Release ASUS RT-AC68U Firmware version 3.0.0.4.386.49703 (07-20-2022)

[Mittenz]

Asus forums are saying this is good frimware, is this superior to merlins latest release?

 

[dosborne]

Asus forums are saying this is good frimware, is this superior to merlins latest release?

"Superior" is subjective. Merlin adds options not found in stock. Stock may have security fixes Merlin doesn't yet. Etc

 

[L&LD]

And RMerlin may have fixes stock doesn't yet too.

 

[wouterv]

Good or bad (firmware) depends on your expectations and desires.
Stock firmware in general is reliable and stable for straight forward use of the router, that is using the consumer router for the basic functions as 3-in-one "router + network switch + wireless access point" for home usage. Or even shorter: to enable wireless Internet in your home [period].

The previous firmware version did run on my router without a reset or any interaction from the release / installation date to the date of installation of the next firmware release (over 100 days).
The extra add-ons (everything listed below Network Map in General) and custom configuration through SSH usually cause the troubles.
Asuswrt-Merlin adds more features to the user interface and supports further customization beyond the stock firmware features.

 

[SwampKracker]

Good or bad (firmware) depends on your expectations and desires.
Stock firmware in general is reliable and stable for straight forward use of the router, that is using the consumer router for the basic functions as 3-in-one "router + network switch + wireless access point" for home usage. Or even shorter: to enable wireless Internet in your home [period].

View attachment 43263

The previous firmware version did run on my router without a reset or any interaction from the release / installation date to the date of installation of the next firmware release (over 100 days).
The extra add-ons (everything listed below Network Map in General) and custom configuration through SSH usually cause the troubles.
Asuswrt-Merlin adds more features to the user interface and supports further customization beyond the stock firmware features.

Nice diagram. Is that your creation?

 

[wouterv]

Nice diagram. Is that your creation?

Yes it is, be aware it is just a visualized functional diagram.

 

[andyross]

Just a minor note that the correct spelling would be "Antennas", without the apostrophe. With it, it is implying it is something that belongs to Antenna.

 

[wouterv]

Corrected.

 

[Wychwood]

After deploying the 3.0.0.4.386.49703 firmware update manually through the web interface, the router locked up and failed to restart successfully. Could I have left it longer? Perhaps. But with no feedback on progress after the normal firmware update time and being patient thereafter, it certainly seemed to have failed.

I manually reset and restored to factory defaults, then reapplied the update.
It eventually completed the update and I restored my settings, but I ended up with problems with DDNS not connecting to NoIP automatically and OpenDNS failing to function anymore. The router locked up again once afterwards and that was enough for me, combined with the DDNS and OpeVPN issues to decide to roll back to the previous firmware release which was stable and functioning well.

Manually re-entering all the settings is not an option I am willing to perform except as an absolute last resort.
IMHO, if a successfully applied firmware update cannot accept the previous firmware's router settings, either during the update process, or from a restore file, then I would consider it a flawed implementation.

 

[OzarkEdge]

After deploying the 3.0.0.4.386.49703 firmware update manually through the web interface, the router locked up and failed to restart successfully. Could I have left it longer? Perhaps. But with no feedback on progress after the normal firmware update time and being patient thereafter, it certainly seemed to have failed.

I manually reset and restored to factory defaults, then reapplied the update.
It eventually completed the update and I restored my settings, but I ended up with problems with DDNS not connecting to NoIP automatically and OpenDNS failing to function anymore. The router locked up again once afterwards and that was enough for me, combined with the DDNS and OpeVPN issues to decide to roll back to the previous firmware release which was stable and functioning well.

Manually re-entering all the settings is not an option I am willing to perform except as an absolute last resort.
IMHO, if a successfully applied firmware update cannot accept the previous firmware's router settings, either during the update process, or from a restore file, then I would consider it a flawed implementation.

Suggested reading:

Reset FAQ
Reset via Reset button/webUI Restore/node removal - clears settings in NVRAM; reboot restores fw defaults from CFE (fw defaults)
Hard Reset via WPS button/webUI Restore+Initialize - also clears data logged in /jffs partition (fw defaults+clear logs)
Restoration - uploads fw in Rescue Mode

OE

 

[bbunge]

After deploying the 3.0.0.4.386.49703 firmware update manually through the web interface, the router locked up and failed to restart successfully. Could I have left it longer? Perhaps. But with no feedback on progress after the normal firmware update time and being patient thereafter, it certainly seemed to have failed.

I manually reset and restored to factory defaults, then reapplied the update.
It eventually completed the update and I restored my settings, but I ended up with problems with DDNS not connecting to NoIP automatically and OpenDNS failing to function anymore. The router locked up again once afterwards and that was enough for me, combined with the DDNS and OpeVPN issues to decide to roll back to the previous firmware release which was stable and functioning well.

Manually re-entering all the settings is not an option I am willing to perform except as an absolute last resort.
IMHO, if a successfully applied firmware update cannot accept the previous firmware's router settings, either during the update process, or from a restore file, then I would consider it a flawed implementation.

Using a restore file from a prior firmware version is not recommended at all. Go back to the prior firmware to use the restore file or bite the bullet and reset with initialize and fumble finger the settings back in.

 

[wouterv]

You shall really (yes REALLY) keep other means of your configuration besides the backup file.
As said restoring the backup file only (yes ONLY) works with the firmware version it was made with.
Make screenshots of the configuration changes you make from Factory Defaults or even better to my opinion: keep a simple text file with all the changes you make.
Like this:

RT-AC68U

2.4 GHz Network Name: xxx1
Network Key: yyy1
5 GHz Network Name: xxx2
Network Key: yyy2

Advanced Settings - Wireless - General - 2.4 GHz
Channel bandwidth: 20 MHz
Control Channel: 6 (or 1 or 11)

Advanced Settings - Wireless - WPS
Enable WPS: OFF

Advanced Settings - Wireless - Professional - 2.4 GHz
Modulation Scheme: Up to MCS 7 (802.11n)
Airtime Fairness: Disable
Explicit Beamforming: Disable
Universal Beamforming: Disable

Advanced Settings - Wireless - Professional - 5 GHz
Airtime Fairness: Disable
Universal Beamforming: Disable

Advanced Settings - LAN - LAN IP
Host Name: router1
IP Address: 192.168.1.1

Advanced Settings - LAN - DHCP Server
IP Pool Starting Address: 192.168.1.3 (3 because there is a Media Bridge with address 2)

Advanced Settings - WAN - Internet Connection
Enable UPnP: No

Advanced Settings - Administration - System
USB Mode: USB 2.0
Time Zone: (GMT+1:00) Amsterdam, Berlin, Bern
DST time zone changes starts: month = 3, weekday = 5th Sun, 2 hours
DST time zone changes ends: month = 10, weekday = 5th Sun, 3 hours

USB application - Servers Center - Media Server
Enable UPnP Media Server: OFF

USB application - Servers Center - Media Server
Network Printer Server
Network Printer Server: OFF

 

[gabox]

Hi mates, I successfully upgraded without reset. But I notice that if I activate the parental control, I notice a drop in my internet speed. Is there a bug or something I am doing wrong. ? thanks.

 

[L&LD]

You need to provide more specifics.

 

[muddymind]

You shall really (yes REALLY) keep other means of your configuration besides the backup file.
As said restoring the backup file only (yes ONLY) works with the firmware version it was made with.
Make screenshots of the configuration changes you make from Factory Defaults or even better to my opinion: keep a simple text file with all the changes you make.
Like this:

RT-AC68U

2.4 GHz Network Name: xxx1
Network Key: yyy1
5 GHz Network Name: xxx2
Network Key: yyy2

Advanced Settings - Wireless - General - 2.4 GHz
Channel bandwidth: 20 MHz
Control Channel: 6 (or 1 or 11)

Advanced Settings - Wireless - WPS
Enable WPS: OFF

Advanced Settings - Wireless - Professional - 2.4 GHz
Modulation Scheme: Up to MCS 7 (802.11n)
Airtime Fairness: Disable
Explicit Beamforming: Disable
Universal Beamforming: Disable

Advanced Settings - Wireless - Professional - 5 GHz
Airtime Fairness: Disable
Universal Beamforming: Disable

Advanced Settings - LAN - LAN IP
Host Name: router1
IP Address: 192.168.1.1

Advanced Settings - LAN - DHCP Server
IP Pool Starting Address: 192.168.1.3 (3 because there is a Media Bridge with address 2)

Advanced Settings - WAN - Internet Connection
Enable UPnP: No

Advanced Settings - Administration - System
USB Mode: USB 2.0
Time Zone: (GMT+1:00) Amsterdam, Berlin, Bern
DST time zone changes starts: month = 3, weekday = 5th Sun, 2 hours
DST time zone changes ends: month = 10, weekday = 5th Sun, 3 hours

USB application - Servers Center - Media Server
Enable UPnP Media Server: OFF

USB application - Servers Center - Media Server
Network Printer Server
Network Printer Server: OFF

I made a Javascript in tampermonkey to automate all configs. Honestly as a software engineer I find appalling how a proper parser to/from nvram settings hasn't been made. With the ammount of router reboots it really is a pain in the butt and takes forever to finish the configuration.

I'm currently working on a game and imagine save files not working between versions.... It all just smells lasiness and lack of care for the user experience.

There are new settings? Load default values for those ones.
There are deprecated settings? Just ignore them.
There are settings that changed? Transform those to the new schema. Not possible? Load default ones.
At the end inform the user of what was not mapped entirely.

This would take care of almost all settings and a huge QoL for the user.

Sorry for the rant but some developers are just plainly incompetent and it irks me to no end.

 

[RMerlin]

There are new settings? Load default values for those ones.
There are deprecated settings? Just ignore them.
There are settings that changed? Transform those to the new schema. Not possible? Load default ones.

Asuswrt does all of these, including converting old format to new format when required. It happened when they enabled password encryption, or when the storage format for Parental Control schedulers got changed. This is done by the code inside router/rc/format.c in the source code.

A lot of the "a factory default reset fixed my issue" are more akin to "A Windows format and reinstall resolved my issue". It`s just a quick way to resolve an issue rather than spending hours investigating the root cause. They often come from users having customized their wireless settings to "optimize" them, and those settings are causing problems down the road. The factory default reset mostly simply sets all of these back to their original values, resolving the issue at the same time. Sometimes, it`s because users have filled up the available nvram (increasingly common with the RT-AC68U), and the factory default reset gets rid of old settings that are no longer being used.

I never reset my own router settings. My RT-AX88U had close to two years long of firmware updates installed one of top of the other, without any issue.

If a factory default reset is truly needed (for instance when Asus changes the size of the nvram storage), it is explicitly documented in their changelog.

 

[wouterv]

Regarding Backups in general. The router config backup is like a PC imagine, suitable for the machine it is made from.
Another router backup method (text based) is described here.

To restore to a different machine you need source files and you better document how to build a system from scratch.
I have run into situations where the client made daily backups (other than the prescribed method which was to tape) of his huge VMS system and was very proud of his automated mechanism, until the moment we needed a to restore a backup and the backup failed to restore, turning out that the backup procedure was wrong and produced useless backup files.
Then you are glad to know how to build up systems from scratch and the lesson learned was: when you make backups, also test that you can restore those backups.
I was always very keen to write down in detail a proven working build-from-scratch procedure and told clients be my guest to find a smarter way, but my method works for sure when followed to the letter.

 

[Justinh]

If a factory default reset is truly needed (for instance when Asus changes the size of the nvram storage), it is explicitly documented in their changelog.

And I'd hope this is automated in some way also. Not everyone will read the changelog or understand it. If you have auto update on, this would be a blocker. I say that because if a reset is required, the upgrade should force it, then your router would suddenly not be working. This is a tricky scenario.

 

[RMerlin]

And I'd hope this is automated in some way also.

I doubt you mean here that the factory default reset should be done automatically...

If you have auto update on, this would be a blocker.

Their auto-update mechanism allows them to specify a minimal version. That way, if for example 388_20000 required a factory default reset when coming from 386_xxxxx, they could set the minimal version for an auto-update to 388_10000, requiring you to manually update it if ou haven't already upgraded to a 388 build.

Mandatory factory default resets are very, very rare in their case. Sometimes they do have a required minimal version tho.

 

[Justinh]

Yep, I mean the reset, and that is why it is would be a blocker (to auto updating). And if fixes require a reset to actually apply the fixes, I don't see how an update could not force a reset. That could spell bad news for ASUS in many ways.

But, ASUS can't rely on users to manually update. As long as the router is working, they won't touch it. And if they set auto-update on, they think they are always good, up-to-date, and safe with new fixes, right? This could leave people vulnerable for a long time.