- 01 March 2024 (2 messages)
-
@droplister I have seen you in the XCP universe a few times lately (the youtube series about previous fork war etc) and I just found this Medium post by you and I really wish that I found this post when you made it. Great read. Thanks https://medium.com/@droplister/counterparty-development-101-2f4d9b0c8df3Counterparty Development 101Counterparty is NOT a blockchain.
-
I do a little blogging - 02 March 2024 (11 messages)
-
You are also a mod. Please feel free to remove spammers in the future -
Joined. -
If you feel motivated maybe put the specs in a blog post -
Can be a short post 😝
- 03 March 2024 (10 messages)
-
I don't really get the appeal of Pi's these days. They're not much cheaper that a fully decked-out NUC - 04 March 2024 (108 messages)
-
I'm going to nuke my AWS instance and wait till there is a stable path forward to test this... I cannot get past this error -
pip 22.0.2 from /usr/lib/python3/dist-packages/pip (python 3.10)
-
this is the cmd that I cannot get past
-
cargo install maturin
-
happy to test later just do not want to pay for something that does not work, once we know the steps I will stand it up again and install
-
it is all good, I do not *need* this at the momet but would love to help test and have one running at some point for my projects
-
ubuntu@ip-172-31-52-115:~$ python3 --version
Python 3.10.12
ubuntu@ip-172-31-52-115:~$ -
ubuntu@ip-172-31-52-115:~$ python3.11 --version
Python 3.11.8
ubuntu@ip-172-31-52-115:~$ -
ubuntu@ip-172-31-52-115:~$ python --version
Command 'python' not found, did you mean:
command 'python3' from deb python3
command 'python' from deb python-is-python3
ubuntu@ip-172-31-52-115:~$ -
Aliases are the issue here. Python and python3 run diff versions and scripts call python instead.
-
sorry dumb question time (dumber) how do I revert?
-
git update?
-
I was doing this step
-
Note: Python 3.11 is currently required.
sudo apt install python3-pip
git clone https://github.com/CounterpartyXCP/counterparty-lib.git
brew install leveldb # macOS-only
CFLAGS="-I/opt/homebrew/include -L/opt/homebrew/lib" # macOS-only
cargo install maturin
pip3 install -e . -
ok, have to jump on a call but will ping back
-
ok
-
it is weird I cannot use a thumbs up in this chat
-
OSX
-
Your system has both versions installed and is having trouble deciding which version to use when. Lookup aliases for python.
-
so I basically need to get python to point to the 3.11 version, if CP will support 3.10 I might just revert. I would like to have a version that is easy to rebuild so I can make a video and doc on the steps to get a node up and running
-
Send me a link to the instructions you are using when you get a minute.
-
This is what I was following:
-
-
I think did this to get 3.11
-
these instructions should work: https://ubuntuhandbook.org/index.php/2022/10/python-3-11-released-how-install-ubuntu/
-
no worries, nothing to be sorry about, will look to try this
-
I still get errors trying to cargo install Maturin
-
no
-
cargo install maturin
-
from the github
-
efaulting to user installation because normal site-packages is not writeable
Collecting maturin
Downloading maturin-1.4.0-py3-none-manylinux_2_12_x86_64.manylinux2010_x86_64.musllinux_1_1_x86_64.whl (10.6 MB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 10.6/10.6 MB 71.5 MB/s eta 0:00:00
Collecting tomli>=1.1.0
Downloading tomli-2.0.1-py3-none-any.whl (12 kB)
Installing collected packages: tomli, maturin
Successfully installed maturin-1.4.0 tomli-2.0.1 -
that seemed to work
-
ok
-
ok just doing the pip3 install again and I think I missed the "." and once done will do the bootstrap / start
-
ubuntu@ip-172-31-52-115://home/ubuntu$ counterparty-server bootstrap
counterparty-server: command not found
ubuntu@ip-172-31-52-115://home/ubuntu$ -
happy to try kickstart
-
just cannot find when it is installed
-
looked in -lib/-lib and -cli
-
ubuntu@ip-172-31-52-115://home/ubuntu$ ls
addrindexrs bitcoin-26.0 counterparty-cli counterparty-lib screenlog.0
ubuntu@ip-172-31-52-115://home/ubuntu$ -
back in 5
-
ubuntu@ip-172-31-52-115:/$ cd ~/.local/
ubuntu@ip-172-31-52-115:~/.local$ ls
bin lib share
ubuntu@ip-172-31-52-115:~/.local$ cd bin
ubuntu@ip-172-31-52-115:~/.local/bin$ ls
b58 block coinc counterparty-client counterparty-server coverage coverage-3.10 coverage3 flask keychain ku maturin msg normalizer py.test pytest tx
ubuntu@ip-172-31-52-115:~/.local/bin$ -
yup
-
counterparty-server kickstart?
-
ubuntu@ip-172-31-52-115:~/.local/bin$ counterparty-server kickstart
Traceback (most recent call last):
File "/usr/lib/python3/dist-packages/pkg_resources/__init__.py", line 573, in _build_master
ws.require(__requires__)
Requirement.parse('setuptools-markdown==0.2'), {'counterparty-cli'})
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "/home/ubuntu/.local/bin/counterparty-server", line 33, in <module>
sys.exit(load_entry_point('counterparty-cli', 'console_scripts', 'counterparty-server')()) -
shortened the error
-
yes
-
Warnings:
- Ensure addrindexrs is running and up to date.
- Ensure that bitcoind is stopped.
- The initialization may take a while.
Proceed with the initialization? (y/N)🙃️️ -
should I stop my btc node?
-
ok running kickstart, thanks for the help
-
hitting the gym
-
will check post workout
-
I have been dumping stuff in this chat as I will use to doc what I have done, sorry to all others in here
-
Release v9.61.2 · CounterpartyXCP/counterparty-core
9.61.2 Release Notes This is a hotfix release: Fix integer overflow in dispensers. Invalidate broadcast with malformed text. Fix Logging for Destructions with Invalid Asset. Upgrade Commands fedn...
-
I do like that there is momentum to remove all the unnecessary components. And I’m all for simplification.
But I understand what Jeremy is saying. We HAVE been relying on the fednode as the ONLY way to install counterparty.
With the current change in the repo, the “official” fednode is now broken. -
should probly throw a note in the readme for counterparty-lib with a link to counterparty-core as well
-
There in no counterparty-lib… right? Maybe it does work by redirect? Not sure about those details so maybe is not really broken???
-
there is, its just a placeholder tho
-
Yeah is going to be broken
-
I trust the development (I should still verify), but breaking the way we are used to installing without much notice is not good.
But so few people install nodes…
At least the github.com/CNTRPRTY/federatednode is still there. Now more relevant! 🤓GitHub - CNTRPRTY/federatednode: Federated Node Build System - Core Version - Don't trust, verifyFederated Node Build System - Core Version - Don't trust, verify - CNTRPRTY/federatednode
-
Right
- 05 March 2024 (88 messages)
-
I’m on debian for these reasons. PITA but more direct.
-
Hi @joebarba what brings you to this chat?
-
Correct me if I'm wrong. I interpret the statement as a request for pre-support of a future contentious fork
-
I interpreted as I responded.
As one of the few running production nodes, I know first hand that the fednode WAS the only way to install CP.
That is changing, being able to do it directly or with a much simpler docker setup. These are good changes.
BUT we should be a decentralized protocol. So changes like these should have more notice.
As of now, there is no clear way to install the CURRENT active version of the protocol from the “official” repo. -
Is about the detail of the repo rename. The fednode relies on the repository names.
And yes, all this should be solved with some changes to the fednode which shouldn’t be too hard -
Yep if the main protocol is improved enough to not need the fednode I can agree to deprecate it.
-
Oh yeah! I would reduce it even more lol
I’m in the mindset that the api endpoints should be application specific. The scope of the protocol to be reduced to consensus and block parsing mostly, with just a basic example api.
But that is just my personal opinion based on the products I use the protocol for -
What does it signify?
-
Yeah I didn’t think so either that’s why I asked
-
Name change is nice, i think -core is better than -lib, just might make sense to add a link to -core from -lib so old links will lead you to the right place
-
Including no counterblock and no counterwallet?
-
Package them together, sure I understand that, bitcoin core comes with a wallet. Counterparty does not. I dont have an opinion on counterblock at all but it would be a shame to not have an official counterparty wallet.
-
CLI will work great with images and stamps and mass adoption. /s
-
For a time it was, 0 developers were running it directly. Mainly because of how outdated it was…
So just clarifying the realities we were in -
This is nice!
-
Respectfully, the counterparty.io website only mentions counterwallet as a wallet for counterparty.
-
They are not part of the protocol. I would leave the “official” repo as slim as possible
-
No official wallet will leave this community ripe for fraud.
-
You don’t have to say anything?
-
#2. I understand not wanting to packaging them (bitcoin core comes with a gui wallet though so?)
An official wallet could be offered NOT bundled. It would also not be bundled if fednode is abandoned. -
#3 TBD after the goal posts stop moving.
-
The entire codebase?
-
But thats the way we always did it.
-
Good point 🤣
-
I think we are ALL very happy with the path the protocol is in. Thank you
-
I have tried engaging about the technical reasons behind counterblock and counterwallets problems here a few times, to have the same conversation without making progress. I’ll wait or engage on github later if there seems to be any light at the end of the counterwallet tunnel. If you know the technical reasons other than (counterblock broke) please enlighten me. Thanks.
-
How are data entities that do update (like orders and dispensers) going to be represented in the messages table?
I thought that even though table rows would not be updated, the data model (as in its non-time-series representation) itself would still be updatable… -
Using github today, I see why the lib/core rename is hitting hard. Dead URL’s everywhere.
-
Manually change the url from -lib/ to -core/ and you can still load a broken github URL FYI.
- 06 March 2024 (82 messages)
-
Kickstart and Bootstrap causing errors while trying to run 1st time
-
ubuntu@ip-172-31-52-115:~$ counterparty-server kickstart
[2024-03-06 10:58:06][INFO] Running v1.1.5 of counterparty-server.
[2024-03-06 10:58:06][INFO] Running v9.61.1 of counterparty-lib.
Warnings:
- Ensure addrindexrs is running and up to date.
- Ensure that bitcoind is stopped.
- The initialization may take a while.
Proceed with the initialization? (y/N): y
[OK] Connecting to addrindexrs...
[OK] Getting last known block hash...
[OK] Checking database state...
[OK] Backing up database...
[OK] Initialising database...
[OK] Starting blocks parser from block 284185...
[OK] Cleaning up...
Last parsed block: 289999
Kickstart done in: 97.351s
[2024-03-06 10:59:50][ERROR] Unhandled Exception
Traceback (most recent call last):
File "/home/ubuntu/.local/bin/counterparty-server", line 8, in <module>
sys.exit(server_main())
File "/home/ubuntu/counterparty-lib/counterparty-cli/counterpartycli/__init__.py", line 18, in server_main
server.main()
File "/home/ubuntu/counterparty-lib/counterparty-cli/counterpartycli/server.py", line 172, in main
server.kickstart(
File "/home/ubuntu/counterparty-lib/counterparty-lib/counterpartylib/server.py", line 507, in kickstart
kickstarter.run( -
and the end this...
-
counterpartylib.lib.check.ConsensusError: Incorrect txlist_hash hash for block 290000. Calculated d86938404e91f0c70c52b66beedc996a1601cabd10c18f3f052ad6110216af3a but expected c15423c849fd360d38cbd6c6c3ea37a07fece723da92353f3056facc2676d9e7
ubuntu@ip-172-31-52-115:~$ /usr/lib/python3.10/multiprocessing/resource_tracker.py:224: UserWarning: resource_tracker: There appear to be 80 leaked shared_memory objects to clean up at shutdown
warnings.warn('resource_tracker: There appear to be %d '
counterparty-server kickstacd .. -
(Hoped you knew what I meant, as blocks represent time. Anyway here rephrasing that part.)
“How are data entities that do update (like orders and dispensers) going to be represented in the messages table?
I thought that even though table rows would not be updated, the data model (as in its “static” initial representation) itself would still be updatable…”
Orders and dispensers are defined by an initial transaction. Whatever related event happens in the following blocks (order matches, dispenses) update the state of these initial transactions. And messages show an update for these.
Having a new append only DB approach (no row updates), how are entity “state updates” handled?
In the messages table new orders are shown as inserts, while new states for the previously inserted show as updates.
And it all makes sense and didn’t think this would change…
(Hope what I meant is clearer now.) -
So the messages table will only show inserts?
The messages table was cohesive in its inserts + updates design. As an application you could rely only on messages to know state.
It feels like the elegant simplicity of the protocol is being broken, as now applications that previously trusted the messages, will now have to make calculations to detect updates?
What is the purpose of the messages table now? -
I think there should be a separation between the database schema (which is now append only), and the “normalized” representation of the data entities. And the messages table should be this representation. Just like it was before changing to the append-only DB
-
if you can tell me where the db is I should be able to post it to a s3 bucket
-
I’ve used it twice to build explorer sites.
You can just ingest the feed and build out your tables with it. Or you can index just the messages and use the category column for filtering on it.
If the APIs get a lot better that’s less important. It’s a nice log of events imo.
Without a log like this are a lot of implicit updates happening, how could you audit those updates?
I don’t care either way, I’ll just do the new thing. -
For both xcpdev and bitSTART I use the messages table
-
You can go block by block detecting any data of interest.
Without it, you would have to know all tables and query each one separately per block. -
Big difference. Only one query vs n tables, per block
-
What's nice about the messages table is that if you've coded something to index Counterparty, you have a little less work to do with supporting new features because so far to date its just a new message category.
At least that's why I just used that approach with XCP.io (which is barely developed) so that it would be more flexible to changes being made. -
You can make it work however you want, people will adapt. But the message table has like that bindings json column which acts like a meta field which makes it flexi whereas the tables are all column-y.
-
I wouldn’t mess with the messages table and how it worked. I would be very disappointed if this changes.
The messages table may be the main table I’ve been using for my products. Changing its nature will require substantial rebuild.
And I just personally like that table. It made perfect sense as it existed, is simple to understand making the protocol approachable. -
Do the professional developer things you want to do
-
i def understand removing it since its a duplication of data
-
no reason you couldnt just have teh get_messages call in the api converted to something that assembles the same data from the other tables
-
I wouldn’t want to be that server
-
Then that’s the only alternative if removed, or not?
-
I like when counterparty registers a 0 on a dividend because someone had balance of the token before but doesn’t now. For financial data the explicitness is nice I think.
-
Keeping it is not an alternative?
-
you could make it optional
-
I would want it to continue working as it does today. A single table where all relevant state changes are recorded per block. With difference between new data entities, and the updates to such entities
-
historical data?
-
what happened in X block
-
thats standard block explorer functionality
-
ZeroMQ, unless Counterparty is using it, you have to re-implement Counterparty to use it as a means of listening. I could see web wallets choosing to do that.
Reading tables, the clunky fednode docker setup has been an impediment to me to do that right now. I know xcp.dev has figured that out though and it makes sense as an approach. I don't use docker in any of my day-to-day, using stuff like Vercel, so skill issue.
APIs, it sounds like the best route where the main issue is a small gap in releases. I think what could make APIs a lot better is normalizing quantities and names or always returning the longname or whether an asset is divisible along with quantity data. -
ZeroMQ is not replacing the messages table. It shouldn’t be in the list
-
There is a messages hash.
-
One of the foot guns that Bitcoin Runes has that I don't think they've thought enough about is that they're going to let people define to how many places their assets are divisible and it's a range of like 0 to 18 or something like that. It's going to be so annoying normalizing.
-
annoying enough just having divisibility
-
The specific message written in the transaction could be separated in some way.
In xcp.dev, I label it as the “main message” to stand out from the rest of the messages generated by a transaction.
Is there a term for the distinction between these messages? -
Nice counterparty-server UI improvements.
-
yes saw it, will be working on getting db tonight or tomorrow.
-
I was just trying to copy to local (osx) and getting
-
ubuntu@172.31.52.115: Permission denied (publickey).
-
trying to scp
-
I used a .pem file to ssh but do not think this is the issue
-
I was going to do the s3 bucket path but did that more for the ORD index as it was many gigs and wanted to be able to share this db is only 281m
-
nvmd just using filezilla
-
do you have a place I can post this for you to review?
-
here is a s3 bucket link that will die in 5 hrs
-
this is 12 hours, I will be driving back to camp tomorrow but can post again in the am if this does not work
-
-
there was a .old file in the same directory but I assumed that is not needed
- 07 March 2024 (136 messages)
-
The removal of the messages table means the removal of the messages hash.
It is more than a protocol change, it is changing the very nature of what is consensus.
My understanding of the social agreement of the “dev team” and the rest of the developers is that the next release is about optimizing and consensus fixes.
Let's keep it like this. Already 3 developers with explorers have expressed our reliance on the messages table. It is very much used.
I still believe we should keep it. But I can also see how it can be removed (like adding issuance's issuer to the ledger).
I believe, we should not confuse the trust we have given to the founders to renew the protocol with a “free pass” to do *anything*. -
I think that’s right.
You can think about the code and protocol in insolation, but I think it helps to consider the “federated network” and the ripples that changes have across that network.
You can weaken the federation if the wave is bigger than can be handled.
The more used frontends are basically re-implementing Counterparty to work how they do today.
So, on paper a big change within the counterparty node could have the same outputs and be “no different” but only if you ignore how existing services in the network interface.
This message table is an interface, effectively. Clearly not by design.
Because of how things are today, I would heavily weigh how xchain and freewallet interface and work with counterparty. These services have the most economic weight.
It sounds like removing this messages table is a preference but not a necessary change for your updates at this point. Treat it like a logger.
I would think you could get rid of it in tandem with improving the APIs so that people don’t need to re-implement counterparty and have an alternative to the messages table as an interface. -
Unless you’re gonna deliver an explorer (ord bakes one in) and a working wallet in tandem with these changes.
-
Sir get_messages is part of the api
-
That would break a lot if you removed all the get_[table] commands
-
That’s what that issue says
-
Basically since the API is not perfect, the get_[table] commands are necessary to route around it
-
So let me know when the rest of the API is perfect and I’ll use that lol
-
Same, my suggestion would be to perfect the rest of the API in one update with a plan to deprecate in the following update
-
Gives everyone a full cycle to make changes
-
I will say that having the ability to query tables directly allows people to be more creative and flexible as different uses cases come up that weren’t necessarily thought of or planned for prior
-
Perfect example is I can filter though get issuances with description wild cards to find asset data almost an order of magnitude faster than get_asset_info
-
Sure but descriptions weren’t necessarily as important for assets when the api was written
-
So rather than wait for the next api update when new queries are needed it’s flexible enough to allow creating more custom calls
-
Removing the flexibility makes users much more dependent on an active dev team addressing every need promptly
-
But having a flexible api made that less of an issue is my point
-
And maybe the solution is to just build a better api with some flexibility built in that’s more “controlled”
-
Well it’s hard to comment on the non existent new api
-
Maybe it solves all these problems, I guess we shall see
-
I think before even discussing a transition period we need to see what’s proposed as the replacement
-
Is this the script that did not play well with the mass volume of numerics on xchain? If so, this change would benefit you.
-
I use get_messages in memepool when viewing txs by block, this is standard block explorer behavior and I’m not opposed to replacing it with TBD new api call / combination of calls in the future but there’s certainly a use case for it
-
Wrt get_[table] calls, these are 90% of the calls I use throughout wallets and explorer and were used because they were most performant, also not opposed to replacing with TBD new calls but i would hope we don’t need to sacrifice performance for the sake of making the api “better”
-
I had to speed run changing out xchain API calls for counterparty API calls in January so I know I can get it done in about a week if necessary 😆
-
I can document all the calls I’m using and why and maybe that would help in working appropriate replacements
-
I think the biggest problem is filtering by each parameter isn’t available in a lot of the non get_[table] calls
-
So with get_asset_info I can’t say here’s a bunch of assets but I only need the results to give me ones with “STAMP:” prefix in the description
-
But I could take the same asset list and run with that filter in get_issuances
-
Sure it just becomes a performance issue without it since I have to filter results rather than get filtered results in the first place, but I get what you’re saying
-
what is the arg is it just kickstart --database-file? Like just remove the db so just recreates it?
-
sorry for being dumb so the --database-file is used like "--database-file newcp.db"
-
I just have not used a lot of linux before, almost all my stuff is GUI based
-
-
these are the only reactions I can make in this TG, not sure if this affects others or not
-
This is osx
-
I only have this issue in this TG
-
Will this compatibility extend to continuing seeing updates in messages?
I believe we should have a clear distinction between the database SQL schema (which is now append only) vs the normalized entities.
AND I also believe that CP developers that want to create their own api to the DB should be embraced. Obviously, with schema refactoring, these power users will need to update their queries. But DB schema changes should not be common for a mature protocol.
I literally remember the day I saw there was a hidden sql api endpoint and how I would have preferred to use that rather than having to be limited by the predefined api which has(had?) many limits and inefficiencies. -
You can appeal to what things should be, but it’s not a greenfield.
-
And people are giving feedback
-
I missed some scrolls maybe
-
Scroll right within that box for more.
-
The way I see things is different. And I can understand being misunderstood.
For me, Counterparty is the hashes, which are generated by a single database.
The API is split between write and read. For the write API, I am fully in agreement that it should not be bypassed.
But for the READING of the data, I rather build the queries myself based on the needs of my application. And I believe this is a better way to effectively use the protocol data.
I’m very sure the critics to this approach would have a different opinion if they had applications.
I would literally lower the scope of the READ API to some basic examples, and then delegate to app developers the READing interface to the database.
For THIS SOFTWARE, which basically everything ends in an SQLite DB, it is a POSITIVE for developers to just query directly. Why I need to wait for a “dev team” to build the query for me? It feels like there is an “elitism” where only a select few decide how to READ the data? I don’t buy it. -
I KNOW is not “standard practice”. I’m saying, it is the MOST direct interface to Counterparty.
The day CP stops being a SQLite DB, then my approach might not work.
I don’t understand the resistance to being more efficient lol. -
I’m a power user. I don’t need anyone telling me how to READ a database
-
When did this channel become a sock drawer?
-
Processing all this… and I think it is pretty evident that the high usage by CP product developers of the messages and get_<table> proves the point that a customized READ api in the reference protocol implementation is kind of pointless. Developers just want the data.
But I can still understand having some basic read api, for “backward compatibility” (which in many cases translates to quite inefficient queries) of such basic api consumers. -
I don’t think you screwed up.
-
Interesting description…
-
weird, that works, I guess it resets for a new TG?
-
OK] Connecting to addrindexrs...
[OK] Getting last known block hash...
[OK] Checking database state...
[OK] Initialising database...
[OK] Starting blocks parser from block 278269...
[OK] Cleaning up...
Last parsed block: 289999
Kickstart done in: 1872.533s
[2024-03-07 15:44:33][ERROR] Unhandled Exception
Traceback (most recent call last):
File "/home/ubuntu/.local/bin/counterparty-server", line 8, in <module>
sys.exit(server_main())
File "/home/ubuntu/counterparty-lib/counterparty-cli/counterpartycli/__init__.py", line 18, in server_main
server.main()
File "/home/ubuntu/counterparty-lib/counterparty-cli/counterpartycli/server.py", line 172, in main
server.kickstart(
File "/home/ubuntu/counterparty-lib/counterparty-lib/counterpartylib/server.py", line 507, in kickstart
kickstarter.run(
File "/home/ubuntu/counterparty-lib/counterparty-lib/counterpartylib/lib/kickstart/__init__.py", line 380, in run
tx_index = parse_block(kickstart_db, cursor, block, block_parser, tx_index)
File "/home/ubuntu/counterparty-lib/counterparty-lib/counterpartylib/lib/kickstart/__init__.py", line 282, in parse_block
blocks.parse_block(kickstart_db, block['block_index'], block['block_time'])
File "/home/ubuntu/counterparty-lib/counterparty-lib/counterpartylib/lib/blocks.py", line 201, in parse_block
new_txlist_hash, found_txlist_hash = check.consensus_hash(db, 'txlist_hash', previous_txlist_hash, txlist)
File "/home/ubuntu/counterparty-lib/counterparty-lib/counterpartylib/lib/check.py", line 261, in consensus_hash
raise ConsensusError(error_message)
counterpartylib.lib.check.ConsensusError: Incorrect txlist_hash hash for block 290000. Calculated d86938404e91f0c70c52b66beedc996a1601cabd10c18f3f052ad6110216af3a but expected c15423c849fd360d38cbd6c6c3ea37a07fece723da92353f3056facc2676d9e7
/usr/lib/python3.10/multiprocessing/resource_tracker.py:224: UserWarning: resource_tracker: There appear to be 31 leaked shared_memory objects to clean up at shutdown
warnings.warn('resource_tracker: There appear to be %d ' -
same error
-
but I did not update any code so I guess expecting the same result is expected?
-
from which dir
-
I think you renamed right after I pulled
-
when do I run that cmd
-
ubuntu@ip-172-31-52-115:~$ git log
fatal: not a git repository (or any of the parent directories): .git -
commit f941aea0cf36ec10789a82c1deefa038f4e9aeb8 (HEAD -> develop, origin/develop)
Merge: 2d8d5f15 8164faff
Author: Adam Krellenstein <adam@krellenstein.com>
Date: Sun Mar 3 07:42:25 2024 +0100
Merge pull request #1465 from CounterpartyXCP/wp/python-3-10
Support Python 3.10 -
I could pipe to a file if you needed
-
Last parsed block: 289999
-
mine was 289999
-
LIVe right now. Runes developers FWIW https://www.youtube.com/live/XMu2GPlQg0UOrdinals Coding Club (3/7/24)
Come hang out on the Ordicord and watch live coding, issue + PR review. Join the event live! https://discord.gg/ordinals
-
My favorite quote so far ‘dockers bs’…
-
Point of differentiation to Runes, the max "multi send" size is 6. It's a lot higher in Counterparty. They can go beyond 6, but only if they are sending each address the same amount.
- 08 March 2024 (37 messages)
-
I assume is based on the opreturn limit? Would they support more if is not a standard tx?
-
Yes
-
That’s cool
-
Ord dev thinks standardness will change and op return limit will go away or rise.
-
@jp_janssen is this organically connected to you? I do not know enough about the economics or politics of this niche at the moment. https://openstamp.io/explorer/jpja What is the use case?Link
Features SRC20 & Stamps Marketplace, Minting Service, Launchpad, and an Explorer for SRC20 & Stamps asset tracking.
-
Not related to me. -
I’m just sharing what I’m seeing.
-
I just virtualized a server and can now repurpose that hardware for a new node. Does anyone have any tips, ideas or thoughts about spinning up a new testnet bitcoin and counterparty node on debian or ubuntu? I’m leaning towards debian, and open to any os really.
Does counterparty play well with testnet? -
Can we pin this?
-
so I do the git pull again
-
Should that last command reflect -core/ instead of -cli/?
-
Is there a hack for getting XCP assets into a testnet wallet? Is there an active testnet explorer with dispensers, btcpay, oracles etc? I’m pretty sure I will need to burn test btc for testnet xcp.
-
I asume I can delete the old folders?
-
so just have couterparty-core
-
and not the others
-
yup
-
just can remove -cli and -lib
-
ok so after this run kickstart
-
should I nuke the db files also?
- 09 March 2024 (43 messages)
-
ok re running ks
-
I am taking a vm snapshop for cloning and rollback as I reach certain mile markers with initial node setup. Let me know where in that processes you would snapshot, if anyone has input on development related node backup/snapshots and why. Thanks.
-
I do this with ec2 volumes
-
Create snapshots whenever you want. I stop everything and label them with the last block
-
ubuntu@ip-172-31-52-115:~/.local/share/counterparty$ counterparty-server kickstart
[OK] Block 334999 parsed in 0.004s [56730/554873 blocks parsed - tx_index: 127538 - 0:53:10/8:40:07 (7:46:56)]
[OK] Cleaning up...
Last parsed block: 334999
Kickstart done in: 4725.921s
[2024-03-09 01:45:13][ERROR] Unhandled Exception
Traceback (most recent call last):
File "/home/ubuntu/.local/bin/counterparty-server", line 8, in <module>
sys.exit(server_main())
File "/home/ubuntu/counterparty-core/counterparty-cli/counterpartycli/__init__.py", line 18, in server_main
server.main()
File "/home/ubuntu/counterparty-core/counterparty-cli/counterpartycli/server.py", line 173, in main
server.kickstart(
File "/home/ubuntu/counterparty-core/counterparty-lib/counterpartylib/server.py", line 525, in kickstart
kickstarter.run(
File "/home/ubuntu/counterparty-core/counterparty-lib/counterpartylib/lib/kickstart/__init__.py", line 353, in run
tx_index = parse_block(kickstart_db, cursor, block, block_parser, tx_index)
File "/home/ubuntu/counterparty-core/counterparty-lib/counterpartylib/lib/kickstart/__init__.py", line 278, in parse_block
blocks.parse_block(kickstart_db, block['block_index'], block['block_time'])
File "/home/ubuntu/counterparty-core/counterparty-lib/counterpartylib/lib/blocks.py", line 204, in parse_block
new_txlist_hash, found_txlist_hash = check.consensus_hash(db, 'txlist_hash', previous_txlist_hash, txlist)
File "/home/ubuntu/counterparty-core/counterparty-lib/counterpartylib/lib/check.py", line 262, in consensus_hash
raise ConsensusError(error_message)
counterpartylib.lib.check.ConsensusError: Incorrect txlist_hash hash for block 335000. Calculated 9c584d9407bdf34ea4a3de44f1a5854ba47065a5e13aa1a3dca8239fcdc0e226 but expected 437d9507185b5e193627edf4998aad2264755af8d13dd3948ce119b32dd50ce2 -
I'm running a EC2 instance and it is costing around 335$ usd a month with the testing I have been doing, more just fyi is someone was looking to stand one up
-
yes develop I think
-
when I run git checkout I get the following from -core
-
ubuntu@ip-172-31-52-115:~/counterparty-core$ git checkout develop
Already on 'develop'
Your branch is up to date with 'origin/develop'.
ubuntu@ip-172-31-52-115:~/counterparty-core$ -
ubuntu@ip-172-31-52-115:~/counterparty-core$ git branch
* develop
master
ubuntu@ip-172-31-52-115:~/counterparty-core$ -
should I just start from a fresh instance at this point?
-
rule out red herrings?
-
ok
-
link to db
-
-
because the file is getting bigger you can use the awscli to send files from an E2C to S3
-
just have to setup a user and grant permission
-
without it would be hard for me up at camp with starlink on the upload
XCP Dev Chat
Public archive of Telegram messages.
- 01 Mar 2024 (2)
- 02 Mar 2024 (11)
- 03 Mar 2024 (10)
- 04 Mar 2024 (108)
- 05 Mar 2024 (88)
- 06 Mar 2024 (82)
- 07 Mar 2024 (136)
- 08 Mar 2024 (37)
- 09 Mar 2024 (43)
- 10 Mar 2024 (11)
- 12 Mar 2024 (5)
- 13 Mar 2024 (3)
- 15 Mar 2024 (17)
- 16 Mar 2024 (13)
- 18 Mar 2024 (7)
- 19 Mar 2024 (14)
- 20 Mar 2024 (82)
- 21 Mar 2024 (37)
- 22 Mar 2024 (12)
- 23 Mar 2024 (14)
- 26 Mar 2024 (57)
- 27 Mar 2024 (35)
- 28 Mar 2024 (151)
- 29 Mar 2024 (59)
- 30 Mar 2024 (187)
- 31 Mar 2024 (8)