v0.10.9.x Help test RC2 forking issues

Status
Not open for further replies.
yidakee said:
MN1 - EC2 North Virginia - 54.88.23.198 - Shows up instantly on chaeplin's list as :1 ... but remote daemon reports it twice (!!) - local daemon reports just once.
address 0 ;
mpv3jJQi9otaVTDhheHyKP7k8toLd73ott

ubuntu@ip-172-31-23-208:~$ darkcoind masternode list | grep 198
"54.198.191.99:19999" : 1,
"54.198.109.108:19999" : 1,
"54.198.145.174:19999" : 1,
"54.198.252.150:19999" : 1,
"54.88.23.198:19999" : 1,
"54.88.23.198:19999" : 1
Click to expand...
Click to expand...

I can confirm that your IP appears twice, using different pubkeys - this maybe a flaw in the masternode protocol. Can you write down the steps to reproduce this?

Code: 
~/darkcoin/src$ ./darkcoind masternode list | grep 198
    "54.198.191.99:19999" : 1,
    "54.198.109.108:19999" : 1,
    "54.198.145.174:19999" : 1,
    "54.198.252.150:19999" : 1,
    "54.88.23.198:19999" : 1,
    "54.88.23.198:19999" : 1
~/darkcoin/src$ ./darkcoind masternode list pubkey | grep 198
    "54.198.191.99:19999" : "myVSR4kptFPTSFjARc3DD8RkFvkLQ2CJ8c",
    "54.198.109.108:19999" : "n13k8bjboMdfkFCozoJW3Cwq9ZtcD4bFVP",
    "54.198.145.174:19999" : "mrXi3JeVASCew48xZCRnHq1L2M4FLvJDhT",
    "54.198.252.150:19999" : "mnwaBzVu4uMr5KEpQXYohSRxXwvgF5okiy",
    "54.88.23.198:19999" : "mvdVPpgaLxZGv1ddWTKeAsELc7uxAhHuZ6",
    "54.88.23.198:19999" : "mpv3jJQi9otaVTDhheHyKP7k8toLd73ott"

Evan, please have a look at this, it should not be possible to bind a IP twice!
 
Last edited by a moderator:
Humm... I just downloaded darkcoind of the official site, where it asks everyone to please update to 10.10.1 and yet, the download was 10.9.13
 
TanteStefana said:
Humm... I just downloaded darkcoind of the official site, where it asks everyone to please update to 10.10.1 and yet, the download was 10.9.13
Click to expand...

For me all is fine

Code: 
:~$ wget http://www.darkcoin.io/downloads/rc/darkcoind
--2014-06-14 20:19:28--  http://www.darkcoin.io/downloads/rc/darkcoind
Resolving www.darkcoin.io (www.darkcoin.io)... 162.252.83.46
Connecting to www.darkcoin.io (www.darkcoin.io)|162.252.83.46|:80... connected.
HTTP request sent, awaiting response... 200 OK
Length: 54246314 (52M)
Saving to: ‘darkcoind’

100%[======================================>] 54,246,314  2.13MB/s   in 8.4s

2014-06-14 20:19:41 (6.13 MB/s) - ‘darkcoind’ saved [54246314/54246314]

ubuntu@ip-10-65-166-48:~$ strings darkcoind | grep v0.1
v0.10.10.1-beta

Please retry
 
flare said:
I can confirm that your IP appears twice, using different pubkeys - this maybe a flaw in the masternode protocol. Can you write down the steps to reproduce this?

Code: 
~/darkcoin/src$ ./darkcoind masternode list | grep 198
    "54.198.191.99:19999" : 1,
    "54.198.109.108:19999" : 1,
    "54.198.145.174:19999" : 1,
    "54.198.252.150:19999" : 1,
    "54.88.23.198:19999" : 1,
    "54.88.23.198:19999" : 1
~/darkcoin/src$ ./darkcoind masternode list pubkey | grep 198
    "54.198.191.99:19999" : "myVSR4kptFPTSFjARc3DD8RkFvkLQ2CJ8c",
    "54.198.109.108:19999" : "n13k8bjboMdfkFCozoJW3Cwq9ZtcD4bFVP",
    "54.198.145.174:19999" : "mrXi3JeVASCew48xZCRnHq1L2M4FLvJDhT",
    "54.198.252.150:19999" : "mnwaBzVu4uMr5KEpQXYohSRxXwvgF5okiy",
    "54.88.23.198:19999" : "mvdVPpgaLxZGv1ddWTKeAsELc7uxAhHuZ6",
    "54.88.23.198:19999" : "mpv3jJQi9otaVTDhheHyKP7k8toLd73ott"

Evan, please have a look at this, it should not be possible to bind a IP twice!
Click to expand...

What is extra weird is that both those wallet addresses are mine.

Hold on! Just found the problem. I accidentally used the same EC2 IP for both local wallet config file.

Singapore node
address 0 = mvdVPpgaLxZGv1ddWTKeAsELc7uxAhHuZ6
masternodeprivkey=91fw5HuZ9mHhAQtGfr6cobEYzwqyMECDePbb8jNmzuCwbPuXvZu
masternodeaddr=54.88.23.198:19999
Click to expand...

North Virginia node

address 0 = mvdVPpgaLxZGv1ddWTKeAsELc7uxAhHuZ6
masternodeprivkey=927Wgp1G5B8eZNcu2XSvyoHZqcFNR2bW3QFv3tnUJBF4aMTgYwq
masternodeaddr=54.88.23.198:19999 <-- THIS is incorrect IP, should have been 54.255.178.13
Click to expand...

But to make matters worse... Remote reports "successfully starterted masternode"... tried stopping and starting, still successfull.
Tried rebooting and if fixed it

ubuntu@ip-172-31-3-203:~$ darkcoind masternode debug
Missing masternode input, please look at the documentation for instructions on masternode creation

(then local masternode start)

ubuntu@ip-172-31-3-203:~$ darkcoind masternode debug
successfully started masternode
Click to expand...

So there is a bug it seems. I pointed two local nodes, each with its own masternodeprivkey, activated two nodes, one with identical key, the other with different key, and it got listed twice with the original correct masternodeprivkey IP.
Anyways, apparently fixed, but cannot see my second node - 54.255.178.13
 
Hummm, well, I just deleted temp internet files... tried downloading again and got 10.9.13 so I did wget http://www.darkcoin.io/downloads/rc/darkcoind and again, I get 10.9.13 when I check with getinfo. I actually don't remember where I downloaded the 13.10.1 on my local machine? You all are getting the correct version of the darkcoin.io site? I'm just going to copy it off my local and put it on my remote....
 
If you downloaded the file "darkcoind" in the past it is saved on the root dir. Any future downloads will be "darkcoind.1" , 2, 3 etc
(if you want to run it use: ./darkcoined.1)
I think you keep running the old version.
Delete/rename the old darckcoind before you download a new one.

If you tried to download it few times in the past..... by now you might be at darkcoind.10
 
Tante, if you leave darkcoind in the folder, and do wget, it will download darkcoind, but save it as darkcoind.1

do

ls -la
Click to expand...

and check for darkcoind, and darkcoind.1 or even darkcoind.2 - wget will not overwrite over previous darkcoind
It may be you're downloading a new version, but actually installing an old one.

That is why I always wget, chmod +x, move to /usr/bin/ then immediately delete the download
 
yidakee said:
So there is a bug it seems. I pointed two local nodes, each with its own masternodeprivkey, activated two nodes, one with identical key, the other with different key, and it got listed twice with the original correct masternodeprivkey IP.
Click to expand...

Others would call it a feature: That way you can setup a "multi ticket node": prepare 5 local clients with 1000DRK and masternodeprivkey each, and point all five to the one hot masternode --> voila, you have your 5000DRK masternode, which has 5 entries in the masternode list --> 5 times the votes.

Will give it a try :grin:
 
flare said:
Others would call it a feature: That way you can setup a "multi ticket node": prepare 5 local clients with 1000DRK and masternodeprivkey each, and point all five to the one hot masternode --> voila, you have your 5000DRK masternode, which has 5 entries in the masternode list --> 5 times the votes.

Will give it a try :grin:
Click to expand...

I know!! But that doesnt make sense. Then what is the logic of having remote server with the local masternodeprivkey then? Might as well either

a) leave it blank so remote accept multiple tickets
b) insert masternodeprivkey into config file, so it rejects any other masternodeprivkey

Right now it seems neither/or - please do have a go at it flare! In any case I will be trashing this node. Singapore has horrible horrible ping even with broadband. SSH is a draaaaaaaaaag.... Moving to Ireland ;)
They have Guinness on tap too, love that beer.
 
yidakee said:
I know!! But that doesnt make sense. Then what is the logic of having remote server with the local masternodeprivkey then? Might as well either

a) leave it blank so remote accept multiple tickets
b) insert masternodeprivkey into config file, so it rejects any other masternodeprivkey
Click to expand...

As far as i know the masternodeprivkey is only used at startup.

- Start the remote with key1
- Kill the remote
- Replace the key to key2
- Start the remote with key2

could work out. We'll see...
 
Exactly... which is what is puzzling me.

- both remote have brand new IP
- both local trashed testnet3 = new wallet = new address 0
- both local generated brand new masternodeprivkey

by accident, I inserted the same EC2 IP in both local conf files. Fired up the first one (the correct config), and then the second one (wrong config). So, with that reasoning, I could very easily do the opposite. So if this is true, I could

a) handshake successfully masternodeprivkey A_local - A_remote
b) activate extra 1k ticket from B_local - A_remote (in this case B_local key is ignored)
c) invert sequence, handshake B_local - B_remote
d) activate extra 1k A_local - B_remote (in this case, A_local key is ignored)

In essence, falsely doubling the deposit. And its a squared law right? So with 3k, you could make 9k. Wow!! LOL, we should keep this a secret between us testnetters ;) haha
 
flare said:
As far as i know the masternodeprivkey is only used at startup.

- Start the remote with key1
- Kill the remote
- Replace the key to key2
- Start the remote with key2

could work out. We'll see...
Click to expand...

Works :grin: - actually i am not even using different servers, its all on one machine, you only need to ensure that there is only one wallet running at a time :smile:

Code: 
ubuntu@ip-10-167-172-12:~/darkcoin/src$ ./darkcoind masternode list pubkey | grep 196
    "184.73.179.196:19999" : "mkLN57jC9sYqMkp8RBcJF2H9hvhz6vabwQ",
    "184.73.179.196:19999" : "n38QabvJH1WVnzUkp2AfvbPK1mBHK74Ck8",
ubuntu@ip-10-167-172-12:~/darkcoin/src$ ./darkcoind masternode list  | grep 196
    "184.73.179.196:19999" : 1,
    "184.73.179.196:19999" : 1,

It's up to Evan to decide, wether this is a bug or a feature. Anyway, can't be exploited, because as soon as you move the 1000DRK from the wallet, the enty in the list will be removed.
 
yidakee said:
I could

a) handshake successfully masternodeprivkey A_local - A_remote
b) activate extra 1k ticket from B_local - A_remote (in this case B_local key is ignored)
c) invert sequence, handshake B_local - B_remote
d) activate extra 1k A_local - B_remote (in this case, A_local key is ignored)

In essence, falsely doubling the deposit. And its a squared law right? So with 3k, you could make 9k. Wow!! LOL, we should keep this a secret between us testnetters ;) haha
Click to expand...

I don't think this will work - no perpetuum mobile possible :smile:
 
flare said:
I don't think this will work - no perpetuum mobile possible :smile:
Click to expand...

Then how did my B_remote get activated? I dont get it. It started successfully but shaking hands with A_remote using B_local masternodeprivkey? So what happened, inactive B_remote was pinging the network and picked up its masternodeprivkey how?

To recap, accidentally both my A_local and B_local wallets were configed with independent masternodeprivkey, but same masternodeaddr=A_remote
So how on earth did

1) my B_remote server get "successfully started masternode" ?
2) why did the IP of A_remote appear twice with both my A_local_address0 & B_local_address0 ?

Flare, are you going to let them run to check of votes? Seems like a quirky solution, but hey, if it works, it saves on EC2 setups - severely reducing the actual number of nodes, which is a bad thing.
 
Last edited by a moderator:
yidakee said:
Then how did my B_remote get activated?
Click to expand...

How do you determine that the startup of B_remote was indeed sucessful? The "sucessfully started" message may the wrong.

yidakee said:
To recap, accidentally both my A_local and B_local wallets were configed with independent masternodeprivkey, but same masternodeaddr=A_remote
So how on earth did

1) my B_remote server get "successfully started masternode" ?
Click to expand...
The message may be wrong - as long as the IP does not show in 'global' masternodelist there is no use in this message. It will not make you get voted, the masternodelist entry is what counts.

yidakee said:
2) why did the IP of A_remote appear twice with both my A_local_address0 & B_local_address0 ?
Click to expand...

Because you started two clients with distinct masternodeprivkey but same remote IP. Exactly the setup i am running on one machine now.

- wallet1(privkey1 + wallet.dat1)
- wallet2(privkey2 + wallet.dat2)

start wallet1
start masternode wallet1
stop wallet1
start wallet2
start masternode wallet2
stop wallet2

result: Two entries, same IP, different pubkey/walletadress
 
flare said:
How do you determine that the startup of B_remote was indeed sucessful? The "sucessfully started" message may the wrong.
Click to expand...

Well, because;

1) stopped both EC2 instances to generate new IP, Remote_A and Remote_B
2) deleted both local testnet3 folders, so brand new address 0 local, so no mix-up in network
3) Updated all to 10.10.1
4) configured both local configs by accident pointing to Remote_A IP
5) started both Remote Masternodes first, masternode debug was "waiting for input"
6) started local A & B, masternode start ... but both with masternodeaddr=Masternode_A:19999
7) Both Remote_A & Remote_B report successfully started, when there was no direct handshake to Remote_B

So how did it become active?

Mind testing this?

I have since rebooted my Remote_B, and it was no longer stuck with "successfully started".... Fixed my Local_B config to correctly point to Remote_B. Fired it up and both sides report local/remote say "successful" - but the Remote_B IP does not show up anywhere.

Through daemon, Remote_A IP is still showing twice with :1 - on chaeplin's list it only shows up once. I've triple checked configs. Seems like a bug to me, but hey, I'm just a geek amateur.

This is Remote_A
ubuntu@ip-172-31-23-208:~$ darkcoind masternode list | grep 54.88.23.198
"54.88.23.198:19999" : 1,
"54.88.23.198:19999" : 1,
Click to expand...

Remote_B is
54.255.178.13

So it seems I locked both my A & B address 0's to Remote_A, even though I've stopped Local_B and redirected to masternodeaddr=Remote_B ...

ubuntu@ip-172-31-23-208:~$ darkcoind masternode list pubkey | grep 54.88.23.198
"54.88.23.198:19999" : "mvdVPpgaLxZGv1ddWTKeAsELc7uxAhHuZ6",
"54.88.23.198:19999" : "mpv3jJQi9otaVTDhheHyKP7k8toLd73ott",
Click to expand...
 
Last edited by a moderator:
Yes, I'd be honored. If it passes the Tante Stefana test, it's fool proof ('cause I'm realizing all the truly stupid spastic things I've been doing, LOL. Before I had my darkcoind in the home directory, now it's in /usr/bin, and that was causing me to be unable to delete darkcoind and install the new version, LOL. I'd delete a version in /home/Ubuntu but it wasn't the one I was executing, LOL.

Ok, I'll go through the tutorial one at a time and see if I have any hiccups (aside from the strawberry daiquiri I'm about to make, LOL)
 
flare

Results are in. Multicket works. Getting payed to both addressed for the same Masternode_A

http://23.23.186.131:1234/address/mpv3jJQi9otaVTDhheHyKP7k8toLd73ott
http://23.23.186.131:1234/address/mvdVPpgaLxZGv1ddWTKeAsELc7uxAhHuZ6

So since I've rebooted Masternode_B, and started succeffully local/remote, it just doesnt show up on the list, something is "locking out" my Masternode_B.
I will try stopping it again but generate a new masternodeprivkey. If it doesnt I'll try a new local wallet with new address 0
 
Last edited by a moderator:
More good stuff to report. I nuked testnet3 again, new genkey, new IP on masternode_B setup.
Not only did the the previous Local_B - Masternode_A drop the pubkey, the new Masternode_B got instantly picked up and is listed correct.

Still a bit tough to wrap my head around why all that nonsesne happened, but hey, testnet seems really well oiled up !!

EDIT: new love for Singapore. Record breaking payout. 30 minutes in I got payed - vive la variance!!
http://23.23.186.131:1234/address/mtSQDmxDEY69Kn31uo6aRJHxD1nYdFFzvo
 
Last edited by a moderator:
Status
Not open for further replies.
Back
Top