From 64e2cf069b5dc415d9497652503d46f4b2d1a2c5 Mon Sep 17 00:00:00 2001
From: gaolifeng <sklxz@qq.com>
Date: Fri, 14 Oct 2022 12:25:31 +0800
Subject: [PATCH] =?UTF-8?q?=E5=8A=A0=E5=85=A5truenasseeker=E7=AB=A0?=
 =?UTF-8?q?=E8=8A=82=E5=92=8C=E5=90=AF=E5=8A=A8=E9=A1=BA=E5=BA=8F=E8=B0=83?=
 =?UTF-8?q?=E6=95=B4=E7=AB=A0=E8=8A=82?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

---
 README.md                                     |   3 +-
 ...truenasseeker.py脚本自动扫描nfs服务并挂载.md |  40 ++++++
 .../设置docker-compose等待nfs挂载后再启动.md  |   2 +
 files/truenasseeker.py                        | 120 ++++++++++++++++++
 4 files changed, 164 insertions(+), 1 deletion(-)
 create mode 100644 details/使用truenasseeker.py脚本自动扫描nfs服务并挂载.md
 create mode 100644 files/truenasseeker.py

diff --git a/README.md b/README.md
index 8edbbd0..36116c8 100644
--- a/README.md
+++ b/README.md
@@ -333,7 +333,8 @@ TrueNas：
 服务主机ubuntu：
 
 - [使用setup_aquar.sh脚本进行环境初始化](./details/使用setup_aquar.sh脚本进行环境初始化)
-- 设置docker-compose等待nfs挂载后再启动
+- [设置docker-compose等待nfs挂载后再启动](./details/设置docker-compose等待nfs挂载后再启动.md)
+- [使用truenasseeker.py脚本自动扫描nfs服务并挂载](./details/使用truenasseeker.py脚本自动扫描nfs服务并挂载.md)
 
 Windows虚拟机：
 
diff --git a/details/使用truenasseeker.py脚本自动扫描nfs服务并挂载.md b/details/使用truenasseeker.py脚本自动扫描nfs服务并挂载.md
new file mode 100644
index 0000000..214722f
--- /dev/null
+++ b/details/使用truenasseeker.py脚本自动扫描nfs服务并挂载.md
@@ -0,0 +1,40 @@
+如果你遇到了搬家或者更换路由器的情况，你网络中所有的设备都会失去原有的IP地址，这会导致Aquar中各个子系统间原本建立起来的关联会全部失效。
+
+truenasseeker.py脚本就是用来解决这个问题的。它在每次开机时会检查系统是否有效地挂载了`/opt/aquar/storages/aquarpool`目录，如果这个目录没有正常挂载，则会载局域网中扫描truenas主机，并修改ubuntu的nfs配置为最最新地址，然后重新挂载这个这个地址。
+
+如果你已经部署了truenasseeker.py但在网络变动后发现docker应用中的存储池数据仍然不见了，你可以在ubuntu控制台上以超级用户执行一次`aqserv restart`命令,或者干脆重启一次机器，再看存储池内容是否恢复了。
+
+1.`sudo -i`切换超级用户。
+
+2.安装依赖的软件包。
+
+```shell
+apt install -y nmap
+pip install scapy python-nmap pyfunctional
+```
+
+3.将truenasseeker.py脚本拷贝至`/opt/aquar/src`路径下。
+
+你可以尝试执行`python3 /opt/aquar/src/truenasseeker.py`命令运行一次看是否会出错，正常情况下它会输出"truenas seeker is no need to do anything."
+
+4.创建`/usr/lib/systemd/system/trunas-scan.service`配置文件,配置内容如下：
+
+```
+# [Unit]
+# Description=scan local network for truenas
+# After=docker.service opt-aquar-storages-aquarpool.mount
+# 
+# [Service]
+# Type=simple
+# User=root
+# ExecStart=python3 /opt/aquar/src/truenasseeker.py
+# 
+# [Install]
+# WantedBy=multi-user.target
+```
+
+5.执行`systemctl daemon-reload`重载系统服务。
+
+6.执行`systemctl enable trunas-scan.service`将脚本设置为开机启动。
+
+7.执行`systemctl status trunas-scan.service`查看服务状态是否为enabled。如果是，则代表脚本正常部署了。
\ No newline at end of file
diff --git a/details/设置docker-compose等待nfs挂载后再启动.md b/details/设置docker-compose等待nfs挂载后再启动.md
index 795d5ee..3fd1398 100644
--- a/details/设置docker-compose等待nfs挂载后再启动.md
+++ b/details/设置docker-compose等待nfs挂载后再启动.md
@@ -1,5 +1,7 @@
 设置docker-compose等待nfs挂载后再启动
 
+默认情况下docker-compose中的服务和nfs挂载动作的执行是不区分先后顺序的，这会导致开机以后docker服务已经启动但nfs挂载点未就绪，于是在docker应用中就无法正常的看到存储池中的文件了。所以这一步是必不可少的。
+
 1.执行`sudo -i`进入管理员角色
 
 2.执行`systemctl list-units | grep mount`寻找与fstab中配置对应的信息
diff --git a/files/truenasseeker.py b/files/truenasseeker.py
new file mode 100644
index 0000000..4c6c7a6
--- /dev/null
+++ b/files/truenasseeker.py
@@ -0,0 +1,120 @@
+#! /usr/bin/env python
+# vim: set fenc=utf8 ts=4 sw=4 et :
+# apt install -y nmap
+# pip install scapy python-nmap pyfunctional
+# -------/usr/lib/systemd/system/trunas-scan.service--------
+# [Unit]
+# Description=scan local network for truenas
+# After=docker.service opt-aquar-storages-aquarpool.mount
+# 
+# [Service]
+# Type=simple
+# User=root
+# ExecStart=python3 /opt/aquar/src/truenasseeker.py
+# 
+# [Install]
+# WantedBy=multi-user.target
+
+
+from __future__ import absolute_import, division, print_function
+import logging
+import scapy.config
+import scapy.layers.l2
+import scapy.route
+import math
+import os
+import getopt
+import sys
+import shutil
+import re
+import nmap
+from functional import seq
+
+logging.basicConfig(format='%(asctime)s %(levelname)-5s %(message)s', datefmt='%Y-%m-%d %H:%M:%S', level=logging.DEBUG)
+logger = logging.getLogger(__name__)
+nm = nmap.PortScanner()
+
+def long2net(arg):
+    if (arg <= 0 or arg >= 0xFFFFFFFF):
+        raise ValueError("illegal netmask value", hex(arg))
+    return 32 - int(round(math.log(0xFFFFFFFF - arg, 2)))
+
+
+def to_CIDR_notation(bytes_network, bytes_netmask):
+    network = scapy.utils.ltoa(bytes_network)
+    netmask = long2net(bytes_netmask)
+    net = "%s/%s" % (network, netmask)
+    if netmask < 16:
+        logger.warning("%s is too big. skipping" % net)
+        return None
+
+    return net
+
+
+def scan():
+    if os.geteuid() != 0:
+        print('You need to be root to run this script', file=sys.stderr)
+        sys.exit(1)
+
+    for network, netmask, _, interface, address, _ in scapy.config.conf.route.routes:
+
+        # skip loopback network and default gw
+        if network == 0 or interface == 'lo' or address == '127.0.0.1' or address == '0.0.0.0':
+            continue
+
+        if netmask <= 0 or netmask == 0xFFFFFFFF:
+            continue
+
+        # skip docker interface
+        if interface.startswith('docker') or interface.startswith('br-') or interface.startswith('tun'):
+            logger.warning("Skipping interface '%s'" % interface)
+            continue
+
+        net = to_CIDR_notation(network, netmask)
+
+        nfsAddress = None
+        if net:
+            ipInfo = nm.scan(hosts=net, arguments='-p 22,80,111,139,443,445,8006,8008,8081,8384')
+            resList = seq(list(ipInfo['scan'].items())).map(lambda i: (i[0],seq(list(i[1]['tcp'].items())).filter(lambda p: p[1]['state'] == "open").map(lambda pi: pi[0] )))
+            for res in resList:
+                logger.info("%s: %s" % (res[0],res[1]))
+                if 22 not in res[1] and 80 in res[1] and 111 in res[1] and 139 in res[1] and 443 in res[1] and 445 in res[1]:
+                    nfsAddress = res[0]
+
+        return nfsAddress
+
+def checknfs():
+    return os.path.ismount('/opt/aquar/storages/aquarpool')
+
+
+def updatefstab(nfsAdress):
+    shutil.copy('/etc/fstab', '/etc/fstab.bak')
+    fstabFile = open("/etc/fstab", "r+")
+
+    configText = fstabFile.read()
+    prepart = re.split("##\[aquar config start\]##", configText)[0]
+    postpart = re.split("##\[aquar config end\]##", configText)[1]
+    nfsConfig = "\n##[aquar config start]##\n %s:/mnt/aquarpool /opt/aquar/storages/aquarpool nfs defaults,_netdev 0 0\n##[aquar config end]##\n" % nfsAdress
+    logger.info("new nfs config:")
+    logger.info(nfsConfig)
+    newConifg = prepart + nfsConfig+ postpart
+    logger.info("fstab now is:")
+    logger.info(newConifg)
+
+    fstabFile.seek(0)
+    fstabFile.write(newConifg)
+    fstabFile.truncate()
+    fstabFile.close()
+
+
+if __name__ == "__main__":
+    print('truenas seeker starting.')
+    if not checknfs(): 
+        print('truenas seeker find nfs directory is broken, trying to find truenas in local network.')
+        nfsAddress = scan()
+        updatefstab(nfsAddress)
+        print('truenas seeker updated fstab succesfully.')
+        os.system("mount -a")
+        print('truenas seeker has remounted fstab.')
+    else:
+        print('truenas seeker is no need to do anything.')
\ No newline at end of file