v0.10.16 - Onyx v2

GermanRed+

Active Member
Aug 28, 2014
299
109
113
May I ask why Onyx v2 MN is not backward compatible with Onyx? There should not be any technical issues preventing that? Right?
 

yidakee

Well-known Member
Foundation Member
Apr 16, 2014
1,812
1,168
283
Updated my nodes. Will be checking to see if they drop
 

fernando

Powered by Dash
Dash Core Team
Moderator
Foundation Member
May 9, 2014
1,527
2,058
283
OK, MN:s dropped about after one hour, not good if need restart every hour.
Same here. I think they've been up less than an hour. At least I got a payment in that time :)
 

donho

Member
Masternode Owner/Operator
Apr 16, 2014
96
20
58
I'm running a hot/cold MN. "masternode debug" shows: "masternode started remotely"
MN list shows ~220MNs but I don't see my own IP
Have it running for ~18hours now on 16.6 and no payment so far (not that unusual)
Do I need to restart it because it's not in the MNlist or do I just have to wait?
 

yidakee

Well-known Member
Foundation Member
Apr 16, 2014
1,812
1,168
283
I'm running a hot/cold MN. "masternode debug" shows: "masternode started remotely"
MN list shows ~220MNs but I don't see my own IP
Have it running for ~18hours now on 16.6 and no payment so far (not that unusual)
Do I need to restart it because it's not in the MNlist or do I just have to wait?
Yes... is "darkcoind masternode list | grep xx.xx.xx.xx" does not bring back your IP with :1 - then it dropped.
 
  • Like
Reactions: fernando

djcrypto

Member
May 27, 2014
180
94
88
I'm just watching the masternode count. If it starts dropping again there's a problem
Any info how fast they drop off the list?
Mine are still online since I restarted them at 3:30am.
Sleep now...
 

donho

Member
Masternode Owner/Operator
Apr 16, 2014
96
20
58
I'm running a hot/cold MN. "masternode debug" shows: "masternode started remotely"
MN list shows ~220MNs but I don't see my own IP
Have it running for ~18hours now on 16.6 and no payment so far (not that unusual)
Do I need to restart it because it's not in the MNlist or do I just have to wait?
Fixed it with this: "shutdown all, delete peers.dat on local and remote, startup all, send masternode start twice a few seconds apart"
All credit goes to moocowmoo!
thanks again mate ;)
 

GermanRed+

Active Member
Aug 28, 2014
299
109
113
Give people/pools time to upgrade.
Why is there a rush to upgrade for v16 when v15 had a long waiting period to the enforcement? There is no reason to suddenly force people to upgrade immediately when v15 took 21 builds to become stable. The protocol should also be backward compatible until the v16 becomes stable.
 

yidakee

Well-known Member
Foundation Member
Apr 16, 2014
1,812
1,168
283
2 hours in and all 5 nodes are down :/

EDIT: Reverted first node back to 15.21, and debug shows started, and is :1 on the list without local restart
EDIT2: Just realised the node I've updated got payed a little while ago after updating to 16.6, dont know if before of after dropping though.

EDIT3:

Its cosmetic !! doing masternode list on v.15.21 show all my 16.6 MN as :1, issuing the same command under 16.6 daemon returns empty. Reverted 2/5 to 15.21 and will leave the rest and report back
 
Last edited by a moderator:

fernando

Powered by Dash
Dash Core Team
Moderator
Foundation Member
May 9, 2014
1,527
2,058
283
Why is there a rush to upgrade for v16 when v15 had a long waiting period to the enforcement? There is no reason to suddenly force people to upgrade immediately when v15 took 21 builds to become stable. The protocol should also be backward compatible until the v16 becomes stable.
Because a problem was discovered in the Darksend protocol that could lead to transactions being traced through fees.
 
  • Like
Reactions: GermanRed+

UdjinM6

Official Dash Dev
Dash Core Team
Moderator
May 20, 2014
3,638
3,538
1,183
Because a problem was discovered in the Darksend protocol that could lead to transactions being traced through fees.
Yes and No.

Yes, we should hurry as Darksend is core feature (now).

No, we should test more, just a few more hours of testing wouldn't hurt anyone but they would reveal:
1. Memory issues and crashes
2. Protocol issues and MNs delistings
and this would help us to avoid a lot of confusions

There is a HUGE amount of work Evan has done for these few days and there is a lot of changes
https://github.com/darkcoin/darkcoin/commit/8833d12a3af7d13b29079b6202ef2455971c8e34
I really respect it so why ruin it by pushing not properly tested code to mainnet?
 
  • Like
Reactions: thelonecrouton

fernando

Powered by Dash
Dash Core Team
Moderator
Foundation Member
May 9, 2014
1,527
2,058
283
Yes and No.

Yes, we should hurry as Darksend is core feature (now).

No, we should test more, just a few more hours of testing wouldn't hurt anyone but they would reveal:
1. Memory issues and crashes
2. Protocol issues and MNs delistings
and this would help us to avoid a lot of confusions

There is a HUGE amount of work Evan has done for these few days and there is a lot of changes
https://github.com/darkcoin/darkcoin/commit/8833d12a3af7d13b29079b6202ef2455971c8e34
I really respect it so why ruin it by pushing not properly tested code to mainnet?
I didn't have time to get into testnet because it is true that it was super fast, but I guess Evan felt confident about the code. Besides, not everything can be seen in testnet because the sizes and number of players are so different.
 
  • Like
Reactions: moli

Lariondos

Well-known Member
Foundation Member
Apr 8, 2014
89
61
158
I'm curious about the enforcement. It has been said that the changes to Darksend would not require a hard fork. So what's the point of turning enforcement off?
 

yidakee

Well-known Member
Foundation Member
Apr 16, 2014
1,812
1,168
283
Continue getting :1 from list grep v.15.21, but blank on v.16.6.
 

UdjinM6

Official Dash Dev
Dash Core Team
Moderator
May 20, 2014
3,638
3,538
1,183
I didn't have time to get into testnet because it is true that it was super fast, but I guess Evan felt confident about the code. Besides, not everything can be seen in testnet because the sizes and number of players are so different.
That's true, testnet differs from mainnet in terms of size and number of players. 1200 MNs vs less then 100 tMN in last test for example. But that shouldn't be "contra", that should be "pro". You can simulate any kind of (inappropriate) behavior there (forks, different kind of enforcements and even 51% attack), you can force smaller time periods for pings and other events to speed up protocol checks - however that would require to clean most of hardcoded values;) but that's a good thing anyway. In this situation for example dseep could be sent in 20 sec and tMNs could be removed in 70 sec of inactivity instead of 20 and 70 minutes respectively and so on.
 

AjM

Well-known Member
Foundation Member
Jun 23, 2014
1,335
571
283
Finland
Maybe there should be 2 week period when only bugs get fixed.
Jira issues is growing pretty fast.
 

AjM

Well-known Member
Foundation Member
Jun 23, 2014
1,335
571
283
Finland
this issue with darksend fees was most important, and its not like those 34 bugs are very important - wallet works great most of the time. and evan already made it clear that he needs more developers. so lets just be happy he gets so much work done and not tell him what to do next
Sure, important bugs first i agree, but this is just bump to remember.
What is the point to make jira issues if they never get fixed?