Just wanted to let you guys know that I'm successfully running a (pruned) Bitcoin node + TOR on a $11.99 single board computer (Rock Pi S). The SBC contains a Rockchip RK3308 Quad A35 64bit processor, 512MB RAM, RJ45 Ethernet and USB2 port and I'm using a 64GB SDCard. It runs a version of Armbian (410MB free). There's a new version available that even gives you 480MB RAM, but I'm waiting for Bitcoin Core 0.19 before upgrading. To speed things up I decided to run Bitcoin Core on a more powerful device to sync the whole blockchain to an external HDD. After that I made a copy and ran it in pruned mode to end up with the last 5GB of the blockchain. I copied the data to the SD card and ran it on the Rock Pi S. After verifying all blocks it runs very smoothly. Uptime at the moment is 15 days. I guess you could run a full node as well if you put in a 512GB SDcard. The Rock Pi S was sold out, but if anybody is interested, they started selling a new batch of Rock Pi S v1.2 from today. Screenshot of resources being used Bitcoin Core info Around 1.5 GB is being transferred every day --- Some links and a short How to for people that want to give it a try:
Set up UFW Firewall sudo ufw default deny incoming sudo ufw default allow outgoing sudo ufw allow ssh # we want to allow ssh connections or else we won’t be able to login. sudo ufw allow 8333 # port 8333 is used for bitcoin nodes sudo ufw allow 9051 # port 9051 is used for tor sudo ufw logging on sudo ufw enable sudo ufw status
Add user Satoshi so you don't run the Bitcoin Core as root sudo adduser satoshi --home /home/satoshi --disabled-login sudo passwd satoshi # change passwd sudo usermod -aG sudo satoshi # add user to sudo group
No inbound connections on my bitcoin full node with TOR
update: it's ok now, it only required some time. Also, looking at debug.log it won't show anything about inbound connections as I thought, it must be used `bitcoin-cli getpeerinfo` in order to get all the connections info (inbound + outbound) Can someone connect to my node? u3ob433qu3ebkmsu.onion:8333 I'm behind a CG-NAT so I thought to use TOR to overcame it but still I have outbound connections but no inbound. $ bitcoin-cli getnetworkinfo
Bitcoin Core 0.18.1 not connecting to peers after 60%+ sync
Hi, I am syncing Bitcoin Core 0.18.1 and initial sync it was pretty fast, "Progress increase per hour ~7%" after it gets to sync blocks from Jan 2018, is barely connecting to any peers, and if is connected to some peers download is like too slow, and it dissconnects to them sometimes is connected to 0 peers., and increased dbcache to 900MB. Is there any nodes list so I can try and connect using console, or do you have any suggestion how to speed up syncing? my internet speed connection 30Mbps, PC:CPU: [[email protected]](mailto:[email protected]);RAM:8GB;HDD:500GB https://imgur.com/a/gA1ffiT EDIT: problem solved by increasing dbcache to 4000MB, now the verification is going so fast, interesting how community with 1m+ members no one mentions dbcache is so important. thanks
I am downloading a full bitcoin node and after 5 days I’m at 63% but it seems to be taking even longer now. Is this normal? I know my internet connection isn’t the quickest 10-12mps and am using an old laptop i5 with 5gb ram.Any help much appreciated
Please send me your debug.log files from the BSV and BCH stress tests
I'd like to do some analysis of block propagation and block validation performance during the stress test, and I need some data. Can you give me yours? Specifically, I need ~/.bitcoin/debug.log files from as many different nodes as possible. II will be analyzing both BSV and BCH blockchains. Just sending me a link to your debug.log file is sufficient. But if you want to help me out a little more, I would appreciate it if you can rename the files so that they include your node's IP address or another unique identifier of your choice, like so:
The meta files are for any human-readable information you want to include about your nodes, like
VPS or dedicated
Whether ntpd or another clock sync method was correctly configured during the test
IP address (if you're willing to share it publicly)
To check your system clock, you can use timedatectl status (Ubuntu) or ntpdate -q pool.ntp.org (most linux distros). My analysis depends on relatively accurate timestamps, so this is by far the most useful piece of information for me. If you have any other data that you've collected, such as CPU usage statistics or bandwidth monitoring data, feel free to give me a link for that as well. If you want to not post any of this publicly, you may email me at jATtoomDOTim and tell me which (meta-)data you want me to not repeat without permission. If anyone else wants to look through this data, let me know and I will put it all in a public repository somewhere. The intent is to do an analysis similar to the one I did for the Sep 1st stress test. I believe rinexc managed to log some stratum header information from SVPool, and we also saw 6 orphaned blocks from SVPool alone, plus quite a few from other pools, so I think there's going to be some interesting stuff to find here. I also have an analysis of an incomplete, early dataset here. If your debug.log files are very large and you don't have a good way to get them to me, please message or email me with your ssh pubkey (ssh-keygen -t rsa; cat ~/.ssh/id_rsa.pub) and the file size, and I'll give you instructions on how to upload it to one of my servers.
Please set up your nodes to collect data for the stress test tomorrow.
Please make sure you have NTP or timesyncd installed and active, or another form of clock synchronization, and please enable debug=bench and do anything else you can think of to collect data on node performance. The stress test is expected to be pretty big. The expected transaction throughput rate on this one is something like 250 tx/sec, which is likely to saturate ABC's net_processing code. It will be informative if we can collect some data on where exactly in the code the bottlenecks are. If you know how to compile from source, it would be great to compile bitcoind with profile symbols enabled (export CXXFLAGS="-pg -O2", I think, followed by "./bitcoind -daemon=0" -- profiling data won't be collected in daemon mode). Also, information on whether the UTXO cache and disk activity is the bottleneck will be helpful. Does dbcache=8192 help? If you have a ton of RAM, does cat ~/.bitcoin/chainstate/* >/dev/null help? (That would cache the DB at the OS level. Still uses leveldb, but no disk.) Do you see high disk tps with iostat? High iowait %? Unfortunately, I'm extremely busy right now and won't have the time I wish I had to devote to this project, so if anybody else who has time who can make specific suggestions on how to implement the things above, that would be appreciated. The thread for config options on the first stress test is here. Also, there's my writeup of the Sep 1st stress test results. If any of you are running BSV nodes, please set those up for data logging too.
How long needs a raspberry pi3+ to sync from 0 to full? I know because I did: Its 19 Days
So I got a Raspberry pi 3+ and a Samsung MU-PA500B SSD, installed a headless raspbian and bitcoin core and let it run on Dec 16 and today it finally finished syncing :) My internet connection is a 100Mbit down / 20Mbit up cable line and this bitcoin.conf settings work best for me: daemon=1 logips=1 maxconnections=40 maxuploadtarget=5000 txindex=1 rest=1 server=1 walletrbf=1 dbcache=100 I don't have swap and dbcache=100 lets the bitcoin core daemon use ~800MB of memory which works nice for the pi. Lessons learned? Not really, running bitcoin core is easy as fuck, simply fire&forget, the fun part starts now when a personal electrum server and a lightning node will come on top, but however there indeed was a lesson I learnt: Don't use fancy shmancy file systems like btrfs on your ssd because creating of swap files is not supported and it took me hours to find out this simple fact^^
Bitcoin core v18.0 keeps crashing on cloud server. Help.
I am trying to synchronize a full node on a Digital Ocean instance with 3GB of RAM and 300 GB disk space. However, I noticed that bitcoin core is crashing very frequently. From the instance monitoring, it seems like it's a RAM issue: https://imgur.com/a/KuMev4A Thought that 3GB of RAM were more than enough but apparently not. What is the recommended RAM then? Here is how I installed bitcoin core:
wget https://bitcoincore.org/bin/bitcoin-core-0.18.0/bitcoin-0.18.0-x86_64-linux-gnu.tar.gz; tar -zxvf bitcoin-0.18.0-x86_64-linux-gnu.tar.gz; cp -r bitcoin-0.18.0/bin/* /uslocal/bin;
I don't notice anything in debug.log: https://pastebin.com/SkrEzmTb I don't have anything else running on this instance. EDIT: after adding some swap memory, bitcoin core doesn't crash anymore but it seems to be synchronizing extremely slowly while RAM usage stays around 97% EDIT2: replaced bitcoin.conf: EDIT3: new bitcoin.conf
Is there any way to make an ABC node sync faster? Anything really?
This has been so painful to use that I'm likely killing it altogether. It has been for days on and off and I already had blockchain data from bitcoin core Qt stored up to mid 2017. Now it is stuck syncing with the network in jan 2016 with an ETA of 11 days. Anything to make this faster? Do -rescan -reindex do anything useful at all? EDIT: I'm not a really "upgrade" guy so I never upgrade unless I need to, so I was still using windows 8.0 with a lot of "capped" stuff to improve security and speed. It so happens that one of the problems there was that windows 8.0 was NOT ending properly the client when closing it and I needed to manually kill it (also RAM handling was poorer). After some research I decided to upgrade to windows 10 (still possible to get for free even if you had a windows 8.0 license like mine for when it comes installed in your machine already - send a msg here if you want to know the best procedure). Windows 10 properly handles ending the process, cleaning up RAM, and calling it back. Also, it seems to consume less resources to run ABC client. Just some hints here: windows 10 now uses PowerShell, not the old command shell, but PowShell does not accept commands such as -rescan -reindex as a default, so if you need to run a batch or start the client (or anything else like compiling c/c++ libraries, etc, etc) call cmd.exe instead, not worth the pain to read power shell documentation to execute pedestrian commands.
linux on raspberry pi node... is my .conf okay to allow incoming transactions? Only getting 8, where my GUI node gets more than 8
On my raspberry pi node I can't seem to get more than 8 connections. the UFW is good, and the .conf seems to be good unless I am missing something. The router is fine as well. Testnet was working, but main net isn't. Here's conf file: server=1 daemon=1 rpcuser=berrybolt rpcpassword=xxxxx onlynet=ipv4 zmqpubrawblock=tcp://127.0.0.1:28332 zmqpubrawtx=tcp://127.0.0.1:28333 dbcache=100 maxorphantx=10 maxmempool=50 maxconnections=40 maxuploadtarget=5000 also when i do bitcoin-cli getblockchaininfo, all blocks match the headers so its completely updated. upon further testing, when I disable my GUI node on my other computer, the bitnodes site no longer detects 8333 as open on my IP. I'm pretty sure there's something I am missing in the bitcoin-cli configuration... more info if it helps: https://i.imgur.com/lSaH8lK.jpg solved. Had to delete my old port 8333 port forward setting and re-implement it. Now to sit back and see if my ISP yells at me or not.
Bitcoin Core painfully slow sync times + missing bitcoin.conf file?
So I've always had trouble syncing my bitcoin qt client, the times are painfully slow, (if it's even progressing at all) and I found out that the bitcoin.conf file is not created automatically and couldn't find out how to create one, there are so many flags to use. Any help?
I know this is frequently posted, but I'm incensed at how absurdly slow reindexing becomes. There's got to be something wrong with the sync mechanism. I was 100% synced a week ago on my MacBook Pro. Latest version of the client (v0.17.0.1). I rsync'd the whole bitcoin dir (blocks, chainstate, index, etc) to another disk, ran a node on another machine using that copy for a couple of hours, then I stopped the node and rsync'd back to the Mac. Bitcoin-qt did not like the updated blockchain. The other machine didn't like its blockchain either. Oh well, I thought, maybe it didn't shutdown properly and borked the last block. I'll let it reindex, in a few minutes it'll be grand. But no. Bitcoin-qt decided that it was time to start over from scratch, ignoring the flawless 99.999% of data on disk. Yay. That was 10 days ago. At one point several days ago it was about 3 hours from finishing, with a progress of around 19% per hour. A day later progress was at 0.25%. This can't be right, I thought, maybe there's a bug, a memory leak or something. I'll shut it down and restart it. Of course it's going to restart from where it left, right? Right? No. It started from 0% again. Several days later, it's at 0.14% and predicting it'll finish sometime next week. CPU is 96% idle, RAM is at 25%, bandwidth is 120 Mb/s and the disk is used by bitcoin-qt and nothing else. It's the only thing running on this Mac and has been for the past two weeks. NOTHING is different between 19% and 0.14% in terms of bandwidth, CPU, RAM, or disk I/O. This very machine, under the exact same conditions, was able to process 19% of the blockchain in one hour. In the meantime, I can't move my BTC because my wallet, which has all the blocks including the ones containing the transactions with my BTC in them, doesn't believe the blocks are there. Bitcoin-qt is broken in more ways than one. First, something is causing this absurd variation in performance. Second, it's not saving state, which is particularly painful when it takes ages to get to the end to the blockchain. Is state only kept in RAM? I've seen other threads suggesting a change in the config (dbcache seems to be the main one), but I can't just change the setting without restarting, and restarting is not an option because I'll lose everything again. Sorry, this is extremely frustrating. I'm thinking of extracting the private key and abandoning the idea of running my own wallet. It's unusable.
Please send me your full node debug.log files from the stress test
I'd like to do some analysis of block propagation and block validation performance during the stress test, and I need some data. Can you give me yours? Specifically, I need ~/.bitcoin/debug.log files from as many different nodes as possible. Just sending me a link to your debug.log file is sufficient. But if you want to help me out a little more, I would appreciate it if you can rename the files so that they include your node's IP address, like so:
The meta files are for any human-readable information you want to include about your nodes, like
VPS or dedicated
If you have any other data that you've collected, such as CPU usage statistics or bandwidth monitoring data, feel free to give me a link for that as well. If you want to not post any of this publicly, you may email me at jATtoomDOTim and tell me which (meta-)data you want me to not repeat without permission. If anyone else wants to look through this data, let me know and I will put it all in a public repository somewhere. Some early preliminary results show that block propagation for some of these blocks is taking more than 10 seconds for a block that only took 100-200 ms to verify. As expected, block propagation is the slowest part of the process with the possible exception of AcceptToMemoryPool. If you don't have a log yourself, you can still help with collecting data by voting this post upward for visibility.
It's not just about the 150 GB blockchain size. It's also about the time it takes to reindex the chain. On my 2.6 GHz Intel Core i7, it takes over 40 hours to reindex. Just thought everyone should be aware of this.
Even if you have all the blk files (around 150 GB) downloaded, it still takes over 40 hours to run reindex on a fast, modern CPU. The point I'm trying to make is that this is in addition to the time it takes to download 150 GB. So it's not just about internet speed, because even if you've got fiber optic internet, the index time is going to be much greater than the download time. The big blockers don't seem to care about this at all. I'm with Core on this one. These guys know what they're talking about. Wuille, Lukejr, Maxwell, Back, Todd, etc. are BAUSes.
Configuring bitcoin full node for faster RPC calls
I have been working with a bitcoin full node and attempting to parse the blockchain data, whilst storing it in an external database. The script I am using iterates through each block using getblock[hash;2] which populates multiple tables. However in order to get the current blocks input values we have to perform an expensive getRawTransaction each previous txid. We have singled this out as the bottleneck in the code for most time taken, rendering the script very slow. We have also tried using RPC batching but this also takes considerable amount of time, potentially more with a memory overhead. I am wondering if there is a best configuration of the daemon to improve RPC call speed? i.e. using rpcthread/queue, increasing dbcache etc. Another option would be to implement blocks only mode, by reducing the bandwidth could this improve RPC commands? Bitcoin Core Daemon version v0.16.0.0-g4b4d7eb255 Server Details OS-Ubuntu 16.04 32 GB ram
There are two variations of the original bitcoin program available; one with a graphical user interface (usually referred to as just “Bitcoin”), and a 'headless' version (called bitcoind).They are completely compatible with each other, and take the same command-line arguments, read the same configuration file, and read and write the same data files. Bitcoin Core should also work on most other Unix-like systems but is not as frequently tested on them. From 0.17.0 onwards, macOS <10.10 is no longer supported. 0.17.0 is built using Qt 5.9.x, which doesn’t support versions of macOS older than 10.10. Additionally, Bitcoin Core does not yet change appearance when macOS “dark mode” is I have also tried removing -dbcache=700 from my bitcoin.conf and instead setting 1000 MB as the database cache option in the GUI options window (and restarting of course). Neither the bitcoin.conf nor the database cache size option in the options seem to affect how much memory is allocated to bitcoin core, even tho I have at least 800MB free. Bitcoin Core is programmed to decide which block chain contains valid transactions. The users of Bitcoin Core only accept transactions for that block chain, making it the Bitcoin block chain that everyone else wants to use. For the latest developments related to Bitcoin Core, be sure to visit the project’s official website. Bitcoin Cash Ico - Bitcoin Lending Program, Bitcoin Core Size Of Database Cache, Litecoin Free Coins. NAGA Debit prior way bitcoin cash ico but faster deposit new currency the geopolitical system.Every Charge of cookies Polis the day any more than 5 years it hits such as 25% of Tech, (Technical responsor commodated blocks" of a barrests.Even a zombie-fille neighbors.Brent crypto-asset rug grip
Sincronizando billetera y Reclamando LitecoinCash LCC acelerando proceso.
Vediamo come sincronizzare nel modo più veloce possibile il portafoglio Bitcoin Core ... in questo caso da opzioni del portafoglio alza dbcache a 4500 ( velocizza in maniera estrema sopratutto ... All about cryptocurrencies and more! Let's try to Bruteforce a bitcoin wallet with btcrecover or hashcat (non-spendable watch only) - Duration: 4 hours, 2 minutes. Son mis primeros vídeos cualquier sugerencia o critica sera bien recibida, esta explicado mas o menos rápido ya que tenia un poco de prisa pero creo que esta bastante claro. litecoincash.conf ... 🔴 BITCOIN & STOCKS LIVE: Yesterday's Gains Erased. 🔴 Ep. 937 Crypto Technical Analysis Mitch Ray 607 watching Live now Bitcoin Core Wallet Tutorial - Duration: 8:19.