[RT-AC3200] Router crashes on occasion - is it good enough for fiber?

[semmelbroesel]

Hi.

I have two questions in one.

My router will occasionally crash - meaning the WiFi signal is still up, but the internet connection is gone, both wired and wireless. The cable modem is fine.

Restarting the router typically fixes it - sometimes I have to do it multiple times, sometimes I have to resort to the usual spiel of "power off on modem and router, plugin modem after 30 sec, plugin router after 30 sec" that I grew up with in my tech support days.

I have about 40 devices on my network - four 24/7 streaming security cams, many phones and tablets, video streaming, some online gaming, and I do the occasional big file download/upload.

I'm currently on the official Asus firmware 3.0.0.4.380_3479 - I haven't updated in a while.

So first question - what causes a router to drop internet?

The System Log doesn't look very helpful - there are no entries for a month until my latest restart a couple of days ago:

Aug 21 10:34:29 rc_service: ntp 726:notify_rc restart_upnp
Aug 21 10:34:29 miniupnpd[727]: shutting down MiniUPnPd
Aug 21 10:34:29 miniupnpd[737]: version 1.9 started
Aug 21 10:34:29 miniupnpd[737]: HTTP listening on port 39155
Aug 21 10:34:29 miniupnpd[737]: Listening for NAT-PMP/PCP traffic on port 5351
Aug 21 10:34:29 kernel: * Make sure sizeof(struct sw_struct)=160 is consistent
Aug 21 10:34:30 rc_service: ntp 726:notify_rc restart_diskmon
Aug 21 10:34:30 disk_monitor: Finish
Aug 21 10:34:30 kernel: sizeof forward param = 160
Aug 21 10:34:32 disk monitor: be idle
Aug 21 10:34:35 rc_service: udhcpc 548:notify_rc start_firewall
Aug 21 10:34:35 dhcp client: bound xxxxx via xxxxxx during 282533 seconds.
Aug 21 10:34:35 start_nat_rules: apply the nat_rules(/tmp/nat_rules_eth0_eth0)!
Aug 21 10:34:36 rc_service: udhcpc 919:notify_rc start_firewall
Aug 21 10:34:36 rc_service: waitting "start_firewall" via  ...
Aug 21 10:34:38 WAN Connection: ISP's DHCP did not function properly.
Aug 21 10:34:38 DualWAN: skip single wan wan_led_control - WANRED off
Aug 21 10:34:38 start_nat_rules: apply the nat_rules(/tmp/nat_rules_eth0_eth0)!
Aug 21 10:34:38 wan: finish adding multi routes
Aug 21 10:34:38 stop_nat_rules: apply the redirect_rules!
Aug 21 10:34:38 rc_service: udhcpc 919:notify_rc stop_upnp
Aug 21 10:34:38 rc_service: waitting "start_firewall" via udhcpc ...
Aug 21 10:34:39 rc_service: udhcpc 919:notify_rc start_upnp
Aug 21 10:34:39 rc_service: waitting "stop_upnp" via udhcpc ...
Aug 21 10:34:39 miniupnpd[737]: shutting down MiniUPnPd
Aug 21 10:34:41 miniupnpd[1136]: version 1.9 started
Aug 21 10:34:41 miniupnpd[1136]: HTTP listening on port 55242
Aug 21 10:34:41 miniupnpd[1136]: Listening for NAT-PMP/PCP traffic on port 5351
Aug 21 10:34:41 miniupnpd[1136]: upnp_event_recv: recv(): Connection reset by peer
Aug 21 10:34:44 WAN Connection: WAN was restored.
Aug 21 10:34:44 start_nat_rules: apply the nat_rules(/tmp/nat_rules_eth0_eth0)!
Aug 21 10:34:45 rc_service: udhcpc 919:notify_rc start_firewall
Aug 21 10:34:45 dhcp client: bound xxxxxx via xxxxxx during 282517 seconds.
Aug 21 10:34:46 start_nat_rules: apply the nat_rules(/tmp/nat_rules_eth0_eth0)!
Aug 21 10:34:59 crond[479]: time disparity of 2660614 minutes detected
Aug 21 10:36:19 hour monitor: ntp sync fail, will retry after 120 sec
Aug 21 11:00:32 disk_monitor: Got SIGALRM...
Aug 21 23:00:32 disk_monitor: Got SIGALRM...

I'm also wondering what the SIGALRM for disk_monitor means...

Do I have too many devices with too much traffic on that router? Are there any devices that I should not keep on the network?

Right now, I am using only a few percent of CPU and about 30% RAM.

The second question is: Given that it crashes, is this router good enough for 1Gbps fiber internet?

I'm thinking of upgrading to the RT-AC5300 router, but first I'd like to know your thoughts.

Thanks in advance, and let me know if you need more information.

 

[ColinTaylor]

The disk_monitor messages are normal and can be ignored.

Unless there's something in the log prior to you rebooting it it's difficult to guess what the problem might be. It's possible the power adapter is failing.

Given that your firmware is over three years old I suggest that you install the current version and follow that with a factory reset. Then configure your router again manually (do not reload an old saved settings file). That might be enough to stabilise things.

I wouldn't regard the RT-AC5300 as a particularly significant upgrade over what you already have. They were both released about 5 years ago and are based on very similar hardware. If I were buying a new router I think I would investigate the RT-AC86U or the RT-AX... models.

 

[semmelbroesel]

Thanks for the reply.

Since I already had a major internet outage going on (cable provider had issues), I figured I'd tackle this, so now I'm on the newest firmware, and I made screen shots of all the settings, reset the router to factory settings, and manually entered all settings again.

Let's hope that does it.

Also thanks for the thoughts on the router upgrade.

Would you consider my router as adequate for 1Gbps fiber internet, or would it be worth upgrading to a newer router?

 

[ColinTaylor]

Would you consider my router as adequate for 1Gbps fiber internet, or would it be worth upgrading to a newer router?

Yes it should be adequate unless you want to use a router feature that disables its hardware acceleration (i.e. QoS or VPN client).

 

[Ronald Schwerer]

During your dropped internet access: if the LAN side works (wired & wireless) AND the Modem lights indicate it is WAN connected, it makes me think the router is having trouble re-acquiring a lease on a WAN IP. This smells like an ISP issue. I'm sure someone knows the ssh commands or a way to force the router to release and re-acquire the WAN IP.

AFA 1Gb, it should be fine. Even my old 68U (OC'd to 1200MHz) with Merlin & running Diversion and Skynet comes close (a little over 900Mbps) to getting my theoretical max (around 940Mbps) FIOS speed. Oddly, my UL speed is always about 10% faster than DL. I just got the fiber service last week was wondering if I needed a more powerful router (RAM and CPU) and suspect I'd do better with an 86U. But for the time being, I happy.

 

[semmelbroesel]

Colin, so you're saying QoS slows the router down? I currently have it enabled, but I'm game to disable it if it slows me down...

Ronald, if you (or anyone else) happen to come across a command to release the WAN IP, please let me know, I'm more than willing to try that. The theory that it is my cable provider's fault sound very convincing (and based on that, I have a suspicion many will know which provider I speak of . . . .)

Also thanks for confirming that my router is fast enough for Gb

And in general: A big thanks for your responses!

 

[ColinTaylor]

Yes using the VPN client on the router will restrict your maximum internet speed because a) the router can't use its hardware acceleration and therefore will be limited by its CPU, and b) the VPN encryption requires significant processing power putting a further load on the CPU.

That is why VPN users would want something like the RT-AC86U which has encryption built into the hardware. Even so they won't get gigabit speeds through a VPN.

EDIT: Sorry I misread your post and thought you said VPN, not QoS.

 

[semmelbroesel]

I'm not using VPN, only QoS is turned on.

The description sounds as if it could be useful in limiting unimportant traffic to get important stuff through, but then if I have high bandwidth, this probably won't make a difference, and if the router is fast enough to handle all my traffic, again the feature is probably useless.

Does that sound about right?

 

[ColinTaylor]

Sorry I misread your post and thought you said VPN, not QoS.

But yes using QoS may restrict your speeds. I know using Traditional QoS will severely limit it. I think that Adaptive QoS is much better but I don't know how fast it can go.

IMHO it's unlikely you would need to use QoS with a gigabit connection.

 

[semmelbroesel]

OK, that helps Thanks again so much!

And if someone comes across an ssh command to reset the WAN IP, I'm still interested :-D

 

[ColinTaylor]

And if someone comes across an ssh command to reset the WAN IP, I'm still interested :-D

The first thing I would check when the problem happens is log into the router and on the Network Map page click on the "internet globe". Look at the Internet status panel and make a note of the information, then toggle the Internet Connection button off and on.

 

[semmelbroesel]

I wrote that down (along with my current settings) and will try it next time. Hopefully, there won't be a next time after the firmware upgrade and factory reset.

Thanks!

You guys are wonderful!

 

[semmelbroesel]

Hi. It just happened again.

I ran the troubleshooting you gave me - basically turning off the connection to the modem, waited a while, turned it back on - no luck.

The screen on the right (Network Map - click Globe) still showed all the correct information, e.g. WAN IP and my external DNS servers (OpenDNS, I think, it's been a while).

From the router web interface, I was able to ping my WAN IP, and I was able to ping the Gateway IP, but NOT my DNS server or anything with a domain name. Doesn't that imply that everything on my end is actually good and the issue is with my provider?

I checked their network status, and they didn't see a problem. Their automated troubleshooting restarted my modem remotely, and that still didn't fix it.

Then a few minutes later, all of a sudden, internet connection was back.

Does this also sound like a provider issue to you?

Any other troubleshooting steps I could do?

Also, is there a button to turn the internet off (as in Network Map - Globe) on the ASUS mobile app? I couldn't find it.

Thanks!

 

[ColinTaylor]

This does sound more like an ISP issues to be honest. Can you log into your modem and see its log?

 

[semmelbroesel]

I don't know if I can log into the modem... To be honest, I wouldn't know how to through the router.

Here's the router log just in case.

Around 9:30, we went offline, and I've messed around with the mobile app and then the web interface right after that. This is on the newest firmware and after a factory reset and manual restore of settings.

Aug 30 07:22:56 miniupnpd[1415]: upnp_event_recv: recv(): Connection reset by peer
Aug 30 09:30:03 rc_service: httpd 281:notify_rc start_webs_update
Aug 30 09:32:23 rc_service: httpd 281:notify_rc restart_wan_if 0
Aug 30 09:32:31 rc_service: httpd 281:notify_rc restart_wan_if 0
Aug 30 09:32:34 rc_service: udhcpc 31076:notify_rc start_firewall
Aug 30 09:32:35 miniupnpd[1415]: shutting down MiniUPnPd
Aug 30 09:32:35 wan: finish adding multi routes
Aug 30 09:32:35 rc_service: udhcpc 31076:notify_rc stop_upnp
Aug 30 09:32:35 rc_service: waitting "start_firewall" via udhcpc ...
Aug 30 09:32:35 nat: apply nat rules (/tmp/nat_rules_eth0_eth0)
Aug 30 09:32:36 miniupnpd[31123]: version 1.9 started
Aug 30 09:32:36 miniupnpd[31123]: HTTP listening on port 50652
Aug 30 09:32:36 miniupnpd[31123]: Listening for NAT-PMP/PCP traffic on port 5351
Aug 30 09:32:36 rc_service: udhcpc 31076:notify_rc start_upnp
Aug 30 09:32:36 rc_service: waitting "stop_upnp" via udhcpc ...
Aug 30 09:32:36 miniupnpd[31123]: shutting down MiniUPnPd
Aug 30 09:32:37 miniupnpd[31127]: version 1.9 started
Aug 30 09:32:38 miniupnpd[31127]: HTTP listening on port 54163
Aug 30 09:32:38 miniupnpd[31127]: Listening for NAT-PMP/PCP traffic on port 5351
Aug 30 09:32:38 miniupnpd[31127]: upnp_event_recv: recv(): Connection reset by peer
Aug 30 09:32:48 rc_service: httpd 281:notify_rc restart_wan_if 0
Aug 30 09:32:54 BWDPI: fun bitmap = 83
Aug 30 09:32:55 A.QoS: qos_count=0, qos_check=0
Aug 30 09:32:58 rc_service: udhcpc 31076:notify_rc start_firewall
Aug 30 09:32:58 dhcp client: bound xxxxxx.67/255.255.252.0 via xxxxxx.1 for 313571 seconds.
Aug 30 09:32:59 miniupnpd[31127]: shutting down MiniUPnPd
Aug 30 09:32:59 nat: apply nat rules (/tmp/nat_rules_eth0_eth0)
Aug 30 09:33:00 miniupnpd[32065]: version 1.9 started
Aug 30 09:33:00 miniupnpd[32065]: HTTP listening on port 34983
Aug 30 09:33:00 miniupnpd[32065]: Listening for NAT-PMP/PCP traffic on port 5351
Aug 30 09:33:03 WAN Connection: WAN was exceptionally disconnected.
Aug 30 09:33:03 DualWAN: skip single wan wan_led_control - WANRED off
Aug 30 09:33:04 nat: apply redirect rules
Aug 30 09:33:06 rc_service: httpd 281:notify_rc restart_wan_if 0
Aug 30 09:33:09 rc_service: udhcpc 1168:notify_rc start_firewall
Aug 30 09:33:10 miniupnpd[32065]: add_filter_rule() : chain FUPNP not found
Aug 30 09:33:10 miniupnpd[32065]: Failed to redirect 3074 -> 192.168.0.237:3074 protocol UDP
Aug 30 09:33:10 miniupnpd[32065]: add_filter_rule() : chain FUPNP not found
Aug 30 09:33:10 miniupnpd[32065]: Failed to redirect 9091 -> 192.168.0.97:80 protocol TCP
Aug 30 09:33:10 miniupnpd[32065]: add_filter_rule() : chain FUPNP not found
Aug 30 09:33:10 miniupnpd[32065]: Failed to redirect 9444 -> 192.168.0.97:443 protocol TCP
Aug 30 09:33:10 miniupnpd[32065]: shutting down MiniUPnPd
Aug 30 09:33:10 wan: finish adding multi routes
Aug 30 09:33:10 nat: apply nat rules (/tmp/nat_rules_eth0_eth0)
Aug 30 09:33:10 rc_service: udhcpc 1168:notify_rc stop_upnp
Aug 30 09:33:10 rc_service: waitting "start_firewall" via udhcpc ...
Aug 30 09:33:10 miniupnpd[1296]: version 1.9 started
Aug 30 09:33:11 miniupnpd[1296]: HTTP listening on port 49426
Aug 30 09:33:11 miniupnpd[1296]: Listening for NAT-PMP/PCP traffic on port 5351
Aug 30 09:33:11 rc_service: udhcpc 1168:notify_rc start_upnp
Aug 30 09:33:11 rc_service: waitting "stop_upnp" via udhcpc ...
Aug 30 09:33:11 miniupnpd[1296]: shutting down MiniUPnPd
Aug 30 09:33:12 miniupnpd[1419]: version 1.9 started
Aug 30 09:33:13 miniupnpd[1419]: HTTP listening on port 41152
Aug 30 09:33:13 miniupnpd[1419]: Listening for NAT-PMP/PCP traffic on port 5351
Aug 30 09:33:25 WAN Connection: WAN was restored.
Aug 30 09:33:32 BWDPI: fun bitmap = 83
Aug 30 09:33:32 A.QoS: qos_count=0, qos_check=0
Aug 30 09:33:35 A.QoS: qos rule is less than 22
Aug 30 09:33:35 A.QoS: restart A.QoS because set_qos_conf / set_qos_on / setup rule fail
Aug 30 09:33:36 A.QoS: qos_count=0, qos_check=1
Aug 30 09:33:40 A.QoS: qos rule is less than 22
Aug 30 09:33:40 A.QoS: restart A.QoS because set_qos_conf / set_qos_on / setup rule fail
Aug 30 09:33:41 A.QoS: qos_count=1, qos_check=1
Aug 30 09:33:45 A.QoS: qos rule is less than 22
Aug 30 09:33:45 A.QoS: restart A.QoS because set_qos_conf / set_qos_on / setup rule fail
Aug 30 09:33:46 A.QoS: qos_count=2, qos_check=1
Aug 30 09:33:49 A.QoS: qos rule is less than 22
Aug 30 09:33:49 A.QoS: restart A.QoS because set_qos_conf / set_qos_on / setup rule fail
Aug 30 09:33:49 rc_service: udhcpc 1168:notify_rc start_firewall
Aug 30 09:33:50 dhcp client: bound xxxxx.67/255.255.252.0 via xxxxx.1 for 313536 seconds.
Aug 30 09:33:50 miniupnpd[1419]: shutting down MiniUPnPd
Aug 30 09:33:51 nat: apply nat rules (/tmp/nat_rules_eth0_eth0)
Aug 30 09:33:51 miniupnpd[3926]: version 1.9 started
Aug 30 09:33:51 miniupnpd[3926]: HTTP listening on port 48086
Aug 30 09:33:51 miniupnpd[3926]: Listening for NAT-PMP/PCP traffic on port 5351

(There's more log data after this, but the forum won't let me post that much text.)

My modem is called:

MOTO JRSB6121

I'll try to find out of I can remote into it.

 

[semmelbroesel]

Ha, I got in! Not sure how useful this will be.

Aug 30 2020 08:37:59    5-Warning    Z00.0    MIMO Event MIMO: Stored MIMO=-1 post cfg file MIMO=-1;CM-MAC=94:cc:b9:af:56:21;CMTS-MAC=00:76:86:45:06:b4;CM-QOS=1.1;CM-VER=3.0;
Jan 01 1970 00:00:16    6-Notice    N/A    Cable Modem Reboot due to IP loss ;CM-MAC=94:cc:b9:af:56:21;CMTS-MAC=00:00:00:00:00:00;CM-QOS=1.1;CM-VER=3.0;
Aug 24 2020 13:15:00    3-Critical    T05.0    SYNC Timing Synchronization failure - Loss of Sync;CM-MAC=94:cc:b9:af:56:21;CMTS-MAC=00:76:86:45:06:b4;CM-QOS=1.1;CM-VER=3.0;
Aug 23 2020 11:34:55    3-Critical    R05.0    Started Unicast Maintenance Ranging - No Response received - T3 time-out;CM-MAC=94:cc:b9:af:56:21;CMTS-MAC=00:76:86:45:06:b4;CM-QOS=1.1;CM-VER=3.0;
Aug 23 2020 11:34:50    5-Warning    Z00.0    MIMO Event MIMO: Stored MIMO=-1 post cfg file MIMO=-1;CM-MAC=94:cc:b9:af:56:21;CMTS-MAC=00:76:86:45:06:b4;CM-QOS=1.1;CM-VER=3.0;
Jan 01 1970 00:34:00    3-Critical    R02.0    No Ranging Response received - T3 time-out;CM-MAC=94:cc:b9:af:56:21;CMTS-MAC=00:76:86:45:06:b4;CM-QOS=1.1;CM-VER=3.0;
Jan 01 1970 00:00:16    6-Notice    N/A    Cable Modem Reboot due to power reset ;CM-MAC=94:cc:b9:af:56:21;CMTS-MAC=00:00:00:00:00:00;CM-QOS=1.1;CM-VER=3.0;
Jan 01 1970 00:02:53    3-Critical    R01.0    No Maintenance Broadcasts for Ranging opportunities received - T2 time-out;CM-MAC=94:cc:b9:af:56:21;CMTS-MAC=00:76:86:45:06:b4;CM-QOS=1.1;CM-VER=3.0;
Jan 01 1970 00:02:45    5-Warning    T202.0    Lost MDD Timeout;CM-MAC=94:cc:b9:af:56:21;CMTS-MAC=00:76:86:45:06:b4;CM-QOS=1.1;CM-VER=3.0;
Jan 01 1970 00:02:43    3-Critical    R01.0    No Maintenance Broadcasts for Ranging opportunities received - T2 time-out;CM-MAC=94:cc:b9:af:56:21;CMTS-MAC=00:76:86:45:06:b4;CM-QOS=1.1;CM-VER=3.0;
Jan 01 1970 00:02:24    3-Critical    T05.0    SYNC Timing Synchronization failure - Loss of Sync;CM-MAC=94:cc:b9:af:56:21;CMTS-MAC=00:76:86:45:06:b4;CM-QOS=1.1;CM-VER=3.0;
Jan 01 1970 00:02:22    3-Critical    R02.0    No Ranging Response received - T3 time-out;CM-MAC=94:cc:b9:af:56:21;CMTS-MAC=00:76:86:45:06:b4;CM-QOS=1.1;CM-VER=3.0;
Jan 01 1970 00:01:41    3-Critical    R04.0    Received Response to Broadcast Maintenance Request, But no Unicast Maintenance opportunities received - T4 time out;CM-MAC=94:cc:b9:af:56:21;CMTS-MAC=00:76:86:45:06:b4;CM-QOS=1.1;CM-VER=3.0;
Jan 01 1970 00:01:36    5-Warning    T202.0    Lost MDD Timeout;CM-MAC=94:cc:b9:af:56:21;CMTS-MAC=00:76:86:45:06:b4;CM-QOS=1.1;CM-VER=3.0;
Jan 01 1970 00:01:12    3-Critical    T05.0    SYNC Timing Synchronization failure - Loss of Sync;CM-MAC=94:cc:b9:af:56:21;CMTS-MAC=00:76:86:45:06:b4;CM-QOS=1.1;CM-VER=3.0;
Jan 01 1970 00:01:09    3-Critical    R02.0    No Ranging Response received - T3 time-out;CM-MAC=94:cc:b9:af:56:21;CMTS-MAC=00:76:86:45:06:b4;CM-QOS=1.1;CM-VER=3.0;
Jan 01 1970 00:00:16    6-Notice    N/A    Cable Modem Reboot due to power reset ;CM-MAC=94:cc:b9:af:56:21;CMTS-MAC=00:00:00:00:00:00;CM-QOS=1.1;CM-VER=3.0;
Aug 23 2020 10:28:00    3-Critical    T05.0    SYNC Timing Synchronization failure - Loss of Sync;CM-MAC=94:cc:b9:af:56:21;CMTS-MAC=00:76:86:45:06:b4;CM-QOS=1.1;CM-VER=3.0;
Aug 23 2020 10:27:50    6-Notice    N/A    MDD Recovery following MDD Loss;CM-MAC=94:cc:b9:af:56:21;CMTS-MAC=00:76:86:45:06:b4;CM-QOS=1.1;CM-VER=3.0;
Aug 23 2020 10:27:13    5-Warning    T202.0    Lost MDD Timeout;CM-MAC=94:cc:b9:af:56:21;CMTS-MAC=00:76:86:45:06:b4;CM-QOS=1.1;CM-VER=3.0;

 

[Ronald Schwerer]

The screen on the right (Network Map - click Globe) still showed all the correct information, e.g. WAN IP and my external DNS servers (OpenDNS, I think, it's been a while).
.....
From the router web interface, I was able to ping my WAN IP, and I was able to ping the Gateway IP, but NOT my DNS server or anything with a domain name. Doesn't that imply that everything on my end is actually good and the issue is with my provider?

I checked their network status, and they didn't see a problem. Their automated troubleshooting restarted my modem remotely, and that still didn't fix it.

Then a few minutes later, all of a sudden, internet connection was back.

OpenDNS had a system-wide outage for several hours this morning. It was reported in EU and US. Symptoms were - everything looks good in the router and modem, but no external sites reachable. Fix was to change to another DNS. I first went with my ISP's DNS. When that fixed the problem I changed over to Quad9. I guess it would be a good idea to split the DNS1 and DNS2 IPs to different services to prevent total confusion next time.

Probably a trial run of the planned DDOS coming in November. I guess I need my own caching DNS server.

 

[semmelbroesel]

Ah, that outage could be it... So I guess I'll set my secondary DNS to my provider's.

... except my internet only went out for 10 minutes or so this time. We were streaming TV until minutes before it went out and again immediately after it came back.

Am I correct in assuming that the two DNS servers I enter do not have to be the same DNS provider, i.e. the first could go to OpenDNS and the second to my provider?

 

[Ronald Schwerer]

That seems to be the case. Seems logical-it is just a service running at a specific IP address. I never really saw the need to the backup until today.

 

[semmelbroesel]

Sweet, thanks so much!