Asuswrt-Merlin 3.0.0.4.374.33 is out

[RogerSC]

There's currently an issue with Comcast and IPv6 where it fills up your routing tables. I suspect this might be leading to resource exhaustions, followed by a router reboot.

Comcast blames Asus, and Asus blames Comcast. People in the middle of them blame the old kernel used by Asuswrt. So at this point, there is no real solution if that's the same thing affecting you.

For this prolblem, the following script in /jffs/scripts/init-start works for me (I'm also on Comcast using the RT-N66U):

#!/bin/sh
echo 512 > /proc/sys/net/ipv6/neigh/default/gc_thresh1
echo 1024 > /proc/sys/net/ipv6/neigh/default/gc_thresh2
echo 2048 > /proc/sys/net/ipv6/neigh/default/gc_thresh3
touch /tmp/init-start.ran

I was getting "neighbor table overflows" regularly until I started using the above script, and now I don't. So if this is the problem causing your reboots, you might try the above script and see if it helps. You can increase the values for the threshholds that are used up to a point, this one is experimental.

I'm not seeing this problem on the Netgear R7000 wihen I have IPv6 enabled.

 

[RMerlin]

I'm not seeing this problem on the Netgear R7000 wihen I have IPv6 enabled.

The R7000 is based on kernel 2.6.36.4 (like the AC56/AC68), so it could be a fix at the kernel level.

Has anyone ever reported the same problem with the AC56/AC68U, or only on the older models based on 2.6.22.19?

 

[RogerSC]

The R7000 is based on kernel 2.6.36.4 (like the AC56/AC68), so it could be a fix at the kernel level.

Has anyone ever reported the same problem with the AC56/AC68U, or only on the older models based on 2.6.22.19?

Yes, with the R7000, it could still be happening, but not reported in the log...the logging is pretty sparse compared with the RT-N66U. Or it could be a fix in the networking code in the kernel, or something altogether different *smile*. Hard to tell.

 

[LiFePO4]

Parental Control Working!

Hi RMerlin:

I am a new owner for the RT-N66R and ran into the parental control issue (where it kicked my daughter off on Saturday even though she was allowed). I searched far and wide on this and converged on this forum.

I came across a post that said disabling the 11PM-12AM Sat block fixed it (which it did) and then I found your custom firmware (and the other threads discussing this issue). Seeing that you were trying a fix (that Makkie2002 did) in your firmware, I loaded your latest (34.2) and have verified it fixes the Parental Control issue.

Seeing that not many have replied back to you on whether this fix works, I am replying to let you know it does. I am also loving your FW and will send a donation. Thanks!!!

P.S. Done - 9D33540740173544F

 

[RMerlin]

I came across a post that said disabling the 11PM-12AM Sat block fixed it (which it did) and then I found your custom firmware (and the other threads discussing this issue). Seeing that you were trying a fix (that Makkie2002 did) in your firmware, I loaded your latest (34.2) and have verified it fixes the Parental Control issue.

Seeing that not many have replied back to you on whether this fix works, I am replying to let you know it does. I am also loving your FW and will send a donation. Thanks!!!

Thank you - much appreciated!

I was disappointed that Asus chose not to include the fix, but apparently they have more major changes in mind for Parental Control for the future. Might have been nice to have at least a fix right now, but at least I can offer an alternative that works for the time being.

 

[DrTeeth]

At least it is nice to know that Asus has long term plans for their router's firmware. This is in complete contrast to the BIOS updates for their motherboards where support seems to get dropped as soon as the sucker is purchased .

 

[RMerlin]

At least it is nice to know that Asus has long term plans for their router's firmware. This is in complete contrast to the BIOS updates for their motherboards where support seems to get dropped as soon as the sucker is purchased .

Aside from their recent release of OpenVPN and Time Machine, I can tell you that I see quite a few other features in the source code that they are working on. One of them is persistent syslog (which they actually enabled without mentionning it in 374_339). There is also code being laid down that will allow eventually to send log and other notifications through Email. And there is also something else on which I have no details yet, as I only see some bits of it appearing here and there in the code since 374_306. All I can tell is that it seems related to a third party service provider...

BIOSes are usually only meant to initialize the hardware. Once a BIOS is stabilized, there is little need to keep updating it, beside adding new cpuids if Intel releases new models.

 

[DrTeeth]

BIOSes are usually only meant to initialize the hardware. Once a BIOS is stabilized, there is little need to keep updating it, beside adding new cpuids if Intel releases new models.

I use software that relies on a properly working BIOS*. The quality of BIOSes has dropped like a brick since OSs took over their functions.
* Partitioning and boot management software - BIBM by Terabyte Unlimited

 

[RMerlin]

I use software that relies on a properly working BIOS*. The quality of BIOSes has dropped like a brick since OSs took over their functions.
* Partitioning and boot management software - BIBM by Terabyte Unlimited

There's been a major shift in the past two years too with the migration from the rusty old 16-bit BIOS to 64-bit EFI. It wouldn't surprise me if it took a few more years for everyone to be able to deal with this appropriately.

Not to mention that by being far more complex, I fear that EFI will become increasingly buggy, like operating systems and any regular applications in general, making my life as a technician even more painful than it already is

It took me a few weeks with settings fiddling and noting down every single changes over the course of two weeks to track down why my new PC build would refuse to go to sleep on the first attempt (because usually it took a few hours of uptime to be able to reproduce the problem again). In the end I tracked it down to ASPM that required to be enabled in the EFI.

 

[DrTeeth]

Damn computers! It is so important not to let the buggers win!

Cheers

DrT

 

[steve7929]

does this firmware version or router (N66U) have issues with 3 tb Seagate drives connected to it for media server purposes? It seems to be detecting only some of my folders on the external drive and not the others

 

[bha]

When you say "must reconfigure" rather than save/reload settings, what exactly does this mean?

I am coming from an older version of asuswrt-merlin, so I am required to reconfigure according to instructions.

I plan to take screenshots of all of the web UI pages, and save them. Then I will need to re-enter all of these settings once I have loaded the new firmware.

But a few questions:
-Saw something about 'clearing the nvram'. Does this need to be done?

-What about things like scripts in the /jffs/scripts and /jffs/configs folders? Will these persist, or do I need to back them up and recreate them?

-What about cron jobs (I have one set up to reboot the router nightly)

Thanks!!

 

[RMerlin]

But a few questions:
-Saw something about 'clearing the nvram'. Does this need to be done?

This is the same thing as reverting to Factory Defaults.

-What about things like scripts in the /jffs/scripts and /jffs/configs folders? Will these persist, or do I need to back them up and recreate them?

As noted on the FIrmware upgrade page, JFFS is never guaranteed to survive a firmware upgrade, especially on routers where jffs must share the same 32 MB as the firmware itself (RT-N16 and RT-N66). You should always backup these before flashing a new firmware.

-What about cron jobs (I have one set up to reboot the router nightly)

Cron jobs are located in RAM, so they wouldn't survive even just a reboot. They must be recreated at boot time by a JFFS script.