Pages:
Author

Topic: Someone is spamming the blockchain - page 4. (Read 7813 times)

donator
Activity: 452
Merit: 252
July 10, 2013, 08:54:59 AM
#21
we were going to run into this problem sooner or later, script allows for data to be stored in the blockchain, however since there is currently no incentive to hold the blockchain (fees being paid to hold it), eventually we'll run into serious storage problems enhanced by using the blockchain for storage, my companies design is looking at that seriously and we're currently trying to figure out a method that will keep bloat down to a minimum while maintaining a DDOS proof account ledger.
legendary
Activity: 1890
Merit: 1086
Ian Knowles - CIYAM Lead Developer
July 10, 2013, 08:46:17 AM
#20
When signing with ECDSA you need to include a random number, which makes the signature different.

Okay - so random number rather than timestamp - the point is you *can* embed information then (and if random number then easier probably as perhaps if it was a timestamp it might be checked for range).

Of course I assume it is not that big so not much information can be sent with such an approach (yes - too lazy to check the exact details as I don't think it matters very much - if the wish is to send messages then Bitmessage would be a much better option).
legendary
Activity: 1792
Merit: 1111
July 10, 2013, 08:42:28 AM
#19
There is no timestamp in transaction

So what is the thing that makes an identical tx different each time you sign it (I have tested this so I know it to be a fact)?


When signing with ECDSA you need to include a random number, which makes the signature different.

There is no timestamp in transaction. Check the protocol.
hero member
Activity: 574
Merit: 500
July 10, 2013, 08:39:33 AM
#18
If the sender was to pay a fee he would probably refrain from this behavior

Of course - so the question is whether there something wrong with current fee rules that are permitting this?


Indeed. If a malicious person in control of a bot-net started making thousands of these swap transactions per hour then that would take down the network surely.
legendary
Activity: 1890
Merit: 1086
Ian Knowles - CIYAM Lead Developer
July 10, 2013, 08:37:17 AM
#17
If the sender was to pay a fee he would probably refrain from this behavior

Of course - so the question is whether there something wrong with current fee rules that are permitting this?
hero member
Activity: 752
Merit: 500
bitcoin hodler
July 10, 2013, 08:31:17 AM
#16
this can only be prevented by mining pools not adding the no fee transactions to the blockchain. If the sender was to pay a fee he would probably refrain from this behavior
legendary
Activity: 1890
Merit: 1086
Ian Knowles - CIYAM Lead Developer
July 10, 2013, 08:29:35 AM
#15
There is no timestamp in transaction

So what is the thing that makes an identical tx different each time you sign it (I have tested this so I know it to be a fact)?
legendary
Activity: 1792
Merit: 1111
July 10, 2013, 08:27:58 AM
#14
It's not spamming they are passing messages through timestamps

There is no timestamp in transaction
legendary
Activity: 1890
Merit: 1086
Ian Knowles - CIYAM Lead Developer
July 10, 2013, 08:27:09 AM
#13
Curious as to why they would use 300 BTC for an experiment when 5 would do the trick just the same...

It certainly is a lot of BTC to be doing this with - am guessing the owner must have quite a lot of BTC to spare.
hero member
Activity: 574
Merit: 500
July 10, 2013, 08:24:19 AM
#12
Why don't they just use BitMessage?

Indeed - if wanting to send information that would make a lot more sense - perhaps it is just an experiment (although using 300 BTC for such fun could be rather expensive if something goes wrong).


Curious as to why they would use 300 BTC for an experiment when 5 would do the trick just the same...
legendary
Activity: 1890
Merit: 1086
Ian Knowles - CIYAM Lead Developer
July 10, 2013, 08:21:43 AM
#11
Why don't they just use BitMessage?

Indeed - if wanting to send information that would make a lot more sense - perhaps it is just an experiment (although using 300 BTC for such fun could be rather expensive if something goes wrong).
legendary
Activity: 1212
Merit: 1037
July 10, 2013, 08:17:08 AM
#10
It's not spamming they are passing messages through timestamps

Why don't they just use BitMessage?
legendary
Activity: 1890
Merit: 1086
Ian Knowles - CIYAM Lead Developer
July 10, 2013, 07:53:45 AM
#9
How do I do this. I want to store my backups with the blockchain.

Hmm... at a rate of a few bytes per 10 minutes (assuming one tx per block which this spammer isn't even achieving btw) I would guess you would not want to be backing up anything very big (unless you are willing to wait years for your backup to complete).
sr. member
Activity: 406
Merit: 250
ALGORY.io Crowdsale starts on 8/12/2017
July 10, 2013, 07:41:35 AM
#8
Whatever, the downside is obvious: making the block chain data larger than necessary.
legendary
Activity: 1890
Merit: 1086
Ian Knowles - CIYAM Lead Developer
July 10, 2013, 07:17:48 AM
#7
Which would make sense, if the sender had some control over the timestamps.

I assume timestamps is referring to what is stored in a tx itself (not the block timestamp as that is provided by the miner).

Note that if you sign the same raw tx multiple times (by repeating the sign command) you end up with a different signed raw tx each time - this is due to a timestamp field in each tx (AFAIA).
kjj
legendary
Activity: 1302
Merit: 1026
July 10, 2013, 07:13:21 AM
#6
The messages are encoded in the time stamps.

Which would make sense, if the sender had some control over the timestamps.
legendary
Activity: 1890
Merit: 1086
Ian Knowles - CIYAM Lead Developer
July 10, 2013, 07:08:28 AM
#5
It matches the free transaction criteria.

Either someone is bored and wrote a script to keep sending the coins back and forth, or it's passing data. I think the latter is more likely.

I wonder whether the free tx criteria is too lenient (it could become quite a problem if we end up with 100s or even 1000s of bots doing the same thing).

The possibility of it being a (rather slow) method of sending data (via the timestamp part of the tx) does indeed make some sense - the address that I linked to is constant (it keeps being sent out from and back to that address) so that is also interesting as it may well be possible to work out *who* it is that is doing this.

Of course it is always possible that the computer that this bot is running on might crash and burn with the 300 BTC being lost forever. Grin
vip
Activity: 1316
Merit: 1043
👻
July 10, 2013, 06:58:59 AM
#4
It matches the free transaction criteria.

Either someone is bored and wrote a script to keep sending the coins back and forth, or it's passing data. I think the latter is more likely.
newbie
Activity: 19
Merit: 0
July 10, 2013, 06:54:44 AM
#3
So they are using the second/microsecond/etc as a way to encode a message? mmm did not think of that.

Will these transactions get verified? There is no fee included?

legendary
Activity: 1890
Merit: 1086
Ian Knowles - CIYAM Lead Developer
July 10, 2013, 12:46:29 AM
#2
I think it's clearer from this address: https://blockchain.info/address/15Z4XmorKSN51ndyPrZ2EtL7Nnksb88888 you can see that each tx is between 1 and 3 blocks after the previous tx.

It is a little surprising that these tx's are being mined without fee (although that is of course up to miners), however as they only each have the one UTXO they are all very small and of course the amount isn't considered as "dust" (maybe it's some sort of a protest?).
Pages:
Jump to: