What's new

Beta Asuswrt-Merlin 386.2 Beta is now available

  • SNBForums Code of Conduct

    SNBForums is a community for everyone, no matter what their level of experience.

    Please be tolerant and patient of others, especially newcomers. We are all here to share and learn!

    The rules are simple: Be patient, be nice, be helpful or be gone!

Status
Not open for further replies.
No change at all to the IPv6 code between beta 1 and 2. In fact, the only change that touches the WAN interface code in beta 2 is that I reverted Asus's change that was applying static routes for DNS servers, and those were for IPv4.

You will have to look at the system log to determine what is failing exactly.
My AX86U mesh nodes disappear when upgrading my main AX88U router to B2. IPv6 also fails for me too.

Mar 21 07:10:26 dnsmasq[2455]: bad IPv6 address at line 25 of /etc/dnsmasq.conf
Mar 21 07:10:26 dnsmasq[2455]: FAILED to start up
Mar 21 07:10:27 kernel: FPM Pool 1: invalid token 0x206d0000 freed
Mar 21 07:10:32 dnsmasq[2468]: bad IPv6 address at line 25 of /etc/dnsmasq.conf
Mar 21 07:10:32 dnsmasq[2468]: FAILED to start up
Mar 21 07:10:53 crond[1160]: time disparity of 1513565 minutes detected
Mar 21 07:11:02 dnsmasq[2574]: bad IPv6 address at line 25 of /etc/dnsmasq.conf
Mar 21 07:11:02 dnsmasq[2574]: FAILED to start up
Mar 21 07:11:32 dnsmasq[2681]: bad IPv6 address at line 25 of /etc/dnsmasq.conf
Mar 21 07:11:32 dnsmasq[2681]: FAILED to start up

Are there any clues here?

Rolling back to B1 the nodes reappear and IPv6 is working again.

Thanks
HB
 
Last edited:
Updated to Beta 2 and working fine.
Uptime: 0 days 12 hour(s) 31 minute(s) 13 seconds

One Aimesh node connected fine and working.

Thank you for new beta build.
 
My AX86U mesh nodes disappear when upgrading my main AX88U router to B2. IPv6 also fails for me too.

Mar 21 07:10:26 dnsmasq[2455]: bad IPv6 address at line 25 of /etc/dnsmasq.conf
Mar 21 07:10:26 dnsmasq[2455]: FAILED to start up
Mar 21 07:10:27 kernel: FPM Pool 1: invalid token 0x206d0000 freed
Mar 21 07:10:32 dnsmasq[2468]: bad IPv6 address at line 25 of /etc/dnsmasq.conf
Mar 21 07:10:32 dnsmasq[2468]: FAILED to start up
Mar 21 07:10:53 crond[1160]: time disparity of 1513565 minutes detected
Mar 21 07:11:02 dnsmasq[2574]: bad IPv6 address at line 25 of /etc/dnsmasq.conf
Mar 21 07:11:02 dnsmasq[2574]: FAILED to start up
Mar 21 07:11:32 dnsmasq[2681]: bad IPv6 address at line 25 of /etc/dnsmasq.conf
Mar 21 07:11:32 dnsmasq[2681]: FAILED to start up

Are there any clues here?

Rolling back to B1 the nodes reappear and IPv6 is working again.

Thanks
HB
It seems like a bug in beta 2. If you accidentally left any ipv4 addresses in "LAN -> DNSFilter -> Custom (user-defined) DNS x", these thing will end up in /etc/dnsmasq.conf like somthing as "dhcp-option=dnsf8,option6:23,[1.0.0.1]", which will cause this issue.
It's not the case in beta 1.
 
It seems like a bug in beta 2. If you accidentally left any ipv4 addresses in "LAN -> DNSFilter -> Custom (user-defined) DNS x", these thing will end up in /etc/dnsmasq.conf like somthing as "dhcp-option=dnsf8,option6:23,[1.0.0.1]", which will cause this issue.
It's not the case in beta 1.
I did indeed have some custom DNS addresses! Having removed them and reinstalled B2 everything seems to be fine again - thank you!! IPv6 is working and the nodes appeared this time.

Cake seems to be working very well. My bufferbloat measurement on Waveform has gone from a grade B with no QoS to an A+ with Cake on a 33/8 connection - impressive.

HB
 
Is there a command that can force an ntp sync ?
Restarting ntpd/chronyd (whichever one you use) should force a sync.

Edit: if you have the scMerlin-addon installed just hit restart timeserver.
 
Restarting ntpd/chronyd (whichever one you use) should force a sync.

Edit: if you have the scMerlin-addon installed just hit restart timeserver.
Thanks, but the time is still not updating for some reason, I have factory reset etc and ntp does seem to be running:

3759 bob 3748 S /usr/sbin/ntp -t -S /sbin/ntpd_synced -p pool.ntp.org
5100 bob 3396 R grep ntp

Sat, May 05 06:33:17 2018
* Daylight savings time is enabled in this time zone.
 
Mmm what does point to:

C:\Users\waama>w32tm /monitor /computers:no.pool.ntp.org
no.pool.ntp.org[37.44.185.42:123]:
ICMP: error IP_REQ_TIMED_OUT - no response in 1000ms
NTP: error ERROR_TIMEOUT - no response from server in 1000ms


But connected via 4g on my phone:

C:\Users\waama>w32tm /monitor /computers:no.pool.ntp.org
no.pool.ntp.org[162.159.200.123:123]:
ICMP: 59ms delay
NTP: -1.0937481s offset from local clock
RefID: (unknown) [0x1308150A]
Stratum: 3

I am wondering if this coincidental and actually it is my ISP having problems.

I dont think this is the firmware as have now downgraded and reset again but same issue.
 
Last edited:
That would not be a valid test, because then I would not saturate by ISP's link. That means my modem's buffers for instance would never get filled, as the modem would then be able to send data faster than what the router behind it is set to.
I believe the point of the bufferbloat craze is to do exactly that: ensure the router is in control of the bottleneck instead of the upstream devices.

But no matter. You have plenty of volunteer testers!
 
It seems like a bug in beta 2. If you accidentally left any ipv4 addresses in "LAN -> DNSFilter -> Custom (user-defined) DNS x", these thing will end up in /etc/dnsmasq.conf like somthing as "dhcp-option=dnsf8,option6:23,[1.0.0.1]", which will cause this issue.
It's not the case in beta 1.
OK - this condition matches my situation - I had IPv4 addresses listed in Custom DNSFilter inputs from when I used the "Custom" setting but left them in place after switching to "Router" mode. I'll clear these out and try again. Great find.

As a side note, my Rogers Ignite IPTV devices all seem to be fine with IPv6 disabled, so I will likely leave my config as is as I have no other need for IPv6 at this time.

EDIT: Curious if this issue/condition is specific to AX88U's in any way or is it just a co-incidence that only those of us with AX88U Routers have seen this?
 
Last edited:
Thank you very much RMerlin for this.
Installed 386.2 beta 2 about 4 hours ago on RT-AC86U.
Nothing to report. (FYI, I always have IPv6 disabled, and no AIMesh.)
Bufferbloat at A, as usual with cake (cake-qos or the built-in cake of 386.2).

About:
If you accidentally left any ipv4 addresses in "LAN -> DNSFilter -> Custom (user-defined) DNS x", these thing will end up in /etc/dnsmasq.conf like somthing as "dhcp-option=dnsf8,option6:23,[1.0.0.1]", which will cause this issue.
Not sure I understand the context but I'm sharing my data as additional reference point. I am using DNSFilter. I have all three "Custom (user-defined) DNS" fields defined in LAN/DNSFilter. "Custom 1" is assigned to 3 clients. I didn't make any specific check but there are no connectivity loss, everything seems OK. For reference, part of my dnsmasq.conf:
Code:
XXX@AC86:/tmp/etc# more /etc/dnsmasq.conf | grep dhcp-option
dhcp-option=lan,3,192.168.17.1        
dhcp-option=lan,252,"\n"
dhcp-option=br1,3,192.168.101.1
dhcp-option=br2,3,192.168.102.1
XXX@AC86:/tmp/etc#
192.168.17.1 is my router
 
About:

Not sure I understand the context but I'm sharing my data as additional reference point. I am using DNSFilter. I have all three "Custom (user-defined) DNS" fields defined in LAN/DNSFilter. "Custom 1" is assigned to 3 clients. I didn't make any specific check but there are no connectivity loss, everything seems OK. For reference, part of my dnsmasq.conf:
I think the piece missing from your situation is IPv6 being enabled. If it's disabled, the upgrade seems to be fine with those entries in place. Clear the entries with IPv6 enabled and it's OK. The "trigger" looks to be IPv6 enabled and custom DNSFilter entries present, from what those of us that have experienced this have seen. Still only a select few.
 
Last edited:
Mmm what does this point to:

C:\Users\waama>w32tm /monitor /computers:no.pool.ntp.org
no.pool.ntp.org[37.44.185.42:123]:
ICMP: error IP_REQ_TIMED_OUT - no response in 1000ms
NTP: error ERROR_TIMEOUT - no response from server in 1000ms


But connected via 4g on my phone:

C:\Users\waama>w32tm /monitor /computers:no.pool.ntp.org
no.pool.ntp.org[162.159.200.123:123]:
ICMP: 59ms delay
NTP: -1.0937481s offset from local clock
RefID: (unknown) [0x1308150A]
Stratum: 3

I am wondering if this coincidental and actually it is my ISP having problems.

I dont think this is the firmware as have now downgraded and reset again but same issue.
Okay so the above works if I connect the vpn client and run from a machine routed through that. Odd.
 
Yes, I use DNSFilter in "Router" mode to send all DNS traffic to my Pi-holes.

EDIT: So far, only @vdemarco and I seem to have seen this behaviour. We both have AX88U's as our Router. Not sure if that's of any assistance.
i am also having the router do dnsfiltering.

So i had custom entries (the default items in the router field) i just cleared them out, updated to B2 and its all working now.
 
Last edited:
The website https://www.asuswrt-merlin.net/ is still saying the last one is Beta 1.
Possibly for the router you are looking for there isn't a beta 2 available, yet my two routers are working without issues with beta 2.
RMerlin's Download page

Edit: Although RMerlin's Front Page doesn't reflect his current offerings, many people race for the chance to upload the latest/greatest fw, so "someone" is always posting the latest fw drops. Possibly not updated, in case fw is pulled or RMerlin is too busy.

Stay Safe Everyone
 
Last edited:
Possibly for the router you are looking for there isn't a beta 2 available, yet my two routers are working without issues with beta 2.
RMerlin's Download page

Stay Safe Everyone
I think he’s referring to the homepage
FE773FEE-64B1-4DA6-8F85-63D016E02259.jpeg
 
Thank you for this link I think this test will replace the DSL reports for me when anyone asks for a site to test bufferbloat.

So I tested my rt-ac86u and received a "C", site mentions SQM. How do we turn that on in Merlin? I can not seem to find the SQM setting in Merlin
 
Status
Not open for further replies.

Sign Up For SNBForums Daily Digest

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