ASUS RT-AC68U Firmware version 3.0.0.4.385.20253

[jc816]

Anyone else experiencing constant > 80% CPU levels after upgrade to this version? GUI is really sluggish. Downgraded and CPU went back to normal 0-20%.

(If you haven't done so already, and if you feel compelled to do so) Could you email your experience and issuses regardimg the firmware to Asus? I have and encourage everyone affected to do so. Thank you.

 

[jc816]

Almost 100% cpu usage after upgrade.

(If you haven't done so already, and if you feel compelled to do so) Could you email your experience and issuses regardimg the firmware to Asus? I have and encourage everyone affected to do so. Thank you.

 

[Grisu]

You dont need to write 3 seperate replies, you could combine them into just one answer.
Click reply, copy everything between start and end {quota}, click next reply and paste previous, repeat these steps for all quotas.
Then you can answer each serperately after each end-quota or at the end for all.

 

[RMerlin]

You dont need to write 3 seperate replies, you could combine them into just one answer.
Click reply, copy everything between start and end {quota}, click next reply and paste previous, repeat these steps for all quotas.
Then you can answer each serperately after each end-quota or at the end for all.

There is nothing wrong in separately answering each message. For instance, it's the only thing you can do when using Taptalk.

 

[Grisu]

He answered all 3 with the exactly same text. That doesnt make any sense to me, but if you like to read it three times there are other ways to do so.

 

[jc816]

He answered all 3 with the exactly same text. That doesnt make any sense to me, but if you like to read it three times there are other ways to do so.

Thank you for your feedback. I am new to SNB (I have never used this service until today), and any forum for that matter, and don't know all the commenting tricks yet. I posted the same message to three commentors who have been experiencing the performance issues, have reverted their firmware back to a previous version, and were unclear in their comments on whether or not they contacted Asus. I was encouraging them to do so in efforts to help everyone out who is experiencing this problem--the more "squeaky wheels," the greater the likelihood that they resolve this issue. It has been a very frustrating week for me with working from home and having wireless performance issues (granted, given the current global situations, I have a lot to be grateful for). My intentions were just to encourage everyone to help each other out. Thanks again for your feedback though and I hope you have a wonderful day.

 

[Pedro]

stay with 384 for the moment.

which .384 is the most stable in ur opinion?

 

[Grisu]

You could try Merlin 384.13 known to be quite stable for most.

 

[delaveganz]

Running as a media bridge - I switched from Merlin 384.15 to Asus 3.0.0.4.385.20253 a few days ago due to loss of connectivity to the main router from time to time.

I now noticed that my RT-AC68U CPU utilisation was at 85-90%. Switched to Merlin 384.16 and its down to less than 5%. Hope it can reliably keep the connection though.

 

[VANT]

my 68u is set up as a node and have high cpu usage (top command under ssh). downgrade to 252 solve this issue.

 

[kozman]

my 68u is set up as a node and have high cpu usage (top command under ssh). downgrade to 252 solve this issue.

Did you already use the Feedback tool in the UI to send a note to Asus that it's a bug? The more we pester them, maybe the faster they will fix it.

 

[VANT]

ofcourse

 

[opq]

+1 on having this issue, which was also resolved by reverting the firmware. Scary that we all had to roll back to FW without the kr00k vulnerability fix.

 

[RMerlin]

Scary that we all had to roll back to FW without the kr00k vulnerability fix.

Don't lose sleep over Kr00k. The NVD gives it a risk score of 3.1 out of 10...

Just an overblown issues by the media.

 

[kozman]

Don't lose sleep over Kr00k. The NVD gives it a risk score of 3.1 out of 10...

Just an overblown issues by the media.

Maybe. Maybe not. They said the same about the umpteen vulns related to SPECTRE/MELTDOWN and others of its ilk. Might take a while before it's weaponized easily but give it time, the basement elves are hard at work I'm sure. There are millions of unpatchable WiFi routers out there just waiting to be hacked.

 

[Ronald Schwerer]

Unless those "basement elves" move-in next door, you don't need to worry.

 

[kozman]

Unless those "basement elves" move-in next door, you don't need to worry.

Because they would *never* sit outside of some corporate target's parking lot.

 

[RMerlin]

They said the same about the umpteen vulns related to SPECTRE/MELTDOWN and others of its ilk.

Spectre has a risk score of 5.6, which was higher. And it was deemed more dangerous for shared servers, particularly in virtualized hosting, where strangers are sharing the same CPUs, and therefore it could gain them access to the data from other customers.

Might take a while before it's weaponized easily but give it time, the basement elves are hard at work I'm sure.

Read up on how NVD assigns a score to a security issue. It's not about whether or not it's currently actively exploited. They do not triple a security issue's score just because suddenly an exploit appears in the wild.

 

[wouterv]

Spectre has a risk score of 5.6, which was higher. And it was deemed more dangerous for shared servers, particularly in virtualized hosting, where strangers are sharing the same CPUs, and therefore it could gain them access to the data from other customers.



Read up on how NVD assigns a score to a security issue. It's not about whether or not it's currently actively exploited. They do not triple a security issue's score just because suddenly an exploit appears in the wild.

Can you shed some more light on the new security related Asus Healing System, with downloaded signature files?
This seems to be automatically enabled, with no user intervention and no visible GUI settings.

 

[hancox]

@RMerlin you're your usual humble self, but wouldn't moving to your excellent firmware fix both the CPU issues with this build (albeit by a gpl rollback of sorts), as well as provide a Kr00k patch? Just a thought for those concerned...