Can someone explain to me the "small output" problem?

[johny1976]

hello,

this is my transaction:



Still 0% propagation and 0 confirmations. Fee is 0.0002 which I think is enough

I just don't get it.

[shorena]

You have two outputs, maybe from change?

One of them is 0.00000438 BTC and thus below the dust limit, which makes your TX "non standart" and bitcoin-core clients >0.9 will not relay it AFAIK.

Edit: Which wallet did you use? Did it not warn you about this?

[johny1976]

You have two outputs, maybe from change?

One of them is 0.00000438 BTC and thus below the dust limit, which makes your TX "non standart" and bitcoin-core clients >0.9 will not relay it AFAIK.

Edit: Which wallet did you use? Did it not warn you about this?

I used blockchain.info wallet and it didn't warn me.. But had I other option? If I wanted to send exactly that amount there was no other chance to do it without this problem, right?

[Newar]

You have two outputs, maybe from change?

One of them is 0.00000438 BTC and thus below the dust limit, which makes your TX "non standart" and bitcoin-core clients >0.9 will not relay it AFAIK.

Edit: Which wallet did you use? Did it not warn you about this?

I used blockchain.info wallet and it didn't warn me.. But had I other option? If I wanted to send exactly that amount there was no other chance to do it without this problem, right?

It depends what is available to the wallet logic to choose from to begin with.

blockchain.info has the "Custom Send" option where you can select the addresses the unspent outputs are chosen from. Another option could have been to export the private keys to another wallet software which may chosen differently since it may have been programmed differently, again, also depending on what would have been available to the wallet logic to choose from to begin with.

[shorena]

-snip-
It depends what is available to the wallet logic to choose from to begin with.

blockchain.info has the "Custom Send" option where you can select the addresses the unspent outputs are chosen from. Another option could have been to export the private keys to another wallet software which may chosen differently since it may have been programmed differently, again, also depending on what would have been available to the wallet logic to choose from to begin with.

Its confirmed now (> 100 blocks), but I just realized one of the inputs was 438 satoshi as well (from here [1]), which makes me wonder if bc.i does select inputs at all or just uses all from a certain address that roughly fits the requirements.


[1] https://blockchain.info/tx/692f64ffdbe724bcf202ac827353e27239cdea9e3d2f3a98e9e1904c74e34908
which again looks like it was change.