$ sudo apt update && sudo apt upgrade -y
查看系统版本
$ lsb_release -a
No LSB modules are available.
Distributor ID: Ubuntu
Description: Ubuntu 20.04.3 LTS
Release: 20.04
Codename: focal
重启
sudo shutdown -r now
查看显卡信息，这里有两块GeForce RTX 2080 Ti显卡
$ lspci -vnn | grep VGA
12:00.0 VGA compatible controller [0300]: ASPEED Technology, Inc. ASPEED Graphics Family [1a03:2000] (rev 41) (prog-if 00 [VGA controller])
21:00.0 VGA compatible controller [0300]: NVIDIA Corporation TU102 [GeForce RTX 2080 Ti Rev. A] [10de:1e07] (rev a1) (prog-if 00 [VGA controller])
61:00.0 VGA compatible controller [0300]: NVIDIA Corporation TU102 [GeForce RTX 2080 Ti Rev. A] [10de:1e07] (rev a1) (prog-if 00 [VGA controller])
打开NVIDIA Driver Downloads，选择对应显卡的驱动程序进行下载，这里下载的事目前最新的版本470.63.01
$ mkdir ~/dist && cd ~/dist
$ wget <https://cn.download.nvidia.com/XFree86/Linux-x86_64/470.63.01/NVIDIA-Linux-x86_64-470.63.01.run>
$ ls -hl NVIDIA-Linux-x86_64-470.63.01.run
-rw-r--r-- 1 sol sol 259M Sep 3 10:10 NVIDIA-Linux-x86_64-470.63.01.run
安转依赖包
sudo apt install build-essential libglvnd-dev pkg-config -y
禁用默认的Nouveau驱动
$ lsmod | grep nouveau
nouveau 1986560 0
mxm_wmi 16384 1 nouveau
i2c_algo_bit 16384 2 ast,nouveau
drm_ttm_helper 16384 2 drm_vram_helper,nouveau
ttm 73728 3 drm_vram_helper,drm_ttm_helper,nouveau
drm_kms_helper 237568 5 drm_vram_helper,ast,nouveau
video 49152 1 nouveau
wmi 32768 2 mxm_wmi,nouveau
drm 548864 7 drm_kms_helper,drm_vram_helper,ast,drm_ttm_helper,ttm,nouveau
$ sudo vim /etc/modprobe.d/blacklist-nvidia-nouveau.conf
blacklist nouveau
options nouveau modeset=0
$ sudo update-initramfs -u
$ sudo shutdown -r now
安装Nvidia驱动，一直确认即可
sudo bash ~/dist/NVIDIA-Linux-x86_64-470.63.01.run
通过nvidia-smi指令查看驱动信息
$ nvidia-smi
Fri Sep 3 11:13:32 2021
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 470.63.01 Driver Version: 470.63.01 CUDA Version: 11.4 |
|-------------------------------+----------------------+----------------------+
| GPU Name Persistence-M| Bus-Id Disp.A | Volatile Uncorr. ECC |
| Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute M. |
| | | MIG M. |
|===============================+======================+======================|
| 0 NVIDIA GeForce ... Off | 00000000:21:00.0 Off | N/A |
| 0% 34C P0 61W / 260W | 0MiB / 11019MiB | 1% Default |
| | | N/A |
+-------------------------------+----------------------+----------------------+
| 1 NVIDIA GeForce ... Off | 00000000:61:00.0 Off | N/A |
| 0% 31C P0 51W / 280W | 0MiB / 11019MiB | 0% Default |
| | | N/A |
+-------------------------------+----------------------+----------------------+
+-----------------------------------------------------------------------------+
| Processes: |
| GPU GI CI PID Type Process name GPU Memory |
| ID ID Usage |
|=============================================================================|
| No running processes found |
+-----------------------------------------------------------------------------+
如果想查看GPU使用率，还可以安装nvtop指令
$ sudo apt install nvtop -y
$ nvtop
如果要使用GPU则必须要安装CUDA(并行计算框架)，打开CUDA Toolkit 11.4 Update 1 Downloads页面根据自己的版本进行安装，如果你用的不是这个版本，其他历史版本可以在CUDA Toolkit Archive处找到。
$ cd ~/dist/
$ wget <https://developer.download.nvidia.com/compute/cuda/11.4.1/local_installers/cuda_11.4.1_470.57.02_linux.run>
$ sudo sh cuda_11.4.1_470.57.02_linux.run
===========
= Summary =
===========
Driver: Installed
Toolkit: Installed in /usr/local/cuda-11.4/
Samples: Installed in /home/sol/, but missing recommended libraries
Please make sure that
- PATH includes /usr/local/cuda-11.4/bin
- LD_LIBRARY_PATH includes /usr/local/cuda-11.4/lib64, or, add /usr/local/cuda-11.4/lib64 to /etc/ld.so.conf and run ldconfig as root
To uninstall the CUDA Toolkit, run cuda-uninstaller in /usr/local/cuda-11.4/bin
To uninstall the NVIDIA Driver, run nvidia-uninstall
Logfile is /var/log/cuda-installer.log
安装完成之后我们需要配置一些环境变量
$ vim ~/.bashrc
......
export CUDA_HOME=/usr/local/cuda-11.4
export PATH=/usr/local/cuda-11.4/bin:$PATH
export LD_LIBRARY_PATH=/usr/local/cuda-11.4/lib64:$LD_LIBRARY_PATH
最好重启一下系统吧，如果不重启就加载一下环境变量
source ~/.bashrc
参考文章从源码构建Solana(索拉纳)。
如果你使用的是有GPU的版本，在编译之前需要生成solana-perf-libs相关的文件
$ git clone -b v1.7.10 <https://github.com/solana-labs/solana.git> ~/dist/solana-src-v1.7.10
$ git clone <https://github.com/solana-labs/solana-perf-libs.git> ~/dist/solana-perf-libs
$ cd ~/dist/solana-perf-libs
$ make -j$(nproc)
把文件copy到Solana仓库中，然后继续编译安装即可
make DESTDIR=~/dist/solana-src-v1.7.10/target/perf-libs/cuda-11.4 install
首先我们需要连接到Devnet群集
solana config set --url <http://api.devnet.solana.com>
确保可以访问到Devnet群集，国内网络环境可能会经常出现超时等情况，可以多尝试几次
$ solana transaction-count
1603626215
查看集群中的其他节点，ctrl+c可以退出
solana-gossip spy --entrypoint devnet.solana.com:8001
增加UDP缓冲区
sudo bash -c "cat >/etc/sysctl.d/20-solana-udp-buffers.conf <<EOF
# Increase UDP buffer size
net.core.rmem_default = 134217728
net.core.rmem_max = 134217728
net.core.wmem_default = 134217728
net.core.wmem_max = 134217728
EOF"
sudo sysctl -p /etc/sysctl.d/20-solana-udp-buffers.conf
增加内存映射文件限制
sudo bash -c "cat >/etc/sysctl.d/20-solana-mmaps.conf <<EOF
# Increase memory mapped files limit
vm.max_map_count = 1000000
EOF"
sudo sysctl -p /etc/sysctl.d/20-solana-mmaps.conf
编辑/etc/systemd/system.conf文件[Service]段增加下面的内容
LimitNOFILE=1000000
[Manager]段增加下面的内容
DefaultLimitNOFILE=1000000
重新加载
sudo systemctl daemon-reload
增加文件描述符数量
sudo bash -c "cat >/etc/security/limits.d/90-solana-nofiles.conf <<EOF
# Increase process file descriptor count limit
* - nofile 1000000
EOF"
退出终端重新登陆，查看limit限制
$ ulimit -n
1000000
这里创建的是fs钱包
mkdir -p ~/deploy/sol
solana-keygen new -o ~/deploy/sol/validator-keypair.json
solana config set --keypair ~/deploy/sol/validator-keypair.json
每次最多只能领取10个SOL，但是可以重复领取
solana airdrop 10
检查余额
solana balance
在Solana上运行验证器节点，需要创建一个投票帐户
solana-keygen new -o ~/deploy/sol/vote-account-keypair.json
solana create-vote-account ~/deploy/sol/vote-account-keypair.json ~/deploy/sol/validator-keypair.json
我们创建一个systemd服务来运行Solana，首先需要创建一个运行脚本
$ mkdir ~/deploy/sol/scripts -p
$ vim ~/deploy/sol/scripts/validator.sh
#!/bin/bash
solana-validator \\
--identity ~/deploy/sol/validator-keypair.json \\
--vote-account ~/deploy/sol/vote-account-keypair.json \\
--ledger ~/deploy/sol/validator-ledger \\
--rpc-port 8899 \\
--entrypoint devnet.solana.com:8001 \\
--limit-ledger-size \\
--cuda \\
--log ~/deploy/sol/solana-validator.log
$ chmod +x ~/deploy/sol/scripts/validator.sh
创建/etc/systemd/system/sol.service文件
$ sudo vim /etc/systemd/system/sol.service
[Unit]
Description=Solana Validator
After=network.target
Wants=solana-sys-tuner.service
StartLimitIntervalSec=0
[Service]
Type=simple
Restart=always
RestartSec=1
User=sol
LimitNOFILE=1000000
LogRateLimitIntervalSec=0
Environment="PATH=/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/home/sol/.local/share/solana/install/active_release/bin:/usr/local/cuda-11.4/bin"
Environment="CUDA_HOME=/usr/local/cuda-11.4"
Environment="LD_LIBRARY_PATH=/usr/local/cuda-11.4/lib64:$LD_LIBRARY_PATH"
#Environment="SOLANA_METRICS_CONFIG=host=https://metrics.solana.com:8086,db=devnet,u=scratch_writer,p=topsecret"
ExecStart=/home/sol/deploy/sol/scripts/validator.sh
[Install]
WantedBy=multi-user.target
加载systemd文件
sudo systemctl daemon-reload
运行sol并设置开机启动
sudo systemctl enable --now sol
查看运行状态
sudo systemctl status sol
查看运行日志
tail -f ~/deploy/sol/solana-validator.log
默认情况下日志文件会越来越大，所以我们需要通过logrotate进行日志切割
$ sudo vim /etc/logrotate.d/sol
/home/sol/deploy/sol/solana-validator.log {
rotate 7
daily
missingok
postrotate
systemctl kill -s USR1 sol.service
endscript
}
$ sudo systemctl restart logrotate
我们可以把validator信息发布到链上，这样其他用户也可以查看到我们的一些信息
$ solana validator-info publish --keypair ~/deploy/sol/validator-keypair.json "Dark Devnet Validator"
Publishing info for Validator 6QpnVQNT1EhPFnCseSwSXbfZWhXniduMRVvEDxnJAhg6
Success! Validator info published at: 88dkzpAQVue79AhCpvg5r8fnnt5rgXzrRUyYQMN2yt75
2KuewbB9FnGDszwqV4ifnhntGwzmLA4YzQvFWgzkPsMNe3zpn1VsZUqyxQRiBbYupQ2LNFja3xCwopZGm5kH8PEE
查看节点信息
$ solana validator-info get
# 这里返回的是一个列表，需要过滤找一下自己对应节点的ID
Validator Identity: 6QpnVQNT1EhPFnCseSwSXbfZWhXniduMRVvEDxnJAhg6
Info Address: 88dkzpAQVue79AhCpvg5r8fnnt5rgXzrRUyYQMN2yt75
Name: Dark Devnet Validator
等待一段时间后，可以检查下是否追上网络的其他节点，此过程可能需要很久，取决于你的网络情况
solana catchup ~/deploy/sol/validator-keypair.json
查看日志文件是否有错误信息
grep 'error|warn' ~/deploy/sol/solana-validator.log
验证我的节点是否已经在集群上了
solana-gossip spy --entrypoint entrypoint.devnet.solana.com:8001
# 过滤88dkzpAQVue79AhCpvg5r8fnnt5rgXzrRUyYQMN2yt75
监控节点信息，Ctrl+C可以退出
solana-validator --ledger ~/deploy/sol/validator-ledger monitor
显示我的节点的块生产和跳过的插槽
solana block-production | grep $(solana address)
将我当前时代的领导者日程表导出到文本文件：
solana leader-schedule | grep $(solana address) \\
> ~/leader-schedule-epoch-$(solana epoch)-$(solana address).txt
查看epoch相关信息
$ solana epoch-info
Block height: 77586723
Slot: 78709787
Epoch: 182
Transaction Count: 1623421128
Epoch Slot Range: [78624000..79056000)
Epoch Completed Percent: 19.858%
Epoch Completed Slots: 85787/432000 (346213 remaining)
Epoch Completed Time: 9h 31m 29s/1day 23h 53m 47s (1day 14h 22m 18s remaining)
查看所有的validators
solana validators