Author

Topic: NXT :: descendant of Bitcoin - Updated Information - page 2194. (Read 2761637 times)

full member
Activity: 378
Merit: 104
basically, the chain can be rolled back 720 block if i remember correctly, so if we all agree to reload files from that date, we can roll everything back.

BUT if DfB is asleep for that long, we might be fucked, right?

There are .bak files from about an hour ago, before it all happened.

I have the next files at block 24167 before the hell happened.  I had noticed the block running long and thought I might be on a fork.  Saved them off to re-download the chain.  {I have verified at that block.}

But, we really don't need the blockchain at any level.  Most likely, there will be a new client to fix the issue.   That will result in the chopping-off off of the bad chain.  Essentially, the system starts at block 24168 again.

The important thing for now is that dgex.com stop deposits (and maybe trading) until the new client comes out.  Then, all will be fine with a new client.

agree.
i think the fix will be quite easy, it's just to invalidate these "ghost" transactions
full member
Activity: 378
Merit: 104
Hey guys, I am running 0.4.4 and my blocks aren't updating. If I delete the NXT folder, and redownload it will work, but then the blocks stop updating.

Anyone know whats up?

try to delete transactions.nxt and blocks.nxt, restart the java process, wait 2-3 mins and unlock your account
legendary
Activity: 1320
Merit: 1007
Hey guys, I am running 0.4.4 and my blocks aren't updating. If I delete the NXT folder, and redownload it will work, but then the blocks stop updating.

Anyone know whats up?
hero member
Activity: 808
Merit: 1011
It would be great to see you all at https://nextcoin.org for a better networking.
Register and support the community with your knowledge and communicate issues and suggestions to get NXT better and bigger.

And take a look into the NXT-Wiki!

Thanks!
hero member
Activity: 644
Merit: 500
basically, the chain can be rolled back 720 block if i remember correctly, so if we all agree to reload files from that date, we can roll everything back.

BUT if DfB is asleep for that long, we might be fucked, right?
It seems to be bigger problem than some growing fork. Someone find trigger to anti-coins, I suppose.

Developers should put to client some checkpoints for blockhain. There is no checkpoints.
Cos from very start creator wanna network to rise trully decentalized.

Anyway, the result is that the blockchain after the corrupted block is invalid
Exactly, I think. Let's shutdown until we'll get fix. Good night.
nov
sr. member
Activity: 433
Merit: 251
Independent crypto developer
I downloaded latest NXT wallet.

I see there just one NXT  there in web wallet, although there were 301 NXT.

In block explorer it shows that I should have 302 NXT.

What is wrong that I do not see the balance correctly?

And the confirmation amount is more than 1440.

But it does not show that it is forging?

I use latest nxt044.zip client version.
newbie
Activity: 43
Merit: 0
I've found an issue in peer address sharing mechanism. I forgot to uncomment code that checks peer address before sending it to other peers. This increases time required to find legit nodes coz the attacker floods the network with IPs that look valid but don't host NRS. I'll release a new version in a couple of days, now I'm going to sleep. If something goes wrong, don't panic, I'll fix it next morning.

GOOD WORK
legendary
Activity: 1151
Merit: 1003
Developers should put to client some checkpoints for blockhain. There is no checkpoints.
legendary
Activity: 1372
Merit: 1000
Quote
What is the appropriate amount of Nxt to keep in a digitalocean 2GB hallmarked account?

Quote
I use 100NXT, but the more the more weight the node has, and under DDoS attack, the more likely it will be targeted.

Yep, I'm trying to figure out the appropriate amount of DDoS to soak up with a digitalocean 2GB VPS.
sr. member
Activity: 378
Merit: 250
Anyway, the result is that the blockchain after the corrupted block is invalid, but the current client code does not know that. Once the bug is fixed, and the blockchain rescanned, it should reject the block with the invalid transaction and start a new fork. Hopefully BCNext and CfB can provide a fix soon. In the meantime, I would advise everybody not to do any transactions - because they will all be recorded in a chain that is going to be orphaned after the fix.

Is the blockchain rescanned on startup normally? I guess not, but may be made configurable in a fixed version.

full member
Activity: 266
Merit: 100
NXT is the future
   
24188      2584657662098653454      25-12-2013 01:49:12   
0      0 + 0      0 B   
1      15847357259617408334      261 %   
member
Activity: 98
Merit: 10
Yes, looks like somebody found an integer overflow bug. Account.balance is long, but Transaction.amount and Transaction.fee are integers. My guess is that somewhere in the transaction validation code those are subtracted in a way that gives a value < Integer.MIN_VALUE, which overflows and becomes a positive. So a transaction with fee exceeding the account balance is wrongly considered valid. Or something along those lines.

Anyway, the result is that the blockchain after the corrupted block is invalid, but the current client code does not know that. Once the bug is fixed, and the blockchain rescanned, it should reject the block with the invalid transaction and start a new fork. Hopefully BCNext and CfB can provide a fix soon. In the meantime, I would advise everybody not to do any transactions - because they will all be recorded in a chain that is going to be orphaned after the fix.


Yeah, such is the growing pains of a completely new code base.  They did say it was alpha.  I am going to bed.

Yeah, we're all alpha testers.

Then we get to be beta testers.

Then we all become early adopters Smiley
full member
Activity: 126
Merit: 100
What is the appropriate amount of Nxt to keep in a digitalocean 2GB hallmarked account?

I use 100NXT, but the more the more weight the node has, and under DDoS attack, the more likely it will be targeted.
legendary
Activity: 1232
Merit: 1001
Yes, looks like somebody found an integer overflow bug. Account.balance is long, but Transaction.amount and Transaction.fee are integers. My guess is that somewhere in the transaction validation code those are subtracted in a way that gives a value < Integer.MIN_VALUE, which overflows and becomes a positive. So a transaction with fee exceeding the account balance is wrongly considered valid. Or something along those lines.

Anyway, the result is that the blockchain after the corrupted block is invalid, but the current client code does not know that. Once the bug is fixed, and the blockchain rescanned, it should reject the block with the invalid transaction and start a new fork. Hopefully BCNext and CfB can provide a fix soon. In the meantime, I would advise everybody not to do any transactions - because they will all be recorded in a chain that is going to be orphaned after the fix.


Yeah, such is the growing pains of a completely new code base.  They did say it was alpha.  I am going to bed.
legendary
Activity: 1232
Merit: 1001
lol...
newbie
Activity: 21
Merit: 0
mastercoin appeared in coinmarketcap on 3d place ^^
legendary
Activity: 1092
Merit: 1010
I´ll get the drinks in, then  Cheesy
sr. member
Activity: 392
Merit: 250
Yes, looks like somebody found an integer overflow bug. Account.balance is long, but Transaction.amount and Transaction.fee are integers. My guess is that somewhere in the transaction validation code those are subtracted in a way that gives a value < Integer.MIN_VALUE, which overflows and becomes a positive. So a transaction with fee exceeding the account balance is wrongly considered valid. Or something along those lines.

Anyway, the result is that the blockchain after the corrupted block is invalid, but the current client code does not know that. Once the bug is fixed, and the blockchain rescanned, it should reject the block with the invalid transaction and start a new fork. Hopefully BCNext and CfB can provide a fix soon. In the meantime, I would advise everybody not to do any transactions - because they will all be recorded in a chain that is going to be orphaned after the fix.
full member
Activity: 126
Merit: 100
Since CfB already slept, can laowai or any other with technical knowledge answer me this question.

I am connected to the network using my local computer without hallmark. I don't care about forging benefit but I would like to support the network as much as I can. Does my computer help the network ? or does it just cause more stresses to some of your hallmarked nodes ?
If my node does not help the network, I would like to close my computer now so I can concentrate on some writing for alias system and asset exchange.

I am not exactly sure, but I think writing for the alias system and asset exchange are much more useful right now Smiley

Local computer helps the network only minimally as far as I can deduce, maybe even hurting at the time of DDOS, which is going on right now and it's big.

Yes I think we came to the conclusion last night that un hallmarked nodes are going to put more strain on the network during a DDoS attack. That said if you can hallmark a node, great, if you can't still feel free to setup an un-hallmarked one....this is your network too!
member
Activity: 98
Merit: 10
basically, the chain can be rolled back 720 block if i remember correctly, so if we all agree to reload files from that date, we can roll everything back.

BUT if DfB is asleep for that long, we might be fucked, right?

There are .bak files from about an hour ago, before it all happened.
Jump to: