OpenWRT compatible Router for 2024 in India

This is the exact reason why I shifted to x86 router.
Those consumer grade routers can't do SQM beyond 100-150Mbps. The CPU power ain't enough.
And SQM is helpful in load scenarios where your ping shoots up and causes bufferbloat.
Yeah, even the ARM chip on the RPi4 is enough for 1 gbps, I guess. But I already get an A grade on the waveform bufferbloat test, with only 14ms of additional latency. So since I'm not playing any competitive fps games *while* fully saturating my connection with downloads, I don't think it matters. And even then 14ms is not that bad.
 
I want to put my ISP's existing ont+router in bridge mode because their firmware is most probably dogshit in terms of security and use this as my router instead.
Putting ONT in bridge mode comes with its own set of issues unique to each service provider. Tbh, almost all such routers have poor security but it doesn't matter to a typical home connection as nobody is going to DDoS a home connection & nobody would do network intrusion either when there are equally less unsecure but much more financially rewarding small business networks out there. Tplink itself is known to stop providing firmware updates after 3-4 versions/2-3 years depending on the model's popularity/selling targets. Of course the ISP ONT router is typically poor performing so just connect your good router to ONT via ethernet & place it in DMZ zone of ONT & then turn off wifi of ISP ONT & connect all your devices to your own router wifi & use its features like NAT/port forwarding etc.
 
Putting ONT in bridge mode comes with its own set of issues unique to each service provider. Tbh, almost all such routers have poor security but it doesn't matter to a typical home connection as nobody is going to DDoS a home connection & nobody would do network intrusion either when there are equally less unsecure but much more financially rewarding small business networks out there. Tplink itself is known to stop providing firmware updates after 3-4 versions/2-3 years depending on the model's popularity/selling targets. Of course the ISP ONT router is typically poor performing so just connect your good router to ONT via ethernet & place it in DMZ zone of ONT & then turn off wifi of ISP ONT & connect all your devices to your own router wifi & use its features like NAT/port forwarding etc.
That risk calculation is not at all that simple. I've been hacked before and I work in a field where it can be very lucrative to target me. I had it happen that my Google accounts were compromised, despite using a 32 character password. Still not sure how that happened. I was travelling internationally, staying in hotels. So maybe I made a mistake while there. Though it seems unlikely I would fall for a MIM fake web portal attack, since I was using https and even a vpn most of the time. But I later found out that my dad's laptop was infected, and it was just there on my home network. So that seems the most likely cause.

Either way, I've ditched Google and now keep everything important either self-hosted or on pen and paper. As a result of that, I need to harden my home network too. Openwrt also lets me do things like isolate my wifi devices from devices on my ethernet LAN, or create a seperate VLAN etc. All of this is very important from my perspective of preventing another attack.

I did have to do the DMZ thing because the bridge mode wasn't working with my ISP. But I'm going to give it another try here soon.
 
I've been hacked before and I work in a field where it can be very lucrative to target me.
That already exclude you from the definition of "typical home user". :)

I had it happen that my Google accounts were compromised, despite using a 32 character password.
More than password length, what matters is which 2FA you used because without that even a 64 character password might not help.

Though it seems unlikely I would fall for a MIM fake web portal attack, since I was using https and even a vpn most of the time. But I later found out that my dad's laptop was infected, and it was just there on my home network. So that seems the most likely cause.
I hope it wasn't some free/Chinese vpn. Also, nowadays win 10/11 default security updates are quite good enough & assuming you follow the standard security guidelines on your pc in the network then someone who could bypass those from your father's laptop would also be very likely capable enough to bypass whatever router security you use in any typical consumer grade hardware available.
 
That already exclude you from the definition of "typical home user". :)


More than password length, what matters is which 2FA you used because without that even a 64 character password might not help.


I hope it wasn't some free/Chinese vpn. Also, nowadays win 10/11 default security updates are quite good enough & assuming you follow the standard security guidelines on your pc in the network then someone who could bypass those from your father's laptop would also be very likely capable enough to bypass whatever router security you use in any typical consumer grade hardware available.
I had 2FA with Google Auth turned on for all of my accounts. Password length matters in the context of how long someone running hashcat on a GPU cluster will take to crack it, but that obviously wasn't the vector they used to get me.

As for vpn, no, it was mullvad. Believing Windows Defender was 'good enough' is what got me into that mess. His laptop was fully updated and it still got infected. I later learned that he had possibly clicked on a spam email attachment. Obviously someone on this forum is unlikely to do that but for the average person who can make a mistake like that once in a while, you need more aggressive heuristic based detection and monitoring of network traffic, which defender doesn't do. I have Bitdefender installed on my dad's laptop now because that was the AV that finally detected it.

And no, it doesn't follow that just because someone was able to infect my dad's PC, they would therefore also be able to compromise my router. Look into the Swiss Cheese Model Of Security. Someone having access to an exploit that compromises one part of your network does not mean they have exploits for the others. Also, Openwrt has thousands of eyes looking over its source code, which in turn uses the Linux kernal, which possibly has tens of thousands of eyes on it. That is why it is *secure* for all the intents and purposes I care about.
2FA is good to have but that's not fool-proof.
You don't even need to crack passwords or 2FA when you can just steal authentication/access tokens by some malicious browser extension.

That is probably how @variablevector's google accounts were compromised.
This is a very good point, damn. I had probably logged into my gmail account on his laptop weeks ago and never logged out. It also explains how they were able to circumvent the 2FA.
 
Back
Top