I’m a bit tired of waiting. The advertising promise was OpenWRT 23.05, apparently most of the problems are now solved with that version. I wish the communication regarding the roadmap was clearer.
Firmware delivery on the 23.05 line would also be appreciated. Like others, I switched over to OpenWRT and spent 1/2 a day getting the configuration working. Backed up now, so not a big deal. You only take the pain once with OpenWRT thankfully. Will probably create some script files.
How do you block YouTube ads? Thanks.
Not possible anymore. Only on end-device.
It is possible, set a VPN client and connect to Albania
and then set the VPN policy to domain name and add the YouTube domains, all the YouTube traffic will go through the VPN and you will block or avoid all ads because of the country, legally of course
and then enable AdGuard to have the perfect combo. This is not possible on v4.5.
I think this was the reason:
This isn’t a real solution for blocking ads, tbh.
Yeah, might work - but not because it’s blocking ads, but because in Albania there are no ads (due to reasons, maybe law)
That's the point... Obviously not a native blocking system if that's what you'd like to have, I had a similar setup for a long time on Vanilla OpenWrt and it worked great, so to me personally it is a solution, based on the final results
This information was even removed in the MT6000 announcement: Flint 2 (GL-MT6000) | High-Performance VPN Router - GL.iNet
If we had known that there would be so many problems with routers and that the developers didn’t care about delivering the product as promised, none of us would have purchased products from GL.inet
Hopefully we can see a new version of 23.05+ from the main branch soon aswell ![]()
Since my issue got solved due to interference of my usb stick its running almost flawlessly now, only when multicast applications turn around over wifi then i will notice seq time out messages and even a crash.
But i guess that might be fixed soon as well, but on the other side the main github branch is a total different beast than just 23.05, since a higher gcc is required to compile OpenWrt main somehow these build work faster and a higher kernel 6.6 (i run this kernel).
Mt76 improved from miles ahead and made really big steps compared to mt6000 release ![]()
Next time you flash openWRT would you record it and post a youtube video on a quick setup of it please so those who find it very confusing could maybe follow a small guide.
well sure but I need to know what is confusing ![]()
though the type of images I create are generaly only for my lab environment with pre configuration which I use to fastly check patches and other things without breaking too much of my network ;-), currently my last image is kind of dated since I was testing the multicast crash.
these config files are here: openwrt-flint2-testing/files/etc at main · xize/openwrt-flint2-testing · GitHub this reflects /etc/config for when I build a image, it’s almost a 1:1 configuration what I run just now in OpenWrts runtime.
though if you want to try cleaner images I want you to recommend images from pesa:
always use the sysupgrade ones this will give you the option to preserve the gl-inet u-boot (your recovery OS), you can just use the images in the gl ui and flash it also backwards.
always make sure to flash without your current configuration, because if you come from a MTK image the driver references are different and it may break the boot process.
@xize11 could this fix be related to the problem?
https://git.openwrt.org/?p=openwrt/openwrt.git;a=commit;h=45a8e962a591dfcac252b0de6324319abd080469
jup that is a probability currently I’m testing this patch since it is upstreamed, though i might need to test some more (I still crash
), currently my multi psk crashes because of allmulticast mode is active and my mi smart clock freezes, I got the same suspicioun that multicast packets are either corrupted frames or something else is going on, and then the crash can be very device specific or cause different issues, the crash was caused by a Intel 210AX chip but my mi smart clock uses a mediatek chip but just freezes the touch screen (probably due to flooding).
there is a interesting discussion here:
so I might need to check the patches by nbd168 for each in this discussion
currently my work around is adding option multicast_to_unicast '0' to br-lan but that is not feasible if you need something like iptv over multicast.
This is a temporary solution, but it is also not the right way.
The device has been on the market for almost half a year and the manufacturer has absolutely no way of getting the device to work as advertised.
Would Mercedes sell its cars like this according to the motto, the main thing is to sell them first “We’ll correct mistakes later - if at all” How many Mercedes do you think would be sold?
To be honest, it’s an absolute shame that only emergency solutions come and the router doesn’t work as advertised.
As a German, I can tell you: They do. All companies do. Some more, some less.
Yes, that’s true. However, they still try to improve problems as quickly as possible. With the Flint 2 you have the feeling that with each firmware you go 2 steps back and only one step forward.
Personally, I don’t really like having a 3-year-old Openwrt 21 firmware on it either.
I bought the Flint 2 because of all the promises that were made.
Mercedes is a nice example; Consumer Reports ranks their reliability as 29th worst out of 30.
Another way to deal with this is to lock down the firmware with limited basic functions, like most routers are sold, and update the firmware annually for a year or two. My Subaru was coming off lease by the time the radio worked.
We thought about this before. Unfortunately the UI and its underlaying logic is not targeting at compatibility. Otherwise it will just be a frontend of LuCI, which is designed exactly for this purpose. Actually some vendors does develop by modifying LuCI. But that may create other issues, causing LuCI itself does not work propoerly.
But I would also say that we as a forum community cannot judge whether many or few people are dissatisfied.
Even if 50 people here (and I think that number is already too high) complain about e.g. old OpenWrt or bugs in Flint 2 or similar, then this is in relation to the sales figures - which we don’t know.
If you had only sold 50 devices, it would be catastrophic. If there are 10,000 or more, the “complaint noise” is almost negligible. ![]()