WaterLooDown
Legendary
Offline
Activity: 924
Merit: 1000
|
|
September 13, 2014, 08:43:20 PM |
|
This isn't an exploit with hashrate. In fact, if this is the exploit I think it is, you don't need a lot of hashrate to do it. ...knip...Anyone else noticing the extremely long lag on the blockchain right now? It was like 2 hours of no updates, then all of a sudden we jumped 6 blocks. My wallet, the seeds page, and the exchanges all had the same issue. -Fuse EDIT- Looks like the total network hashrate halved, and the block times are coming in faster than normal. Not saying this is what it is, but I've seen this with coins being attacked with exploit attacks. I would suggest pool ops keep an eye on your pools. EDIT- Looks like we're cracking blocks every 5 seconds. This isn't normal.EDIT- https://explorer.guldencoin.com/#/address/Gf7wGAwJGDLfoHcNCRLKZqpk2EQU5ixA6c - our mystery wallet claiming all these blocks right now. We're watching this closely right now.. What seems to have happened is that a jumppool/multipool mined until KGW changed the difficulty to 535 (block 116955). After that the multipool left and the remaining miners had a hard time mining blocks. A few hours later two blocks were mined and the difficulty dropped to 354 (block 116957). At that point the multipool started mining NLG again. Since about 2:30 hours of blocks were missing the KGW kept difficulty low so the pool could mine a lot of blocks very fast. Other miners also got some blocks, but G7f got most of them because they have a lot of hashing power. We're now at block 117047 with diff 483. The last two blocks were not mined by the multipool, it seems they have left again. The main problem here is the huge spike in block-time which causes delay in transactions being confirmed. This is unacceptable and we're doing extensive research on the problem and are looking at some solutions. But don't expect anything too soon, it's important to make rational decisions and not jump too quickly in a situation like this. People that know me know I'm the last person to make stupid panicked decisions. That being said, I think a form of Digishield, or similar, needs to be added to the roadmap ASAP. That, or you just ask the pool to stop mining NLG. Pool OPs are usually pretty respectful of coin devs. The other option is to flat out block the IP range of the pool in the wallet code, and effectively just shut them out. 12 second block times aren't normal, regardless. Especially with a difficulty of 300+. -Fuse Just want to add that we are taking this very seriously and the investigation into other protocols is our top prio now. It's not a matter of if we will implement but which one. And of course this will have to be planned carefully. +100 Regarding the algo change I think the best choices are between Digishield and DGW2.
|
|
|
|
Jero
|
|
September 13, 2014, 08:43:59 PM |
|
Since today Guldencoinweb is responsive and a real improvement of reading on your smartphone. Also added a donation page which explain how donations will and are spend so far. For those interested, only available in Dutch: http://www.guldencoinweb.nl/doneer.phpAll donations will be used to promote Guldencoin and to donate Guldencoin initiatives. I appreciate all (negative and positive) feedback.
|
|
|
|
veertje
Legendary
Offline
Activity: 952
Merit: 1000
|
|
September 13, 2014, 08:56:11 PM |
|
Since today Guldencoinweb is responsive and a real improvement of reading on your smartphone. Also added a donation page which explain how donations will and are spend so far. For those interested, only available in Dutch: http://www.guldencoinweb.nl/doneer.phpAll donations will be used to promote Guldencoin and to donate Guldencoin initiatives. I appreciate all (negative and positive) feedback. Great website and up-to-date all the time with updates and news about NLG. Nice improvement. Donated some.
|
|
|
|
c_e_d
Member
Offline
Activity: 100
Merit: 10
|
|
September 13, 2014, 09:17:50 PM |
|
Regarding the algo change I think the best choices are between Digishield and DGW2.
With the slingshield mod to increase hash power needed for an attack (by ~20%)? Sounds the best solution available
|
|
|
|
Jero
|
|
September 13, 2014, 09:21:47 PM |
|
Since today Guldencoinweb is responsive and a real improvement of reading on your smartphone. Also added a donation page which explain how donations will and are spend so far. For those interested, only available in Dutch: http://www.guldencoinweb.nl/doneer.phpAll donations will be used to promote Guldencoin and to donate Guldencoin initiatives. I appreciate all (negative and positive) feedback. Great website and up-to-date all the time with updates and news about NLG. Nice improvement. Donated some. Thnx! A very generous and kind donation which will be used to support the growth of Guldencoin! We have a great community and a great coin. I am so happy these days, can't sleep any more... thanks again!
|
|
|
|
Jero
|
|
September 13, 2014, 09:23:04 PM |
|
Regarding the algo change I think the best choices are between Digishield and DGW2.
With the slingshield mod to increase hash power needed for an attack (by ~20%)? Sounds the best solution available Sound smart... something new or proven technology?
|
|
|
|
c_e_d
Member
Offline
Activity: 100
Merit: 10
|
|
September 13, 2014, 10:25:59 PM |
|
Regarding the algo change I think the best choices are between Digishield and DGW2.
With the slingshield mod to increase hash power needed for an attack (by ~20%)? Sounds the best solution available Sound smart... something new or proven technology? It's already used in some coins. I think coders can explain it better than me how it works. It reads like the mod jumps in if something smells like a 51% attack and checks a number of blocks back in the chain for a possible fork. Maybe WaterLooDown can explain/correct me?
|
|
|
|
/GeertJohan
|
|
September 13, 2014, 10:51:20 PM |
|
... knip ...
It's already used in some coins. I think coders can explain it better than me how it works. It reads like the mod jumps in if something smells like a 51% attack and checks a number of blocks back in the chain for a possible fork.
Maybe WaterLooDown can explain/correct me?
I don't believe the problem here is 51% attacks. Chain rewrites do occur in the NLG network. This is expected with the low blocktime and varying hashrate, and is handled fine (just as in bitcoin and pretty much any other coin). I've inspected all chain rewrites since the the beginning of august and in 90% of the rewrites by G7f, G7f rewrites a block that only contains a generation transaction to G7f itself. So that doesn't really sound like a coordinated attack on other miners. Do you maybe have some links to slingshot code? I can't really find it in Google. edit: nevermind i've found it.
|
|
|
|
CIG
Newbie
Offline
Activity: 44
Merit: 0
|
|
September 13, 2014, 11:49:29 PM Last edit: September 15, 2014, 04:13:39 PM by CIG |
|
PS: I'm fairly confident I got this right, but someone feel free to double check me.. i only got into crypto a few months ago so maybe I'm misinterpreting what these logs represent.
Crawled through some numbers again. Yeah markanth is definitely right: Gf7 is Clevermind! Gf7w (tag Clevermind) deposits the coins on Bittrex GQsU I can see is its Bittrex because GQsQ appeared as an input in some of my withdraws from Bittrex.
|
|
|
|
Dutchyyy
Legendary
Offline
Activity: 1197
Merit: 1001
|
|
September 14, 2014, 03:29:29 AM |
|
From now on I am only buying NLG on the weekends. Last weekend it dipped to 850 and it will do the same again this weekend. Best time to get in cheap!
|
|
|
|
Bluestreet
Legendary
Offline
Activity: 988
Merit: 1000
|
|
September 14, 2014, 03:53:09 AM |
|
From now on I am only buying NLG on the weekends. Last weekend it dipped to 850 and it will do the same again this weekend. Best time to get in cheap!
Guldencoins price will slowly rise in time until the first block reward halving which is a long way away. I believe there will be half the coins mined out when this happens. At this point if user base continues to grow at the current rate and merchants joining at same pace as August when the block reward halving does occur or just before is when you will see you biggest price increase. Over 576000 coins being produced a day could mean the price going back to 500 or even 100 if buyers dry up. For this to occur the coin devs would need to be silent for over 2 months.
|
|
|
|
Halofire
|
|
September 14, 2014, 06:06:29 AM |
|
Hey Community, I started the sustainability discussion here: https://forum.guldencoin.com/index.php?topic=521.0Check it out and add some feedback, all is revealed there.
|
OC Development - oZwWbQwz6LAkDLa2pHsEH8WSD2Y3LsTgFt SMC Development - SgpYdoVz946nLBF2hF3PYCVQYnuYDeQTGu Friendly reminder: Back up your wallet.dat files!!
|
|
|
SEGA99
Member
Offline
Activity: 98
Merit: 10
|
|
September 14, 2014, 06:17:42 AM |
|
I have added some feedbacks.Thanks for the working!!We just need you people!
|
|
|
|
Halofire
|
|
September 14, 2014, 06:29:35 AM |
|
I have added some feedbacks.Thanks for the working!!We just need you people! Thanks, I appreciate your words!
|
OC Development - oZwWbQwz6LAkDLa2pHsEH8WSD2Y3LsTgFt SMC Development - SgpYdoVz946nLBF2hF3PYCVQYnuYDeQTGu Friendly reminder: Back up your wallet.dat files!!
|
|
|
markanth
|
|
September 14, 2014, 06:46:45 AM |
|
... knip ...
People that know me know I'm the last person to make stupid panicked decisions. That being said, I think a form of Digishield, or similar, needs to be added to the roadmap ASAP. That, or you just ask the pool to stop mining NLG. Pool OPs are usually pretty respectful of coin devs. The other option is to flat out block the IP range of the pool in the wallet code, and effectively just shut them out. 12 second block times aren't normal, regardless. Especially with a difficulty of 300+.
-Fuse
If you have 100 GH/s, then 12 seconds at 300+ diff is a joke. As said, we're looking into solutions (edit: yes, maybe digishield). Blocking an IP range is not on our list, it's not a solution. If it is clevermining, maybe make a request to clevermining operator to not mine NLG like this? or at least slow down how often they hit NLG? worth a shot? pooled hashrate is up to 200 GH/s according to clevermining.com.
|
|
|
|
LTEX
Legendary
Offline
Activity: 1023
Merit: 1000
ltex.nl
|
|
September 14, 2014, 07:11:59 AM |
|
From now on I am only buying NLG on the weekends. Last weekend it dipped to 850 and it will do the same again this weekend. Best time to get in cheap!
Guldencoins price will slowly rise in time until the first block reward halving which is a long way away. I believe there will be half the coins mined out when this happens. At this point if user base continues to grow at the current rate and merchants joining at same pace as August when the block reward halving does occur or just before is when you will see you biggest price increase. Over 576000 coins being produced a day could mean the price going back to 500 or even 100 if buyers dry up. For this to occur the coin devs would need to be silent for over 2 months. Doesn't the blok reward halving mean less coins enter the market, creating more scarcity resulting in price rise? And I don't think Dev's are willing to silence up for two months, not in coins best interest
|
A fool will just look at the finger, even if it points to paradise!
|
|
|
bram_vnl
Legendary
Offline
Activity: 1148
Merit: 1000
|
|
September 14, 2014, 07:14:09 AM |
|
|
|
|
|
BioMike
Legendary
Offline
Activity: 1658
Merit: 1001
|
|
September 14, 2014, 07:48:43 AM Last edit: September 14, 2014, 08:01:00 AM by BioMike |
|
This isn't an exploit with hashrate. In fact, if this is the exploit I think it is, you don't need a lot of hashrate to do it. ...knip...Anyone else noticing the extremely long lag on the blockchain right now? It was like 2 hours of no updates, then all of a sudden we jumped 6 blocks. My wallet, the seeds page, and the exchanges all had the same issue. -Fuse EDIT- Looks like the total network hashrate halved, and the block times are coming in faster than normal. Not saying this is what it is, but I've seen this with coins being attacked with exploit attacks. I would suggest pool ops keep an eye on your pools. EDIT- Looks like we're cracking blocks every 5 seconds. This isn't normal.EDIT- https://explorer.guldencoin.com/#/address/Gf7wGAwJGDLfoHcNCRLKZqpk2EQU5ixA6c - our mystery wallet claiming all these blocks right now. We're watching this closely right now.. What seems to have happened is that a jumppool/multipool mined until KGW changed the difficulty to 535 (block 116955). After that the multipool left and the remaining miners had a hard time mining blocks. A few hours later two blocks were mined and the difficulty dropped to 354 (block 116957). At that point the multipool started mining NLG again. Since about 2:30 hours of blocks were missing the KGW kept difficulty low so the pool could mine a lot of blocks very fast. Other miners also got some blocks, but G7f got most of them because they have a lot of hashing power. We're now at block 117047 with diff 483. The last two blocks were not mined by the multipool, it seems they have left again. The main problem here is the huge spike in block-time which causes delay in transactions being confirmed. This is unacceptable and we're doing extensive research on the problem and are looking at some solutions. But don't expect anything too soon, it's important to make rational decisions and not jump too quickly in a situation like this. People that know me know I'm the last person to make stupid panicked decisions. That being said, I think a form of Digishield, or similar, needs to be added to the roadmap ASAP. That, or you just ask the pool to stop mining NLG. Pool OPs are usually pretty respectful of coin devs. The other option is to flat out block the IP range of the pool in the wallet code, and effectively just shut them out. 12 second block times aren't normal, regardless. Especially with a difficulty of 300+. -Fuse Just want to add that we are taking this very seriously and the investigation into other protocols is our top prio now. It's not a matter of if we will implement but which one. And of course this will have to be planned carefully. +100 Regarding the algo change I think the best choices are between Digishield and DGW2. KWG seems to be the problem. How about removing/deactivating KWG and making the code more simple? <edit> If completely removing the code isn't an option: While we are calculating both values anyway. Couldn't we make the network to accept the network difficulty in all odd block numbers and the KWG difficulty in the even block numbers? </edit>
|
|
|
|
bram_vnl
Legendary
Offline
Activity: 1148
Merit: 1000
|
|
September 14, 2014, 09:24:08 AM |
|
good to see the facebook page of guldencoin advertising his post
|
|
|
|
Youghoor
|
|
September 14, 2014, 09:58:11 AM |
|
This isn't an exploit with hashrate. In fact, if this is the exploit I think it is, you don't need a lot of hashrate to do it. ...knip...Anyone else noticing the extremely long lag on the blockchain right now? It was like 2 hours of no updates, then all of a sudden we jumped 6 blocks. My wallet, the seeds page, and the exchanges all had the same issue. -Fuse EDIT- Looks like the total network hashrate halved, and the block times are coming in faster than normal. Not saying this is what it is, but I've seen this with coins being attacked with exploit attacks. I would suggest pool ops keep an eye on your pools. EDIT- Looks like we're cracking blocks every 5 seconds. This isn't normal.EDIT- https://explorer.guldencoin.com/#/address/Gf7wGAwJGDLfoHcNCRLKZqpk2EQU5ixA6c - our mystery wallet claiming all these blocks right now. We're watching this closely right now.. What seems to have happened is that a jumppool/multipool mined until KGW changed the difficulty to 535 (block 116955). After that the multipool left and the remaining miners had a hard time mining blocks. A few hours later two blocks were mined and the difficulty dropped to 354 (block 116957). At that point the multipool started mining NLG again. Since about 2:30 hours of blocks were missing the KGW kept difficulty low so the pool could mine a lot of blocks very fast. Other miners also got some blocks, but G7f got most of them because they have a lot of hashing power. We're now at block 117047 with diff 483. The last two blocks were not mined by the multipool, it seems they have left again. The main problem here is the huge spike in block-time which causes delay in transactions being confirmed. This is unacceptable and we're doing extensive research on the problem and are looking at some solutions. But don't expect anything too soon, it's important to make rational decisions and not jump too quickly in a situation like this. People that know me know I'm the last person to make stupid panicked decisions. That being said, I think a form of Digishield, or similar, needs to be added to the roadmap ASAP. That, or you just ask the pool to stop mining NLG. Pool OPs are usually pretty respectful of coin devs. The other option is to flat out block the IP range of the pool in the wallet code, and effectively just shut them out. 12 second block times aren't normal, regardless. Especially with a difficulty of 300+. -Fuse Just want to add that we are taking this very seriously and the investigation into other protocols is our top prio now. It's not a matter of if we will implement but which one. And of course this will have to be planned carefully. +100 Regarding the algo change I think the best choices are between Digishield and DGW2.
|
|
|
|
|