Re: How (and why) to use the Relay Network - page 16.

Paladin45

newbie

Activity: 6

Merit: 0

Quote from: Matt Corallo on October 18, 2014, 03:49:42 PM

Quote from: Paladin45 on October 18, 2014, 12:56:31 PM

I'm trying this with P2Pool and Armory, Linux Mint 17 64-bit. I'm seeing "Unable to resolve relay peer hostname" messages in the terminal window. Are these errors or normal operation?

Sounds like you mis-typed the relay node server (or your DNS is messed up)?

Yep. I missed the "relay" in the address string. Checked it a couple of times before I posted, but missed both times. Thank you!

Matt Corallo

hero member

Activity: 755

Merit: 515

Quote from: Paladin45 on October 18, 2014, 12:56:31 PM

I'm trying this with P2Pool and Armory, Linux Mint 17 64-bit. I'm seeing "Unable to resolve relay peer hostname" messages in the terminal window. Are these errors or normal operation?

Sounds like you mis-typed the relay node server (or your DNS is messed up)?

Paladin45

newbie

Activity: 6

Merit: 0

I'm trying this with P2Pool and Armory, Linux Mint 17 64-bit. I'm seeing "Unable to resolve relay peer hostname" messages in the terminal window. Are these errors or normal operation?

naplam

sr. member

Activity: 252

Merit: 250

Coin Developer - CrunchPool.com operator

yep, -phread after works

Matt Corallo

hero member

Activity: 755

Merit: 515

Quote from: -ck on October 15, 2014, 09:47:30 PM

Oops, I didn't spot the defines. However -pthread has to be after client.cpp for it to work.

Oh after, ok, willdo.

-ck

legendary

Activity: 4088

Merit: 1631

Ruu \o/

Quote from: Matt Corallo on October 15, 2014, 09:43:43 PM

Quote from: -ck on October 15, 2014, 04:25:18 AM

Running of the c++ client now fails to run on linux due to the position of the -pthread linking with the error:

Code:

terminate called after throwing an instance of 'std::system_error'
what(): Enable multithreading to use std::thread: Operation not permitte

Repositioning the -pthread here makes it work:

Code:

g++ -std=c++11 -Wall -I. -I/usr/include ./client.cpp -pthread ./flaggedarrayset.cpp ./utils.cpp ./p2pclient.cpp ./crypto/sha2.cpp -o relaynetworkclient

Probably should at least add -O2 to default build too

Huh? Which branch are you building from???
Master is https://github.com/TheBlueMatt/RelayNode/blob/master/c%2B%2B/make.sh which (unless you run with ./make test) has DEFINES set to -O2 and -flto, and -pthread is before any code?

Oops, I didn't spot the defines. However -pthread has to be after client.cpp for it to work.

Matt Corallo

hero member

Activity: 755

Merit: 515

Quote from: -ck on October 15, 2014, 04:25:18 AM

Running of the c++ client now fails to run on linux due to the position of the -pthread linking with the error:

Code:

terminate called after throwing an instance of 'std::system_error'
what(): Enable multithreading to use std::thread: Operation not permitte

Repositioning the -pthread here makes it work:

Code:

g++ -std=c++11 -Wall -I. -I/usr/include ./client.cpp -pthread ./flaggedarrayset.cpp ./utils.cpp ./p2pclient.cpp ./crypto/sha2.cpp -o relaynetworkclient

Probably should at least add -O2 to default build too

Huh? Which branch are you building from???
Master is https://github.com/TheBlueMatt/RelayNode/blob/master/c%2B%2B/make.sh which (unless you run with ./make test) has DEFINES set to -O2 and -flto, and -pthread is before any code?

Matt Corallo

hero member

Activity: 755

Merit: 515

Quote from: naplam on October 15, 2014, 07:15:04 AM

Are you hitting the maximum sockets limit or something? maybe you could restart them automatically every day or so for the time being

No, the issue y'all are seeing is that, if the thread handling the trusted bitcoind hangs or otherwise dies, it cant do its async connection reaping and, thus, people get disconnected as duplicate connections from the same IP if their connection gets reset somehow...Of course to fix this I need to actually figure out why its hanging/deadlocking/dying...

naplam

sr. member

Activity: 252

Merit: 250

Coin Developer - CrunchPool.com operator

Quote from: Matt Corallo on October 14, 2014, 03:03:37 PM

Quote from: norgan on October 14, 2014, 02:57:17 AM

yup same with Singapore node. both my p2pool nodes have out of cache errors as well.

Seems to be a bug when run for too long they start disconnecting new peers with that...I'm trying to figure out why, but, for now, they should all work atm.

Are you hitting the maximum sockets limit or something? maybe you could restart them automatically every day or so for the time being

-ck

legendary

Activity: 4088

Merit: 1631

Ruu \o/

Running of the c++ client now fails to run on linux due to the position of the -pthread linking with the error:

Code:

terminate called after throwing an instance of 'std::system_error'
what(): Enable multithreading to use std::thread: Operation not permitte

Repositioning the -pthread here makes it work:

Code:

g++ -std=c++11 -Wall -I. -I/usr/include ./client.cpp -pthread ./flaggedarrayset.cpp ./utils.cpp ./p2pclient.cpp ./crypto/sha2.cpp -o relaynetworkclient

Probably should at least add -O2 to default build too

norgan

sr. member

Activity: 308

Merit: 250

Decentralize your hashing - p2pool - Norgz Pool

Quote from: Matt Corallo on October 14, 2014, 10:53:16 PM

Quote from: norgan on October 14, 2014, 09:48:33 PM

Quote from: Matt Corallo on October 14, 2014, 03:03:37 PM

Quote from: norgan on October 14, 2014, 02:57:17 AM

yup same with Singapore node. both my p2pool nodes have out of cache errors as well.

Seems to be a bug when run for too long they start disconnecting new peers with that...I'm trying to figure out why, but, for now, they should all work atm.

it seems to cause p2pool to fall over as well. thats anecdotal of course so can't say for sure but the syptoms seem to line up on both nodes.

This one makes no sense...The way it was integrated into p2pool was to just kick off a thread and let it connect to bitcoind (not even to p2pool). Unless it hung python or hung your bitcoind, I dont see how this is possible.

yeah liek i said anecdotal, i wanted to be clear on that. I'll do some more testing.

Matt Corallo

hero member

Activity: 755

Merit: 515

Quote from: norgan on October 14, 2014, 09:48:33 PM

Quote from: Matt Corallo on October 14, 2014, 03:03:37 PM

Quote from: norgan on October 14, 2014, 02:57:17 AM

yup same with Singapore node. both my p2pool nodes have out of cache errors as well.

Seems to be a bug when run for too long they start disconnecting new peers with that...I'm trying to figure out why, but, for now, they should all work atm.

it seems to cause p2pool to fall over as well. thats anecdotal of course so can't say for sure but the syptoms seem to line up on both nodes.

This one makes no sense...The way it was integrated into p2pool was to just kick off a thread and let it connect to bitcoind (not even to p2pool). Unless it hung python or hung your bitcoind, I dont see how this is possible.

norgan

sr. member

Activity: 308

Merit: 250

Decentralize your hashing - p2pool - Norgz Pool

Quote from: Matt Corallo on October 14, 2014, 03:03:37 PM

Quote from: norgan on October 14, 2014, 02:57:17 AM

yup same with Singapore node. both my p2pool nodes have out of cache errors as well.

Seems to be a bug when run for too long they start disconnecting new peers with that...I'm trying to figure out why, but, for now, they should all work atm.

it seems to cause p2pool to fall over as well. thats anecdotal of course so can't say for sure but the syptoms seem to line up on both nodes.

Matt Corallo

hero member

Activity: 755

Merit: 515

Quote from: DrHaribo on October 14, 2014, 03:51:58 PM

Will new versions of the relay client be announced somewhere? (here maybe?)

Major updates will, yes.

DrHaribo

legendary

Activity: 2730

Merit: 1034

Needs more jiggawatts

Will new versions of the relay client be announced somewhere? (here maybe?)

Matt Corallo

hero member

Activity: 755

Merit: 515

Quote from: spin on October 14, 2014, 08:18:19 AM

So to build a relaynetwork between two nodes one could do the following:

Node 1: 10.0.0.1
bitcoind
relaynetworkserver
relaynetworkproxy 127.0.0.1 10.0.0.2

Node 2: 10.0.0.2
bitcoind
relaynetworkserver 127.0.0.1 8333

On node 2 I could also just run the relaynetworkclient also.
On node 1 (or 2) I could also add a relaynetworkproxy 127.0.0.1 public.REGION.relay.mattcorallo.com to connect to your network?

Questions:
Would the server also updated the local bitcoind with new tx/blocks?
Do I need the terminator at each node also? The servers shows less errors when it is running?
Also let me know if you are not happy with me trying this out.

Why not just run two relay node clients from your different nodes? If they're on the same network (or very close to each other, ie you can just be running a regular bitcoin p2p connection between them, try the branch at https://github.com/TheBlueMatt/RelayNode/tree/for-w, which has code to connect to multiple local mining pool instances and relay between them and to the relay network, with a full bitcoind receiving blocks/txn from the network.

Matt Corallo

hero member

Activity: 755

Merit: 515

Quote from: norgan on October 14, 2014, 02:57:17 AM

yup same with Singapore node. both my p2pool nodes have out of cache errors as well.

Seems to be a bug when run for too long they start disconnecting new peers with that...I'm trying to figure out why, but, for now, they should all work atm.

spin

sr. member

Activity: 362

Merit: 262

So to build a relaynetwork between two nodes one could do the following:

Node 1: 10.0.0.1
bitcoind
relaynetworkserver
relaynetworkproxy 127.0.0.1 10.0.0.2

Node 2: 10.0.0.2
bitcoind
relaynetworkserver 127.0.0.1 8333

On node 2 I could also just run the relaynetworkclient also.
On node 1 (or 2) I could also add a relaynetworkproxy 127.0.0.1 public.REGION.relay.mattcorallo.com to connect to your network?

Questions:
Would the server also updated the local bitcoind with new tx/blocks?
Do I need the terminator at each node also? The servers shows less errors when it is running?
Also let me know if you are not happy with me trying this out.

PatMan

hero member

Activity: 924

Merit: 1000

Watch out for the "Neg-Rep-Dogie-Police".....

EU node trotting along nicely.......

norgan

sr. member

Activity: 308

Merit: 250

Decentralize your hashing - p2pool - Norgz Pool

Quote from: Matt Corallo on October 13, 2014, 06:34:30 PM

Quote from: naplam on October 13, 2014, 04:16:05 PM

Same here with us-east now. Us-west seems to work fine at this time.

Now?

yup same with Singapore node. both my p2pool nodes have out of cache errors as well.

Topic: Re: How (and why) to use the Relay Network - page 16. (Read 695 times)