关于AC87U双电信100M光纤双拨后掉速严重的问题

garyjenny

设备清单：
      光猫1：华为8245 四口千兆  刷原厂固件 包括语音在内 正常使用  IP地址：192.168.1.1
      光猫2：华为8245 四口千兆  刷原厂固件                    正常使用  IP地址：192.168.2.1
      路由： 华硕RT-AC87U   固件：梅林380.57                               IP地址：10.0.1.1
      网线： 所有设备之间全部使用 包尔星克 成品 七类圆线
----------------------------------------------------------------------------------------------------------------------
     从加入第二条电信100M光纤之后，使用路由WAN口连接光猫1，进行PPPOE拨号，路由LAN4连接光猫2，PPPOE拨号
经http://www.speedtest.net测速，可以稳定在210M到217M之间的下行带宽，进行迅雷多线程下载，可以稳定在25M+的
下载速度。
     经过几天的使用期（期间未动过路由），突然有一天，发现测速的时候丢速严重，speedtest.net测速的时候，丢速很厉害
非常的不稳定，但是只要关掉其中一个光猫，就可以稳定测速，两个光纤，每一个单测都可以达到107左右的下行带宽，双光猫
只要一同时开，马上就会出现丢速，掉速的现象。
    经联系客服，重新刷了官方最新固件，刚刷完固件后，测速可以达到两条光纤的正常下载速度，能跑满两条下行带宽，但是
路由运行时间只要长一点（半小时到一小时左右），丢速，掉速现象又会出现。
   后来经过多次重刷固件（官方最新固件或是梅林的最新固件），发现一个很奇怪的现象，每次重刷后，可以稳定运行一小会儿，
只要路由运行时间稍长一点，老问题就会出现。
-----------------------------------------------------------------------------------------------------
单开光猫1的测速
   

-----------------------------------------------------------------------------------------------------
单开光猫2的测速



--------------------------------------------------------------------------------------------------
双拨截图



--------------------------------------------------------------------------------------------------------
双拨的设置图



-------------------------------------------------------------------------------------------------
双拨后进行测速



--------------------------------------------------------------------------
双拨后PING 百度



------------------------------------------------------------------------------------------------------------------------------------------------
以上是问题介绍，恳请有知道原因的朋友，或是有什么好解决方案的朋友告知一下。

固件问题？ AC87U硬件问题？

微信扫一扫，阅读更方便^_^

garyjenny

还是AC87U的吞吐量不行？

迪安

speedtest测速异常的时候，迅雷多线程下载的速度是多少？

garyjenny

迪安 发表于 2016-2-15 15:10
speedtest测速异常的时候，迅雷多线程下载的速度是多少？

我去测试一下 再来回复你

garyjenny

martianfly 发表于 2016-2-15 16:08
看看系统记录下的有没有报什么异常

Aug  1 08:00:39 rc_service: waitting "start_firewall" via  ...
Feb 15 17:53:48 rc_service: ntp 970:notify_rc restart_upnp
Feb 15 17:53:48 miniupnpd[965]: shutting down MiniUPnPd
Feb 15 17:53:48 miniupnpd[994]: HTTP listening on port 41554
Feb 15 17:53:48 miniupnpd[994]: Listening for NAT-PMP/PCP traffic on port 5351
Feb 15 17:53:49 zcip client: configured 169.254.198.208
Feb 15 17:53:49 dnsmasq[446]: read /etc/hosts - 5 addresses
Feb 15 17:53:49 dnsmasq[446]: read /etc/hosts.dnsmasq - 4 addresses
Feb 15 17:53:49 dnsmasq-dhcp[446]: read /etc/ethers - 6 addresses
Feb 15 17:53:49 dnsmasq[446]: using nameserver 114.114.114.114#53
Feb 15 17:53:49 dnsmasq[446]: using nameserver 8.8.8.8#53
Feb 15 17:53:49 dnsmasq[446]: ignoring nameserver 10.0.1.1 - local interface
Feb 15 17:53:49 dnsmasq[446]: ignoring nameserver 10.0.1.1 - local interface
Feb 15 17:53:49 rc_service: ntp 970:notify_rc restart_diskmon
Feb 15 17:53:49 rc_service: waitting "start_firewall" via zcip ...
Feb 15 17:53:49 start_nat_rules: apply the nat_rules(/tmp/nat_rules_0_ppp0_vlan2)!
Feb 15 17:53:50 disk_monitor: Finish
Feb 15 17:53:51 disk monitor: be idle
Feb 15 17:53:51 rc_service: ip-up 546:notify_rc start_pptpd
Feb 15 17:53:52 pptpd[1015]: MGR: Config file not found!
Feb 15 17:53:52 pptpd[1015]: MGR: Maximum of 100 connections reduced to 10, not enough IP addresses given
Feb 15 17:53:52 pptpd[1015]: accel-pptpd-0.8.5 compiled for pppd-2.4.5, linux-2.6.36.4
Feb 15 17:53:52 pptpd[1016]: MGR: Manager process started
Feb 15 17:53:52 pptpd[1016]: MGR: Maximum of 10 connections available
Feb 15 17:53:52 start_nat_rules: apply the nat_rules(/tmp/nat_rules_0_ppp0_vlan2)!
Feb 15 17:53:52 bcrelay[1017]: Running as child
Feb 15 17:53:52 pppd[522]: System time change detected.
Feb 15 17:53:57 pppd[527]: System time change detected.
Feb 15 17:53:58 dnsmasq-dhcp[446]: DHCPDISCOVER(br0) 70:48:0f:55:2c:31
Feb 15 17:53:58 dnsmasq-dhcp[446]: DHCPOFFER(br0) 10.0.1.116 70:48:0f:55:2c:31
Feb 15 17:53:58 dnsmasq-dhcp[446]: DHCPDISCOVER(br0) 70:48:0f:55:2c:31
Feb 15 17:53:58 dnsmasq-dhcp[446]: DHCPOFFER(br0) 10.0.1.116 70:48:0f:55:2c:31
Feb 15 17:53:59 dnsmasq-dhcp[446]: DHCPREQUEST(br0) 10.0.1.116 70:48:0f:55:2c:31
Feb 15 17:53:59 dnsmasq-dhcp[446]: DHCPACK(br0) 10.0.1.116 70:48:0f:55:2c:31 pengweine6sPlus
Feb 15 17:54:03 qtn: bootcfg.tgz exists
Feb 15 17:54:04 hour monitor: daemon is starting
Feb 15 17:54:08 smbd[1041]: [2016/02/15 17:54:08.622723,  0] smbd/sesssetup.c:1355(reply_sesssetup_and_X)
Feb 15 17:54:08 smbd[1041]:   reply_sesssetup_and_X:  Rejecting attempt at SPNEGO session setup when it was not negotiated.
Feb 15 17:54:08 smbd[1042]: [2016/02/15 17:54:08.628848,  0] smbd/sesssetup.c:1355(reply_sesssetup_and_X)
Feb 15 17:54:08 smbd[1042]:   reply_sesssetup_and_X:  Rejecting attempt at SPNEGO session setup when it was not negotiated.
Feb 15 17:54:09 dfs: start dfs scan
Feb 15 17:54:09 nodfs_scan: complete
Feb 15 17:54:09 scs: enable scs complete
Feb 15 17:54:10 crond[454]: time disparity of 285713 minutes detected
Feb 15 17:54:11 rc_service: httpd 452:notify_rc start_autodet
Feb 15 17:54:12 kernel: autodet uses obsolete (PF_INET,SOCK_PACKET)
Feb 15 17:54:12 pppoe-relay[517]: PADO packet from 7c:a2:3e:e2:6c:1b on interface vlan2 does not have Relay-Session-Id tag
Feb 15 17:54:12 pppoe-relay[517]: PADO packet from 7c:a2:3e:e2:6c:1b on interface vlan2 does not have Relay-Session-Id tag
Feb 15 17:54:14 pppoe-relay[517]: PADO packet from 28:6e:d4:f1:62:27 on interface vlan2 does not have Relay-Session-Id tag

这是我把两线从救援模式恢复到负载均衡模式后的记录  然后进行测速 不正常，但是记录里并未有新记录

garyjenny

迪安 发表于 2016-2-15 15:10
speedtest测速异常的时候，迅雷多线程下载的速度是多少？

刚进行了测速，有时候能跑到12M左右 有时候在5到6M左右 ，大多数时间是跑不满单光纤的速度

garyjenny

martianfly 发表于 2016-2-15 16:08
看看系统记录下的有没有报什么异常

Feb 15 17:59:12 rc_service: httpd 452:notify_rc restart_wan_if 1
Feb 15 17:59:12 pppd[527]: Hangup (SIGHUP)
Feb 15 17:59:12 pppd[527]: Connect time 5.6 minutes.
Feb 15 17:59:12 pppd[527]: Sent 3794423 bytes, received 107470762 bytes.
Feb 15 17:59:12 pppd[527]: Terminating on signal 15
Feb 15 17:59:12 pppd[527]: Connection terminated.
Feb 15 17:59:12 pppd[527]: Sent PADT
Feb 15 17:59:13 dnsmasq[446]: read /etc/hosts - 5 addresses
Feb 15 17:59:13 dnsmasq[446]: read /etc/hosts.dnsmasq - 4 addresses
Feb 15 17:59:13 dnsmasq-dhcp[446]: read /etc/ethers - 6 addresses
Feb 15 17:59:13 dnsmasq[446]: using nameserver 114.114.114.114#53
Feb 15 17:59:13 dnsmasq[446]: using nameserver 8.8.8.8#53
Feb 15 17:59:13 dnsmasq[446]: ignoring nameserver 10.0.1.1 - local interface
Feb 15 17:59:13 dnsmasq[446]: ignoring nameserver 10.0.1.1 - local interface
Feb 15 17:59:13 pppd[527]: Terminating on signal 15
Feb 15 17:59:13 pppd[527]: Exit.
Feb 15 17:59:13 wan: [wan1_hwaddr] == [AC:9E:17:93:12:E7]
Feb 15 17:59:13 pppd[1081]: Plugin rp-pppoe.so loaded.
Feb 15 17:59:13 pppd[1081]: RP-PPPoE plugin version 3.11 compiled against pppd 2.4.7
Feb 15 17:59:14 pppd[1084]: pppd 2.4.7 started by admin, uid 0
Feb 15 17:59:14 pppd[1084]: PPP session is 30296 (0x7658)
Feb 15 17:59:14 pppd[1084]: Connected to 7c:a2:3e:e2:6c:1b via interface vlan3
Feb 15 17:59:14 pppd[1084]: Using interface ppp1
Feb 15 17:59:14 pppd[1084]: Connect: ppp1 <--> vlan3
Feb 15 17:59:14 pppd[1084]: CHAP authentication succeeded: Authentication success,Welcome!
Feb 15 17:59:14 pppd[1084]: CHAP authentication succeeded
Feb 15 17:59:14 pppd[1084]: peer from calling number 7C:A2:3E:E2:6C:1B authorized
Feb 15 17:59:14 pppd[1084]: local  IP address 114.235.139.116
Feb 15 17:59:14 pppd[1084]: remote IP address 114.235.139.1
Feb 15 17:59:14 pppd[1084]: primary   DNS address 61.147.37.1
Feb 15 17:59:14 pppd[1084]: secondary DNS address 61.177.7.1
Feb 15 17:59:14 rc_service: ip-up 1089:notify_rc start_firewall
Feb 15 17:59:14 dnsmasq[446]: read /etc/hosts - 5 addresses
Feb 15 17:59:14 dnsmasq[446]: read /etc/hosts.dnsmasq - 4 addresses
Feb 15 17:59:14 dnsmasq-dhcp[446]: read /etc/ethers - 6 addresses
Feb 15 17:59:14 dnsmasq[446]: using nameserver 114.114.114.114#53
Feb 15 17:59:14 dnsmasq[446]: using nameserver 8.8.8.8#53
Feb 15 17:59:14 dnsmasq[446]: using nameserver 61.147.37.1#53
Feb 15 17:59:14 dnsmasq[446]: using nameserver 61.177.7.1#53
Feb 15 17:59:15 start_nat_rules: apply the nat_rules(/tmp/nat_rules_0_ppp0_vlan2)!
Feb 15 17:59:16 wan: finish adding multi routes
Feb 15 17:59:25 rc_service: zcip 1221:notify_rc start_firewall
Feb 15 17:59:25 dnsmasq[446]: read /etc/hosts - 5 addresses
Feb 15 17:59:25 dnsmasq[446]: read /etc/hosts.dnsmasq - 4 addresses
Feb 15 17:59:25 dnsmasq-dhcp[446]: read /etc/ethers - 6 addresses
Feb 15 17:59:25 dnsmasq[446]: using nameserver 114.114.114.114#53
Feb 15 17:59:25 dnsmasq[446]: using nameserver 8.8.8.8#53
Feb 15 17:59:25 dnsmasq[446]: using nameserver 61.147.37.1#53
Feb 15 17:59:25 dnsmasq[446]: using nameserver 61.177.7.1#53
Feb 15 17:59:25 zcip client: configured 169.254.198.208
Feb 15 17:59:25 start_nat_rules: apply the nat_rules(/tmp/nat_rules_0_ppp0_vlan2)!

这一段是我把线路2的DNS换回了自动获取

迪安

garyjenny 发表于 2016-2-15 18:11
刚进行了测速，有时候能跑到12M左右 有时候在5到6M左右 ，大多数时间是跑不满单光纤的速度 ...

um~，截张System Log--Routing Table图看一看呢

涵哥

每次都是过一阵子之后才会这样，所以涵哥想问一下
1.你单线的时候，会不会过一阵子也这样？
2，你测速有变化的时候，有没有换个测速网站试试，最好是电信的，结果如何？
3.测速掉速的时候，你的实际下载速度有多少？可以用BT下载工具试一试
4.不排除固件的问题，所以你可以换一个固件试一下，上一版也可以
5.不排除线路问题，你可以尝试使用其他路由器试一试，对比一下

pigpigcatcat

我觉得小型路由的负载均衡算法肯定是有点问题的，否则双线路的企业级路由如何有市场？能充分用好双线路，最好用规则好两条线路是哪台终端使用，用救援模式加终端规划走线路比较合理和稳定。楼主尝试一下，把BT类的终端走一条线，无线或者要求响应的走另外一条线，或许就能充分和稳定使用了。