What's new

[ 386.4 alpha Build(s) ] Testing available build(s)

  • 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.
Hmm... I am surprised that we are seeing yet another round of development without a new GPL drop from Asus being built-in. I am aware that there have been problems with those GPL builds, but still??? It has been a while now...
It's currently mostly component updates, which were already in the pipeline, but couldn't make it into 386.3. Haveged was also something that I wanted to check for a long time, but only got around to it as the jitterentropy-rngd results on pre-HND were less than expected.

I currently don't intend to release any new release until Asus resolves the GPL issues, as I don't have any other major changes planned for 386.4 (that was all done with 386.3, with the OpenVPN code redesign). They're well aware of it, and are actively working on it. The latest discussions regarding this were as recent as yesterday. It's not entirely in their hands at this time. So for now, it will be mostly just housekeeping and bug fixing.
 
So ... to be clear ... you will not release version 386.4 until it includes GPL updates ??? or GPL updates will only emerge with version 386.5 ??
 
I currently don't intend to release any new release until Asus resolves the GPL issues, as I don't have any other major changes planned for 386.4 (that was all done with 386.3, with the OpenVPN code redesign). They're well aware of it, and are actively working on it. The latest discussions regarding this were as recent as yesterday. It's not entirely in their hands at this time. So for now, it will be mostly just housekeeping and bug fixing.
I'm hoping that the GPL issues surround Broadcom moving to a kernel capable of running Wireguard at the kernel level (probably long term support linux-5.10), although it might be nice to have the Paragon NTFS3 full read/write NTFS driver that seems to be coming to 5.15 (also likely an LTS kernel). So many VPNs have adopted Wireguard within the past year or so. Sorry if we're getting off topic, here.
 
Alpha 1, 386.4 installed yesterday on ac5300 with ac68u AiMesh. All is stable so far. dirty upggrade.
 
I'm hoping that the GPL issues surround Broadcom moving to a kernel capable of running Wireguard at the kernel level (probably long term support linux-5.10)
Historically, Broadcom has never upgraded an SDK kernel, because it would require complete revalidation of the whole software stack, and require manufacturers to update their entire firmware stack to work with the new kernel, and require third party developers like Tuxera and Trend Micro to also redesign their own products, and retest everything. From a business point of view it makes absolutely no sense.

Beside, you don`t need 5.10 to run Wireguard, it works just fine with 4.1.

although it might be nice to have the Paragon NTFS3 full read/write NTFS driver that seems to be coming to 5.15 (also likely an LTS kernel).
Why? My personal past experience has shown that the Tuxera NTFS driver Asuswrt uses is way, way better than Paragon, both in terms of performance and reliability. Especially since what is getting merged in the kernel is their open source driver, which doesn't perform as well as the separate commercial driver they also offer, unless they somehow decided to shut down their commercial product, and open sourced all their proprietary work present in the closed source commercial product.

That upstream Paragon driver will be an improvement over NTFS-3G, but it won't match either the commercial version of Paragon NTFS or Tuxera NTFS.
 
Not arguing, just adding some detail ...

Beside, you don`t need 5.10 to run Wireguard, it works just fine with 4.1.

Ah, found the backport page at wireguard.com and 4.1.52 (what my router is running now), is on the list with a successful build. Maybe that driver can be added, as the kernel version is supposed to perform better than the user space Wireguard.

Why? My personal past experience has shown that the Tuxera NTFS driver Asuswrt uses is way, way better than Paragon, both in terms of performance and reliability. Especially since what is getting merged in the kernel is their open source driver, which doesn't perform as well as the separate commercial driver they also offer, unless they somehow decided to shut down their commercial product, and open sourced all their proprietary work present in the closed source commercial product.

That upstream Paragon driver will be an improvement over NTFS-3G, but it won't match either the commercial version of Paragon NTFS or Tuxera NTFS.

I had read that Paragon was submitting their commercial version, but that might be a reporter's error. Was thinking that if Asus updated the kernel, then maybe Tuxera might have to be re-negotiated (certainly retested as you suggested).
 
Dirty flash over 386.3_2 All working well. Thanks Merlin !!
 
I had read that Paragon was submitting their commercial version, but that might be a reporter's error.
Honestly, I don't know for sure, as I haven't seen any first party information. It would be odd for me for a company to suddenly open source their internal proprietary code, as it's part of their core business, selling that product to various embedded device manufacturers like router or NAS manufacturers.

Asus was using the Paragon driver initially, they switched to Tuxera with either the RT-AC56U or RT-AC68U.
 
As for this change -- "No longer restart nasapps/wsdd2 on WAN start."
On AX88U, immediately after manual WAN disconnection, the router disappears from Windows Network, and I must restart wsdd2 to make it appear again.
Similarly, after WAN restart, I must restart wsdd2 again to make it appear.
WAN type is PPPoE.
 
Last edited:
Here's one small item: The speedtest page is missing in this alpha? The page tab and header info are there, but the control to activate the test, that is missing.
Everything else seems to be working...
_________________________________________________________________________________________________________________________________________________________________________

Did a full hard reset on every device from the modem down through every device. Full functioning restored!
 
Last edited:
It's currently mostly component updates, which were already in the pipeline, but couldn't make it into 386.3. Haveged was also something that I wanted to check for a long time, but only got around to it as the jitterentropy-rngd results on pre-HND were less than expected.

I currently don't intend to release any new release until Asus resolves the GPL issues, as I don't have any other major changes planned for 386.4 (that was all done with 386.3, with the OpenVPN code redesign). They're well aware of it, and are actively working on it. The latest discussions regarding this were as recent as yesterday. It's not entirely in their hands at this time. So for now, it will be mostly just housekeeping and bug fixing.
I figured it was mostly for your component updates. Despite ASUS release speed, for security reasons, I hope others can completely understand why you keep components up-to-date via alpha/beta phase first. God forbid you release something to the main firmware batch that breaks curl, WSD, or any other custom components.
 
Here's one small item: The speedtest page is missing in this alpha? The page tab and header info are there, but the control to activate the test, that is missing.
Everything else seems to be working...
_________________________________________________________________________________________________________________________________________________________________________

Did a full hard reset on every device from the modem down through every device. Full functioning restored!
Looks fine on my AX88U. You may want to force refresh your browser in case there were changes to CSS that is cached.
 
Logged in via PuTTY and ran "u" via AMTM. It reported Diversion and Skynet right away but had a hard stall before posting Unbound. Entware posted OK.
AMTM posted no updates available.

Due to the hard stall, I entered Unbound and ran #1 "update" and it appears to have upgraded "haveged" components.

Anyone else seeing this?
 
As for this change -- "No longer restart nasapps/wsdd2 on WAN start."
On AX88U, immediately after manual WAN disconnection, the router disappears from Windows Network, and I must restart wsdd2 to make it appear again.
Similarly, after WAN restart, I must restart wsdd2 again to make it appear.
WAN type is PPPoE.
By testing the old wsdd2 binary on this alpha FW.

The problem seems from the new wsdd2.
With the old wsdd2, after the WAN restart, I don't have to restart it, and the router appears again in Windows Network.
(With either version of wsdd2, after manual wan disconnection, the router disappears, so maybe this is always the case, which I just never noticed).

It's also weird that, with the new wsdd2, after a while, the router simply disappear from Windows Network, and must restart it (it doesn't crash, still running), and after a couple of hours, disappears again and so on.
 
Last edited:
update after 1 day it would no longer connect to the internet had to go back to 386.3 release to get things working again . strange
 
Status
Not open for further replies.

Latest threads

Sign Up For SNBForums Daily Digest

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