Alert: chain fork caused by pre-0.8 clients dealing badly with large blocks

[MatthewLM]

I had to share this. 

Am I fine using the 0.8 to transact and is something this sort likely to happen in the future? Thanks!

It's only required for miners (Edit: They can use v0.8 under the default settings apparently), providing miners stick to the v0.7 compliant rules.

[lch5556875]

Incredible this

[solex]

Also, I'm afraid it's very easy to say "just test for longer" but the reason we started generating larger blocks is that we ran out of time. We ran out of block space and transactions started stacking up and not confirming (ie, Bitcoin broke for a lot of its users). Rolling back to the smaller blocks simply puts us out of the frying pan and back into the fire.

We are only in this situation because ONE source application is flooding the network with low fee transactions. That source application is abusing the Bitcoin network as a messaging system because it helps their business model run just slightly more conveniently for them.

.....
 1dicegEAr   56000      0.85449 |    62398 |   52977 (0.85686) |    8850 |     571 |   77209.45 |   76593.02 |   400.54 |    616.42 |  99.202
 1diceDCd2   60000      0.91553 |    95911 |   87374 (0.91620) |    7992 |     545 |   70581.58 |   69778.98 |     0.50 |    802.59 |  98.863
 1dice9wVt   64000      0.97656 |    16558 |   15147 (0.97956) |     316 |    1095 |   23993.71 |   23581.20 |   240.19 |    412.51 |  98.281
----------------------------------------------------------------------------------------------------------------------------------------------
           small (bets < 4 BTC) |  3522656 | 1452152           | 2054095 |   16409 |  766592.38 |  752057.27 |   261.23 |  14535.10 |  98.104
            big (bets >= 4 BTC) |   114417 |   52348           |   61777 |     292 | 2635378.11 | 2579381.45 | 23628.70 |  55996.66 |  97.875
----------------------------------------------------------------------------------------------------------------------------------------------
                                |  3637073 | 1504500           | 2115872 |   16701 | 3401970.49 | 3331438.72 | 23889.93 |  70531.76 |  97.927
----------------------------------------------------------------------------------------------------------------------------------------------

SD Profit before fees:      70531.76560380 BTC (2.073%)
Cumulative Fees Paid:        2837.93797500 BTC
SD Profit after fees:       67693.82762880 BTC (1.990%)
Pending Liabilities:          -74.93245984 BTC
Final SD Profit:            67768.76008864 BTC (1.992%)
----
Since Satoshi Dice started, there have been:
Blockchain Tx: 11423662  :  SatoshiDice Tx:  6698542  (58.6%)
Blockchain MB:   4904.6  :  SatoshiDice MB:   2753.4  (56.1%)[/font]

Some of this profit can be fairly spent RIGHT NOW on improvements to internalize transaction flow, or it should be blocked until it does. This is the real choice.

[marcus_of_augustus]

Sweet summary ...  ,

levity is always useful at times like these too ... after all it's just bitcoin, who needs em?

[JoelKatz]

Some of this profit can be fairly spent RIGHT NOW on improvements to internalize transaction flow, or it should be blocked until it does. This is the real choice.

Wasn't one of the advantages of Bitcoin supposed to be that you don't need anyone's permission to use it? Rather than asking someone to stop or, worse, trying to make them stop, the better path is to fix yourself so they're not hurting you.

[bitpop]

If this was a debate you don't have to follow along. You still have that power. But the majority voted. Also, this is still an infant community.

Some of this profit can be fairly spent RIGHT NOW on improvements to internalize transaction flow, or it should be blocked until it does. This is the real choice.

Wasn't one of the advantages of Bitcoin supposed to be that you don't need anyone's permission to use it? Rather than asking someone to stop or, worse, trying to make them stop, the better path is to fix yourself so they're not hurting you.

[Ausversucht]

Also, I'm afraid it's very easy to say "just test for longer" but the reason we started generating larger blocks is that we ran out of time. We ran out of block space and transactions started stacking up and not confirming (ie, Bitcoin broke for a lot of its users). Rolling back to the smaller blocks simply puts us out of the frying pan and back into the fire.

We are only in this situation because ONE source application is flooding the network with low fee transactions. That source application is abusing the Bitcoin network as a messaging system because it helps their business model run just slightly more conveniently for them.

.....
 1dicegEAr   56000      0.85449 |    62398 |   52977 (0.85686) |    8850 |     571 |   77209.45 |   76593.02 |   400.54 |    616.42 |  99.202
 1diceDCd2   60000      0.91553 |    95911 |   87374 (0.91620) |    7992 |     545 |   70581.58 |   69778.98 |     0.50 |    802.59 |  98.863
 1dice9wVt   64000      0.97656 |    16558 |   15147 (0.97956) |     316 |    1095 |   23993.71 |   23581.20 |   240.19 |    412.51 |  98.281
----------------------------------------------------------------------------------------------------------------------------------------------
           small (bets < 4 BTC) |  3522656 | 1452152           | 2054095 |   16409 |  766592.38 |  752057.27 |   261.23 |  14535.10 |  98.104
            big (bets >= 4 BTC) |   114417 |   52348           |   61777 |     292 | 2635378.11 | 2579381.45 | 23628.70 |  55996.66 |  97.875
----------------------------------------------------------------------------------------------------------------------------------------------
                                |  3637073 | 1504500           | 2115872 |   16701 | 3401970.49 | 3331438.72 | 23889.93 |  70531.76 |  97.927
----------------------------------------------------------------------------------------------------------------------------------------------

SD Profit before fees:      70531.76560380 BTC (2.073%)
Cumulative Fees Paid:        2837.93797500 BTC
SD Profit after fees:       67693.82762880 BTC (1.990%)
Pending Liabilities:          -74.93245984 BTC
Final SD Profit:            67768.76008864 BTC (1.992%)
----
Since Satoshi Dice started, there have been:
Blockchain Tx: 11423662  :  SatoshiDice Tx:  6698542  (58.6%)
Blockchain MB:   4904.6  :  SatoshiDice MB:   2753.4  (56.1%)[/font]

Some of this profit can be fairly spent RIGHT NOW on improvements to internalize transaction flow, or it should be blocked until it does. This is the real choice.

Um, is it just me or are they really SUPPORTING the network with the fees they are paying on transactions?  In 16 - 20 years this is going to be our bread and butter tx fees will likely hit 25+ bitcoins per block within that time frame if my math is even close to correct.  (not showing my math here, because it is filled with a lot of conjecture and guesstimation and I really don't feel like having it picked apart.)  If we think competition for blocks is harsh now.  Just wait until we are looking at 10-100k tx per block and fees that are 5x+ what the generated coins are, also consider that bitcoins will likely be in the $100-$1500 range by then (likely higher) as long as the network stays strong.  The adoption is there now.  We just need to keep stability in the network.  We need to get this shit figured out with larger blocks, because eventually with fees being the major source of income, no one will want to LIMIT block size.

[notme]

Um, is it just me or are they really SUPPORTING the network with the fees they are paying on transactions?  In 16 - 20 years this is going to be our bread and butter tx fees will likely hit 25+ bitcoins per block within that time frame if my math is even close to correct.  (not showing my math here, because it is filled with a lot of conjecture and guesstimation and I really don't feel like having it picked apart.)  If we think competition for blocks is harsh now.  Just wait until we are looking at 10-100k tx per block and fees that are 5x+ what the generated coins are, also consider that bitcoins will likely be in the $100-$1500 range by then (likely higher) as long as the network stays strong.  The adoption is there now.  We just need to keep stability in the network.  We need to get this shit figured out with larger blocks, because eventually with fees being the major source of income, no one will want to LIMIT block size.

Miners will certainly want to limit free transactions, since they cost money to store.  There is also the argument that with unlimited block size we will have a tragedy of the commons situation where everyone uses minimal fees.  If there is a limit, transactors will compete for the available slots, driving up fees.  It's not as simple as you seem to think.

[Ausversucht]

Available slots only works for just so long, and then you MUST adjust it.  Minimal fees are fine as long as the network continues to grow and flourish.  If you are collecting 0.2 - 0.5 BTC per block in fees right now that is GREAT!!!  Because if the network increases 100fold that could rise to 20 - 50 BTC per block.  I know there are other factors to consider, but I am hoping that in 16 - 20 years (yes I'm not looking at tomorrow or next year or two) Our reward for solving a block will be higher than the 50 BTC we started with.

[marcus_of_augustus]

Um, is it just me or are they really SUPPORTING the network with the fees they are paying on transactions?  In 16 - 20 years this is going to be our bread and butter tx fees will likely hit 25+ bitcoins per block within that time frame if my math is even close to correct.  (not showing my math here, because it is filled with a lot of conjecture and guesstimation and I really don't feel like having it picked apart.)  If we think competition for blocks is harsh now.  Just wait until we are looking at 10-100k tx per block and fees that are 5x+ what the generated coins are, also consider that bitcoins will likely be in the $100-$1500 range by then (likely higher) as long as the network stays strong.  The adoption is there now.  We just need to keep stability in the network.  We need to get this shit figured out with larger blocks, because eventually with fees being the major source of income, no one will want to LIMIT block size.

Miners will certainly want to limit free transactions, since they cost money to store.  There is also the argument that with unlimited block size we will have a tragedy of the commons situation where everyone uses minimal fees.  If there is a limit, transactors will compete for the available slots, driving up fees.  It's not as simple as you seem to think.

The fees and block limits discussion is long past due, they are intricately connected. Tighter limits on blocks will lead to more competition for scare resources and thus higher bids (fees) paid to process transactions.

User pays.

Expanding block limits to ridiculous sizes will ensure that blockchain maintenance becomes a rich man's only game .. i.e. only Google, Facebook and Citigroup can afford to maintain full blockchains. Conversely it will keep fees relatively lower for longer ....  decisions, decisions. But better we are kept aware of the impending changes before being rushed into them by an employee of Google.

there also exists the "amicable separation" solution and to split into a little persons chain BDB and big guys chain levelDB?

[notme]

Available slots only works for just so long, and then you MUST adjust it.  Minimal fees are fine as long as the network continues to grow and flourish.  If you are collecting 0.2 - 0.5 BTC per block in fees right now that is GREAT!!!  Because if the network increases 100fold that could rise to 20 - 50 BTC per block.  I know there are other factors to consider, but I am hoping that in 16 - 20 years (yes I'm not looking at tomorrow or next year or two) Our reward for solving a block will be higher than the 50 BTC we started with.

I'm not saying to leave the 1MB limit in place.  I'm just saying this statement is incorrect:

no one will want to LIMIT block size.

Expanding block limits to ridiculous sizes will ensure that blockchain maintenance becomes a rich man's only game .. i.e. only Google, Facebook and Citigroup can afford to maintain full blockchains. Conversely it will keep fees relatively lower for longer ....  decisions, decisions. But better we are kept aware of the impending changes before being rushed into them by an employee of Google.

Exactly.  Make the blocks too large too quick and watch the storage of the blockchain become centralized.  There is plenty of room to expand it without too much concern, but just throwing out the limit completely would never achieve miner consensus.

[robbonz]

http://coincompare.com/themes/bitcompare/images/bitcoin-fork.jpg

[Dabs]

Cumulative Fees Paid:        2837.93797500 BTC

What's wrong with the fees paid? Those all went to miners.

[solex]

Cumulative Fees Paid:        2837.93797500 BTC

What's wrong with the fees paid? Those all went to miners.

It is a small fraction of the total revenue, swamped by the block reward, Collectively revenue is doing very handsomely at the moment anyway:
https://blockchain.info/charts/miners-revenue
does Bitcoin need to be forced to have high fees at this time?

The problem is that the SD volume is essentially unbounded and is stress-testing Bitcoin quickly towards a known limit (the 1Mb block size) but now also finding a previously latent limit (default number of db locks) along the way.

I posted this only a few hours before the chain fork situation arose because I was worried about SD accelerating Bitcoin to the 1MB limit.
https://bitcointalk.org/index.php?topic=150493.msg1612165#msg1612165

My suggestion was to slowly push back flooding ("dead puppy") volume with higher fees targeted at that source.
Ultimately SD is under the control of the Bitcoin community. But what it is doing could be replicated in an attack by a malicious group.

[Syke]

Cumulative Fees Paid:        2837.93797500 BTC

What's wrong with the fees paid? Those all went to miners.

           small (bets < 4 BTC) |  3522656

What's wrong is they're spamming the chain with microtransactions with minimal fees causing a lot of dust that will bloat the chain. It's not healthy.

[Smoovious]

Cumulative Fees Paid:        2837.93797500 BTC

What's wrong with the fees paid? Those all went to miners.

           small (bets < 4 BTC) |  3522656

What's wrong is they're spamming the chain with microtransactions with minimal fees causing a lot of dust that will bloat the chain. It's not healthy.

"minimal" fees, which just so happens to follow the rules when it comes to transactions and the inclusion of fees... if you don't like the "minimal" fees on small transactions, then lobby to get the fees changed on small transactions... _ALL_ small transactions... don't just go targetting SD specifically. They are just following the system in place.

-- Smoov

[ProfMac]

Cumulative Fees Paid:        2837.93797500 BTC

What's wrong with the fees paid? Those all went to miners.

           small (bets < 4 BTC) |  3522656

What's wrong is they're spamming the chain with microtransactions with minimal fees causing a lot of dust that will bloat the chain. It's not healthy.

"minimal" fees, which just so happens to follow the rules when it comes to transactions and the inclusion of fees... if you don't like the "minimal" fees on small transactions, then lobby to get the fees changed on small transactions... _ALL_ small transactions... don't just go targetting SD specifically. They are just following the system in place.

-- Smoov

Equilibrium is an old Indian word that means "nobody is happy."

[Smoovious]

Cumulative Fees Paid:        2837.93797500 BTC

What's wrong with the fees paid? Those all went to miners.

           small (bets < 4 BTC) |  3522656

What's wrong is they're spamming the chain with microtransactions with minimal fees causing a lot of dust that will bloat the chain. It's not healthy.

"minimal" fees, which just so happens to follow the rules when it comes to transactions and the inclusion of fees... if you don't like the "minimal" fees on small transactions, then lobby to get the fees changed on small transactions... _ALL_ small transactions... don't just go targetting SD specifically. They are just following the system in place.

-- Smoov

Equilibrium is an old Indian word that means "nobody is happy."

When everybody is happy, liberty has died.

-- Smoov

[Come-from-Beyond]

Some of this profit can be fairly spent RIGHT NOW on improvements to internalize transaction flow, or it should be blocked until it does. This is the real choice.

Sounds like extortion, doesn't it?

[markm]

Some of this profit can be fairly spent RIGHT NOW on improvements to internalize transaction flow, or it should be blocked until it does. This is the real choice.

Sounds like extortion, doesn't it?

Huh? Him exercising his freedom to spam and me exercising my freedom not to include his spam in blocks I create, or even not to pass them along to my neighbors, is extortion?

Or him extorting from me, by spamming, the coding-minutes it'd take to filter him out is extortion?

-MarkM-