Run a Bitcoin Testnet 4 node

[seoincorporation]

How can i run a Bitcoin testnet 4 node?

Most of you know how the V3 is about to die with those 2.4M unconfirmed transactions, and the V4 is on the way, but i see the testen 4 is alive and it already has blocks (https://mempool.space/testnet4) so, how can i run a testnet v4 node?

I would appreciate it if anyone had a hint on this topic because I wasn't able to find information about this.

---UPDATE----

USER: mocacinno packaged everything in a container:
https://bitcointalk.org/index.php?topic=5496494

[achow101]

Probably this PR: https://github.com/bitcoin/bitcoin/pull/29775

You'd have to pull the code in the PR and compile it yourself.

[LoyceV]

Most of you know how the V3 is about to die with those 2.4M unconfirmed transactions, and the V4 is on the way

I can think of several reasons to abandon the current testnet, but the transaction backlog caused by spam isn't one of them. I assume it's caused by Ordinals and other hypes, which makes it trivially easy for them to move to the new testnet. You'll still need to outbid them on fees.

[seoincorporation]

Most of you know how the V3 is about to die with those 2.4M unconfirmed transactions, and the V4 is on the way

I can think of several reasons to abandon the current testnet, but the transaction backlog caused by spam isn't one of them. I assume it's caused by Ordinals and other hypes, which makes it trivially easy for them to move to the new testnet. You'll still need to outbid them on fees.

The reasons behind the attack on the v3 are not the Ordinals, the main fixes for V4 are two:

-Add Testnet4 difficulty adjustment rules fix.
-Add timewarp attack prevention for Testnet4.

I don't understand at all the technical side of the problem, but some people describe it as a block storm https://blog.lopp.net/the-block-storms-of-bitcoins-testnet/

And if you want to follow the current discussion about the migration, you can do it on the next link: https://groups.google.com/g/bitcoindev/c/9bL00vRj7OU/m/9yCPo3uUBwAJ

Looks like devs, aren't happy at all with the fact that Testnet has a value on markets, and they are even thinking about stopping the Testnet and only moving on with the signet. But I'm sure that will not happen, the community needs the testnet even if some people give it a bad use.

3. Is all of the above a waste of time and we should instead deprecate testnet in favor of signet?

[mocacinno]

How to do this, on sles 15 sp4 (but should work on any linux distro)

Part 1: pull the pull request

Code:

git clone https://github.com/bitcoin/bitcoin.git
cd bitcoin 
git fetch origin pull/29775/head:pr-29775
git checkout pr-29775
git merge origin/master

at this point, you'll see a merge conflict... You'll need to edit src/init.cpp manually (open it with an editor and look for <<<<<<<). Some editors (like visual studio code) will make it very easy to solve merge conflicts.

then, after resolving the merge conflict

Code:

git add src/init.cpp
git merge --continue

Now, you can build

Tomorrow i *might* add a how-to build done
edit: fixed a typo (thanks LoyceV)

[jonval21]

Some new devs decide take action into the bitcoin ecosystem and some of them are mining already have millions of Tbtc4 (even satoshi group wouldnt  do such thing), the real reason is centralize testnet . Tbtc3 will be online always, there is no possible way to shut down a blockchain with POW, many people will use Tbtc3 due huge distribution for more than a decade.

BTC is based in distribuid system with no central authority (no coder is the authority in BTC)
be aware of POW system that allow fix the problem of The Byzantine generals problem

the resent attack (spam, campaing etc) to testnet3 is a clear red flag in bitcoin development, but we well see  game theroy taking action in the recents events, after that, we can predict what is going to happens to BTC ecosystem in the future.

BTC was designed for a group of people we huge knowledges in cryptograpy, distribuid system, and economic laws. so is no about just to know code.

some programmners are touching the BTC code and changing it violating the balance in the system ignoring the knoledge about distribuid system and economic laws. and that is a huge concern in the futuere of BTC.

hope some of the first (expert  in (economists  and distribuid system) programmers  be aware of this.

code is easy, what is hard is the science behind BTC.

[pooya87]

some of them are mining already have millions of Tbtc4

Testnet4 is at block #25039 and even if one person had mined all the blocks they would have 1.2 million coins not "millions of tbtc4"! It doesn't matter though, testnet coins have no value and also most probably it will reset when things are decided.
https://mempool.space/testnet4

Tbtc3 will be online always, there is no possible way to shut down a blockchain with POW, many people will use Tbtc3 due huge distribution for more than a decade.

Maybe. But don't forget that we aren't talking about any blockchain. We are talking about a TEST network. When the test environment becomes useless, it will be abandoned sooner or later.

Also have you noticed it is the version 3 already? That means we've already shut down 2 of them successfully...

[mocacinno]

In case you want to run your own testnet4 node: i packaged everything in a container:

I did think it deserved it's own thread, since i guess other people might find it handy, and it's also still a work in progress
https://bitcointalk.org/index.php?topic=5496494

[LoyceV]

You'll need to edit src/init.cpp manually

You missed a "t".

You'll need to edit src/ini.cpp manually (open it with an editor and look for <<<<<<<).

How?

Code:

<<<<<<< HEAD
    // TODO: remove the sentence "Nodes not using ... incoming connections." once the changes from
    // https://github.com/bitcoin/bitcoin/pull/23542 have become widespread.
    argsman.AddArg("-port=<port>", strprintf("Listen for connections on <port>. Nodes not using the default ports (default: %u, testnet3: %u, testnet4: %u, signet: %u, regtest: %u) are unlikely to get incoming connections. Not relevant for I2P (see doc/i2p.md).", defaultChainParams->GetDefaultPort(), testnetChainParams->GetDefaultPort(), testnet4ChainParams->GetDefaultPort(), signetChainParams->GetDefaultPort(), regtestChainParams->GetDefaultPort()), ArgsManager::ALLOW_ANY | ArgsManager::NETWORK_ONLY, OptionsCategory::CONNECTION);
=======
    argsman.AddArg("-port=<port>", strprintf("Listen for connections on <port> (default: %u, testnet: %u, signet: %u, regtest: %u). Not relevant for I2P (see doc/i2p.md).", defaultChainParams->GetDefaultPort(), testnetChainParams->GetDefaultPort(), signetChainParams->GetDefaultPort(), regtestChainParams->GetDefaultPort()), ArgsManager::ALLOW_ANY | ArgsManager::NETWORK_ONLY, OptionsCategory::CONNECTION);
>>>>>>> origin/master

I only kept the first "argsman" line fromt he above. Is that correct?

Tomorrow i *might* add a how-to build

I'd appreciate it. I'm stuck here:

Code:

./autogen.sh 
configure.ac:39: installing 'build-aux/ar-lib'
configure.ac:37: installing 'build-aux/compile'
configure.ac:24: installing 'build-aux/config.guess'
configure.ac:24: installing 'build-aux/config.sub'
configure.ac:27: installing 'build-aux/install-sh'
configure.ac:27: installing 'build-aux/missing'
Makefile.am:7: error: Libtool library used but 'LIBTOOL' is undefined
Makefile.am:7:   The usual way to define 'LIBTOOL' is to add 'LT_INIT'
Makefile.am:7:   to 'configure.ac' and run 'aclocal' and 'autoconf' again.
Makefile.am:7:   If 'LT_INIT' is in 'configure.ac', make sure
Makefile.am:7:   its definition is in aclocal's search path.
Makefile.am: installing 'build-aux/depcomp'
parallel-tests: installing 'build-aux/test-driver'
autoreconf: error: automake failed with exit status: 1

In case you want to run your own testnet4 node: i packaged everything in a container:

LOL. I should have continued reading first

[mocacinno]

Hi LoyceV,

Yup, you were right, you had to keep the top line, the bottom one is no longer needed.

It might be best to just use the container, i have it running on a demo machine and it works just fine... But if you want to try to build it on your local machine, you're always welcome to shoot me a quick PM (local language is fine, doesn't have to be in English)

I'm currently figuring out if i can build some kind of cpuminer in the same container and get it to work properly... cpumining might be long dead, it should still be feasible for testnet4.

PS: thanks for pointing out the typo, i'll fix it right away

[LoyceV]

I'm currently figuring out if i can build some kind of cpuminer in the same container and get it to work properly... cpumining might be long dead, it should still be feasible for testnet4.

I'll try your container next (and I've never used containers before). CPUmining would be fun, I tried it before on testnet, but had to conclude it's impossible. If there's even a single ASIC mining on testnet4, I expect CPUmining to be futile already.

Testnet4 is at block #25039 and even if one person had mined all the blocks they would have 1.2 million coins not "millions of tbtc4"! It doesn't matter though, testnet coins have no value and also most probably it will reset when things are decided.

I can think of a nice use: by the time the block reward gets low again, 1.2 million coins is enough to pay 1 testnet coin in transaction fees every 10 minutes for 20 years. That means miners will "always" get a bit more, even when the block reward is diminishing.

[mocacinno]

--snip--
I'll try your container next (and I've never used containers before). CPUmining would be fun, I tried it before on testnet, but had to conclude it's impossible. If there's even a single ASIC mining on testnet4, I expect CPUmining to be futile already.

The mining container is ready, it works, but (like you suspected) somebody is already running one (or more) asic's on testnet4.... why... i really wonder why.... why would you run an asic on a network that's designed to test ... why won't you let others get some tBTC to play with... At least, if you're nuking the testnet, create a faucet where others can have some tBTC to test out their setups...

If you need help with containers, let me know... But basically, if you trust what i've done, you could basically install docker and docker-compose on your system, make sure the docker daemon is running (on most linux distro's, a systemd service is created for you, you just need to enable and start it), then copy the content of the docker-compose.yml code block from this post (https://bitcointalk.org/index.php?topic=5496494.msg64078200#msg64078200) into a local docker-compose.yml file, edit the paths, ports and passwords and execute "docker-compose up" (or "docker-compose up -d" to start in the background)

If you don't trust what i've done (i don't blame you), you could copy the first code block in  above post in a file called "Dockerfile" (you can check all steps), then run "docker build -t name_of_image:random_tag .". At this point, you'll have an image of your own in your local repo, and you can edit the image: tag in the docker-compose.yml file to point to your local image instead of mine.

[LoyceV]

(like you suspected) somebody is already running one (or more) asic's on testnet4.... why... i really wonder why.... why would you run an asic on a network that's designed to test ... why won't you let others get some tBTC to play with... At least, if you're nuking the testnet, create a faucet where others can have some tBTC to test out their setups...

That's what I expected. I'll just give up already, there's no point to try mining against ASICs.

[LoyceV]

Part 1: pull the pull request
~
then, after resolving the merge conflict

I (finally) got it running! It took some fiddling with Mullvad VPN (for some reason it keeps blocking internet every time I create a new installation) and installing curl, autoconf, pkg-config, libtool, libc6-dev, build-essential and libboost-all-dev (thanks to other people with the same errors for recommending all those packages).
The result is a running bitcoind on testnet4. But there's no bitcoin-qt, which I kinda expected to be part of the default Bitcoin Core source compilation.
The total blocks directory is only 297 MB. This must be how Bitcoin felt in the early days.

[mocacinno]

Part 1: pull the pull request
~
then, after resolving the merge conflict

I (finally) got it running! It took some fiddling with Mullvad VPN (for some reason it keeps blocking internet every time I create a new installation) and installing curl, autoconf, pkg-config, libtool, libc6-dev, build-essential and libboost-all-dev (thanks to other people with the same errors for recommending all those packages).
The result is a running bitcoind on testnet4. But there's no bitcoin-qt, which I kinda expected to be part of the default Bitcoin Core source compilation.
The total blocks directory is only 297 MB. This must be how Bitcoin felt in the early days.

I'm glad you got it working
In case anybody else is having problems: i did make a container running the patched version and i published the Dockerfile... If you run opensuse leap 15.6, you *should* (theoretically) be able to use the Dockerfile as a cookbook...

• ignore the lines starting with #
• if a line starts with COPY, make sure the attached file is copied to the working directory
• if a line starts with RUN, remove the word RUN and execute the rest of the line
• if a line starts with WORKDIR, chdir to the workdir specified
• if a line starts with ENV, set the enviromental variable (either in ~/.bash_rc, ~/.profile,... or in your session by executing "export variable=value"
• if a line starts with ENTRYPOINT or CMD, it's basically the command to start the app
• if a line starts with EXPOSE, it'll usually list the ports your app is listening on (and you can ignore this if you use the Dockerfile as a cookbook)

EDIT: i just found out that v28.0 will merge the testnet4 pr... (or at least, testnet4 will be available). I'm building v28.0rc1 right now, and i'll make it available as Dockerfile and as an image on my dockerhub account.

• You can find the dockerfile on my github account: https://github.com/mocacinno/bitcoin_core_docker/tree/v28.0rc1
• You can find the image here: https://hub.docker.com/r/mocacinno/btc_core/tags

ps. i'm not in the habit of building release candidates... I've made an exception because of testnet4... I'm not completely sure if the v28.0rc1 tag will continue to exist after 28.0 is officially released... If you revisit this thread in a couple of months, and the links to my github and dockerhub are no longer working, just see if it hasn't been replaced by v28.0

[LoyceV]

I'm glad you got it working
In case anybody else is having problems:

Shortly after compiling it myself, I found bitcoincore.org/bin/bitcoin-core-28.0/test.rc1/, which has all the binaries already.