Using 6 RTAC68U , same issue ...since 380.57

[autourdesam]

Hi all ,
I am using 6 different RT-AC68U (work, friends and personnal) since firmware 380.57 ( the first one I used as far as i remember) and one RTAC66U (still with 380.57) . They replaced Linksys E2000 with Tomato Shibby or DDWRT because of the better web interface design of MarlinWRT and the design of stats present in Traffic Analyzer
The RTAC66u has bug only one time in a year , needed to restart it manually . All the RT-AC68U have regulary bug like loosing web interface or worse loosing Internet/WAN connection . They are all used behind an ADSL box , connected on a lan port of the box and they are assigned as the DMZ Zone of these boxes.

Actually they used the fw 380.63_2 from MerlinWRT . Bugs arrived less often.... but still present .

- one bug is the 100% usage of a core . A remote reboot with the button interface lock defintively the router : it never restart except with a physica power off/on . The only way to have it remotly reboot is to have access through ssh ( sometimes you can even not log in... :-( ) and make an emergency reboot :

echo 1 > /proc/sys/kernel/sysrq
echo b > /proc/sysrq-trigger

A visible see of the problem can be the blue leds (wan+lan) blinking extremely quick on the router

- another bug is lost of the http interface/service (local or remote) , even if all other functions seems to work well (switching, access to internet,...)

I never had any stable connection issue with e2000 with TomatoShibby on same network configuration : only RTAC68U is new each time ...

All RTac68u were new hardware , flashed out of the box like that :
1- Power on and then automatic upgrade with the last asus firmware
2- Reset to hardware default
3- Upgrade of the actual merlinwrt fw release
4- configuration ( Stat act ON ,nat/pat, DHCP lease and fix attribution , Wifi SSID , VPN PPTP and OpenVPN server and accounts , remote admin with specific ports on https , etc ... )
5- for each new release of MerlinWRT --> Fw upgrade .


I absolutely need a permanent stable connection on router like I had with Tomato ( but hugly interface and poor stats ) .

Thanks for help to solve problems , or giving advice to improve stability .

Sam

 

[bbunge]

I use John's fork on two AC68U at a small business. Both are up for months at a time till I get around to doing the next firmware update. Both run a script to keep NVRAM cleaned out.

 

[autourdesam]

Can you tell me more about that ?

Both run a script to keep NVRAM cleaned out.

Which script ?

I must recognize that the ones that are less use have only few issues ... the more sollicated ones have more time those issues, particullary the ones with Wifi uses I think ...
P.S. : I forget one... it is 7 , not 6 ;-) ... but this last one is more stable and thinking of it , it is the one where wifi is rarely use . A second one with less wifi usage but hard cable usage meet those troubles every 1 to 2 months : but the most problem is that it is quite far for me , so I need someone in the place to have it reboot , so it is not very useful :-( and make a loss of service.
P.S.2 : please note that the RTac66u which is only one core processor is rarely wifi used

So, a possible wifi issue that over use one processor and make finally the router bug ?

 

[RMerlin]

Most stability issues I have seen were caused by using a USB disk and either sharing through SMB/FTP or running Download Master - both of which can easily fill up the available RAM. Make sure you disable all of these services, see if it helps.

 

[autourdesam]

Most stability issues I have seen were caused by using a USB disk and either sharing through SMB/FTP or running Download Master - both of which can easily fill up the available RAM. Make sure you disable all of these services, see if it helps.

Thanks for your answer and your interest for my case.
No router use USB Stick and I didn't use any AiCloud2.0 options or USB applications .
I only use Menus :
General : Network Map , Traffic Analyzer
Advanced : Wireless (General only on almost and only one with Professional Tx Power adjustement to reduce it) , Lan (all except IPTV), Wan , VPN (servers) , Firewall to enable ping respond, Administration (almost System menu), System Log , and Network Tools to use Wake on Lan function

I use different server/nas behind this router and some other services (webcam, etc ... )
If needed I can send you a backup of the configuration settings of some routers .
I didn't detect in the web interface(Network Map / RAM) more than 68/70MB of the 250MB (28% used only even after a few weeks !)
The web interface is always more slow then asking from the web than from the Lan and maybe sometimes web interfce doesn't answer anymore after multiple interrogation in an hour .
Some routers have few clients Lan->Wan some have few Wan-> Lan , some have lots from Wan to Lan (few servers but many NAT (10 to 20) )
No more than 2 or 3 active VPN clients to Router VPN servers.

I use French language interface .

Hope it helps in diagnostic ...
Thanks a lot

 

[bbunge]

You can read about John's fork here: http://www.snbforums.com/threads/fork-asuswrt-merlin-374-43-lts-releases-v20e9-v22ba.18914/

My routers were seeing a lot of computers come and go on the LAN and the older Asus firmware would not clear out NVRAM entries for computers that were no longer on the LAN. This would, in time, cause problems.

See this post for the fix: http://www.snbforums.com/threads/nmp_client_list-eating-up-all-nvram-space.20491/

Some of this problem was supposed to be fixed with the 380.XX firmware. John's fork works well for me and as the routers are 30 miles away I'm not inclined to make a radical change...yet.

 

[autourdesam]

@bbunge : thanks for these threads . In the second one , I can read :

This will give you a cron job that will run at 0100 every day and clean your routers NVRAM in three areas. My experience is that the values are rebuilt in NVRAM as SMB (Windows or Samba) devices become active and talk with each other and enabling devices to show up on AiCloud. Cleaning out the webdav_smb_pc= area may not be necessary for you and likely will mess up some things like traffic analysis. Your call to use this. As for me it is more important to keep the router running than to see who is surfing where.

For me it is important to have the "story of usage by who" for the last 30 days ... but I didn't mind to "reset" each day the actual list of clients in network map ....
All RTAC68U I use are with a 380.xx fw .... so do you think I need adding this script ?

 

[bbunge]

Keep an eye on the resources being used. NVRAM can build slowly and the newer builds should warn you with a flashing "!" on the main screen when it gets full. RAM usage goes up and down depending upon what processes are working. I have added a swap file on a USB with 256M size to my AC66U_B1 because I have a 640G USB drive attached and use Download Master.
Refreshing the Client Status list may clear out some NVRAM for you. If you are like me you do not want to daily connect to a remote router to check its status (I am a volunteer and do not get paid for most of what I do for churches and faith based organizations). The script works for me and should work for you. No harm in trying it! Best of luck!

 

[wocram]

I have a 256GB Flash Drive connected to the USB 2.0 port which is in use by Download Master (v 3.1.0.95) 24/7.

A 2TB WD HD is attached to the USB 3.0 port of the router for storage and playing through Media Server.

Samba and FTP are configured. I mostly use FTP as SMB is not very satisfying on my Macbook Pro.

Never had any strange behavior and uptime of router is mostly several weeks (reset mainly when updating the firmware and occasionally by village power outage).

My ADSL modem however is configured in Bridge Mode.

 

[autourdesam]

@wocram and the others who have a perfect stable connection :
- do you have enable TrafficAnalyzer>Statistic then click sur ON ???
- do you often manage your router(s) remotly from WAN interface ?
- do you have many clients from Lan interface (more than 40 simultaneously) ?
- do you have many clients from Wan interface (more than 40 simultaneously) ?

Thanks for your help for this diagnostic ... and have an happy new year ;-)

 

[wocram]

1). Yes, Traffic Analyzer is On.
2) Yes, I regularly manage my router over WAN.
3) No, Maximum is around 15 devices connected at the same time.
4) No, Only me personally connect to the router over WAN.

 

[autourdesam]

Hi ,
I actually have my personnal RTac68u with one core at 100% ! FW is 363_2 ... I have saved syslog and netstat ...
RMerlin what can I look/give you to help to diagnose the issue before I restart the router ?
Thanks a lot

 

[autourdesam]

Please note that :
Memory is 78MB Used , 172 MB free (so 31% used)
Core2 is 100% used
I can navigate in web interface from lan side
For SSH , I can give user : admin and password but there is no prompt to go further .
The router is started since 22 days 20 hours 4 minute(s) 47 seconds
and the general log show from last start

Aug 1 02:00:44 ntp: start NTP update
Aug 1 02:00:44 openvpn-routing: Refreshing policy rules for client 1
Dec 28 14:15:00 rc_service: ntp 778:notify_rc restart_upnp
Dec 28 14:15:00 rc_service: waitting "start_vpnserver1" via udhcpc ...
Dec 28 14:15:01 openvpn-routing: Allow WAN access to all VPN clients
Dec 28 14:15:01 openvpn-routing: Refreshing policy rules for client 2
Dec 28 14:15:01 openvpn-routing: Allow WAN access to all VPN clients
Dec 28 14:15:01 openvpn-routing: Refreshing policy rules for client 3
Dec 28 14:15:01 openvpn-routing: Allow WAN access to all VPN clients
Dec 28 14:15:01 openvpn-routing: Refreshing policy rules for client 4
Dec 28 14:15:01 openvpn-routing: Allow WAN access to all VPN clients
Dec 28 14:15:01 openvpn-routing: Refreshing policy rules for client 5
Dec 28 14:15:01 openvpn-routing: Allow WAN access to all VPN clients
Dec 28 14:15:01 rc_service: udhcpc 522:notify_rc stop_pptpd
Dec 28 14:15:01 rc_service: waitting "start_vpnserver1" via udhcpc ...
Dec 28 14:15:03 openvpn[915]: OpenVPN 2.3.12 arm-unknown-linux-gnu [SSL (OpenSSL)] [LZO] [EPOLL] [MH] [IPv6] built on Nov 11 2016
Dec 28 14:15:03 openvpn[915]: library versions: OpenSSL 1.0.2j 26 Sep 2016, LZO 2.08
Dec 28 14:15:03 openvpn[916]: WARNING: using --duplicate-cn and --client-config-dir together is probably not what you want
Dec 28 14:15:03 openvpn[916]: NOTE: your local LAN uses the extremely common subnet address 192.168.0.x or 192.168.1.x. Be aware that this might create routing conflicts if you connect to the VPN server from public locations such as internet cafes that use the same subnet.
Dec 28 14:15:03 openvpn[916]: Diffie-Hellman initialized with 2048 bit key
Dec 28 14:15:03 openvpn[916]: Socket Buffers: R=[122880->122880] S=[122880->122880]
Dec 28 14:15:03 openvpn[916]: TUN/TAP device tun21 opened
Dec 28 14:15:03 openvpn[916]: TUN/TAP TX queue length set to 100
Dec 28 14:15:03 openvpn[916]: do_ifconfig, tt->ipv6=0, tt->did_ifconfig_ipv6_setup=0
Dec 28 14:15:03 openvpn[916]: /usr/sbin/ip link set dev tun21 up mtu 1500
Dec 28 14:15:03 openvpn[916]: /usr/sbin/ip addr add dev tun21 10.8.0.1/24 broadcast 10.8.0.255
Dec 28 14:15:03 openvpn[916]: UDPv4 link local (bound): [undef]
Dec 28 14:15:03 openvpn[916]: UDPv4 link remote: [undef]
Dec 28 14:15:03 openvpn[916]: MULTI: multi_init called, r=256 v=256
Dec 28 14:15:03 openvpn[916]: IFCONFIG POOL: base=10.8.0.2 size=252, ipv6=0
Dec 28 14:15:03 openvpn[916]: Initialization Sequence Completed
Dec 28 14:15:03 miniupnpd[780]: shutting down MiniUPnPd
Dec 28 14:15:03 miniupnpd[779]: shutting down MiniUPnPd
Dec 28 14:15:03 miniupnpd[779]: Failed to remove pidfile /var/run/miniupnpd.pid: No such file or directory
Dec 28 14:15:04 miniupnpd[926]: HTTP listening on port 48454
Dec 28 14:15:04 miniupnpd[926]: Listening for NAT-PMP/PCP traffic on port 5351
Dec 28 14:15:04 rc_service: ntp 778:notify_rc restart_diskmon
Dec 28 14:15:04 disk_monitor: Finish
Dec 28 14:15:05 rc_service: udhcpc 522:notify_rc start_pptpd
Dec 28 14:15:05 rc_service: waitting "stop_pptpd" via udhcpc ...
Dec 28 14:15:06 disk monitor: be idle
Dec 28 14:15:07 pptpd[928]: MGR: Config file not found!
Dec 28 14:15:07 start_nat_rules: apply the nat_rules(/tmp/nat_rules_eth0_eth0)!
Dec 28 14:15:08 kernel: * Make sure sizeof(struct sw_struct)=160 is consistent
Dec 28 14:15:09 kernel: sizeof forward param = 160
Dec 28 14:15:12 rc_service: udhcpc 522:notify_rc start_firewall
Dec 28 14:15:12 dhcp client: bound 192.168.1.252 via 192.168.1.254 during 43200 seconds.
Dec 28 14:15:13 start_nat_rules: apply the nat_rules(/tmp/nat_rules_eth0_eth0)!
Dec 28 14:15:17 crond[438]: time disparity of 742394 minutes detected
Dec 28 14:16:18 rc_service: rc 1183:notify_rc restart_wrs
Dec 28 14:16:24 start_nat_rules: apply the nat_rules(/tmp/nat_rules_eth0_eth0)!
Dec 28 15:00:06 disk_monitor: Got SIGALRM...
Dec 28 23:00:06 disk_monitor: Got SIGALRM...
Jan 6 02:00:10 rc_service: rc 25624:notify_rc restart_wrs
Jan 6 02:00:14 start_nat_rules: apply the nat_rules(/tmp/nat_rules_eth0_eth0)!