[384.11_Alpha - builds] Testing all variants.

[skeal]

What is your WAN test? Mine is a ping of my ISP's first hop router IP address. Do you have a router VPN client? If so, does your WAN test work in both cases?

If you are referring to network monitoring, I don't use dns query or ping. Both are unchecked. You think this could be a clue?

 

[EmeraldDeer]

If you are referring to network monitoring, I don't use dns query or ping. Both are unchecked. You think this could be a clue?

I would not know how the router behaves with regards to WAN online/offline if you do not use either dns query or ping.

 

[skeal]

I just tried using ping and 1.1.1.1 so no dns would be required and it failed the same way. I'm sure it has something to do with the vpn server.
EDIT: That is, it's interfering with the wan up and so no ntp update can take place.

 

[EmeraldDeer]

Notes regarding 384.11_alpha1-g90fb05a944

• For what it's worth, the Network Map listing appears to be working better. Before, only directly connected devices would show up. Devices connected to a switch would not. The Network Map list appears to impact pull down lists for Trend Micro graphs such as Web History.
• Tracked Connections is no longer showing IPv6
• The new BusyBox minimal NTP sever is disabled by default which is good because I am using ntpMerlin
• Perhaps Administration > System > Basic Config > Secondary NTP Server is new
• Administration > System > Service > Permita Login com Senha is Portuguese for Allow Login with Password
  
• Nothing in System Log for IPv6 - perhaps IPv6 is not working at all?

 

[EmeraldDeer]

Notes regarding 384.11_alpha1-g90fb05a944

• For what it's worth, the Network Map listing appears to be working better. Before, only directly connected devices would show up. Devices connected to a switch would not. The Network Map list appears to impact pull down lists for Trend Micro graphs such as Web History.
• Tracked Connections is no longer showing IPv6
• The new BusyBox minimal NTP sever is disabled by default which is good because I am using ntpMerlin
• Perhaps Administration > System > Basic Config > Secondary NTP Server is new
• Administration > System > Service > Permita Login com Senha is Portuguese for Allow Login with Password
  
• Nothing in System Log for IPv6 - perhaps IPv6 is not working at all?

I backed out to 384.10_2

• IPv6 is back
• Stubby actually started. It had not started with alpha and I was not able to determine why. NOTE: I mean start automatically during boot. I was able to start stubby minutes after boot.
• Since Stubby started, Skynet started as well.
• The Network Map is back to just directly connected devices.

 

[skeal]

I backed out to 384.10_2

• IPv6 is back
• Stubby actually started. It had not started with alpha and I was not able to determine why.
• Since Stubby started, Skynet started as well.
• The Network Map is back to just directly connected devices.

I'm hanging in there. VPN's work if manually start them. I'll wait to hear from @RMerlin to see what can be figured out with this dastardly problem....LOL

 

[RMerlin]

@RMerlin were you able to shed any light on this sir?

It's a very early alpha build. Virtually nothing has been tested yet beyond ensuring that "it doesn't crash", so I'm not investigating any reported issue at this time - too much will change during the alpha cycle to be worth the time.

 

[kernol]

It's a very early alpha build. Virtually nothing has been tested yet beyond ensuring that "it doesn't crash", so I'm not investigating any reported issue at this time - too much will change during the alpha cycle to be worth the time.

Which does beg the question ... why a third party kicks off the alpha thread in this forum in the first place???
No offence intended @octopus - just feel it is better that @RMerlin initiate alpha threads when he is good and ready to receive responses via the forum.

Those who subscribe to Github notifications can of course play on the bleeding edge without any early-bird posts here.

 

[skeal]

Observation: 5G wifi and just wifi in general is better with the alpha 1 for the AX88U. Way less problems. Fewer disconnects, less time to bring network up (ism startup), no more fixed channel. I'm overall pleased so far with this early build. Just make sure if you are jumping in to turn the OVPN server off before a reboot and ake sure a OVPN client isn't set to start at reboot, and you should be fine. Seems QOS has an issue starting. I have @FreshJR QOS and after a reboot you have to click the apply button on the QOS page to get the graphs to start. That so far is all the trouble I've had!!

 

[RMerlin]

Which does beg the question ... why a third party kicks off the alpha thread in this forum in the first place???

Nothing wrong with that. I don't start a thread because I don't want to give the impression that I'm going to provide actual support for these alpha builds. But if someone else wants to start one for users to share feedback among themselves, I'm fine with it.

 

[Sanna1967]

Just for info ( i know there is no support for this alpha , and my model is not there either, with the link )
The new "Scheduled check for new firmware availability" button does not stay off
No need to answer, it's just to report

Compiled by myself for my AC86 from the mainline ( commit :169e54e )

Cheer

 

[wpte]

So far it seems that this alpha build with the latest Asus code merge fixed my IPv6 problems I had in 384.10 but didn't have in 384.9
Happy days

 

[L&LD]

RT-AC88U has been added to the alpha!

 

[skeal]

On the AX88U I struggle a bit to keep QOS working. It seems at boot that FreshJR_QOS starts or at least modifies the Classification page but shows no graphs. If I click apply on the QOS page things come around, and QOS works. I woke this morning and the graphs again were gone, and a hit of the apply button, again restored it's function. I know it's an alpha, just some food for thought is all.

Edit: I didn't restart QOS after a reboot last night, thus it was looking down this morning..lol .

 

[skeal]

Also with this alpha on the AX88U the OVPN client 1 Kill Switch doesn't work. If the client is down routed devices (strict) still have access to the internet.

 

[Swistheater]

So far it seems that this alpha build with the latest Asus code merge fixed my IPv6 problems I had in 384.10 but didn't have in 384.9
Happy days

That is awesome! Though I do think asus likes to fix things for some and some how break it for others.

 

[Marin]

Upgraded my AX88U and no issues. Just a reminder that Stubby users should uninstall and stabilize before the upgrade, IMHO. DoT working as we speak. Nice job on the interface @RMerlin !!

Did you uninstall NTPMerlin as well?

384.11 (xx-xxx-2019)
  - NEW: DNS-Over-TLS (also known as DoT) is now supported.
         You can configure it on the WAN -> Internet Connection
         page.  You can manually add your own servers, or chose
         one (or a few) from the preset list.  (themiron)
  - NEW: NTP daemon on the router, to allow your LAN clients to
         synchronize their clocks with it.  Dnsmasq will also
         announce it through DHCP option 42, for clients using
         that parameter.
  - UPDATED: RT-AX88U to 384_5951 GPL.
  - UPDATED: Nano 4.0.
  - UPDATED: Curl 7.64.1.
  - UPDATED: Dropbear 2019.78.
  - CHANGED: Replaced the custom ntpclient with a proper ntpd
             implementation, for reduced memory footprint and
             increased accuracy.
  - CHANGED: Made the secondary NTP server configurable through the
             webui.  Note that ntpd will use BOTH servers, so clear
             the second server if there is one and you don't want
             to use it.
  - CHANGED: Re-designed firmware upgrade page, moving the schedule
             option to that page, and removed support for the Beta
             channel.
  - CHANGED: Removed popup messages showing on the DDNS page when
             a service state change was detected.  Report it within
             the page instead.
  - CHANGED: Report available firmware version within the new firmware
             notification popup.
  - CHANGED: Moved LED control (formerly known as Stealth Mode) to
             the System page.
  - CHANGED: Do not restart whole network whenever changing an IP
             reservation on the Networkmap card.
  - CHANGED: Allow URLs up to 64 chars long on the URL filter.
  - CHANGED: pre-mount user script now receives the filesystem
             as second argument.
  - CHANGED: Moved DNSSEC and DNS Rebind protection settings
             to the WAN page.
  - FIXED: Reboot scheduler would sometime get stuck or corrupt
           plugged USB drives.  Now doing a more thorough
           shutdown of services, should hopefully make it
           more reliable.
  - FIXED: CVE-2019-1543 issue with Chacha20-poly1305 in
           OpenSSL 1.1 (themiron)

 

[skeal]

Did you uninstall NTPMerlin as well?

I didn't have it installed, I also removed the second ntp server field as it had something in it.

 

[Marin]

I didn't have it installed, I also removed the second ntp server field as it had something in it.

Ok [emoji108]. I may uninstall it just to see how things go first after the upgrade.


Sent from my iPhone using Tapatalk

 

[heysoundude]

Did you uninstall NTPMerlin as well?

384.11 (xx-xxx-2019)
  - NEW: DNS-Over-TLS (also known as DoT) is now supported.
         You can configure it on the WAN -> Internet Connection
         page.  You can manually add your own servers, or chose
         one (or a few) from the preset list.  (themiron)
  - NEW: NTP daemon on the router, to allow your LAN clients to
         synchronize their clocks with it.  Dnsmasq will also
         announce it through DHCP option 42, for clients using
         that parameter.
  - UPDATED: RT-AX88U to 384_5951 GPL.
  - UPDATED: Nano 4.0.
  - UPDATED: Curl 7.64.1.
  - UPDATED: Dropbear 2019.78.
  - CHANGED: Replaced the custom ntpclient with a proper ntpd
             implementation, for reduced memory footprint and
             increased accuracy.
  - CHANGED: Made the secondary NTP server configurable through the
             webui.  Note that ntpd will use BOTH servers, so clear
             the second server if there is one and you don't want
             to use it.
  - CHANGED: Re-designed firmware upgrade page, moving the schedule
             option to that page, and removed support for the Beta
             channel.
  - CHANGED: Removed popup messages showing on the DDNS page when
             a service state change was detected.  Report it within
             the page instead.
  - CHANGED: Report available firmware version within the new firmware
             notification popup.
  - CHANGED: Moved LED control (formerly known as Stealth Mode) to
             the System page.
  - CHANGED: Do not restart whole network whenever changing an IP
             reservation on the Networkmap card.
  - CHANGED: Allow URLs up to 64 chars long on the URL filter.
  - CHANGED: pre-mount user script now receives the filesystem
             as second argument.
  - CHANGED: Moved DNSSEC and DNS Rebind protection settings
             to the WAN page.
  - FIXED: Reboot scheduler would sometime get stuck or corrupt
           plugged USB drives.  Now doing a more thorough
           shutdown of services, should hopefully make it
           more reliable.
  - FIXED: CVE-2019-1543 issue with Chacha20-poly1305 in
           OpenSSL 1.1 (themiron)

woah.
Not exactly sure what "proper ntpd implementation" means, but having followed ntpmerlin and kvic's version before it, I'm pretty sure that's the gold standard.
I think @Jack Yaz 's script lets me input more than 2 ntp servers for redundancy. (4-5 if memory serves...maybe more)
in the (hopefully unlikely) event something fairly catastrophic happens, the likelihood of having a reliable time to reference to drops considerably in this case:
most publicly-accessible timeservers are stratum 2, but they commonly access the same stratum 1 servers for their reference.
Suggesting referencing pools invites the possibility for more significant jitter and the resultant offset, because geography/consensus. I'm of a mind that building a constellation of reference clocks to make sure I have a reliable and redundant stream from a variety of stratum 1 servers is probably better, even with rollover times.