Looks like it should be `18081/tcp` from your configs.

You're only allowing p2p and zmq-pub through.

oh... I removed it earlier cause I was using it, but now it seems I need it. let me try adding it

p2pool connects to both RPC and ZMQ pub.

okay, seems to be working I'll have to wait for new blocks to get added to test if it desyncs

finally after an 8 minute block its working

thanks so much for the help guys

the console doesnt spam much anymore after latest update?

how do I now know when its ready?

mightysnowman: not the case for me...

console only wrote info about 2 blocks and thats it

hm

is it functional still?

idk, the miner can always mine right from the start of p2pool but it doesnt count (?)

I usually wait for the big spam to end and then connect xmrig

type "status" into p2pool

doesnt show anything

hm I broke something then

will re-clone and re-build everything

yeah its probably stalled, the console should say its giving shares to n miners too

I can't figure out why when I run p2pool as a different user with systemctl as .service I can't connect to stratum... do different users have different firewall rules?

^ Nevermind

QuickBASIC: what was the solution?

turn it off and on again?

I predict that we will find a block at over 400% effort

Followed by another over 100% effort

.waiger p2pool at 51% by 2022

lulz, no one would find shares

was mentioned that smaller p2pools could be spawned :)

but in the end it's about the same averaged out

but whenever you find a share, you get many payouts at once on average

seems like 49.67.2.64 is just scanning every p2pool peer I have every few minutes

scanning how? What's in the log?

wats the highest loglevel

6

hrm, i wonder how big my logfile is ....

ah only 187 mb. still reasonable

just instant connections and disconnections sech1, all about the same time, every few minutes (so specific port gets banned, but it uses different one each time)

if the IP is banned, your node will disconnect it first

is IP or ip:port set banned?

you need to find it in the log when it was first banned

IP is banned

but anyhow, comes up every few minutes

here one a few seconds ago

2021-09-09 19:24:09.0404 P2PServer peer 49.67.2.64:41581 banned for 600 seconds

2021-09-09 19:13:40.2030 P2PServer peer 49.67.2.64:40300 banned for 600 seconds

so yeah 600 seconds

2021-09-09 19:05:28.6476 P2PServer peer 113.128.175.181:56182 banned for 600 seconds

got same periods in other nodes

uh, wrong one

2021-09-09 19:03:12.6619 P2PServer peer 49.67.2.64:39919 banned for 600 seconds

if you're on log level 5, it should say what this peer did wrong a few lines up

2021-09-09 18:52:40.8541 P2PServer peer 49.67.2.64:41912 banned for 600 seconds

sadly only on 4, can switch to 5

gingeropolous[m]: "QuickBASIC: what was the solution?" Not sure. It just started working or it wasn't broken to begin with.

sech1: "turn it off and on again?" Lol. Blame layer 0 (me, the user).

yep, they try to keep connecting when banned

Should I do anything about this stuff? p2pool seems to be fully functional... (full message at libera.ems.host/_matrix/media/r0/do…13185bccd95d2422171dacd89c85e3f40a0)

> didn't send handshake messages first

so yeah probably whoever said about using nc to connect to people

DataHoarder: can your p2pool-compose run without root? Not familiar with docker.

you can setup docker to be used by non-root users Tonux

that said giving access to docker is akin to giving almost full access

then there is podman

which can run non-root

> <@abberant:matrix.org> Should I do anything about this stuff? p2pool seems to be fully functional... (full message at libera.ems.host/_matrix/media/r0/do…009d2a13703d3830e5113dc6a2459fcd325)

tonux[m]: Can add non-root users to run docker commands: docs.docker.com/engine/install/linux-postinstall

otherwise for docker itself just add your user to "docker" group Tonux

if you want to set it up via podman, sadly haven't tried but there is plenty of documentation online

Yeah, dunno how I feel about giving a user a group if privilege escalation could be possible

then I'd recommend podman indeed

I usually run such setups within a VMs already, given how my cluster works, hopefully it helps some :)

Will look into it, thanks :)

you need at least podman v3.2 for it

"Docker Compose is now supported with rootless Podman (#9169)."

Can't find podman in my distro (Trisquel 9) and I prefer not to build/install packages root permitting packages unless they are GPG signed. However this looks like I may be able to have a non-root user use docker without adding them to privileged groups docs.docker.com/engine/security/rootless

s/packages//, s/GPG/PGP/

sech1 you shortened the commandline in README to just "Run ./p2pool --host 127.0.0.1 --wallet YOUR_WALLET_ADDRESS" rest is now defaults and unneeded now?

even --host can be removed if monerod is on the same machine

all parameters have defaults now (except --wallet)

perfect thank you

280% effort, p2pool wants to stay at 69 blocks for as long as possible :D

if wanted Tonux you can compile it from source, see last version, github.com/containers/podman/releases/tag/v3.3.1 which is GPG signed. However if you find the other docker setup easier/safer then that might be the way for you :)

Oh yeah, I was looking for a signature file but the git tag is signed. Will try docker rootless and if that doesn't go well will try the above. Thanks :)

> <@sethsimmons:monero.social> You need to allocate huge pages:... (full message at libera.ems.host/_matrix/media/r0/do…faa333b1d82263c65004868993b6f97642b)

I thought for sure that we'd move on from block 69 at exactly 7 hours. ;-P

abberant[m]: They are all fine, but may need to restart all 3 procs

Or the host itself sometimes

I did :/

how do huge pages benefit p2pool? maybe its not even worth looking into since it seems to function normally

sech1 any chance at adding something like

--print-time=N so can see status and hashrate in logs when using tail -f or not using p2pool interactively? Or making it so you can run cmds from cmdline like you can with monerod (i.e. like 'monerod status' will give status of current running daemon).

(or an api to query!)

All Your Shares Are Belong To Us

shame the blocks belong to someone else :D

well this is beating previous efforts heh

it wants 369.69%

or WORSE

420.69%

it is even more painful due to having >500% effort here

PPLNS window = 2160 blocks (+62 uncles, 0 orphans)

Your shares = 0 blocks (+0 uncles, 0 orphans)

my poor little Ryzen 5 3600 just isn't getting lucky tonight

340.

369.69

This is it, no more blocks. p2pool is doomed. p2pool death spiral!!!

I think they highest I went through was some 580% or so on MO

blocks seem to be happenin fast. wonder if we're getting a lot of on / off behavior somewhere

I average about 1 share a day on a Ryzen 3600. Its not much but its honest work.

its perhaps the most honest work out there

to hell with honesty, I'm going to become a politician :)

2021-09-09 23:18:35.3290 P2Pool submit_block: BLOCK ACCEPTED at height 2445923 and difficulty = 323465146207

well it made us work for that one :)