Huh, that's a neat example of unintended side effects.

Thanks all for clarifications.

No, that's not it.

The important part is "resets back to the minimum for a single block, after which it returns to its previous value".
There are plenty of such jumps - going between millions then one then millions over one block.
According to the code, the rule is that if the time of new block is over 20 minutes past the time of the previous block, then the difficulty can be set to one for this block and only this block.

What we are seeing on the graph is difficulty being permanently dropped, not just for one block, but for the whole next period, and then goes up slowly in steps of 2016. That is a completely different rule.

Actually, it's quite periodic. Made a graph of testnet difficulty, below.
So, what rule am i missing here?

http://i.imgur.com/UDePepN.jpg

I've noticed a surge of testnet blocks over the last few days, and apparently the difficulty permanently got reset to 1 around block 000000000002949f844e92645df73ce9c093e5aac0d962a0fa13eb076eec835c

I know it should go to 1 temporarily if there are no blocks mined for 20 minutes, but i wasn't aware of it being able to go to 1 permanently like that.

Further looking back reveals that this happens periodically. I.e. the previous one was at block 000000000007682725067b780feccae8143ae2a2b771639630f4d5b44b8548c1.

Is this a specific rule on the testnet - difficulty getting reset to 1 every now and then?

Or is it some sort of an effect from too many 20-minutes-late 1.0 blocks? The latter shouldn't be possible, since difficulty can't change by more than a factor of 4 at a time.

Ok, figured it out.
The segwit parts of my blocks were wrong, so most nodes would reject them.

However, there seem to be some old nodes that accept erroneous blocks like that (being unaware of these parts thanks to backwards compatibility), including that block explorer, but they don't propagate into the more up-to-date part of the network.
That's where the impression of segmentation came from.

I eventually stumbled onto a node that gave me a proper reject message (bad-witness-nonce-size), and it all clarified from there.

I've been trying to mine the Bitcoin testnet (for fun/education) with some obsolete HW, and after about 6 mined block the picture is always the same: The block is accepted and propagated by the nodes i can reach, show up on https://testnet.blockexplorer.com/blocks , but never show up on https://www.blocktrail.com/tBTC

Some time later, new blocks appear at the same height on then blocktrail side, and eventually my block gets orphaned.
Thing is, it's never even in the orphaned list on blocktrail, which makes me wonder if there is some propagation issues.

The blocks from the blocktrail side make it to blockexplorer side, but any block i submit on the blockexplorer never makes it to the blocktrail side.

I've tried to gather getaddr replies from many nodes, and then connected to about 50 live nodes out of the list, all of which would receive the block i mined.
However, it still won't show up in any shape or form on blocktrail side.

I'm confused about what is going on over there.

Earlier today i made a "birthday card" transaction, encoding a simple OP_RETURN message.
The recipient is a Douglas Adams fan, so on a whim i decided to make the transaction id start with a few 42s - 4242a76834b028a61aa94dcfda878a0106ce0a20d4a3fd9234990b1565b6539a
I have written my own Bitcoin client from scratch, including all the crypto stuff, so that was rather easy to do by adding a counter to k in the ECDSA signing function and letting it brute-force the thing for a few minutes.

However, when all was done i wondered how is it usually done, and a quick google failed to reveal anything even remotely related. There is a bit of speculation about it's possibility on this forum, and that's it.

So, the question is - have anyone ever made a "vanity" txid before?
Is there any meaningful demand for a tool that can do this?

How does the PoS work?
If, say, i have an address with 10k XSPEC, does it mean that i should be getting extra 500 per year (5% inflation target), thus a 2.something XSPEC added every other day?
Tried to reverse engineer it, but things don't add up.

Yep, i was trying them via Tor, and all were timing out.

The onion ones from cryptoid.info worked, thank you.

Is the network alive?
I'm trying to connect to the seed nodes with a custom client, and am getting no response from any of them and two block explorers.

Here are the ones i tried, port 37347:

spectre.cash
cryptoid.info
spectreproject.io
node1.spectreproject.io
node2.spectreproject.io
node3.spectreproject.io
node4.spectreproject.io

Alternatively, can someone provide an address of a known-good node?

I understood that. The question is what should i have set differently in a transaction for it to have been marked as RBF?

So, what should i change/set in a tx for it to be marked as RBF enabled?
There is surprisingly little i can google about this.

Greetings.

Lately i've been playing with Bitcoin at protocol level, trying to implement all the basics from scratch as a learning exercise.

Yesterday i made and signed a transaction to move some satoshis from one address completely to another, on mainnet.
At 44.643 sat/B fee it predictably didn't go anywhere, and after a day of poking ViaBTC's "accelerator" and getting "Submissions are beyond limit" i decided to try to send it again with a proper fee.
I read many times that an identical transaction with higher fee should supersede an existing one with a lower fee.

However, when i send it to the network it won't propagate.
After exhausting all possible bugs on my end i tried submitting it via https://coinb.in/#broadcast, and got a "258: txn-mempool-conflict" error.
Apparently it is being considered a double spend or some other sort of a collision.

So, the question is - is it possible to re-send a transaction with a higher fee, and if yes how to do it?

Here is the old transaction:

And the new one: