Fork at Block 58396?
-
Is this what a fork looks like? - Just wondering why the “Forked second block” has Bitcoin addresses in it?
14Qyt99aHa9DffEPT2a6RLTcM1YbXPRndk: 0.48760201
1Mcy6KEU5MGG9uWsVrYpRpkBSFfMMHEx47: 0.50669752
1AveXLG5PmKkDm9b1c2dLcQjX83P4kjiKJ: 0.56004949
1EFhFoVMwdnk8B2vcqbyxxcwgkKDWRxPKJ: 0.56122993
1NDxawGEAaC3ZB1axMDUHZDcPfetUpB42b: 1.00976513
1LT5GK9qreHGnDbTH7fZMaNyPaPykfFqVg: 1.05444895 -
yes this is a fork of 1 block. so not a big issue it happen sometimes.
so lets look at it a bit:
The block has 2 entries in the block chain one with it’s number and another one only known by it’s hash but has the hash of 58395 as previous.since the winning block was made by the attacker it can have been on purposed to orphan the block or an accident.
in normal condition what happen is- both block are mine and distribute to the network each on their side of network
- so miner mine on the first block they received
- when someone find the next block it is broadcasted and the other block becomes orphan as his chain is shorter
in orphan block all address are shifted to not be the normal address, FTC address shifted looks like Bitcoin address but they are not. if you click on them you will see the address and it’s transaction history if it’s not it’s first used. so click on one of the transaction (and note the amount for this address) in it’s history and you will see the real address as the output of the transaction(if more then one output you need to correlate the amount, but should be the top one) you can also check the # that is followed by the output number for the address.
it’s a one block as it has no next. if more then one block it will have next until one as no next. isolated 1-2 block orphan is usually not a big concern, longer need to be looked at a bit more carefully after 5 it’s very very likely an attack(for FTC with difficulty >50). many small fork is also a concern as it can be an attacker racing the chain and. one orphan every 2 to 15 blocks is a concern. 2 orphan isolated in 2 successive blocks have small probability but can happen, but more then 5 in 100 blocks is suspicious, 5 in 10 blocks is probably an attack.
how often we see 1 block fork when not under attack 1-2 time per 500 blocks, possibly a bit more at diff below 95 and sometimes 0 for 500 at diff 114.
-
Thanks for your explanation, of forks, Groll. Very interesting and informative.
-
Hi Groll, I’ll limit myself to 2 questions arising from the fork information.
1. I’ve been thinking of offering to run a daily automatic statistics gathering program on my PC, it could update statistics to the Forum, but have a “moderator” . Say automating some relevant analysis, e.g. number of orphans per block, Rate of Difficulty change, rate of coin production etc
From your explanation of forks, I assume we already run a system to highlight orphans. Could these uploaded to a forum post, so the community could monitor them?
2. Do you not agree it would be useful to have all orphans categorised? and, couldn’t that could be done by the community? As the information extracted could inform community action, like enforcing connection to the auto checkpoint (known good) node?
I personally, as an expert in Data monitoring and analysis, think it is essential collect and analyse this type of data. That information is essential to develop the correct solutions. Feathercoin has the opportunity to investigate all anomalies by community action and that would differentiate Feathercoin as a coin with real involvement and transparency.
I don’t care if we are over cautious and it scares some people off, they will come back when we prove we have solved the problem, not swept it under the carpet.
-
[quote]From your explanation of forks, I assume we already run a system to highlight orphans. Could these uploaded to a forum post, so the community could monitor them?[/quote]
nope we have nothing that is automated. i do it manually, but this would be a great tool
[url=http://2. Do you not agree it would be useful to have all orphans categorised? and, couldn’t that could be done by the community? As the information extracted could inform community action, like enforcing connection to the auto checkpoint (known good) node?] 2. Do you not agree it would be useful to have all orphans categorised? and, couldn’t that could be done by the community? As the information extracted could inform community action, like enforcing connection to the auto checkpoint (known good) node?[/url]
all anomalies: time, orphans are always good to be highlight so someone can look at them instead as searching for them ;) but this should not make a flood of post as only 4-5 persons will look at them on normal condition. Trust me looking at it everyday for long time is boring when nothing happen. and advance checkpoinitng should make malicious orphan something of the past. time will still be a problem. -
[url=http://2. Do you not agree it would be useful to have all orphans categorised? and, couldn’t that could be done by the community? As the information extracted could inform community action, like enforcing connection to the auto checkpoint (known good) node?] 2. Do you not agree it would be useful to have all orphans categorised? and, couldn’t that could be done by the community? As the information extracted could inform community action, like enforcing connection to the auto checkpoint (known good) node?[/url]
all anomalies: time, orphans are always good to be highlight so someone can look at them instead as searching for them ;) but this should not make a flood of post as only 4-5 persons will look at them on normal condition. Trust me looking at it everyday for long time is boring when nothing happen. and advance checkpoinitng should make malicious orphan something of the past. time will still be a problem.
[/quote]I agree, Groll.
I’m only looking at the Blocks and Hash, in depth now, because I believe we / Feathercoin need that information to make the correct development decisions.
I am hoping we make enough fixes so I can stop very soon!!! (PLEASE), and sit back watching my miners…
My analysis also shows the Difficulty change being restricted to 40% has been very successful, if we had restricted it to 25% - (especially down) I don’t think we would still be so concerned.
-
[quote]
[quote name=“wrapper0feather” post=“23858” timestamp=“1375260941”]
[url=http://2. Do you not agree it would be useful to have all orphans categorised? and, couldn’t that could be done by the community? As the information extracted could inform community action, like enforcing connection to the auto checkpoint (known good) node?] 2. Do you not agree it would be useful to have all orphans categorised? and, couldn’t that could be done by the community? As the information extracted could inform community action, like enforcing connection to the auto checkpoint (known good) node?[/url]
all anomalies: time, orphans are always good to be highlight so someone can look at them instead as searching for them ;) but this should not make a flood of post as only 4-5 persons will look at them on normal condition. Trust me looking at it everyday for long time is boring when nothing happen. and advance checkpoinitng should make malicious orphan something of the past. time will still be a problem.
[/quote]I agree, Groll.
I’m only looking at the Blocks and Hash, in depth now, because I believe we / Feathercoin need that information to make the correct development decisions.
I am hoping we make enough fixes so I can stop very soon!!! (PLEASE), and sit back watching my miners…
My analysis also shows the Difficulty change being restricted to 40% has been very successful, if we had restricted it to 25% - (especially down) I don’t think we would still be so concerned.
[/quote]I totally agree, and your charts demonstrate it nicely. Well done!
-
Cheers Kevlar, the information really shows how successful that change was. Any idea who came up with that 40% difficulty change idea as a fix? They need a gold star.
It works lovely against our 2 main problems, hash switching and attacks.
[attachment deleted by admin]
