[384.11_Alpha - builds] Testing all variants.

[L&LD]

I really want to hear Jack Yaz's or Merlin's opinion about this. It seems we are just going in circle with this.

Not 'we'.

You did read the link I gave you from Jack Yaz in post 130, correct.

 

[bluepoint]

Not 'we'.

You did read the link I gave you from Jack Yaz in post 130, correct.

That doesn't mean anything, according to jack, Merlin's ntp advertises its availability. It doesn't mean it provides time automatically to all clients. It's there if you want to use it. This will be my last post on this. I'll wait for Jack or Merlin's opinion, it will be nice to hear from them, I could be wrong.

 

[L&LD]

That doesn't mean anything, according to jack, Merlin's ntp advertises its availability. It doesn't mean it provides time automatically to all clients. It's there if you want to use it. This will be my last post on this. I'll wait for Jack or Merlin's opinion, it will be nice to hear from them, I could be wrong.

You're a hard one to convince.

I like to take a chance here and there and possibly learn something new by being wrong, once in a while.

And in your case, it is easy enough to test this assumption, right?

 

[dave14305]

You're a hard one to convince.

I like to take a chance here and there and possibly learn something new by being wrong, once in a while.

And in your case, it is easy enough to test this assumption, right?

Almost no popular client OSes respect DHCP option 42 for NTP. This was discussed in John's thread after I suggested adding the option after reading an older post by sfx2000 about it. Looks nice, but has no practical purpose for the masses. An NTPFilter iptables rule would be more effective than the DHCP option.

[Fork] Asuswrt-Merlin 374.43 LTS releases (V39E3)

 

[L&LD]

Almost no popular client OSes respect DHCP option 42 for NTP. This was discussed in John's thread after I suggested adding the option after reading an older post by sfx2000 about it. Looks nice, but has no practical purpose for the masses. An NTPFilter iptables rule would be more effective than the DHCP option.

[Fork] Asuswrt-Merlin 374.43 LTS releases (V39E3)

I stand corrected, but still adding to my knowledge!

 

[RMerlin]

Yes, RMerlin's. The options set in dnsmasq/DHCP do that for you and all your other connected clients automatically.

Windows does not heed DHCP option 42.

 

[vdemarco]

Windows does not heed DHCP option 42.

neither do any apple products

 

[jerry6]

this is way above my level today , al i know is i got no connection when i rrboot non of the browsers connct to any dsitre until i disconnect vpmn client andstart over , is there some settingi need to implement in the new fw to get things working on startup ? this is getting complicated

 

[AndreiV]

this is way above my level today , al i know is i got no connection when i rrboot non of the browsers connct to any dsitre until i disconnect vpmn client andstart over , is there some settingi need to implement in the new fw to get things working on startup ? this is getting complicated

It's still experimental , don't turn it on , you don't need it if you are using a VPN.

 

[EmeraldDeer]

I'd suggest to use something like below instead

#!/bin/sh
echo "1" > /proc/sys/net/ipv6/conf/all/accept_ra
echo "1" > /proc/sys/net/ipv6/conf/all/forwarding
echo "1" > /proc/sys/net/ipv6/conf/eth0/accept_ra
echo "0" > /proc/sys/net/ipv6/conf/eth0/forwarding

Here are the settings after an alpha3 upgrade:

HdB342665LNtPFM@RT-AX88U-F098:/jffs/scripts# cat /proc/sys/net/ipv6/conf/all/accept_ra
1
HdB342665LNtPFM@RT-AX88U-F098:/jffs/scripts# cat /proc/sys/net/ipv6/conf/all/forwarding
1
HdB342665LNtPFM@RT-AX88U-F098:/jffs/scripts# cat /proc/sys/net/ipv6/conf/eth0/accept_ra
0
HdB342665LNtPFM@RT-AX88U-F098:/jffs/scripts# cat /proc/sys/net/ipv6/conf/eth0/forwarding
0

After setting eth0 accept_ra to 1, WAN IPv6 Gateway went from empty to what it had been set to in 384.10_2. IPv6 is working now.

Thank you!

EDIT: Please disregard below because I backed out to 384.10_2 and still ICMPv6 is blocked.

I used to get 19/20 on the IPv6 test at http://ipv6-test.com/ with just the 1 problem being the lack of a reverse DNS record which I do not see as a concern.

Now I get a 17/20 due to ICMPv6 being blocked. I double checked that Windows 10 Kaspersky Internet Security Packet Rules still has my rule to allow ICMPv6.

UPDATE: I connected my Windows 10 laptop directly to the Arris SB8200 cable modem and ICMPv6 is still not working. Either it is something new in Windows or more likely with Comcast Internet Service Provider.

UPDATE 2: The ping only tab http://ipv6-test.com/pingtest/ has an ICMPv6 test and it is working to my router. So is the other test just broken or does it do more than echo request?

 

[Howard_inGA]

ALPHA 3 is just humming along with no hitches in the git-a-long!
1 days 2 hours 28 minute(s) 26 seconds
RT-AC5300

 

[Swistheater]

Here are the settings after an alpha3 upgrade:

HdB342665LNtPFM@RT-AX88U-F098:/jffs/scripts# cat /proc/sys/net/ipv6/conf/all/accept_ra
1
HdB342665LNtPFM@RT-AX88U-F098:/jffs/scripts# cat /proc/sys/net/ipv6/conf/all/forwarding
1
HdB342665LNtPFM@RT-AX88U-F098:/jffs/scripts# cat /proc/sys/net/ipv6/conf/eth0/accept_ra
0
HdB342665LNtPFM@RT-AX88U-F098:/jffs/scripts# cat /proc/sys/net/ipv6/conf/eth0/forwarding
0

After setting eth0 accept_ra to 1, WAN IPv6 Gateway went from empty to what it had been set to in 384.10_2. IPv6 is working now.

Thank you!

I used to get 19/20 on the IPv6 test at http://ipv6-test.com/ with just the 1 problem being the lack of a reverse DNS record which I do not see as a concern.

Now I get a 17/20 due to ICMPv6 being blocked. I double checked that Windows 10 Kaspersky Internet Security Packet Rules still has my rule to allow ICMPv6.

Did you leave the remove IPV6 neighbor solicitation on the tools page enabled? if so that might be what is causing you to get a 17.

 

[EmeraldDeer]

Did you leave the remove IPV6 neighbor solicitation on the tools page enabled? if so that might be what is causing you to get a 17.

I had not enabled "Firewall: Drop IPv6 neighbour solicitation broadcasts (default: No)" because the first try at setting eth0 accept_ra to 1 had solved the IPv6 gateway issue.

 

[eclp]

@EmeraldDeer ...

Have you tried this for Windows 10?

cmd

netsh firewall set icmpsetting type=all mode=enable

 

[EmeraldDeer]

@EmeraldDeer ...

Have you tried this for Windows 10?

cmd

netsh firewall set icmpsetting type=all mode=enable

I tried that and still no ICMPv6 ping.
Since my firewall is being managed by Kaspersky, I am not exactly sure what it does in response.
I added rules in Kaspersky in case Comcast passes ICMPv6 again.

 

[LiFePO4]

I tried that and still no ICMPv6 ping.
Since my firewall is being managed by Kaspersky, I am not exactly sure what it does in response.
I added rules in Kaspersky in case Comcast passes ICMPv6 again.
View attachment 17104

I am Comcast and up until today I got 19/20 on ipv6-test but today it has been 17/20 due to ICMP. I have not done anything new so I am wondering if it is Comcast (I tried a couple PCs on my network and the same result). I use McAfee and the setting in there is still set to allow ICMP pings.

 

[Swistheater]

I am Comcast and up until today I got 19/20 on ipv6-test but today it has been 17/20 due to ICMP. I have not done anything new so I am wondering if it is Comcast (I tried a couple PCs on my network and the same result). I use McAfee and the setting in there is still set to allow ICMP pings.

Yea I am only getting 17 out of 20 as well must be.

 

[Swistheater]

Then again it could also be the test that the site is running as well. The only real way to tell is to try manually doing the ping test and see what kind of results you get.

 

[Swistheater]

https://tools.keycdn.com/ipv6-ping
Try this, I used it and I am getting good results.

 

[Swistheater]

The only real world test is to unplug the modem from the router and run the test from the modem plugged into a computer if it still says filtered then it is comcast or the device, if it doesn't then there is filtering being done by the router, but that isn't necessarily a bad thing the router could be only filtering bad stuff.