[Fork] Asuswrt-Merlin 374.43 LTS releases (V44EA)

ColinTaylor

Part of the Furniture
So the error rate is 0.07 %. I have no idea of what is "normal" in any case it's quite low IMHO.
Yes that would be the error rate to the ONT. I'm not that familiar with PPPoE setups but I'd guess that you could also do ifconfig ppp0 to get the stats for that link. (Or do ifconfig -a and look at all the interfaces. ;))

But for your actual question about errors on the LAN connections; thinking about this a bit more I suspect that because br0 isn't a physical interface (it's a bridge) that it will never report any errors. At least I've never seen any. To find the actual interface name look at the output of brctl show. I suspect it will be called vlan1. So you could then look for errors with ifconfig vlan1. But again, it's a virtual interface not a physical one so I doubt it will report anything useful.

So I think the only way to detect errors would be to send traffic down the links and look for lost packets. There aren't really any good tools to do this built into the firmware. Does the RT-N16U have Entware installed? That would allow you to install some more useful tools.
 

Lord Lovaduck

Regular Contributor
Thanks again, Mr Taylor!
Yes, for sure I have entware installed so I can install some testing tools if needed.
So as per your advice, I did run brctl show and ifconfig -a (output below). Only eth0 shows errors. Now, since the switch has its own brains, I am probably missing a lot of the action since on switch control I have NAT acceleration set to YES which would hide some information. Will change and try again all commands.
One thing I noticed (and this is another can of worms I am opening) is that only one interface showing 1480 as the MTU size is ppp0 which worries me about fragmentation. I don't know if that is relevant or if the code adjust the MTU somehow to optimal values.
Best regards from Patagonia....
Code:
[email protected]:/tmp/home/root# brctl show
bridge name     bridge id               STP enabled     interfaces
br0             8000.bcaec5c4f56e       yes             vlan1
                                                        eth1
[email protected]:/tmp/home/root# ifconfig -a
br0       Link encap:Ethernet  HWaddr BC:AE:C5:C4:F5:6E
          inet addr:192.168.17.1  Bcast:192.168.17.255  Mask:255.255.255.0
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
          RX packets:1810479 errors:0 dropped:0 overruns:0 frame:0
          TX packets:796865 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:0
          RX bytes:214421560 (204.4 MiB)  TX bytes:159211414 (151.8 MiB)

eth0      Link encap:Ethernet  HWaddr BC:AE:C5:C4:F5:6E
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
          RX packets:98418575 errors:0 dropped:0 overruns:0 frame:0
          TX packets:98294978 errors:60732 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:1000
          RX bytes:2128880689 (1.9 GiB)  TX bytes:1238959623 (1.1 GiB)
          Interrupt:4 Base address:0x2000

eth1      Link encap:Ethernet  HWaddr BC:AE:C5:C4:F5:6E
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
          RX packets:1116351 errors:0 dropped:0 overruns:0 frame:25480306
          TX packets:1478280 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:1000
          RX bytes:254617002 (242.8 MiB)  TX bytes:618284488 (589.6 MiB)
          Interrupt:3 Base address:0x1000

lo        Link encap:Local Loopback
          inet addr:127.0.0.1  Mask:255.0.0.0
          UP LOOPBACK RUNNING MULTICAST  MTU:16436  Metric:1
          RX packets:1106587 errors:0 dropped:0 overruns:0 frame:0
          TX packets:1106587 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:0
          RX bytes:253690106 (241.9 MiB)  TX bytes:253690106 (241.9 MiB)

ppp0      Link encap:Point-to-Point Protocol
          inet addr:10.81.38.208  P-t-P:10.81.36.1  Mask:255.255.255.255
          UP POINTOPOINT RUNNING MULTICAST  MTU:1480  Metric:1
          RX packets:68995752 errors:0 dropped:0 overruns:0 frame:0
          TX packets:30198494 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:3
          RX bytes:2479782702 (2.3 GiB)  TX bytes:3397336627 (3.1 GiB)

vlan1     Link encap:Ethernet  HWaddr BC:AE:C5:C4:F5:6E
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
          RX packets:1252600 errors:0 dropped:0 overruns:0 frame:0
          TX packets:896030 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:0
          RX bytes:125199011 (119.3 MiB)  TX bytes:167142649 (159.3 MiB)

[email protected]:/tmp/home/root#
Yes that would be the error rate to the ONT. I'm not that familiar with PPPoE setups but I'd guess that you could also do ifconfig ppp0 to get the stats for that link. (Or do ifconfig -a and look at all the interfaces. ;))

But for your actual question about errors on the LAN connections; thinking about this a bit more I suspect that because br0 isn't a physical interface (it's a bridge) that it will never report any errors. At least I've never seen any. To find the actual interface name look at the output of brctl show. I suspect it will be called vlan1. So you could then look for errors with ifconfig vlan1. But again, it's a virtual interface not a physical one so I doubt it will report anything useful.

So I think the only way to detect errors would be to send traffic down the links and look for lost packets. There aren't really any good tools to do this built into the firmware. Does the RT-N16U have Entware installed? That would allow you to install some more useful tools.
 

ColinTaylor

Part of the Furniture
Yes, for sure I have entware installed so I can install some testing tools if needed.
So as per your advice, I did run brctl show and ifconfig -a (output below). Only eth0 shows errors. Now, since the switch has its own brains, I am probably missing a lot of the action since on switch control I have NAT acceleration set to YES which would hide some information. Will change and try again all commands.
One thing I noticed (and this is another can of worms I am opening) is that only one interface showing 1480 as the MTU size is ppp0 which worries me about fragmentation. I don't know if that is relevant or if the code adjust the MTU somehow to optimal values.
I don't think there's anything to worry about on the WAN side. The ppp0 interface is being tunnelled through eth0 so it needs to have a smaller MTU to allow for the tunnel's packet overhead. Clients use PMTUD to discover the maximum packet size to use to avoid fragmentation.

For Entware you could install a better version of ping and then flood ping the IP addresses of each of your APs. Leave it running for a while and then Ctrl-C to break out of it and see how many lost packets you have lost.
Code:
# opkg install iputils-ping
Code:
# ping -f 192.168.1.10
PING 192.168.1.10 (192.168.1.10) 56(84) bytes of data.
.^
--- 192.168.1.10 ping statistics ---
168034 packets transmitted, 168034 received, 0% packet loss, time 18934ms
rtt min/avg/max/mdev = 0.046/0.072/6.747/0.046 ms, ipg/ewma 0.112/0.077 ms
Remember to wiggle the cables around at each end to make sure there aren't any intermittent connections. ;)
 

rreiner

New Around Here
(long-time Merlin and Merlin-LTS user, first time poster :) )

The current issue is with the latest LTS release (374.43_44EAj9527) on an RT-AC66U (original MIPS device, not the B1).

The problem is that regardless of the setting of LAN / DHCP Server / Enable multicast DNS (Avahi mDNS), the mDNS service does not start.

I know the MIPS routers use mDNSResponder rather than Avahi, and I see the mDNS* binaries under /usr/sbin.

But the service never runs (ps w | grep -i mdns shows no processes).

I also see syslog entries like these:

616:Jul 24 12:36:04 (none) user.notice rc_service: service 612:notify_rc start_mdns
617:Jul 24 12:36:04 (none) user.notice rc_service: INFO: rc notified of unrecognized or custom event 'mdns'


Am I missing something, or is this a bug?

Thanks,
R.
 

john9527

Part of the Furniture
Am I missing something, or is this a bug?
Well, the current bug is that I should not show that option on the MIPs routers.

On the MIPs routers, there is no separate mDNS service for mDNSresponder. It's tied to the Media Server and only starts/stops with the Media Server (always been that way).
Do you think you have a need to run it independently?
 

rreiner

New Around Here
Yes.. I don't need the media server, but I'd like to have an mDNS reflector (to copy mDNS multicast frames across network segments).
 

Mythos

New Around Here
Greetings,
I have the RT-N66U ASUS Wireless Router with, Asuswrt-Merlin firmware "RT-N66U_380.70_0.trx" installed. I find it to be very stable, so no complaints.

I am curious though... does the latest release of this Merlin fork have any benefits in direct comparison, or should I skip it if I don't have any problems with my current firmware? Thanks guys.

Best regards,
// Mythos
 

ColinTaylor

Part of the Furniture
I am curious though... does the latest release of this Merlin fork have any benefits in direct comparison, or should I skip it if I don't have any problems with my current firmware?
This fork contains many bug fixes and enhancements over 380.70, but if you're not currently effected by any of those the main advantage would be the security fixes. So if you're using the N66U as your internet gateway device I would strongly recommend migrating to this firmware.

EDIT: Just realised I gave you the same answer back in December.
 

Mythos

New Around Here
This fork contains many bug fixes and enhancements over 380.70, but if you're not currently effected by any of those the main advantage would be the security fixes. So if you're using the N66U as your internet gateway device I would strongly recommend migrating to this firmware.

EDIT: Just realised I gave you the same answer back in December.
Yes i've been on the fence, worried about potential drawbacks migrating to this fork as everything here has been working great for a long time. Much obliged.
 

Raul_77

New Around Here
Hey guys, I need some help sorting out a problem.
So I use PiHole for my DNS and I have ENABLED DNS Filtering, so ALL DNS inquiries are sent to PiHole IP (since some of my devices are hard-coded to use Google DNS) anyway, the issue I am trying to resolve is find a way to have backup in case PiHole is down.
I was thinking of this:
In LAN settings, DNS 1 = PiholeIP / DNS 2 = Router IP
In WAN Settings, DNS 1 = PoholeIP /DNS 2 = 8.8.8.8
The idea is, IF my Pihole is down, my device tries DNS 2, then router tries PiHole, and since that is down, it then tries Google

My question is :
A) is the above logic correct? would it work?
B) if so, I need to add the Router MAC address to the DNS Filtering list to NO Filtering, How do I get the Mac Address?

Appreciate your time.
 

ColinTaylor

Part of the Furniture
Hey guys, I need some help sorting out a problem.
So I use PiHole for my DNS and I have ENABLED DNS Filtering, so ALL DNS inquiries are sent to PiHole IP (since some of my devices are hard-coded to use Google DNS) anyway, the issue I am trying to resolve is find a way to have backup in case PiHole is down.
I was thinking of this:
In LAN settings, DNS 1 = PiholeIP / DNS 2 = Router IP
In WAN Settings, DNS 1 = PoholeIP /DNS 2 = 8.8.8.8
The idea is, IF my Pihole is down, my device tries DNS 2, then router tries PiHole, and since that is down, it then tries Google

My question is :
A) is the above logic correct? would it work?
B) if so, I need to add the Router MAC address to the DNS Filtering list to NO Filtering, How do I get the Mac Address?

Appreciate your time.
There was a similar discussion here. So that would be:

WAN Settings:
DNS 1 = 8.8.8.8
DNS 2 = empty

LAN - DHCP settings:
DNS 1 = PiholeIP
DNS 2 = empty
Advertise router's IP in addition to user-specified DNS = Yes (or set to No and put the router's address in LAN DNS 2)

DNSFilter:
Enable DNS-based Filtering = On
Global filter mode = Custom #1
Custom DNS #1 = router's IP address (e.g. 192.168.1.1)
Client list: Pihole = No Filtering (Only necessary if you don't want the PiHole to forward through the router's DNS server.)
 

dave14305

Part of the Furniture
Hey guys, I need some help sorting out a problem.
So I use PiHole for my DNS and I have ENABLED DNS Filtering, so ALL DNS inquiries are sent to PiHole IP (since some of my devices are hard-coded to use Google DNS) anyway, the issue I am trying to resolve is find a way to have backup in case PiHole is down.
I was thinking of this:
In LAN settings, DNS 1 = PiholeIP / DNS 2 = Router IP
In WAN Settings, DNS 1 = PoholeIP /DNS 2 = 8.8.8.8
The idea is, IF my Pihole is down, my device tries DNS 2, then router tries PiHole, and since that is down, it then tries Google

My question is :
A) is the above logic correct? would it work?
B) if so, I need to add the Router MAC address to the DNS Filtering list to NO Filtering, How do I get the Mac Address?

Appreciate your time.
I think it could work if you add strict-order to /jffs/configs/dnsmasq.conf.add so that the router always prefers dns1 if available. But you have an equal chance of Pi-Hole getting queries from the real client IP or the router IP for any client’s query, so any special client features in use on pi-hole would be negated half the time.
 

Lord Lovaduck

Regular Contributor
Hi back! I did what you suggested Colin, thanks.
I got a 5% packet drop to one of the targets using ping. Not sure why such a high number of packets were dropped; could be in part cable issues in part that the cheap repeater/access point I am using not being able to respond quick enough to the flood of pings. Will keep checking on this, isolating the components to see which one is the main culprit. Got the tools now, it's a simple test setup that works. Thanks a lot for your help!

I don't think there's anything to worry about on the WAN side. The ppp0 interface is being tunnelled through eth0 so it needs to have a smaller MTU to allow for the tunnel's packet overhead. Clients use PMTUD to discover the maximum packet size to use to avoid fragmentation.

For Entware you could install a better version of ping and then flood ping the IP addresses of each of your APs. Leave it running for a while and then Ctrl-C to break out of it and see how many lost packets you have lost.
Code:
# opkg install iputils-ping
Code:
# ping -f 192.168.1.10
PING 192.168.1.10 (192.168.1.10) 56(84) bytes of data.
.^
--- 192.168.1.10 ping statistics ---
168034 packets transmitted, 168034 received, 0% packet loss, time 18934ms
rtt min/avg/max/mdev = 0.046/0.072/6.747/0.046 ms, ipg/ewma 0.112/0.077 ms
Remember to wiggle the cables around at each end to make sure there aren't any intermittent connections. ;)
 

KGB7

Very Senior Member
@john9527

Just switched from Merlin 384.16 to Fork 44EA. Router RT-AC68. Couple of things I would like to share with you.

1) Hotspot connected via USB port, detection is still a bit buggy. Router sees the Hotspot Brand and make, but wont get internet connection, even though it gets an IP. Disconnecting the hotspot from USB port, Router wont release the WAN IP that was assigned by a hotspot.
Had to reboot the router a couple of times. That didn't solve the issue. Power cycled the hotspot as well.
Switched settings from USB Hotspot to Android, that didn't work either. Rebooted the router few more times and it started working.
Hotspot detection and communication with the router is still spotty, but its an improvement from few versions ago.

2) Took 3 tries to install Diversion. Installation process kept hanging at random stages. So I installed Skynet and then Diversion installation worked flawlessly. Not sure if it was coincidence or just luck. This wasn't an issue in the past.

3) Since im using Hotspot connected via a USB port on the router, I have WAN set to USB (in Dual-WAN Tab) and im not using dual WAN setting. Im seeing an odd bug. Lets say im on the main screen on the router, if I click on the WAN tab on the right side, it will take me to a Internet Connection window that shows standard WAN connection page, not the USB Modem / USB Tethering window. (see images bellow.). So I have to click on the Dual want tab or any tab and then click on Internet Connection (first tab on top under WAN) and then router shows me the USB Modem / USB Tethering options.
This bug can be observed on Mac and on Windows computer. On Safari browser and on Edge Browser.

4) I mainly use Mac laptop for everything and Safari as main web browser. With 44EA firmware, router asks me to re-enter username and password the second time after 10sec, even though im already logged in to the router. This is not an issue with Edge Browser on Mac or in Windows OS.

5) Safari on Mac wont save login credentials for the router when using Fork firmware, but Edge browser does. When using Merlin firmware, Safari browser does save login credentials for the router.
I cleared cache in the Safari, even changed LAN IP in the router.
EDIT: I deleted cookies in Safari on Mac (login credentials) for the router it self and now, im able to login to the router with out being asked for user name and password. Very odd. I'll test it in the next few days, but I call that a win for now.
EDIT 2: Router asked me for a password 10sec later, even though im already logged in to the router . Safari V13.1.12.
EDIT 3: Router asked me to re-enter password again after 10sec.

Cant think of anything else at this time, since I mainly use the router for WiFi, Diversion and Skynet.
Over all, 44EA firmware is working well for my daily needs.

Thanks again for your hard work.
Cheers.

EDIT 4:
6) I dont know if this is a new feature in the 44EA firmware but I havent seen it before and I havent seen it in Merlin firmware. I havent used Asus OEM firmware in years.
In the past, if I log in to a router from a device "A"(laptop, cellphone, etc). If I dont log out and try to log in with device "B", I would get a message that I need to log out with device "A".
To bypass the message, all I had to do, is manually turn off the router and im able to login with device "B"(computer, tablet, cellphone, etc) in to the router.
But now, turning off the router, removing the power cord, no longer logs out device "A". I still have to manually log out of the router using device "A", before I can log in to the router with device "B".
 

Attachments

Last edited:

KGB7

Very Senior Member
Quick question. Running Firmware 44EA on AC68 router.

Out of curiocity. How accurate is the Traffic Monitor on the router?
SpeedTest App by Ookla shows different down and up speeds.

Thanks.

Edit.
Once I chose the Wireless 5Ghz tab, the speeds are more aligned with the Speed Test App. Just kind of odd, that Internet Tab has different speed results. Since there is no other activity on the network.

Its not the most important thing, just personal curiosity.
 

Attachments

Last edited:

KGB7

Very Senior Member
@john9527

Asus AC68, firmware 44EA.

Unfortunately, firmware 44EA is not ready to be used with standalone hotspots, that require connection via USB port to the router (USB2 or via USB3).
After couple of hours of using 44EA firmware, things went sideways.

Speed test, still showed consistent speeds. An average of 100Mb/s down, 18Mb/s up.

But web surfing became difficult. Many websites would simply not load (Snbforums, CNN, CapitalOne, Pepco, Merlin firmware website, etc). Web Pages that eventually did load, were missing images/pics. I was not able to log in to my Amazon account.
Apps, on iOS devices that require internet connection, would not load or would not log in to online accounts; bank apps, map apps, calendar app, email apps.
I was not able to access webpages to access online accounts for: water bill, electric, bill, gas bill, etc. Webpages, accounts via apps, that I can easily access using cellular network on my iPhone via apps or via web browser, work with no issues on a cellphone using cellular network. (not connected to home wifi AC68 router).

Unfortunately, I have no choice but to go back to Merlin firmware. AT&T Hotspot, is my main internet access I use at this time time, since im grandfathered in to unlimited data. I do have an option for Comcast/Xfinity home internet, but I would be paying up to x3 per month for similar plan.

I have two hotspots, ZTE Velocity 2 and Netgear Aircard 781S. Both hotspots experience same issues when using 44EA firmware on AC68 router.

None of these issues are present on Merlin firmware 384.14/15/16/17/18.

John.

Thank you again for your hardwork. I hope you find a solution in the near future for people as my self, who use hotspots. Let me know if you have any questions.

EDIT:
I mainly use the router for Wifi in my home. Its running Skynet and Diversion. Nothing is connected to LAN ports, I have no media servers. Boring vanilla, home network, even my TV has no internet connection.
 
Last edited:

jrmwvu04

Very Senior Member
@KGB7 The issues you see with Safari, especially the password related ones, are limitations/quirks of Safari. Nothing you can do about those really.
 

Sign Up For SNBForums Daily Digest

Get an update of what's new every day delivered to your mailbox. Sign up here!
Top